j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 ; : 1 e1 Available online at www.sciencedirect.com journal homepage: www.elsevier.com/locate/jmrt Original Article An assessment of pyrite thin-film cathode characteristics for thermal batteries by the doctor blade coating method Trang-Le Thi Thu a, Thuy-Le Thi Thu b, Tran Dinh Manh c,**, Trung-Truong Tan b, Jeng-Kuei Chang d,* a Institute of Materials Science and Engineering, National Central University, Taoyuan, 32001, Taiwan Institute of Research and Applied Technological Science, Dong Nai Technology University, Dong Nai, 810000, Viet Nam c Institute of Applied Technology, Thu Dau Mot University, Tran Van on Street, Phu Hoa Ward, Thu Dau Mot City, Binh Duong, 820000, Viet Nam d Department of Materials Science and Engineering, National Chiao Tung University, Hsinchu, 30010, Taiwan b article info abstract Article history: Using FeS2 (pyrite) as an active material for cathodes in the thermal battery has received Received 27 January 2021 much more attention due to its abundant natural resources, cheapness, and excellent ef- Accepted 10 May 2021 ficiency Nevertheless, scientists take a large internal resistance issue of the FeS2 cathodes Available online 21 May 2021 into urgent consideration, decreasing the electrochemical efficiency of typical LieSi/FeS2 thermal batteries In this study, we surveyed the effect of binders, thin-film thicknesses, Keywords: and the addition of conductive carbonaceous additives such as carbon black (CB), super P Pyrite thin-film (SP) and activated carbon (AC) on the FeS2 thin film cathode fabrication by applying the Thermal batteries blade coating method To obtain the FeS2 homogeneous slurry, we utilized the ball-milling Blade coating process to reduce the FeS2 particle size from 8.7 mm to 0.9 mm Subsequently, the FeS2 thin Adhesion test film with different thicknesses, expected to elevate mass loading causing higher capacity Cathodes of thermal batteries, by means of the doctor blade was successfully fabricated from the homogeneous slurry comprising ball-milled FeS2 active material, polyvinylidene fluoride (PVDF) binder or sodium silicate (Na2SiO3) binder, accompanied with conductive carbonaceous additive; even without conductive carbonaceous additive, thin films were still capable of being produced Among these samples, the thin-film type, mass loading from 1.4 to 3.6 mg/cm2 (corresponding to the doctor blade thicknesses in a range of 150e300 mm), manufactured from the slurry consisting of 80 wt% of ball-milled FeS2, 15 wt% of conductive carbonaceous additive (CB/SP/AC), and wt% of PVDF binder will promisingly contribute to increasing electrochemical efficiency of thermal batteries, possibly on account of high mechanical durability © 2021 The Authors Published by Elsevier B.V This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) * Corresponding author ** Corresponding author E-mail addresses: manhtd@tdmu.edu.vn (T.D Manh), jkchang@nctu.edu.tw (J.-K Chang) https://doi.org/10.1016/j.jmrt.2021.05.014 2238-7854/© 2021 The Authors Published by Elsevier B.V This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/) 1140 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 ; : 1 e1 Introduction Thermal batteries (TBs) or molten salt batteries are power sources that use solid electrolytes and operated at a high temperature (between around 350 and 500 C) [1] Similar to other batteries, the single-cell structure of the thermal battery consists of an electrolyte placed between an anode (LieSi) and a cathode (FeS2) [2e5] Specifically, the use of the heat source is to allow the thermal battery to reach operating temperatures TBs are mainly employed as energy sources for major military applications (guided missiles, rockets, guided bombs, and mines) [6,7], radar and electronic guidance, emergency backup power, etc due to their excellent mechanical robustness, longevity, short activation time, high reliability along with high energy density and power density [3e5,8e10] Besides, TBs are often used in safe and urgent cases Several of the largest TBs are mainly incorporated into the hydraulic systems as backup power sources in the scope of military aircraft Potential of the Li and Li-alloy/FeS2 couple was demonstrated in the past [11e13] instead of the previously Ca/CaCrO4 system, since the Li and Li-alloy/FeS2 system possessed a powerful combination with well-characterized and foreseeable chemical reactions as well as being eco-friendly than Ca/ CaCrO4 system, no presence of Cr(VI) as a carcinogen and FeS2 being available in nature Some scientists reported FeS2, having electrical conductivities in a range between 0.03 and 333 S/cm [14] and an energy band gap approximately 0.92 eV [15,16] at ambient temperature, as a good material for both ntype and p-type semiconductor [17e19] More importantly, its electrical conductivity will increase with increasing temperature, which promotes it be perfect for TBs However, the FeS2 cathode still exists some minor issues such as medium thermal stability, voltage transients on activating the battery, and the remarkable ability to dissolve in the molten-salt electrolytes This phenomenon is willingly resolved by lithiation [20] Li2O and Li2S are popular lithiation agents, added limited amounts around 1e2 w/o to inhibit voltage transients In addition, notable solubility of FeS2 in molten-salt electrolytes turns into a problem when the battery in the state of the open circuit for prolonging time or a very small load as < 20 mA/cm2 is applied to the battery The diffusion of dissolved FeS2 into the separator can have reactions with anodic species to produce Fe and Li2S, which may cause battery capacity loss [21] Electrochemical efficiency of LieSi/FeS2 thermal batteries was reduced due to high internal resistance of the pyrite cathodes [6] Almost all the efforts of researchers in recently years have reported that reducing of the internal resistance of lithium and Ni-MH batteries was conducted by adding conductive carbonaceous materials, such as carbon blacks (CBs) and carbon nanotubes (CNTs) [11,12,22e24] but improving the electrochemical efficiency of LieSi/FeS2 thermal batteries has not yet been reported Recently, Choi et al [25] have demonstrated that adding conductive carbonaceous materials to pyrite electrode enhanced electrochemical performance of LieSi/FeS2 thermal batteries In addition, the effectiveness of binders in the electrolyte (e.g ceramic materials) has been studied [10] As a result, fumed silicas was considered to be more effective because it was in need of less material, around w/o than the previous electrolyte material required a large concentration around 35e50 w/o Most of the previous methods in thermal battery manufacturing generated thicker electrodes resulting in reducing electrode usages, consequently reducing energy density of the thermal battery Therefore, thinning of the cathode electrode must ensure mechanical strength and optimal electrochemical efficiency of the electrode is essential and can be carried out by tape casting process with doctor blade method which is expressed via previous study [8,26,27] In this study, we assessed the effect of binders, the addition of conductive carbonaceous additives such as carbon black (CB), super P (SP) and activated carbon (AC), and thin-film thicknesses of the fabricated FeS2 thin film cathodes via assessment of forming material characterization Materials and methods 2.1 Preparation of active materials Raw FeS2 powder (99%, LinYi, China) was undergone the ballemilling process attachment with Zirconium (Zr) ball (3 mm diameter) and anhydrous EtOH (99.5%) operated at a constant velocity of 250 revolutions per minute (rpm) during 24 h for particle size reduction Finally, the smaller-particle FeS2 was obtained after drying the milled FeS2 collected by filtration with general meshwork 2.2 Slurry preparation The slurry was prepared by mixing the compositions of the above-milled FeS2 as an active material, carbonaceous material (SP/CB/AC, 99.9%, SigmaeAldrich) as an additive material, and a binder The binder used is Polyvinylidene fluoride (PVDF) or liquid sodium silicate (Na2SiO3) with sodium tripolyphosphate (STP)/lignosulfonic acids (LSAS)/sodium hexametaphosphate (SHMP) as a dispersant in a weight ratio in N-Methyl-2pyrrolidone (NMP) or deionized (DI) water as a solvent, respectively PVDF was absolutely dissolved in NMP under stirring, or DI water and STP/LSAS/SHMP and Na2SiO3 were stirred till all solid dissolved, then gradually added FeS2 and SP/CB/AC The as-prepared mixture in the sealed container was stirred at 500 rpm by using a magnetic stirrer for h to obtain the homogeneous slurry All of samples were conducted with different procedures as described in Table and done in triplicate 2.3 Electrode coating The as-prepared electrode homogeneous slurry was cast onto the ethanol-washed aluminum foil to remove oxide on the surface The wet coated thin film was obtained after the autocoating machine flattened the slurry with 150e300 mm thickness of the doctor blade with the size of 10 cm 15 cm Finally, this samples was then dried under vacuum at 100 C for h 2.4 Assessment of materials characterization To assess characterization of tested materials, the measurements of the particle sizes and the structure of material of 1141 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 ; : 1 e1 Table e All of the conducted samples with different procedures Number 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Active material FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 FeS2 Additive Binder Dispersant Solvent Substrate None None CB SP AC CB SP AC CB SP AC CB SP AC None None None None None None None None None None None None None None None None None None None None CB SP AC CB SP AC SP AC CB SP AC PVDF PVDF PVDF PVDF PVDF PVDF PVDF PVDF PVDF PVDF PVDF PVDF PVDF PVDF Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 Na2SiO3 None None None None None None None None None None None None None None STP STP STP STP STP STP STP STP STP LSAS LSAS LSAS SHMP SHMP SHMP STP STP STP STP STP STP STP STP STP STP STP STP STP STP STP STP NMP NMP NMP NMP NMP NMP NMP NMP NMP NMP NMP NMP NMP NMP DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water DI water Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Al Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Foil Heating process Ratio (wt%) Blade thickness (mm) Result 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 100ºCe3 hrs 50 C (30 min) ¼> 75 C (1 h) ¼> 85 C (30 min) ¼> 100 C (1 h) 99:1 98.5:1.5 83.5:15:1.5 83.5:15:1.5 83.5:15:1.5 80:15:5 80:15:5 80:15:5 80:15:5 80:15:5 80:15:5 80:15:5 80:15:5 80:15:5 91:4:5 87:8:5 85:10:5 80:15:5 78:17:5 75:20:5 70:25:5 65:30:5 60:35:5 90:5:5 80:15:5 70:25:5 90:5:5 80:15:5 70:25:5 85:10:5 80:15:5 75:20:5 65:30:5 60:35:5 30:60:5:5 30:60:5:5 30:60:5:5 45:45:5:5 45:45:5:5 45:45:5:5 60:30:5 60:30:5 70:20:5:5 70:20:5:5 70:20:5:5 150 150 150 150 150 150 150 150 200 200 200 300 300 300 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 150 Good Good Good Good Good Good Good Good Good Good Good Good Good Good Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Exfoliated Good Good Good Good Good Good Good Good Good Good Good 100ºCe3 100ºCe3 100ºCe3 100ºCe3 100ºCe3 100ºCe3 100ºCe3 100ºCe3 100ºCe3 100ºCe3 100ºCe3 hrs hrs hrs hrs hrs hrs hrs hrs hrs hrs hrs FeS2: pyrite, CB: carbon black, SP: super P, AC: activated carbon, PVDF: polyvinylidene fluoride, Na2SiO3: sodium silicate, STP: sodium tripolyphosphate, LSAS: lignosulfonic acids, SHMP: sodium hexametaphosphate, NMP: N-Methyl-2-pyrrolidone and DI water: deionized water FeS2 powder before and after the ballemilling process by dynamic light scattering (DLS) and X-ray diffraction (XRD), respectively were carried out The PVDF or Na2SiO3 binder, raw FeS2 powder, ball-milled FeS2 powder, and ball-milled FeS2 film using PVDF or Na2SiO3 binder (shaved the active thin-film layer off the copper foil substrate) were subjected to thermal-gravimetric analysis (TGA), to determine thermal stability properties The durability of the formed thin films after coated using PVDF or Na2SiO3 binder was assessed by the adhesion test Electrochemical efficiency was also predicted by mass loading 2.4.1 Dynamic light scattering (DLS) In this research, the method of DLS, one of the most popular measurement technique for measuring small particles in solution, was used to determine the particle size distribution profile of the raw FeS2 and ball-milled FeS2 powder DLS measurements were carried out using a Malvern Zetasizer software, version 7.12 We used ethanol 99.5% as a dispersant because the FeS2 samples well dispersed in it Each sample would be repeated three times across the scanning session, corresponding to recognition of a regular rhythm 1142 2.4.2 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 ; : 1 e1 Thermo-gravimetric analyzer (TGA) Because thermal batteries must be operated at high temperature, FeS2 powder before and after ball milling process, binders, and the powder scraped off from the cathode thin films were treated with TGA analysis, PerkinElmer TGA7 series, in order to prove their thermal stability The heating range was set from 40 to 600 C at 10 C/min rate operated under nitrogen Two Na2SiO3 binder samples were prepared as guided below: (1) Liquid sample: original Na2SiO3 binder; (2) Powder sample: liquid Na2SiO3 binder was dried in the oven at 100 C for h to become solid binder, then heated up to around 300 C for approximately 10 to become powder binder after grinding by mortar and pestle 2.4.3 X-ray diffractometer (XRD) The structural changes of FeS2 active materials after the ballmilling process will be recognized by X-ray diffraction (XRD) According to that, XRD analysis was performed to identify the phases, composition, and structure of the raw and ball-milled FeS2 powder sample for comparison The sample was put on the sample holder which works as a substrate and then pressed flatly by using the measure glass to perform a better baseline and clearer peaks in the XRD result graph, followed by placing inside the XRD (Bruker) chamber operated with Cu Ka radiation (l ¼ 0.15418 nm) at 40 kV and 40 mA Data was recorded between 20 and 70 at a scan rate of around 2.4 /min 2.4.4 Adhesion test The adhesion test was conducted to check the substrateactive material connection as follow two methods Method one: The thin films were punched to produce the circleshaped electrodes (1.2 cm in diameter) Examining adhesion by seeing through the edge of the circle whether the active material layer was still glued smoothly to the substrate or exfoliated from the substrate Method two: Rectangularshaped pieces (3 cm cm) were cut from the thin films Dimensions of these pieces depend on the tape sizes used to the test We pasted the tapes to the rectangular-shaped pieces’ surfaces and applied the same pressing force to all the pieces from the surfaces, then peeling the tapes off The distributions of amounts of material over the tapes reflect the adhesion Results and discussion 3.1 Particle sizes of active material The FeS2 particle size was controlled by the ball-milled process and diminishes when the ball-milled time increases According to previous studies, the ball-milling process was conducted to the raw FeS2 powder for 24 h Fig 1a and b present the results of DLS analysis, the raw FeS2 particle size is 8.7 mm and 0.9 mm for the ball-milled FeS2 particle size The ball-milled FeS2 size is around times smaller than the raw FeS2 particle size Since different particle sizes of FeS2 own different thermal decomposition rates at the high operating temperature of the thermal battery, the electrochemical Fig e Intensity distribution analysis of (a) raw FeS2 powder and (b) ball-milled FeS2 powder performance of the thermal battery differs from the particle sizes of the FeS2 active material There is a certain size range for FeS2 particle, where thermal stability and electrochemical efficiency are maintained As the FeS2 particle size is too small, the discharge capacity of the thermal battery becomes lower due to the higher thermal decomposition rate of the FeS2 active electrode material However, if the particle size of FeS2 is too large, the homogeneous slurry as prepared for the electrode coating process can not be obtained and a large number of pores in the electrode will also exist, causing a rise in internal resistance [8] 3.2 TGA analysis In order to investigate the thermal stability of the electrode material used in this research, which is related thermal decomposition temperatures, the thermal-gravimetric analysis of the PVDF binder, raw FeS2 powder, ball-milled FeS2 powder, and ball-milled FeS2 film using PVDF binder (shaved the active thin-film layer off the copper foil substrate) was conducted The TGA data shows in Fig that PVDF binder has excellent thermal stability due to a low mass reduction rate 1143 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 ; : 1 e1 the excellent thermal stability of solid Na2SiO3 with a mass reduction rate around 4% at 500 C was indicated by TGA data of solid Na2SiO3 A large amount of solvent contains in liquid Na2SiO3 due to the obvious disparity between two states of Na2SiO3 This was observed as an explanation of the poor FeS2 active material layer-copper foil substrate contact when using Na2SiO3 as a binder 75 50 PVDF Raw FeS2 Ball-milled FeS2 Ball-milled FeS2 film 3.3 25 150 300 450 Temperature (°C) 600 Fig e TGA traces of PVDF binder, raw FeS2 powder, ballmilled FeS2 powder, and ball-milled FeS2 film under 1% at 450 C In addition, thermal stability of ball-milled FeS2 powder and ball-milled FeS2 film with a mass reduction rate under 10% at 450 C are acceptable, compared to that of raw FeS2 powder which possesses the low mass reduction rate under 2% at 450 C Basically, the discharge capacity of thermal batteries increases when internal resistance decreases, which results from the good connection between the active material particles in the electrode as lowering the active material particle size [28] On the other side, the thermal battery will be operated at the high temperature (350 C - 500 C), thus electrochemical efficiency greatly increases if the electrode material has high thermal decomposition temperature As mentioned before, in case the FeS2 particle size is too small, out of the certain size range, electrochemical efficiency of thermal batteries using FeS2 as an active electrode material will be declined because of high thermal decomposition rate of FeS2 active electrode material The thermal-gravimetric analysis results of the Na2SiO3 sample are shown in Fig The TGA data of liquid Na2SiO3 confirmed its poor thermal stability due to the high mass reduction rate of approximately 60% at 500 C In comparison, XRD analysis Any changes in the crystal structure of FeS2 will be verified by XRD analysis because chemical and physical damage may occur during the ballemilling process The XRD patterns of raw FeS2 powder and ballemilled FeS2 film using PVDF binder (shaved the active thin-film layer off the copper foil substrate) are presented in Fig 4a and b, respectively As shown, the reflections of the ball-milled FeS2 film are sharper and stronger in comparison with the raw FeS2 powder, proving that it possesses higher crystallinity A comparison of the XRD peaks of raw FeS2 powder and ballemilled FeS2 film proved that there is no change in the crystal structure before and after the ball milling process, demonstrating the 24 h ballemilling process did not change the phase of the FeS2 active material 3.4 FeS2 thin film coating and adhesion test Ball-milled FeS2 was used as an active material for the thinfilm coating process (blade coating method) after having characteristic test results to ensure its ability There were types of binders applied to produce FeS2 homogeneous slurry 3.4.1 With PVDF binder The first coating process was performed from the slurry consisting of FeS2 active material, small amounts of binder and no additive to maximize an amount of active material in an electrode The slurry was prepared by following formulations as shown in Table (e.g number to 2) with different ratios of PVDF binder and their appearances after coated and vacuum dried present in Fig It is clearly seen that their surfaces are uniformly smooth The adhesion test result illustrated in Fig (a) (b) Weight (%) 100 Liquid Na2SiO3 75 Solid Na2SiO3 Intensity (a.u.) Weight (%) 100 50 150 300 450 Temperature (°C) Fig e TGA traces of Na2SiO3 600 30 40 50 (degree) 60 70 Fig e XRD patterns of (a) raw FeS2 powder and (b) ballmilled FeS2 film 1144 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 ; : 1 e1 Fig e The appearances of the FeS2 electrode thin films produced from the slurry with the different PVDF binder ratios and no additive Fig e The adhesion test result of the FeS2 electrode thin films produced from the slurry with the different PVDF binder ratios and no additive Fig e The adhesion test result of the FeS2 electrode thin film produced from the slurry with the different PVDF binder ratios and CB added was conducted on these samples to make sure that the FeS2 electrode thin films are durable enough for cutting into the coin cell shape and during the electrochemical test As a result, the sample with 1.5% of PVDF binder has better adhesion This PVDF binder ratio will be applied for the next experiments The second coating process was performed from the slurry composed of FeS2 active material, 1.5 wt% of PVDF binder, and a carbonaceous additive (SP/CB/AC) in order to increase discharge capacities The slurry was prepared by following formulations as shown in Table (e.g number to 5) with the different additives added and their appearances after coating and drying under vacuum in the oven display in Fig The analogous result with the previous samples fabricated in the first coating process was obtained, which means good adhesion and smooth surface Moreover, the charge transfer resistance of samples added carbonaceous additives is reduced when the amount of a carbonaceous additive increases As a consequence of the formation of conductive networks j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 ; : 1 e1 1145 Fig e The appearances of the FeS2 electrode thin films produced from the slurry with 1.5 wt% of PVDF binder and different additives Table e The real thickness results from the slurry with wt% of PVDF binder and 15 wt% of a carbonaceous additive added after coating by the different doctor blade thicknesses FeS2: CB/SP/AC: PVDF ¼ 80 : 15: (wt%) Doctor blade thickness (mm) 150 200 300 Real thickness (without Al foil) (mm) CB SP AC 108 120 138 100 118 136 110 120 140 FeS2: pyrite, CB: carbon black, SP: super P, AC: activated carbon, PVDF: polyvinylidene fluoride between FeS2 particles and their high electrical conductivity, high charge transfer rate and conductivity density can be relieved [25] The next process is to improve the adhesion and the adhesion test results are shown in Fig After punching to manufacture the coin cell shapes, the FeS2 thin film with wt % of PVDF binder had better visual appearance than the sample with 1.5 wt% of PVDF binder which was cracked on the edges of the circles Because of that, coming samples were prepared by following wt% of PVDF binder ratio The last coating process was conducted to upgrade the mass loading of the electrode which may affect electrochemical efficiency due to increasing the amount of active material of an electrode Different thicknesses of FeS2 thin films were fabricated by using the doctor blade with the thickness range from 150 mm to 300 mm as describable in Table (e.g number to 14) The slurry containing 80 wt% of FeS2, 15 wt% of a carbonaceous additive (CB/SP/AC), and wt% of PVDF binder was coated on the ethanol-washed copper foil with the different doctor blade thicknesses and the real thickness results show in Table 3.4.2 With Na2SiO3 binder The FeS2 thin-film electrode using Na2SiO3 binder could not fabricate without a carbonaceous additive since different binder ratios, different dispersants, and different heating process (Table 1, e.g number 15 to 34) was tried and the result was the loss connection between the active material layer and the copper foil causing by exfoliated after vacuum dried in the oven as illustrated in Fig The thermal stability of carbonaceous material is low, so the amount of activated carbon was gradually reduced (60e20 wt%) to get the minimum amount of carbonaceous additive added, the slurry formulations as shown in Table (e.g number 35 to 45) When the amount of carbonaceous additive added was expected to diminish to 10 wt%, the exfoliated film was collected as also illustrated in Fig The appearance results of samples had CB added are illustrated in Fig 10 It can demonstrate that a good connection between the active material layer and the copper foil was obtained due to the utility of the matrix structure of carbonaceous material where FeS2 particles can be better confined Fig e The appearance of the FeS2 thin film formed from the slurry using 5% Na2SiO3 binder and no carbonaceous additive 1146 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 ; : 1 e1 Fig 10 e The appearance of the FeS2 thin film formed from the slurry using Na2SiO3 binder and CB added with different ratios (a) 60%, (b) 45%, (c) 30%, (d) 20% Fig 11 e The adhesion test result of the FeS2 electrode thin films with the PVDF and Na2SiO3 binder 3.5 Comparison of adhesion between PVDF and Na2SiO3 binder In order to compare mechanical durability between samples using PVDF and Na2SiO3 binder, the adhesion test would be carried out Fig 11 indicated that the FeS2 thin film using PVDF as a binder is more durable than the FeS2 thin film using Na2SiO3 as a binder As a result, the FeS2 thin film using PVDF binder is suggested to investigate in further researches 3.6 Electrode size As proved in the above part, the FeS2 thin film using a PVDF binder was supposed to be undertaken in the further detailed groundwork The sufficient size and mechanical strength of electrodes are required to ensure the performance of TBs In Fig 12, the PVDF-binder FeS2-active material electrode with a thickness of 138 mm, cut into discs from the prepared thin film, was objectively measured the size (around cm in diameter) (a) and mechanical strength (b) It was bent for durability testing, admittedly, the material layer still adhered to the substrate, without any materials peeling off In other words, great mechanical stability and flexibility of the electrode were visually demonstrated We can confirm that it met the mandatory requirements of the thermal-battery manufacture 3.7 Mass loading According to the previous session, there were some samples had been obtained successfully when using PVDF or Na2SiO3 binder for the slurry to produce FeS2 thin films As following preferences [8,29,30], the thickness of the conventional pellet cathode is around several hundreds of micrometers and decreasing the thickness of the cathode to a certain level has several exceptional advantages such as better ion-diffusion across the electrode and lower electrode resistance That means a certain-level thickness of the cathode, in case too 1147 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 ; : 1 e1 loading, the data as showed in Table for PVDF-binder samples and in Table for Na2SiO3-binder samples As a result, the mass loading increased with increasing the amount of FeS2 active material, more clearly proven in Na2SiO3-binder samples in Table 4, the amount of FeS2 from 30 to 70 wt% possessed the mass loading from 1.7 to 7.4 mg/cm2 in the stable state of the real thickness around 108 mm (Fig 14) Moreover, the mass loading also increased with the increase of the film thickness, it could be clearly seen in PVDF-binder samples in Table 3, various real thicknesses from 108 to 138 mm corresponding to the doctor blade thicknesses in a range of 150e300 mm (Fig 13), the mass loading was elevated from 1.4 to 3.6 mg/cm2 In comparison with Na2SiO3-binder, preparation of battery slurry using PVDF-binder provided a better connection between the active materials layer and the substrate (is more durable), along with a large-enough mass loading, which simply increased the mass of active materials and could lead to increase electrochemical efficiency (can increase electrochemical efficiency) of thermal batteries Therefore, the thin-film type with PVDF-binder will promisingly contribute to increasing electrochemical efficiency of thermal batteries Fig 12 e Image of the thermal-battery electrode: (a) Size examination, and (b) Mechanical stability inspection thin cathodes, it may cause reduce the electrochemical efficiency due to the sharp decrease of active-material mass loading Understandingly, some samples were achieved with the best results on this research, which were thoroughly measured the real thickness (without substrate) and the mass Conclusions In this research, using the blade coating method applied to the FeS2 homogenous slurry for the production of smooth and uniform thin films has been possessed It is such a result of the Table e FeS2 mass loading of the successfully coated samples with PVDF binder FeS2 (wt%) 80 80 80 CB/SP/AC (wt%) PVDF (wt%) Blade thickness (mm) Real Thickness (mm) Mass loading (mg/Cm2) 15 15 15 5 150 200 300 108 119 138 1.4 2.6 3.6 FeS2: pyrite, CB: carbon black, SP: super P, AC: activated carbon, PVDF: polyvinylidene fluoride Table e FeS2 mass loading of the successfully coated samples with Na2SiO3 binder FeS2 (wt%) 30 45 60 70 CB/SP/AC (wt%) Na2SiO3 (wt%) STP (wt%) Blade thickness (mm) Real thickness (mm) Mass loading (mg/Cm2) 60 45 30 20 5 5 5 5 150 150 150 150 108 108 108 108 1.7 3.8 4.9 7.4 FeS2: pyrite, CB: carbon black, SP: super P, AC: activated carbon, Na2SiO3: sodium silicate, STP: sodium tripolyphosphate Fig 13 e Image of the ratio of FeS2: CB: PVDF ¼ 80 : 15: wt% compositions with doctor blade thickness: a) 150 mm, b) 200 mm, c) 300 mm 1148 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 ; : 1 e1 FeS2 : AC : Na2SiO3 : STP = 30 : 60 : : FeS2 : AC : Na2SiO3 : STP = 60 : 30 : : FeS2 : AC : Na2SiO3 : STP = 45 : 45 : : FeS2 : AC : Na2SiO3 : STP = 70 : 20 : : Fig 14 e Image of the ratio of different compositions with a 150 mm doctor blade thickness ball-milling process to produce FeS2 powder with the smaller to 0.9 mm particle size According to the XRD result, there is no phase change after the ball-milling process of FeS2 The area of the FeS2 thin film is large enough with a size of 10 cm 15 cm, which is favorable for cutting to become electrodes of thermal batteries The FeS2 thin films fabricated using PVDF binder have outstanding advantages such as higher mechanical durability, simple and easy-to-make procedure than using Na2SiO3 binder Among the samples prepared with PVDF binder, the FeS2 thin film samples manufactured from the slurry consisting of 80 wt% of FeS2, 15 wt% of a carbonaceous additive (CB/SP/AC) and wt% of PVDF binder will be promising electrodes for thermal batteries with high electrochemical performance owing to the high electrical conductivity and good trap networks of carbonaceous materials In addition, real thicknesses of the tested sample from 108 to 138 mm fabricated from the doctor blade thicknesses in a range of 150e300 mm have the mass loading which is elevated from 1.4 to 3.6 mg/ cm2 maybe resulting in increasing capacity of TBs These samples reached the excellent coating results with smooth and uniform surfaces Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper references [1] Ko J, Kang SH, Cheong HW, Yoon YS Recent progress in cathode materials for thermal batteries J Kor Chem Soc 2019;56(3):233e55 [2] Kim D, Jung HM, Um S Theoretical analysis of the timedependent temperature evolution for thermal runaway prevention in multi-layered LiCl-LiBr-LiF thermal batteries J Kor Phys Soc 2009;55(6):2420e6 [3] Guidotti RA, Masset PJ Thermally activated (“thermal”) battery technology: Part IV Anode materials J Power Sources 2008;183(1):388e98 [4] Masset PJ, Guidotti RA Thermal activated (“thermal”) battery technology: Part IIIa: FeS2 cathode material J Power Sources 2008;177(2):595e609 [5] Masset PJ, Guidotti RA Thermal activated (“thermal”) battery technology: Part IIIb Sulfur and oxide-based cathode materials J Power Sources 2008;178(1):456e66 [6] Masset P Iodide-based electrolytes: a promising alternative for thermal batteries J Power Sources 2006;160(1):688e97 [7] Masset P, Schoeffert S, Poinso JY, Poignet JC LiF-LiCl-LiI vs LiF-LiBr-KBr as molten salt electrolyte in thermal batteries J Electrochem Soc 2005;152(2):A405e10 [8] Ko J, Kim IY, Jung HM, Cheong H, Yoon YS Thin cathode for thermal batteries using a tape-casting process Ceram Int 2017;43(7):5789e93 [9] Masset P, Guidotti RA Thermal activated (thermal) battery technology: Part II Molten salt electrolytes J Power Sources 2007;164(1):397e414 [10] Guidotti RA, Masset P Thermally activated (“thermal”) battery technology: Part I: an overview J Power Sources 2006;161(2):1443e9 [11] Cairns EJ Batteries In: Molten salt technology Boston, MA: Springer; 1982 p 287e321 [12] Bernardi D Mathematical modeling of lithium (alloy), iron sulfide cells and the electrochemical precipitation of nickel hydroxide Ph.D Thesis University of California; 1986 [13] Bernardi D, Newman J Mathematical modeling of lithium (alloy), iron disulfide cells University of California; 1986 [14] Sasaki A On the electrical conduction of pyrite Mineral J 1955;1(5):290e302 [15] Finklea SL, Cathey L, Amma EL Investigation of the bonding € ssbauer effect and X-ray mechanism in pyrite using the Mo crystallography Acta Crystallogr Sect A Cryst Phys Diffr Theor Gen Crystallogr 1976;32(4):529e37 j o u r n a l o f m a t e r i a l s r e s e a r c h a n d t e c h n o l o g y 2 ; : 1 e1 [16] Gupta VP, Ravindra NM, Srivastava VK Semiconducting properties of pyrite J Phys Chem Solid 1980;41(2):145e8 nchez C “n” type semiconductivity in natural [17] Echarri AL, Sa single crystals of FeS2 (pyrite) Solid State Commun 1974;15(5):827e31 [18] Caban-Acevedo M, Liang D, Chew KS, DeGrave JP, Kaiser NS, Jin S Synthesis, characterization, and variable range hopping transport of pyrite (FeS2) nanorods, nanobelts, and nanoplates ACS Nano 2013;7(2):1731e9 € hrich J, Tributsch H Growth of [19] Oertel J, Ellmer K, Bohne W, Ro n-type polycrystalline pyrite (FeS2) films by metalorganic chemical vapour deposition and their electrical characterization J Cryst Growth 1999;198:1205e10 [20] Aselage TL, Hellstrom EE Multicomponent phase diagrams for battery applications: II Oxygen impurities in the battery cathode J Electrochem Soc 1987;134(8):1932e8 [21] Guidotti RA, Reinhardt FW, Smaga JA Self-discharge study of Li-alloy/FeS/sub 2/thermal cells In: Proceedings of the 34th international power sources symposium IEEE; 1990, June p 132e5 [22] Garche J, Dyer CK, Moseley PT, Ogumi Z, Rand DA, Scrosati B, editors Encyclopedia of electrochemical power sources Newnes; 2013 [23] Guidotti RA, Reinhardt FW Anodic reactions in the Ca/ CaCrO4 thermal battery 1985 Sandia National Laboratories report SAND83-2271, [September] 1149 [24] Szwarc R, Walton RD, Dallek S, Larrick BF Discharge characteristics of lithium-boron alloy anode in molten salt thermal cells J Electrochem Soc 1982;129(6):1168e73 [25] Choi Y, Cho S, Lee YS Effect of the addition of carbon black and carbon nanotube to FeS2 cathode on the electrochemical performance of thermal battery J Ind Eng Chem 2014;20(5):3584e9 [26] Jung HJ, Na ES, Choi SC The fabrication and electrical properties of PMS-PZT ceramics using a tape casting method J Kor Chem Soc 2001;38(9):860e5 [27] Park JS, Lee SM, Han YS, Hwang HJ, Ryu SS Effects of debinding atmosphere on properties of sintered reactionbonded Si3N4 prepared by tape casting method J Kor Chem Soc 2016;53(6):622e7 [28] Oh I, Cho J, Kim K, Ko J, Cheong H, Yoon YS, et al Poly (imide-co-siloxane) as a thermo-stable binder for a thin layer cathode of thermal batteries Energies 2018;11(11):3154 [29] Wang X, Wang G, Chen J, Zhu X, Tian J, Jiang C, et al Pyrite thin films prepared for thermal batteries via sulfuring electrodeposited iron sulfide films: structure and physical properties Mater Lett 2013;110:144e7 [30] Hu J, Chu Y, Tian Q, Wang J, Li Y, Wu Q, et al Film cathode for thermal batteries using a screen-printing process Mater Lett 2018;215:296e9 ... batteries, the single-cell structure of the thermal battery consists of an electrolyte placed between an anode (LieSi) and a cathode (FeS2) [2e5] Specifically, the use of the heat source is to allow the. .. density of the thermal battery Therefore, thinning of the cathode electrode must ensure mechanical strength and optimal electrochemical efficiency of the electrode is essential and can be carried... temperature of the thermal battery, the electrochemical Fig e Intensity distribution analysis of (a) raw FeS2 powder and (b) ball-milled FeS2 powder performance of the thermal battery differs from the