Native minerals boost performance of coal-derived hard carbon at a low pyrolysis temperature

PAPER • OPEN ACCESS Native minerals boost the performance of coalderived hard carbon at low pyrolysis temperatures To cite this article: Piotr Madajski et al 2026 J. Phys. Energy 8 015008 View the article online for updates and enhancements. You may also like OsPd bimetallic dimer pushes the limit of magnetic anisotropy in atom-sized magnets for data storage Jan Navrátil, Michal Otyepka and Piotr Boski - Large scale characterization and calibration strategy of a SiPM-based camera for gamma-ray astronomy C. Alispach, J. Borkowski, F.R. Cadoux et al. - Site environment characterization for Southern Wide-field Gamma-ray Observatory Vlastimil Jílek, Alena Bakalová, Ladislav Chytka et al. - This content was downloaded from IP address 90.177.128.125 on 03/12/2025 at 11:24 J. Phys. Energy 8(2026) 015008 https://doi.org/10.1088/2515-7655/ae1e2b Journal of Physics: Energy OPEN ACCESS RECEIVED 5 September 2025 REVISED 3 November 2025 ACCEPTED FOR PUBLICATION 11 November 2025 PUBLISHED 3 December 2025 Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. PAPER Native minerals boost the performance of coal-derived hard carbon at low pyrolysis temperatures Piotr Madajski1, Ievgen Obraztsov2,∗, Anita Cymann-Sachajdak3,4, Monika Wilamowska-Zawłocka4, Aristides Bakandritsos2,5, Myroslav Sprynskyy6 and Grzegorz Trykowski1,∗ 1Department of Chemistry of Materials, Adsorption and Catalysis, Faculty of Chemistry, Nicolaus Copernicus University in Torun, Gagarina 7, Toru´ n 87-100, Poland 2Regional Centre of Advanced Technologies and Materials (RCPTM), Czech Advanced Technology and Research Institute (CATRIN), Palack´ y University Olomouc, ˇ Slechtitel˚ u 27, Olomouc 77 900, Czech Republic 3Hanse-Wissenschaftskolleg, Institute for Advanced Study, Lehmkuhlenbusch 4, 27 753 Delmenhorst, Germany 4Department of Energy Conversion and Storage, Faculty of Chemistry, Gdansk University of Technology, Narutowicza 11/12, 80-233 Gda´ nsk, Poland 5Nanotechnology Centre, Centre of Energy and Environmental Technologies, Vˇ SB-Technical University of Ostrava, 17. listopadu 2172/15, 708 00 Ostrava-Poruba, Czech Republic 6Department of Environmental Chemistry and Bioanalytics, Faculty of Chemistry, Nicolaus Copernicus University in Torun, Gagarina 7, Toru´ n 87-100, Poland ∗Authors to whom any correspondence should be addressed. E-mail: [email protected] and try[email protected] Keywords: sodium-ion, lithium-ion, potassium-ion, battery, gagat, hard carbon, mineral doping Supplementary material for this article is available online Abstract In response to the growing demand for energy-efficient and sustainable industrial processes, this study investigates the use of gagat—a naturally occurring, coal-like material, as a low-cost and scalable precursor for hard carbon (HC) anodes in sodium-, lithium-, and potassium-ion batteries. HC was synthesized using a one-step, low-temperature pyrolysis process, eliminating the need for chemical purification and improving energy efficiency and environmental friendliness. The resulting gagat-derived HC (GHC) exhibits tunable nanoporosity, governed by the pyrolysis temperature, and incorporates electrochemically active mineral phases that enhance performance. The GHC anodes deliver competitive electrochemical performance and high cycling stability across Naand Li-ion systems. Due to its abundance, simple processing, and residual minerals that enhance the plateau capacity by ∼43% compared to washed material, gagat shows strong potential for nextgeneration energy storage. 1. Introduction The rapid expansion of green energy technologies and electric vehicle manufacturing has intensified the need for scalable, secure, and cost-effective energy storage systems [1,2]. Li-ion batteries dominate the market, but their growth is limited by the uneven and insecure supplies of critical materials, such as lithium, cobalt, and graphite [3,4]. The EU is especially vulnerable, relying heavily on imports—over 78% of global graphite comes from China, whereas the EU produces only 0.042%, primarily in Austria and Germany [5,6]. This reliance has spurred research into alternative battery chemistries, notably Na-ion (SIBs) and K-ion batteries (KIBs), which leverage more abundant and geographically diversified elements such as sodium and potassium [7]. Among the candidate anode materials, carbon remains a leading choice due to its chemical stability, affordability, and high electrochemical efficiency [8]. Yet, natural graphite has limited applicability for SIBs and KIBs due to its narrow interlayer spacing and severe volume changes during cycling (up to 60%–100%) that compromise performance [9,10]. In contrast, hard carbons (HCs) exhibit disordered structures with larger interlayer spacing, making them better suited for Na+and, to a lesser extent, K+storage. They also offer improved rate performance and negligible volume expansion. These © 2025 The Author(s). Published by IOP Publishing Ltd J. Phys. Energy 8(2026) 015008 P Madajski et al are compatible with heteroatom doping (e.g. N, S, and O), which introduces additional active sites and enhances the electronic conductivity and ion diffusion [11–14]. Although HCs can be used independently or blended with graphite to achieve high-rate capability (up to 6C), the source material and processing route strongly influence their costs and sustainability profile [12,15]. HCs can be derived from numerous sources including biomass (e.g. coconut shells, nutshells, and sugars) [16,17], petroleum and coal derivatives (e.g. pitches, coal, and anthracite) [18,19], and synthetic polymers [20]. Although biomass offers a low-cost, renewable option, it typically suffers from high energy demand. Most scalable HC routes need ⩾1100 ◦C–1500 ◦C to realize closed pores in order to achieve large plateau fractions [21– 26]. Cutting this temperature while avoiding chemical washing would materially reduce the costs and waste of processing anode materials. Other issues include low carbon yield and inconsistent composition [16]. Petroleumand coal-derived carbons provide higher yields (∼50%) and electrical conductivity, but often have narrow interlayer spacing, reducing their compatibility with large alkali ions [27]. Meanwhile, synthetic polymers offer excellent control and reproducibility, but their high cost, environmental impact, and poor sustainability limit their industrial relevance. Therefore, there is a demand for HC anode materials that combine (1) low-energy production processes, (2) compatibility with large alkali ions, and (3) raw material abundance. In this study, we investigate HC derived from gagat—a naturally occurring, lignite-like sedimentary rock using a low-energy synthesis approach. Gagat (G-raw), a second-stage metamorphic coal, is a transitional form between lignite and bituminous coal. It is soft (aiding milling), widely available, and contains ∼70% carbon along with inherent heteroatom doping (S, O, and N) from its geological origin [28, 29]. Its native minerals may enable the structural evolution of gagat from amorphous to turbostratic carbon while promoting limited pore formation, without any preprocessing through a simple pyrolysis process at 500 ◦C–900 ◦C aimed at determining the lowest practical temperature for effective carbon transformation. The resulting gagat-derived HC (GHC) exhibits competitive capacity and rate performance in Naand Li-ion systems, and somewhat limited performance in K-ion systems, with tunable porosity, expanded interlayer spacing, and improved electrochemical behavior enabled by natural S-doping. We hypothesize that native calcite and iron sulfides in gagat catalyze carbon structure evolution during low-temperature pyrolysis, thereby increasing the plateau capacity and rate performance. This work emphasizes the potential of naturally occurring minerals as an efficient source of active materials (AMs) for sodiumand lithium-ion batteries due to their cost-effectiveness, low-energy processing, and inherent structural and compositional advantages, leading to synergistic effects that enhance electrochemical performance. 2. Results and discussion 2.1. Characterization of GHC The morphology of raw and pyrolyzed materials was examined using scanning electron microscopy (SEM). All powders comprise layered particles with irregular surfaces, ranging from 18 to 23 µm in size (figures 1(a)–(c) and S1). Elemental analysis demonstrated that the carbon content of G-raw was 60 wt%, increasing to 70 wt%, 87 wt%, and 90 wt% after pyrolysis at 500 ◦C, 700 ◦C, and 900 ◦C, respectively (table S1). Particle morphology remained largely unchanged at 500 ◦C and 700 ◦C (figure S1), but pyrolysis at 900 ◦C led to the formation of surface-bound nanoparticles (figures 1(c), S1(k), and (l)). Elemental mapping by SEM/EDX revealed the presence of calcium, sulfur, oxygen, and traces of iron (figure 1(g)). Calcium overlapped spatially with sulfur, indicating the likely presence of CaS on the surface of GHC-900, derived from native gagat minerals such as calcite and pyrite [28,29]. To assess the impact of mineral and organic decomposition on microporosity [30], the Brunauer–Emmett–Teller (BET) measurements were performed to determine the specific surface area (SSA). It was ∼3 m2g−1 for G-raw and remained the same for GHC-500, whereas it increased to 34 ±2 m2g−1for GHC-700 and 166 ±2 m2g−1for GHC-900 (table S1), reflecting structural changes at increased pyrolysis temperature. This change is accompanied by the formation of ∼0.6–1.1 nm pores in GHC-900 (figure S2), which are desirable for increasing the plateau capacity and typically achievable only at higher pyrolysis temperatures [31,32]. High-resolution transmission electron microscopy (HR–TEM) was used to assess changes in the nanomorphology of the materials. The structure of GHC-500 remained amorphous (figure 1(d)), whereas GHC-700 and GHC-900 demonstrated increasing carbon order (figures 1(e) and (f)), consistent with higher pyrolysis temperatures. The nanopores of GHC-900 were visible in HR–TEM micrographs, aligned with the BET results (figure 1(f)). Furthermore, the structural evolution of GHC carbon during pyrolysis was monitored by x-ray diffraction (XRD). G-raw and GHC-500 displayed amorphous structure and presence of minerals 2 J. Phys. Energy 8(2026) 015008 P Madajski et al Figure 1. Scanning electron microscopy micrographs and high-resolution transmission electron micrographs of (a), (d) GHC500, (b), (e) GHC-700, and (c), (f) GHC-900. Pseudo-graphitic domains (‘Stacks’) are visible in GHC-700 and GHC-900, whereas GHC-900 additionally exhibits sub-nm voids (‘Pores’); (g) energy dispersive x-ray spectroscopy (EDX) elemental mapping of GHC-900 sample. accompanying G-raw: CaCO3, SiO2, and FeS2(figure 2(a)). Upon pyrolysis at higher temperatures, broad (002) and (100) carbon peaks appeared near 23◦and 44◦, respectively, indicating the formation of low-crystalline carbon. In GHC-700 and GHC-900, CaCO3and SiO2reflections disappeared, whereas the emergence of Fex–1Sx, CaS, FeS2, and Fe0reflections suggests the thermal decomposition of calcite and reduction of iron sulfides. These phases match the elemental distributions observed in the EDX mapping (figure 1(g)), confirming the transformation of inorganic species during thermal treatment. Deconvolution of the broad, asymmetric (002) diffraction peaks of GHC-700 and GHC-900 was performed to quantify the relative contributions of highly disordered and pseudographitic domains (figures 2(a)–(c)). The disordered fraction is typically linked to the sloping region of the discharge curve, favoring alkali-ion adsorption. In contrast, the pseudographitic domains support cation intercalation, contributing to the plateau region of the capacity profile [33]. For GHC-700, the disordered and pseudographitic fractions were 67.0% and 33.0%, respectively. In GHC-900, this ratio shifted to 62.2% and 37.8%, indicating a modest increase in the structural order with higher pyrolysis temperature. The average interlayer spacings (d(002)), calculated using Bragg’s law [34], were 0.390 nm for GHC-700 and 0.389 nm for GHC-900, lying within the optimal 0.37–0.40 nm range associated with high-performance HCs for Na-ion storage (Note S1) [35]. The crystallite size along the c-axis (Lc), calculated from the (002) peak, was 0.91 nm for GHC-700 and 0.95 nm for GHC-900. The corresponding average number of stacked graphene layers was ∼3.3 and ∼3.4, respectively, indicating a small increase in pseudographitic domain development in GHC-900. Raman spectroscopy was used to further probe the nanostructure of the GHC samples. The analysis focused on the G band (∼1590 cm−1), indicative of well-ordered sp2carbon, and the D band (∼1340 cm−1), associated with structural defects and disordered graphene domains (figures 2(d)–(g)). The ID/IGratio increased from 0.70 to 1.01 with rising pyrolysis temperature, suggesting enhanced nanocrystalline domain formation in GHC-900 (figure 2(d)). Spectral deconvolution using the four Voigt functions [36] revealed D1 (∼1350 cm−1), D3 (∼1500 cm−1), D4 (∼1200 cm−1), and G (∼1580 cm−1) 3 J. Phys. Energy 8(2026) 015008 P Madajski et al Figure 2. (a) X-ray diffractogram of G-raw and GHCs; XRD (002) peak fitting of the (b) GHC-700 and (c) GHC-900; (d) Raman spectra for the pyrolyzed samples; deconvoluted D and G bands for the spectrum of the (e) GHC-700 and (f) GHC-900; (g) the deconvoluted Raman peak area ratio, (h) the average crystallite sizes (La) determined from Raman and XRD data, and (i) infrared spectra of the samples. bands (figures 2(e), (f), and S3, table S2). Although XRD reveals modest stacking order in GHC-900, Raman spectroscopy depicts a higher D1 band area, which is associated with defects and disorders in the carbon structure, compared to GHC-700. These observations are consistent with turbostratic carbon, wherein the stacking order and the in-plane disorder induced by heteroatoms within sp2domains can evolve independently [37]. Crystallite sizes (La) were estimated via both XRD and Raman data. Debye– Scherrer analysis yielded Lavalues of 2.05 nm and 3.74 nm for GHC-700 and GHC-900, respectively [34]. In contrast, the Ferrari–Robertson model applied to the Raman data yielded Lavalues of 1.26 nm and 1.35 nm, respectively (figure 2(h), Notes S1 and S2) [38]. These reduced values from Raman analysis reflect the structural disorder associated with the amplified D1 band due to sulfur incorporation into the carbon framework. Infrared (IR) spectra (figure 2(i)) were recorded to investigate the thermal evolution of surface chemistry in the GHC samples. The decreasing intensity of the ∼1600 cm−1band with increasing pyrolysis temperature indicates the loss of labile oxygen-containing functional groups [39,40]. In contrast, the emergence of bands at ∼1130 cm−1and ∼1480 cm−1in GHC-700 and GHC900 corresponds to the formation of C=S and C–S(=O)xsurface groups, respectively [41,42]. Inorganic Caand Fe-based species, which exhibit weak IR activity, were not observed in the spectra. Furthermore, the surface chemistry of GHC-700 and GHC-900 was analyzed using x-ray photoelectron spectroscopy (XPS). The survey spectrum of GHC-700 revealed carbon, oxygen, sulfur, and trace amounts of calcium and silicon (figure 3(a), table S3). In contrast, GHC-900 demonstrated significantly stronger O 1s and Ca 2s/2p signals, along with well-resolved S 2s and S 2p peaks, indicating substantial surface compositional changes. IR spectroscopy displayed a marked decline in organic oxygencontaining groups and is largely insensitive to inorganic surface species. Given the XPS surface sensitivity, the increase in the O 1s signal for GHC-900 is attributed to CaS decomposition under ambient 4 J. Phys. Energy 8(2026) 015008 P Madajski et al Figure 3. (a) X-ray photoelectron spectroscopy (XPS) survey spectra of GHC-700 and GHC-900; high-resolution S 2p spectrum with peak deconvolution for (b) GHC-700 and (c) GHC-900; high-resolution C 1s spectra with peak deconvolution for (d) GHC-700 and (e) GHC-900. conditions [43]. This is supported by the spatial overlapping of S, Ca, and O in SEM/EDX mapping images and the increase in carbonate peak C 1s peak at ∼289.7 eV (figures 1(g) and 3(e)). In GHC700, sulfur was primarily bonded to aromatic carbon (figure 3(b)). In GHC-900, the S 2p spectrum displayed three components: metal sulfide (∼161.5 eV), sulfur bonded to aromatic carbon (∼163 eV), and sulfones (∼169.5 eV), with a peak area ratio of 1:1:2.16, respectively (figure 3(c)). The metal sulfide was identified as Ca and Fe sulfides with SEM/EDX mapping and XRD (figures 1(g) and 2(a)), whereas sulfone and aromatic sulfur species were embedded in the carbon matrix. HR C 1s spectra (figures 3(d) and (e)) revealed dominant sp2carbon (∼284 eV), C-S/C-O (∼286.4–286.5 eV), and C=O (∼287.7 eV) groups (table S4) [44]. The sp3carbon contribution (∼285.3 eV) was minor and decreased at higher pyrolysis temperatures, consistent with structural ordering. 5 J. Phys. Energy 8(2026) 015008 P Madajski et al Figure 4. Rate capability, second-cycle galvanostatic charge-discharge profiles at a C/20 current rate, and stability cycling at a 1C current rate for GHC anodes in (a)–(c) Na-ion, (d)–(f) Li-ion, and (g)–(i) K-ion systems. 2.2. Electrochemical characterization The electrochemical performance of the GHC anodes was evaluated in Na-, Li-, and K-ion half-cells to assess their compatibility with different alkali metal ions and to investigate the impact of minerals on the structure of GHC during pyrolysis on storage behavior (figure 4). Porosity, intrinsic and extrinsic defects, sulfur doping and turbostratic carbon development play critical roles in accommodating the ions of varying sizes and enhancing performance across all three chemistries. For that, the current rate capability and cycling stability were tested for GHCs in Na-ion (figures 4(a) and (b)), Li-ion (figures 4(d) and (e)), and K-ion system (figures 4(g) and (h)) respectively. Cycling stability was assessed at 1C, using a diagnostic protocol of five cycles at C/5 every 100 cycles to distinguish capacity changes due to activation, degradation, or diffusion limitations. 2.2.1. Na-ion First, the electrochemical performance of GHC anodes was evaluated in Na-ion half-cells. GHC-900 exhibited the highest capacity, 268 mAh g−1, at C/20, with only a slight decrease to 256 mAh g−1 at C/10. At higher current rates, the capacities were 214 mAh g−1(C/5), 136 mAh g−1(C/2), and 104 mAh g−1(1C). Notably, GHC-900 fully recovered its original capacity after the rate test, indicating excellent rate stability and structural reversibility. In contrast, GHC-700 delivered lower capacities at all rates: 224 mAh g−1at C/20, 183 mAh g−1at C/10, 147 mAh g−1at C/5, 114 mAh g−1 at C/2, and 90 mAh g−1at 1C. This reflects the differences in graphitization degree and porosity between the samples. GHC-500, which was pyrolyzed at the lowest temperature, exhibited poor capacity (∼100 mAh g−1) regardless of the rate and did not demonstrate a distinct voltage plateau. Initial Coulombic efficiency (ICE) follows a temperature-dependent trend with values of 56.9% (GHC-900), 50.0% (GHC-700), and 32.5% (GHC-500). In subsequent cycles, both GHC-700 and GHC-900 reached >99.4% CE, indicating stable cycling and efficient SEI formation. In contrast, GHC-500 exhibited sluggish CE improvement, reflecting irreversible reduction in the film. All GHCs exhibited excellent cycling stability (figure 4(c)). The ICE of GHC-900 lies at the lower end of the range typically reported for HCs, which is attributed to surface functionalization and relatively high surface area. However, the most comparable materials achieving higher ICE values are obtained only at significantly higher pyrolysis temperatures (table S5). Other approaches to HC preparation for use as Lior Na-ion batteries were reported. 6 J. Phys. Energy 8(2026) 015008 P Madajski et al A two-step process, including the high-pressure pretreatment of the mangrove biomass at 500 ◦C for 1–10 d, followed by a two-stage carbonization at 800 ◦C and 1000 ◦C under argon. The resulting carbon featured uniform ultramicropores (<0.4 nm), and the HC-P6D7 sample exhibited a high reversible lithium storage capacity of 364 mAh g−1[45]. Another approach involved a three-step synthesis of coal-derived HC for sodium-ion batteries, involving the precleaning of bituminous coal, Zn2(OH)2CO3– assisted ball milling to generate uniform pores, and a two-stage heat treatment at 700 ◦C and 1300 ◦C, followed by acid washing. This resulted in abundant closed pores, enhancing the reversible sodium storage capacity to 325 mAh g−1[46]. A simple one-step pyrolysis of subbituminous coal in argon at 1000 ◦C–1500 ◦C and the sample prepared at 1300 ◦C (SHC-1300) demonstrated the best performance with a reversible Na+storage capacity of 291 mAh g−1[47]. These materials show a better or similar performance compared to GHC-900 (table S5); however, such results are obtained following complex multistep synthetic processes and high-temperature pyrolysis. The effect of G-raw composition on transformation of GHCs shows a great potential for energy-efficient material optimization. GHC-500 and GHC-700 anodes maintained their initial capacities over 300 cycles, whereas the GHC900 anode showed a ∼4% increase in capacity, likely due to gradual electrode activation. High stability contrasts with typical Fe-containing HCs, where surface Fe catalyzes electrolyte decomposition [48,49]. HR XPS detected no Fe within the ∼2–5 nm sampling depth, indicating an Fe-free surface. Thus, metallic Fe is encapsulated in the carbon matrix rather than exposed at the electrode-electrolyte interface, isolating catalytic sites and suppressing Fe-driven parasitic reactions, which accounts for the observed long-term stability. Finally, the reproducibility of the GHC anode performance was verified using electrodes fabricated from three independently prepared slurries and films (figure S4). The average capacities at C/20 were 263.0 ±6.6 mAh g−1for GHC-900, 217.2 ±24.9 mAh g−1for GHC-700, and 107.0 ±14.9 mAh g−1for GHC-500, consistent with the trends and values discussed above. 2.2.2. Effect of mineral removal on Na-ion storage The ratio of plateau to sloping capacity in HC anodes is a crucial determinant of the energy density in alkali-ion batteries [50]. This ratio reflects the proportion of pseudographitic versus disordered carbon domains, which is primarily governed by the pyrolysis temperature and the choice of precursor material [51]. Plateau/slope fractions were determined from differential capacity plot as 0.01–0.15 V (plateau) and 0.15–2.7 V (slope) (figure S7(a)). For GHCs, the plateau capacity fraction for Na-ion storage increases markedly with pyrolysis temperature: 4.6% for GHC-500, 22.4% for GHC-700, and 45.2% for GHC-900 (figure S5). Notably, the plateau capacity fraction of GHC-900 exceeds that reported for HC pyrolyzed at 1000 ◦C [51], and the absolute plateau capacity is higher than that of HCs synthesized via the more energy-intensive process [50] or tailored sulfur-doped HC [52]. Moreover, typically, the preparation of HC involves a chemical washing step using organic solvents, acids, or bases to remove inorganic impurities and improve electrochemical performance (table S5). However, these treatments increase production costs and generate chemical waste, making them less favorable for sustainable manufacturing. To evaluate the necessity of such pretreatment steps, we compared the electrochemical performance of GHC-900 with two chemically treated variants: GHC-900A, which was acid-washed with 1M HCl after pyrolysis to remove residual inorganic components and assess the effect of their presence on the electrochemical performance of GHC. The second sample, GHC-900AB, was pretreated with 1M NaOH and 1M H₂SO₄before pyrolysis to evaluate the effects, that the inorganic components have on the GHC during pyrolysis. Untreated GHC-900 demonstrated superior capacity, rate capability, and a higher fraction and absolute value of plateau capacity compared to both treated samples (figure S6). This behavior originates from the synergistic role of native minerals present in the gagat (G-raw), namely CaCO3and FeS2[28, 29], which enables structural transformations typically requiring higher temperatures. First, CaCO3is rapidly decomposed into CaO and CO2when the temperature reaches 750 ◦C [30], leading to increased porosity (table S1). The CO2evolution promotes open pore formation at 800 ◦C due to the etching of carbon with the generation of CO (CO2+C→2CO) [53,54]. Concurrently, FeS2undergoes a reductive transformation in the formed reductive medium (FeS2→FeS →Fe) [55], which was observed in XRD diffractograms (figure 2(a)). Metallic Fe acts as a catalyst promoting local graphitization and the formation of pseudographitic domains at lower temperatures [56]. The GHC-900 sample at C/20 delivers an average plateau capacity of 121.2 ±9.1 mAh g−1over four cycles, compared to 84.4 ±3.7 mAh g−1 for GHC-900AB. That is, the improvement of the GHC-900 plateau capacity over GHC-900AB was ∼43% (figures S6(e) and (f)). The trace amounts of FeS2and Fex–1Sxnanoparticles present in the GHC900 (figure 2(a)) contribute to the sloping capacity via conversion-type reactions near ∼1.5 V (figure S6) [57]. Their contribution was estimated as 11–18 mAh g−1, based on the difference in slope capacity between the washed samples and the pristine GHC-900 voltage profiles (figure S6(f)). Finally, the 7 J. Phys. Energy 8(2026) 015008 P Madajski et al formation of surface sulfones could improve electrode wettability and promote beneficial SEI formation, similar to the effects of sulfonated electrolyte additives [58,59]. 2.2.3. Li-ion Compared to the Na-ion system, the GHCs’ performance in the Li-ion system is substantially different. The GHC-700 capacity was higher than that of GHC-900 being 497.1 mAh g−1at C/20, 384 mAh g−1 at C/10, and 331.3 mAh g−1at C/5, whereas at higher rates, it became equal with GHC-900 and finally the capacity of GHC-900 was higher at 1C (figure 4(d)). Nearly the entire GHC-700 capacity is confined to the sloping region (figures 4(e), S5(c) and (d)) reflecting unequal Li-ion adsorption energy in the disordered structure of HC. In contrast, GHC-900 pyrolyzed at higher temperatures demonstrated both plateau and sloping capacity in the 32.8% to 67.2% ratio, respectively. In the stability test, the anodes showed very good performance. After a mild decrease in capacity during 10 initial cycles, the capacity remains stable for 300 consecutive cycles at 1C. 2.2.4. K-ion In the K-ion system, GHC delivers moderate capacity with pronounced low-voltage hysteresis an evidence of transient K-ion trapping/release. GHC-700 and GHC-900 exhibited similar rate performance, with GHC-900 showing only a slight advantage. GHC-900 delivered a capacity of 189.4 mAh g−1at C/20, which gradually decreased with increasing current rate to 61.6 mAh g−1at 1C (figure 4(g)). The plateau capacity fractions were 31.6% for GHC-700 and 40% for GHC-900, indicating a better pseudographitic character of the latter (figure S5(e) and (f)). Stability cycling revealed a reversible capacity evolution in the GHC-900: initial capacity fade followed by full recovery upon return to low-rate cycling (figure 4(h)). This behavior suggests K-ion trapping/release in the material [60,61]. Differential capacity analysis further confirmed sluggish kinetics in the plateau capacity region, showing pronounced chargedischarge hysteresis (figure S7). This effect is consistent with sluggish K+transport and higher interfacial polarization relative to Na+, reflecting the larger K+radius and its stronger mechanical/kinetic impact on disordered carbons [62,63]. The resulting GHC performance decrease in K-ion systems is likely caused by two factors: (i) transport/structural constraints due to pseudographitic interlayers of 0.389–0.390 nm and (ii) micropores that are beneficial for Na+become more kinetically restrictive for larger K+. 2.3. Kinetic analysis in the Na-ion system Kinetic analysis was conducted for Na-ion storage to evaluate the impact of pyrolysis treatment on the electrochemical behavior of GHC anodes derived from G-raw. The initial cyclic voltammograms (CVs) of all GHC samples exhibit two irreversible cathodic peaks at approximately 1.4 V and 1.2 V during the first cycle, corresponding to the decomposition of ethylene carbonate (EC) and fluoroethylene carbonate (FEC), respectively, and the formation of the solid electrolyte interphase (SEI) [64]. The GHC-500 sample (figure S8(a)) lacks a distinct anodic peak at ∼0.3 V, indicating limited Na+intercalation, likely due to its amorphous carbon structure. Instead, an anodic peak observed at ∼0.43 V is attributed to the conversion reaction of FeS2to metallic Fe and Na2S [65,66]. The corresponding cathodic peak at ∼1.43 V is associated with the formation of Na2FeS2during discharge (figure S8(a)) [67]. In contrast, the GHC-700 (figure S8(b)) and GHC-900 (figure S8(c)) exhibit well-defined redox pairs at ∼0.01 V (cathodic) and ∼0.3 V (anodic), which correspond to the reversible insertion and extraction of Na ions within the carbon matrix [68]. Notably, the CV curves’ shapes of GHC-900 and GHC-700 differ significantly, suggesting distinct Na-ion storage mechanisms arising from structural differences induced by varying pyrolysis temperatures. The distinct CV curves’ profiles of GHC-700 and GHC-900 highlight differences in their Na-ion storage mechanisms [69]. GHC-500, which was subjected to an insufficiently high pyrolysis temperature, remained electrochemically ineffective. To gain a deeper insight into the charge-storage kinetics, multirate CV measurements were conducted (figures 5(a) and (b)), and the corresponding b-values were determined at 0.02 V (plateau) and 0.5 V (slope) to identify the dominant charge-storage mechanism in each region (figure 5(c), Note S3). GHC-700 exhibited b-values of 0.90 and 0.73 for the slope and plateau regions, respectively, indicating surface-controlled domination in slope and mixed control in the plateau region, respectively. In contrast, GHC-900 demonstrated bvalues of 0.82 (slope) and 0.67 (plateau), revealing a shift to a diffusion-controlled charge storage mechanism in the plateau region. To further analyze the electrochemical kinetics, Dunn’s method was used to quantify the fraction of surface-controlled charge storage. At 0.1 mV s−1, the surface-controlled contribution was estimated to be ∼32% for GHC-700 and ∼46% for GHC-900. The surface-controlled charge contribution increased linearly with the scan rate, and at 0.9 mV s−1reached ∼56% and ∼78% for 8