Laser-irradiated coffee waste: a path to eco-friendly anode materials for lithium-ion batteries

Abstract

Hard carbon is being actively explored as a candidate anode material for next-generation batteries, offering ion storage mechanisms distinct from and potentially advantageous to those of graphite. However, conventional synthesis of hard carbon usually relies on high-temperature pyrolysis and chemical activation, which involve high energy consumption and environmental challenges. Spent coffee grounds (SCGs), generated in large amounts worldwide, represent an abundant biomass resource with high carbon content that is often discarded with limited recycling. In this study, SCGs were directly converted into hard carbon and partially graphitized structures using femtosecond laser direct writing (FsLDW) under solvent-free and non-vacuum conditions. Localized photothermal reactions induced by the ultrashort pulses promoted particle consolidation and structural rearrangement, and by varying the laser parameters, the carbonization pathway could be directed to favor either hard carbon or graphene domains. A copper substrate was deliberately employed to spontaneously generate copper compound nanoparticles, which were subsequently etched to create micro-porous carbon with tunable pore characteristics. This laser-based approach provides a controllable and one-step pathway for transforming SCGs into functional carbon anodes, highlighting the potential to selectively prepare materials suitable for both lithium-ion and sodium-ion batteries by adjusting only the laser conditions. While this study focused on lithium-ion battery applications, the tunable control of carbon structure and porosity offers the possibility of extending this strategy to sodium-ion batteries.

Graphical Abstract

Femtosecond laser converts spent coffee grounds into laser-induced hard carbon and graphene (LIH–LIG) composites, enabling eco-friendly anode materials for high-performance lithium-ion batteries

Supplementary Information

The online version contains supplementary material available at 10.1186/s40580-026-00561-1.

Introduction

With the global efforts toward building a sustainable society, the demand for eco-friendly materials and manufacturing technologies is steadily increasing [1, 2]. Among various energy conversion and storage technologies, lithium-ion batteries (LIBs) are widely used in electric vehicles (EVs), portable electronic devices, and energy storage systems (ESSs) because of their high energy density, long cycle life, and environmental compatibility [3]. In LIBs, the anode plays an important role in determining overall battery performance by enabling the insertion, storage, and extraction of lithium ions [4, 5]. Currently, graphite is the most widely used anode material due to its low cost, non-toxicity, and structural stability [6]. However, its reliance on non-renewable resources, the complexity of its production and purification, and associated environmental issues raise concerns about its long-term sustainability [7]. Consequently, the development of sustainable and cost-effective carbon materials derived from natural resources, and producible through low-complexity processes, is being increasingly recognized as a promising alternative.

Coffee is among the most widely consumed beverages worldwide and is the second most traded commodity after petroleum, with a significant and growing global market share [8]. As a consequence, more than ten million tons of spent coffee grounds (SCG) are generated globally each year [9]. SCG are disposed of in landfills, where they can acidify and degrade soils, reduce vegetation, and produce greenhouse gases [10, 11]. Most SCG are disposed of in landfills, where they can acidify and degrade soils, reduce vegetation, and produce greenhouse gases. Recycling of SCG is limited, with current uses mainly as fertilizers, animal feed, and compost [12]. Nevertheless, due to their unique physicochemical properties—being rich in fatty acids, amino acids, lignin, polysaccharides, polyphenols, tannins, and flavonoids—SCG have been explored for use in biodiesel production, antimicrobial devices, biofuels, adsorbents, and fine chemicals [13]. In addition, studies have investigated the carbonization of SCG through pyrolysis combined with catalytic processes for use as anode materials in LIBs [14, 15]. However, these thermos-catalytic processes generally require high temperatures (hundreds to over a thousand ℃) treatment with involving chemical processing, which entails high energy consumption, multiple processing steps, and challenges associated with the disposal of residual alkaline waste [16].

Laser-induced carbonization and graphitization have recently been studied as alternatives that can produce conductive carbon structures under non-contact, maskless, single-step conditions without complex pre-treatment [17]. The laser-induced graphene (LIG) fabrication technique, which converts non-contact carbon-precursor-containing materials into conductive graphene/graphite structures through laser irradiation, was initially developed for polyimide (PI) substrates, but has since been extended to various lignocellulosic materials such as wood, fallen leaves, paper, and cork [18–20]. However, most studies to date have been limited to solid substrates, and there have been no reported cases in which dispersed lignocellulosic waste powders—particularly coffee powder with high porosity and large specific surface area—have been directly irradiated with a laser to form carbon structures for use as anode materials. Previous LIG research has largely focused on forming electrodes on predefined substrates containing carbon precursors for use in electronics. The direct laser carbonization of lignocellulosic biomass powders, which combine high carbon content with morphological flexibility, for application as battery anodes remains largely unexplored.

In this study, SCGs were directly used as a carbon precursor to produce hard carbon and graphene structures in a single step under solvent-free, non-vacuum conditions using femtosecond laser direct writing (FsLDW). Using a high-repetition-rate femtosecond laser (1040 nm, 255 fs, 201.5 kHz), localized photothermal reactions were induced on the powder surface, enabling both particle consolidation and rearrangement of the carbon structure. By adjusting the laser power, the ratio and morphology of hard carbon and graphene could be selectively controlled. Furthermore, a copper substrate was deliberately used so that copper compound nanoparticles were formed within the hard carbon during laser irradiation without the use of additional copper powders or solutions; their content and morphology could also be tuned by the laser conditions. Subsequent etching of these copper compounds produced hard carbon with micro-pores, and the pore size and distribution were likewise controlled by the laser parameters. This ability to modulate structure and composition within the same platform introduces several important design principles. Varying the laser conditions enables a controllable balance between graphitization and pore formation, where graphitic domains improve conductivity and initial capacity, while hard-carbon frameworks with hierarchical porosity favor diffusion-limited storage and long-term stability. The substrate-derived Cu nanoparticles act as transient active species that can initially contribute to capacity but are best removed to achieve stable performance through additional porosity creation. Finally, from a processing standpoint, FsLDW provides a one-step alternative to conventional pyrolysis or furnace-based approaches, eliminating the need for extreme temperatures, long processing times, or controlled atmospheres, while offering direct, on-demand control of carbon phase composition and pore architecture in waste biomass powders. In particular, this method offers novelty by directly producing copper-containing functional carbon structures without additional processing and, through simple modulation of laser parameters, opens the possibility of selectively producing hard-carbon– and graphene-based anodes suitable for both LIBs and sodium-ion batteries (SIBs).

Experimental

Sample preparation

Spent coffee grounds (SCGs) were obtained from commercially purchased various coffee beans. The SCGs were collected after coffee brewing using coffee machine (De’Longhi, ECAM22.110.B). After collection, the SCGs were not washed prior to drying. The resulting SCGs were oven-dried at 80 °C for 12 h to eliminate residual moisture. The dried powders were ground in a planetary ball mill at 400 rpm for 3 h using stainless steel balls (75 mm diameter) with a ball-to-powder mass ratio of 10:1 (Figure S1). After milling, the powders were sieved through a 75 μm mesh, and only the fine fraction (≤ 75 μm) was collected. The sieved powders were evenly spread onto copper foil substrates, and an applicator was used to adjust the powder layer to ~ 100 μm thickness, producing a compact and uniform film suitable for laser irradiation.

Laser processing

Laser irradiation of the prepared SCG films was conducted using a ytterbium-doped fiber femtosecond laser system (Lasernics, FUPL-250-6) operating at a near-IR wavelength of 1040 nm, with a pulse duration of 255 fs, a repetition rate of 201.5 kHz, and a maximum output power of 7 W. The laser beam was delivered and scanned across the sample surface using a galvano scanner (PSXA 104 A) under ambient conditions. During processing, the laser power was tuned between 2.0 and 3.2 W to control the degree of carbonization and graphitization within the powder layer.

Post-treatment process

Laser-treated samples were treated in 1 M HCl solution at 50 °C for 3 h to remove residual Cu species (mild etching). For strong etching, treatment was extended to 5 h. After etching, all samples were washed with deionized (DI) water and dried overnight in an oven at 60 °C, then stirred in 4 M NaOH for 1 h to neutralize the residual acid, followed by repeated DI washing and oven drying.

Characterization

The morphology of CWAM (coffee waste derived anode materials) powders were observed using a field emission scanning electron microscope (FESEM, SU7000, HITACHI, Japan) and high-resolution transmission electron microscope (HRTEM, Tecnai G2 F30 S-TWIN, FEI, USA) operated at 300 kV. The elemental distribution of the CWAM samples was analyzed using energy-dispersive X-ray spectroscopy (EDS, Ultim Max65, Oxford Instruments, UK). X-ray photoelectron spectroscopy (XPS) measurements were carried out on a K-Alpha+ spectrometer (Thermo Scientific, USA) with Al-K excitation. Raman spectra were obtained using a Raman spectrometer (HEDA, WEVE, Korea) with a 532 nm excitation laser. X-ray diffraction (XRD) was performed using a diffractometer (D8 ADVANCE, Bruker, Germany) with a Cu-K radiation source (), a scanning speed of 1°min^-1 , and a 2 range of (10-80°). The analysis was conducted based on the fitting results of the XRD peaks and formulas (1) and (2). The interlayer spacing (d₀₀₂) and the crystallite thickness along the c-axis (Lc) were calculated using the (002) diffraction peak. The interlayer spacing (d₀₀₂) was calculated using Bragg’s law:

1

where “λ = 0.15406 nm " is the wavelength of Cu Kα radiation and the θ is the Bragg angle corresponding to the (002) reflection. Κ is the shape factor, which is 0.89 when Lc is calculated. The crystallite thickness (LC) was calculated using the Scherrer equation:

2

where β is the full width at half maximum (FWHM) of the (002) peak in radians [21, 22].

Electrochemical measurements

All the electrochemical tests of the CWAM samples were characterized in half cells (CR2032). The electrode slurry was prepared by mixing 80 wt% active materials, 10 wt% super P, and 10 wt% poly (vinylidene fluoride) (PVDF) binder in N-methyl-2-pyrrolidone (NMP). The slurry was cast onto a Cu foil current collector and dried at 100°C overnight in a vacuum oven. Electrodes with a diameter of 14 mm (area = 1.54 cm²) were punched, with an active material loading of 1.2–1.5 mg cm^− 2. An electrode sheet was used as the working electrode, lithium metal (Neba corporation, Korea) as the counter electrode, and Celgard2400 (Wellcos, Korea) as the separator. The coin cells were assembled in the argon-filled glovebox with O₂ and H₂O levels maintained below 0.1 ppm. The electrolyte consisted of 1 M LiPF₆ in a 1:1 (v/v) mixture of ethylene carbonate (EC) and ethyl methyl carbonate (EMC) with 2 wt% vinylene carbonate additives (Enchem, Korea). After assembly, cells were rested for 6 h at room temperature. The galvanostatic discharge and charge tests were performed in the voltage range of 0–2.5 V (vs. Li/Li⁺) at room temperature using a battery cycler (WBCS 3000, WonA Tech, Korea). Specific capacities and current densities were based on the mass CWAM active material (1 C = 372 mA g^− 1). The cyclic voltammetry (CV) was conducted under the control of a potentiostat (IVIUM, Netherlands) at a scanning rate of 0.1–2.0 mV s^− 1. Furthermore, the charge storage kinetics were analyzed based on CV curves. The contribution ratio of diffusion- and capacitive-controlled processes was calculated using the equation , where represents the capacitive contribution and corresponds to the faradaic intercalation contribution [21, 22].

Results and discussion

Femtosecond-laser conversion of spent coffee grounds into porous hard carbon anodes

The overall fabrication route is depicted in Fig. 1a. SCGs were first oven-dried to remove residual moisture and subsequently subjected to particle size reduction. As shown in Figure S1, the dried SCGs were ground using a ball milling process operated at 400 rpm for 3 h with a ball-to-powder weight ratio of 10:1. The average particle size decreased significantly from D = 712 ± 287 μm for the raw powders to D = 397 ± 106 μm after milling. To obtain fine and uniform fractions suitable for film preparation, the milled powders were sieved through a 75 μm mesh, and only particles smaller than this threshold (D ≤ 75 μm) were collected for further use. This size control ensured improved homogeneity and reproducibility in the subsequent fabrication steps. After particle size reduction and sieving, SCG powders were uniformly spread onto a recyclable copper substrate and compressed using a roller to form a compact and homogeneous film with a thickness of approximately 100 μm (Figure S2). The rolling step not only improved packing density and surface flatness but also provided a stable powder layer, which is essential for ensuring reproducible interaction with the femtosecond laser during subsequent irradiation. Following film preparation, FsLDW was employed to induce ultrafast, localized, and non-equilibrium laser–material interactions within the compacted SCG films. Under 1040 nm, 255 fs femtosecond laser irradiation, the pulse duration is much shorter than the characteristic heat diffusion time of the SCG powder layer. Therefore, the initial energy deposition cannot be described as uniform equilibrium heating. Instead, the irradiated SCG particles may undergo rapid electronic excitation, nonlinear/multiphoton absorption, and transient plasma formation, followed by electron–phonon coupling, localized heat accumulation, biomass decomposition, carbonization, and rapid quenching. This process simultaneously achieved carbonization and partial graphitization, thereby generating hierarchically porous carbon frameworks consisting of hard carbon and graphene-like domains. As a one-step, solvent-free strategy, FsLDW directly transformed biomass powders into conductive carbon architectures without the need for additional binders, catalysts, or chemical activation. The laser-induced hard carbon films were directly applied as anodes, with their porous structure and high conductivity enabling efficient ion transport and storage. The laser-induced hard carbon films were directly applied as anodes, with their porous structure and high conductivity enabling efficient ion transport and storage. To further clarify the position of the present approach relative to previously reported carbon-anode fabrication strategies, a structured comparison is provided in Table S1 in the Supporting Information. The comparison includes representative reports on laser-induced graphene, laser-carbonized anodes, biomass-derived hard carbon, waste coffee-derived carbon, catalytic graphitization, and porous carbon anodes. Compared with conventional substrate-bound LIG systems and furnace-pyrolyzed biomass hard carbon, the present strategy directly converts dry spent coffee waste powders into powder-form porous hard carbon with graphitic domains under ambient, non-vacuum conditions. In addition, the use of a Cu substrate and post-laser Cu etching provides a distinct route for Cu-assisted pore evolution, differentiating this work from previous coffee-waste-derived or biomass-derived carbon anode studies.

Fig. 1.

a Schematic illustration of the fabrication of laser-induced hard carbon from SCGs and its direct application as an anode in electrochemical cells. b, c SEM images of Spent Coffee Grounds (SCG) powder before laser irradiation. d, e SEM images of SCG powder obtained with a laser power of 2.0 W. f, g SEM images of SCG powder obtained with a laser power of 3.2 W. h, i SEM/EDS mapping showing the distribution of carbon (C), copper (Cu), and oxygen (O) in the SCG powder obtained with laser powers of 2.0 W and 3.2 W

In order to confirm the effect of laser irradiation on the transformation of spent coffee grounds (SCG) powder into carbon anode materials, the microstructural evolution of SCG powder was investigated under different laser powers ranging from 2.0 to 3.2 W. This range was selected as a stable processing window based on preliminary process-window screening, in which SCG powders could be reproducibly converted into carbonaceous anode materials while allowing meaningful control over the relative fractions of laser-induced hard carbon and graphene-like domains (Table S2). Figs. 1b-g illustrate the SEM images, showing surface morphology and microstructural changes of coffee-waste derived anode materials (CWAM) upon laser treatment. The bare coffee powder displayed a smooth and non-porous surface. Conversely, the laser-irradiated CWAM samples exhibited roughened surfaces with numerous pores of various sizes. Particularly, as laser power increased from 2.0 W to 3.2 W, the surface structure became increasingly porous (Figure S3), clearly demonstrating the formation of relatively larger pores under the 3.2 W condition. Such microstructural changes can be attributed to the release of volatile gases (e.g., H₂O, CO, CO₂) during high-temperature laser irradiation, resulting in pore formation within the carbon structure [23, 24]. Additionally, previous studies have reported that increasing carbonization temperature causes micropores to merge and transform into larger pores, aligning well with the larger pore formation observed with increased laser power in this study [25, 26]. Within this processing window, Raman spectra further confirmed the formation and evolution of disordered/graphitic carbon domains, while XRD profile fitting showed that the ordered carbon fraction increased from 11.13% at 2.0 W to 17.12% at 3.2 W, accompanied by a decrease in the disordered carbon fraction from 88.87% to 78.98%. The corresponding process-window screening results are summarized in Table S3 in the Supporting Information. Therefore, increased laser power plays a key role in controlling the microstructure of CWAM, influencing surface porosity and pore size. This indicates potential improvements in ion diffusion pathways and electrochemical reactivity through optimized laser conditions. To further verify the pore characteristics observed in the SEM images, N₂ adsorption–desorption measurements were conducted (Figure S4). The laser-irradiated sample (LSCG-3.2) exhibited hybrid characteristics of Type I and Type IV isotherms with an H4 hysteresis loop with BET surface area of 6.54 m² g^− 1, indicating the presence of a micro–mesoporous carbon structure. In addition, the BJH pore size distribution revealed mesoporous characteristics with pore sizes in the range of 2–50 nm.

Energy-dispersive X-ray spectroscopy (EDS) analysis was conducted on the samples with different intensities (2.0 –3.2 W) to examine surface composition of CWAM samples (Fig. 1h and i, and S5). Both samples primarily showed the presence of C, Cu, and O on the surface. Notably, the 3.2 W sample exhibited stronger signals of Cu and O, suggesting greater presence of Cu or copper oxide (CuO_x) under higher laser power conditions. The presence of Cu and O arises from CWAM samples being cast onto a Cu substrate and subsequently subjected to laser irradiation. Under high-power laser conditions, localized high-temperature environments can cause melting or vaporization of substrate copper, subsequently reacting with oxygen from the air or oxygen generated during thermal decomposition of organic materials, thus forming copper oxide. Consequently, laser irradiation conditions significantly affect both carbon structural formation and metal oxide generation, potentially influencing electrochemical performance directly or indirectly.

Characterization of LIG and LIH

The crystal structure change of SCG powder after laser irradiation was also measured by X-ray diffraction (XRD) analysis (Fig. 2a). The XRD patterns of CWAM samples irradiated at 2.0 and 3.2 W both exhibited a broad peak around 23°, corresponding to the (002) diffraction of amorphous (disordered) carbon. Additionally, a sharp peak appeared near 26.5°, indicative of highly ordered graphitic carbon. These findings confirm that both 2.0 W and 3.2 W samples contain a mixed structure of laser-induced graphene (LIG) and hard carbon. In particular, the 2.0 W sample exhibited a relatively higher proportion of hard carbon and a lower proportion of graphitic carbon compared to the 3.2 W sample. Conversely, the 3.2 W sample showed a higher ratio of graphitic carbon, suggesting the formation of a more developed LIG structure [27]. Therefore, Therefore, as laser power increases from 2.0 to 3.2 W, the pulse energy increases from 9.93 to 15.88 µJ under the same laser configuration, leading to higher local energy deposition and heat accumulation within the compacted SCG powder layer. However, this process should not be regarded as uniform equilibrium heating. Instead, femtosecond laser irradiation creates transient micro-scale reaction zones involving ultrafast energy deposition, localized temperature rise, rapid biomass decomposition, and subsequent rapid cooling. The increased local energy input at 3.2 W promotes partial carbon ordering and graphitization, while the short interaction time and rapid quenching preserve a substantial fraction of disordered hard-carbon-like domains. To further investigate the evolution of the crystalline structure with laser power, profile-fitting analysis was conducted on the XRD peaks between 20° and 30° (Figure S6) [28]. For both 2.0 W and 3.2 W samples, the (002) peak was successfully fitted into a broad peak representing disordered carbon and a sharp peak corresponding to ordered carbon domains. The fitting results revealed that the disordered region accounted for 88.87% and the ordered region for 11.13% in the 2.0 W sample, while the disordered region decreased to 78.98% and the ordered region increased to 17.12% in the 3.2 W sample (Table S3). These results indicate that the higher laser power, the higher the degree of graphitization of the CWAM samples. Additionally, as the laser power increases from 2.0 W to 3.2 W, the d-spacing of CWAM samples decreased from 0.371 nm to 0.363 nm, indicating a more ordered and structurally aligned carbon arrangement.

Fig. 2.

a XRD patterns of SCG powder obtained with laser powers of 2.0 W and 3.2 W. b, c Raman spectra of SCG powder obtained with a laser power of 2.0 W and 3.2 W. d, e Core-level XPS spectra of C 1s for SCG powder obtained at 2.0 W and 3.2 W. f, g Core-level XPS spectra of O 1s for SCG powder obtained at 2.0 W and 3.2 W

Additionally, Cu species originating from the Cu substrate, on which SCG powder was casted, were also detected within the samples during the laser irradiation process, as well as spent coffee grounds (SCG) are converted into laser-induced graphene (LIG) and hard carbon, simultaneously. In both the 2.0 W and 3.2 W samples, diffraction peaks corresponding to metallic Cu are observed at approximately 43.3°, 50.4°, and 74.1°, with the peak intensity increasing at higher laser power. In the 3.2 W sample, additional diffraction peaks associated with CuCO₃ related phases were observed, indicating the partial chemical transformation of substrate-derived Cu species during laser irradiation. The formation of these carbonate-related phases is unlikely to originate solely from atmospheric CO₂ because of its relatively low concentration under ambient conditions. Instead, they are more plausibly attributed to the localized reactive microenvironment generated at the SCG/Cu interface during femtosecond laser treatment [29, 30]. The transient high-temperature conditions induced by laser irradiation can promote rapid thermal decomposition and carbonization of the compacted SCG layer, releasing volatile species such as H₂O, CO, CO₂, and other oxygen-containing fragments derived from cellulose, hemicellulose, lignin, lipids, and other organic components in SCG [31, 32]. These locally generated carbon- and oxygen-containing species can reach concentrations substantially higher than those in the surrounding atmosphere near the heated Cu/C interface. At the same time, laser irradiation can activate the Cu substrate and facilitate the partial incorporation of Cu species into the carbonizing biomass layer. During the rapid heating–cooling process, these Cu species may react with locally generated CO₂/CO-containing and oxygen-containing decomposition products, leading to the formation of crystalline copper carbonate or copper carbonate-related phases [33, 34]. Thus, the observed carbonate-related diffraction peaks can be rationalized by a localized biomass-derived reactive environment rather than by atmospheric CO₂ exposure alone.

Carbon structures of CWAM samples via laser irradiation were further characterized by Raman analysis (Fig. 2b and c, and S7). Measurements were conducted at various positions under 2.0 W and 3.2 W laser conditions. The Raman spectra revealed two characteristics peaks: the D band (~ 1342 cm⁻¹), which corresponds to disordered carbon structures, and the G band (~ 1590 cm⁻¹), which is associated with the stretching vibrations of sp²-hybridized C = C bonds in graphitic carbon domains [35]. For the sample irradiated at 2.0 W, the spectra predominantly indicated an amorphous carbon structure. As the laser power increased to 3.2 W, a noticeable decrease in the intensity ratio of the D to G bands (I_D/I_G) was observed—from an average of 0.82 at 2.0 W (range: 0.74–0.92) to 0.58 at 3.2 W (range: 0.23–0.87). Additionally, at 3.2 W, the emergence of a distinct 2D band around 2700 cm⁻¹, typically associated with stacked graphene layers, was confirmed. These results suggest that increasing laser power promotes a reduction in structural defects and disorder within the carbon matrix, while simultaneously enhancing the formation of graphitized graphene domains—commonly referred to as laser-induced graphene (LIG) [36].

The elemental composition and chemical states of the CWAM samples were investigated using X-ray photoelectron spectroscopy (XPS). As shown in Fig. 2d and e, both the 2.0 W and 3.2 W samples exhibit a main C 1s peak centered at ~ 285 eV, along with characteristic O 1s and Cu 2p peaks at ~ 532 eV and ~ 934 eV, respectively. The high-resolution C 1s spectra (Fig. 2d and e) could be deconvoluted as sp² hybridized carbon (C-C peak), sp³ hybridized carbon (defects peak), carbon and oxygen single bonds (C-O peak), carbon and oxygen double bonds (C = O peak), carboxyl groups (O-C = O peak), Cu carbonate (CuCO₃ peak) for which the corresponding peak positions are at 284, 284.8, 286.3, 287.4, 288.29, 289.34 eV [37–40]. Notably, the relative area of the sp²-hybridized carbon (C–C peak) increased from 6.07% to 7.29% with increasing laser power, and the sp²/sp³ ratio also increased from 0.092 to 0.111, indicating an increase of approximately 20%. This trend suggests an increasing degree of graphitization with higher laser power, which is consistent with the results of Raman and XRD analyses.

The high-resolution O 1s spectra of the 2.0 W and 3.2 W CWAM samples are shown in Fig. 2f and g. The O 1s spectra of both CWAM samples exhibit four deconvoluted peaks corresponding to Cu–O (~ 530.1 eV), Cu–O–C / Cu(OH)₂ / C = O (~ 531.7 eV), C–O (533.1 eV), and O–C = O (534.0 eV) [41]. With increasing laser power, the integrated area of the C–O peak relatively decreases. This is attributed to the replacement of the C–O peak with the C–C/C = C sp² hybridization peak due to the cleavage of oxygenated groups during the photothermal and photochemical conversion, confirming the graphitization of the CWAM samples. Specifically, the Cu–O and Cu–O–C peaks observed at 530.1 eV and 531.7 eV, respectively, suggest that the Cu substrate was partially oxidized or chemically reacted due to localized thermal effects of laser irradiation, leading to the formation of copper oxides (CuO, Cu₂O), copper hydroxide (Cu(OH)₂), or copper carbonate (CuCO₃). Similarly, the high-resolution Cu 2p spectrum (Figure S8) exhibits three peaks located at 932.6 eV, 933.6 eV, and 934.8 eV, corresponding to metallic Cu, copper(II) oxide (CuO), and copper hydroxide (Cu(OH)₂) or copper carbonate (CuCO₃), respectively [42]. These findings, consistent with the O 1s results, confirm the presence of both metallic and oxidized forms of copper, which likely originated from the Cu substrate under high-power laser irradiation. Actually, according to the atomic percentage data presented in Figure S9 and Table S4, the surface Cu content increased from 2.50% in the 2.0 W sample to 4.48% in the 3.2 W sample, corresponding to an increase of approximately 79%. The presence of copper was further confirmed by the XRD analysis and subsequently observed as particles in the TEM images. This suggests that higher laser power enhances the interaction with the Cu substrate, thereby leading to a greater amount of copper species being detected on the sample surface. Although the amount of Cu-derived species increased with laser power, the enhanced graphitic ordering observed in the 3.2 W sample is not considered to originate from a catalytic effect of Cu. Unlike typical graphitization catalysts such as Fe, Co, and Ni, Cu possesses extremely low carbon solubility and therefore cannot effectively promote graphitic carbon formation through dissolution–precipitation or carbide-mediated mechanisms [43, 44]. Instead, the increase in graphitic ordering observed in the XRD, Raman, and XPS analyses is primarily attributed to localized photothermal effects generated at higher laser powers, which facilitate thermal rearrangement of the carbon framework and the development of sp²-bonded domains [26, 28, 45]. Therefore, the increased Cu content is considered a consequence of the laser-induced thermal environment rather than the primary factor governing the evolution from disordered to ordered carbon structures.

Electrochemical properties

Figure 3 presents the comparative electrochemical performance of coffee waste-derived carbon anode materials evaluated in Li-ion half cells under different laser irradiation conditions. Fig. 3a shows the rate performance of CWAM samples irradiated with varying laser powers (2.0–3.2 W). The sample treated with the highest laser power of 3.2 W exhibited the highest specific capacity, whereas the 2.0 W sample showed the lowest at all current rates. At 0.1 C, the 3.2 W sample delivered a capacity of 221 mAh g⁻¹, compared to 161 mAh g⁻¹ for the 2.0 W sample, representing an increase of approximately 37%. Similarly, at 2 C, the 3.2 W and 2.0 W samples exhibited capacities of 105 and 93 mAh g⁻¹, respectively, corresponding to a difference of about 13%. This improved performance of the 3.2 W sample is attributed to additional capacity contributions from multi-step electrochemical conversion reactions between lithium and Cu-containing species, including Cu-carbonate, Cu-hydroxide, and Cu-oxide phases [46–49]. However, this contribution should not be considered as the sole origin of the performance enhancement, because higher laser power simultaneously increases graphitization, modifies porosity, and changes the defect/surface structure of the carbon framework. Therefore, Cu-containing species are interpreted as a contributing component rather than an isolated causal factor. Under the 3.2 W condition, it is assumed that a higher laser power leads to the dissolution of Cu from the substrate, resulting in the formation of a composite structure consisting of Cu nanoparticles embedded within a matrix of laser-induced graphene (LIG) and laser-induced hard carbon (LIH). This structural modification enables the formation of electrochemically active Cu-containing species, including carbonate-, hydroxide-, and oxide-related Cu phases, which further contribute to the overall capacity through conversion-type reactions during lithiation, de-lithiation processes. This structural modification enables the formation of electrochemically active Cu-based metal oxides, which further contribute to the overall capacity through CuO-based conversion reactions. As a result, the 3.2 W sample exhibits enhanced capacity not only due to lithium intercalation/de-intercalation within the carbon matrix, but also from additional conversion-type reactions involving Cu species.

Fig. 3.

a Rate capability of SCG powder electrodes prepared with laser powers ranging from 2.0 to 3.2 W, evaluated by galvanostatic charge/discharge tests at current densities from 0.1 C to 2 C. The de-lithiation capacities in the first cycle are compared. b, c Galvanostatic charge/discharge (GCD) curves of the 2.0 W and 3.2 W samples measured in the voltage range of 0–2.5 V at current densities corresponding to the 2nd, 7th, 12th, 17th, and 22nd cycles (0.1, 0.2, 0.5, 1.0, and 2.0 C, respectively). d, g CV curves of the 2.0 W and 3.2 W samples at a scan rate of 0.1 mV s⁻¹. e, h CV curves of the 2.0 W and 3.2 W samples at scan rates of 0.1, 0.2, 0.5, 1.0, and 2.0 mV s⁻¹ in the voltage range of 0–2.5 V, obtained in the second cycle. f, i Normalized capacitive contribution ratios of the 2.0 W and 3.2 W samples calculated at different scan rates. j, k GCD profiles at 0.1 C under constant-current–constant-voltage (CC–CV) mode for the 2.0 W and 3.2 W samples; cells were discharged at 0.1 C to 0 V and held at 0 V until the current decreased to 0.005 C. l, m Long-term cycling performance of the 2.0 W and 3.2 W samples measured at 0.1 C

Figure 3b and c present the galvanostatic charge/discharge (GCD) profiles of the 2.0 W and 3.2 W samples, respectively. The GCD curve of the 2.0 W sample (Fig. 3b) exhibits a typical lithiation/de-lithiation profile commonly observed in hard carbon anodes [46]. In contrast, the GCD profile of the 3.2 W sample (Fig. 3c) reveals two distinct voltage plateaus in the regions of 1.25–1.5 V and 0.6–0.9 V. These features are attributed to the incorporation of Cu nanoparticles into the carbon matrix under high laser power, which leads to additional capacity contributions through multi-step electrochemical conversion reactions involving Cu-containing species during lithiation/de-lithiation processes. Notably, the two voltage plateaus in Fig. 3c gradually fade with cycling, indicating the depletion of irreversible Cu-related reactions over repeated charge/discharge processes.

CV curves of the CWAM electrodes also confirms the similar trends in the cathodic peaks of GCD. As shown in Fig. 3d, an anodic peak at approximately 0.2 V is attributed to the de-lithiation of lithium, while the cathodic peak near 0 V corresponds to the lithiation process. These redox features are primarily associated with lithium intercalation/de-intercalation within the carbon structure, which is consistent with the low content of Cu-related species under the lower laser power condition (Fig. 2a, S9 and Table S4). This implies that lithium storage in the 2.0 W sample mainly originates from intercalation and capacitive mechanisms within the carbon matrix [46]. In contrast, the CV profile of the 3.2 W sample (Fig. 3g) exhibits two reduction peaks at approximately 1.3 V and 0.7 V, which are associated with the conversion-type reduction reactions of Cu-containing species, including carbonate-, hydroxide-, and oxide-related Cu species. During the subsequent anodic scan, oxidation peaks at approximately 0.2 V, 1.3 V, and 2.5 V are observed, which can be attributed to de-lithiation from the carbon matrix and oxidation processes involving Cu-containing species [47–50]. Furthermore, as shown in Fig. 3h, the Cu-related redox peaks gradually become less distinguishable with increasing scan rates. This indicates sluggish ion diffusion and kinetically limited solid-state conversion reactions of the Cu-containing species at higher scan rates. These results suggest that Cu-based multi-step conversion reactions are kinetically limited, and that capacitive processes become increasingly dominant under fast scan conditions.

To gain a deeper understanding of the lithium-ion storage mechanism of CWAM, a kinetic study based on cyclic voltammetry (CV) was conducted (Fig. 3e and f, 3 h and 3i). Diffusion-controlled contributions were quantitatively estimated from the CV data. As shown in Fig. 3f and i, the 3.2 W sample exhibits consistently higher diffusion-controlled current contributions than the 2.0 W sample at all scan rates. Specifically, at scan rates of 0.1, 0.2, 0.5, 1.0, and 2.0 mV s⁻¹, the diffusion-controlled contributions of the 3.2 W sample are 75%, 68%, 57%, 49%, and 40%, respectively, whereas those of the 2.0 W sample are 70%, 63%, 52%, 43%, and 35%. These results are closely associated with the increased dissolution of Cu from the substrate under higher laser power conditions, which facilitates the formation of electrochemically active Cu-containing species via reactions with the carbon matrix. These Cu-containing species contribute not only through faradaic intercalation reactions but also via conversion-type reactions, thereby significantly enhancing the diffusion-controlled current contribution during the electrochemical process [51]. Therefore, the enhanced diffusion contribution observed in the 3.2 W sample can be attributed to a synergistic lithium storage mechanism involving both the carbon anode materials composed of laser-induced graphene (LIG), laser-induced hard carbon (LIH)—and conversion-type reactions involving Cu-containing species, which are induced by laser power-dependent structural modifications.

Figures 3 L and 3 m show the charge–discharge profiles of the 2.0 W and 3.2 W samples at 0.1 C after 50, 100, and 200 cycles, respectively. During long-term cycling, the 2.0 W sample shows a greater increase in de-lithiation capacity than the 3.2 W sample. Interestingly, the specific capacity of the 2.0 W sample increases from 177.1 to 192.7 and 226.4 mAh g^− 1 at 50, 100, and 200 cycles, respectively, whereas the 3.2 W sample exhibits a more gradual increase from 193.5 to 201.4 and 212.5 mAh g^− 1. Notably, beyond 150 cycles, the 2.0 W sample achieves a higher capacity than the 3.2 W sample during extended cycling. This behavior is attributed to a gradual activation process in the 2.0 W sample, which facilitates the development of additional lithium diffusion pathways and improves lithium-ion mobility [52, 53]. To further explore the lithium storage behavior under low-potential conditions, CC–CV mode was applied by holding the voltage at 0.001 V after constant-current discharge. For the 2.0 W sample, the capacity significantly increased from 184.4 to 268.9 mAh g⁻¹ when the voltage was maintained at 0.001 V, while the 3.2 W sample exhibited a smaller increase from 239.9 to 262.2 mAh g^− 1 (Fig. 3j and k). This capacity enhancement is attributed to the gradual insertion of Li⁺ ions into previously inaccessible interlayer spaces within the graphene-like domain, thereby activating additional intercalation sites [54]. In particular, the 2.0 W sample is interpreted to provide a greater number of lithium storage sites compared to the 3.2 W sample.

Characterization of LIG and LIH after post-treatment process

To further improve the Li ion storage in CWAMs, an additional post-treatment process was applied to the CWAM samples prepared at 2.0 W and 3.2 W (Fig. 4a). A mild acid etching was first conducted using HCl solution, followed by neutralization with NaOH to stabilize the surface structure [54]. The SEM images in Figs. 4b–e show the microstructural changes of the CWAM samples after the etching and neutralization processes. The main carbon frameworks of both samples remained unchanged from their pre-etching state. As observed in energy-dispersive X-ray spectroscopy (EDS; Fig. 4f and g), carbon (C) and oxygen (O) were primarily detected. Compared to the pre-treatment state, Cu signals were no longer detected after post-treatment (Fig. 4h and i), implying removal of Cu and Cu-based compounds. Accordingly, clear differences in porosity inside carbon frameworks were identified. In particular, the 2.0 W sample exhibited a developed porous structure after the post-treatment. This suggests that the post-etching treatment removed Cu-containing species, which may have exposed additional microporous structures within the carbon framework. However, pore formation should not be attributed exclusively to Cu removal, because laser-induced biomass decomposition and volatile gas release can also contribute to the development of porous carbon structures. Therefore, Cu-containing species are considered transient or sacrificial components associated with pore exposure rather than the only pore-forming origin.

Fig. 4.

a Scheme showing Cu etching process of SGC powder. b-e SEM images of Spent coffee grounds (SCG) powder obtained with a laser power of (b, c) 2.0 W and (d, e) 3.2 W. f, g SEM/EDS mapping showing the distribution of carbon (C) and oxygen (O) in the SCG powder obtained with a laser power of 2.0 W and 3.2 W. h, i EDS spectra of Spent coffee grounds (SCG) powder after Cu etching obtained with a laser power of 2.0 W and 3.2 W

To elucidate the formation of pore structures in the CWAM samples, a sequential Cu etching process—consisting of mild and strong etching—was employed. The resulting microstructures were then investigated using transmission electron microscopy (TEM; Fig. 5). As shown in the low-magnification TEM images of samples via mild etching, the carbon framework is composed of wrinkled and interconnected nanosheets, forming a porous structure with embedded Cu particles (Fig. 5a). The HR-TEM images revealed that the microcrystalline structure after mild etching, where laser-induced hard carbon, laser-induced graphene, embedded Cu nanoparticles, and mesopores were observed in Fig. 5b, c and d, and 5e, respectively. Firstly, laser-induced hard carbon showed highly curved graphene-like domains with expanded interlayer spacing (~ 0.37 nm), which are typical features of hard carbon (Fig. 5b). Between these curved layers, distinct mesoporous regions are also identified. These mesopores could be attributed to localized thermal effects and gas evolution during laser irradiation, accompanied by structural rearrangement of carbon and decomposition of organic species [26, 55]. Secondly, in some regions, laser-induced graphene is also identified, exhibiting an interlayer spacing of ~ 0.334 nm (Fig. 5c), consistent with the (002) planes of graphitic carbon from the XRD results [56]. Thirdly, a representative Cu nanoparticle is observed in Fig. 5d, displaying a lattice spacing of 0.206 nm, which corresponds to the (111) plane of metallic Cu. Fourthly, the selective removal of Cu by mild etching may have created voids where Cu particles were originally embedded (Fig. 5e).

Fig. 5.

TEM and HRTEM images of CWAM samples after (a-e) mild and (f-j) strong Cu etching

After the strong Cu etching process, Cu nanoparticles are barely observed in the TEM images, suggesting that Cu was largely removed (Fig. 5f). Except for the removal of residual copper species, the remaining carbon structure (Figs. 5g-j) —including LIH, LIG, and mesopores—remains intact compared to the mildly etched sample. Clearly, hard carbon structures with meso-pores (diameter of 20 nm) are observed (Fig. 5g and h). These meso-pores were created via the removal of Cu nanoparticles, consistent with the TEM images after mild etching (Fig. 5e). In addition, wrinkled LIG creates micropores with 2 nm of diameter as well (Fig. 5i and j). In summary, the etched CWAM samples exhibit a hybrid carbon structure composed of LIH, LIG, meso-pores via Cu etching, and micro-pores via graphene-like wrinkled layers. This complex carbon structure is expected to provide favorable pathways for ion storage and diffusion, thus demonstrating strong potential as a high-performance anode material for Li-ion battery.

Electrochemical properties after post-treatment process

Figures 6a-j present the comparative electrochemical performance of 2.0 and 3.2 W CWAM (coffee waste-derived carbon anode materials) after etching process. Fig. 6a compares their electrochemical rate performance. In contrast to the as-prepared samples via laser-irradiation, the 2.0 W sample exhibited higher capacities than the 3.2 W sample at all C-rates. Specifically, the 2.0 W sample achieved 255.8 mAh g^-1 at 0.1 C and 139.2 mAh g^-1 at 2 C, whereas the 3.2 W sample showed 237.4 mAh g^-1 and 117.6 mAh g^-1, respectively. The first de-lithiation capacity after etching increased to 255.8 mAh g^-1 for the 2.0 W sample and 237.4 mAh g^-1 for the 3.2 W sample (Fig. 6b and c), corresponding to improvements of 95 and 16.5 mAh g^-1, respectively, compared to their pre-etched states. Obviously, the increase of capacity in the 2.0 W samples was attributed to the increase of lithiation reaction below 0.5 V (Fig. 6b). Accordingly, the de-lithiation capacity below 0.5 V increased from 61 mAh g^-1 to 120 mAh g^-1 for the 2.0 W sample, where the capacity contribution in this voltage range rose from 38% to 48% after the etching process. This low-voltage range is primarily associated with lithium intercalation between graphitic-like nanodomains and lithium clustering within closed micropores in hard carbon [27]. Thus, it implies that the dissolution of Cu-related species through the etching process leads to exposure of microporous structures, thereby enabling effective Li⁺ storage and improved electrochemical performance. As direct evidence, just after the hydrochloric acid (HCl) treatment without NaOH neutralization, both the 2.0 W and 3.2 W samples exhibited a distinct voltage drop near 0 V observed at 0.2, 0.5, 1, and 2 C (Figure S10), indicating the remained H⁺ ions block Li ion-accessible pathways in graphitic interlayers and microporous domains [57]. After a NaOH neutralization treatment, such phenomenon was disappeared (Figure S11), where samples showed recovery of the sloping region near 0 V and a remarkable increase in reversible capacity. Therefore, these results suggests that the post-treatment contributes to optimizing the electrode structure and improving lithium storage performance. To examine whether the etching process altered the surface chemistry, additional C 1s XPS analyses were performed on the etched samples (Figure S12 and Table S5). Prior to etching, the 3.2 W sample exhibited higher reversible capacity, ICE, and cycling stability despite having a lower oxygen-functional-group content than the 2.0 W sample. After etching, the opposite trend was observed, where the 2.0 W sample delivered superior electrochemical performance despite exhibiting a higher oxygen-functional-group content than the 3.2 W sample. These observations indicate that oxygen-containing functional groups are not the dominant factor governing the electrochemical behavior in the present system. Previous studies have reported that moderate variations in oxygen-containing functional groups exert only a limited influence on lithium insertion/de-insertion capacity, whereas the electrochemical performance of carbonaceous anodes is more strongly governed by structural characteristics such as ion-accessible storage sites, pore structure, and defect distribution [58, 59]. Therefore, the enhanced electrochemical performance after etching is primarily associated with the structural modifications induced by Cu removal rather than significant changes in surface chemistry.

Fig. 6.

a Rate capabilities of SCG powder samples (2.0 and 3.2 W) after Cu etching and NaOH neutralization treatment, evaluated at current densities ranging from 0.1 C to 2 C. The de-lithiation capacities in the first cycle are compared. b, c Galvanostatic charge/discharge curves of the post-treated 2.0 W and 3.2 W samples measured at 0.1 C in the voltage range of 0–2.0 V during the second cycle. d, g CV curves of the 2.0 W and 3.2 W electrodes recorded at a scan rate of 0.1 mV s⁻¹. e, h CV curves measured at scan rates ranging from 0.1 to 2 mV s⁻¹ in the voltage range of 0–2.5 V. f, i Normalized capacitive contribution ratios of the 2.0 W and 3.2 W electrodes calculated at different scan rates. j Long-term GCD cycling performance of the 2.0 W and 3.2 W samples at 1.0 C

According to the CV analysis shown in Fig. 6d and g, the Cu conversion reaction was no longer observed in either the 2.0–3.2 W samples after the etching process. This indicates that lithium storage occurs only through insertion/extraction processes within the carbon structure. To gain further insight into the lithium-ion storage mechanism of CWAM after the etching, a kinetic study based on cyclic voltammetry (CV) was performed (Fig. 6e, f and h, and 6i). As shown in Fig. 6f and i, the 2.0 W sample exhibited consistently higher diffusion-controlled contributions than the 3.2 W sample across all scan rates. At scan rates of 0.1, 0.2, 0.5, 1.0, and 2.0 mV s^− 1, the diffusion-controlled contributions of the 2.0 W sample were 63%, 55%, 43%, 35%, and 28%, respectively, while those of the 3.2 W sample were 56%, 47%, 36%, 29%, and 22%. This trend is attributed to the relatively higher hard carbon content and lower laser-induced graphene (LIG) ratio in the 2.0 W sample. In contrast, the 3.2 W sample exhibited a greater proportion of surface-controlled (capacitive) contributions, likely due to the higher LIG content, which promotes lithium-ion adsorption on surface sites and structural defects. This behavior is consistent with previous reports, which showed that increasing the carbonization temperature induces long-range ordering and promotes graphitization of the carbon framework [28]. Consequently, the lithium storage behavior of the 2.0 W sample is primarily governed by diffusion-controlled contributions, due to its high hard carbon content and well-preserved microporous structure, while the 3.2 W sample exhibits relatively higher capacitive contributions, which are ascribed to lithium-ion adsorption at defective sites of the LIG structure.

To further investigate the influence of the etching process on the interfacial characteristics and charge-transfer behavior of the electrodes, electrochemical impedance spectroscopy (EIS) measurements were conducted (Figures S13, S14 and Table S6). The Nyquist plots reveal that both the SEI resistance () and charge-transfer resistance () decreased after etching for the 2.0 W and 3.2 W samples. Specifically, the values decreased from 110.0 to 89.5 Ω to 65.0 and 76.9 Ω, respectively, indicating improved electrode/electrolyte interfacial characteristics after the etching treatment. This improvement is attributed to the partial removal of Cu-related species, which likely suppressed undesirable interfacial side reactions and excessive SEI formation [60]. In addition, the values decreased from 103.9 to 98.0 Ω to 70.9 and 83.0 Ω for the 2.0 W and 3.2 W samples, respectively. The reduced charge-transfer resistance suggests enhanced electrochemical reaction kinetics after etching, which can be associated with the formation of additional ion-accessible nanopores and electrochemically active sites following Cu removal. These results are consistent with the improved reversible capacity and lithium-storage behavior observed for the etched samples [61].

The cycling performance of the 2.0 W sample was compared with the commercial hard carbon at 1 C (Fig. 6j). After 250 cycles, the 2.0 W sample and the hard carbon exhibited de-lithiation capacities of 273 and 176.9 mAh g^− 1, respectively, indicating that the 2.0 W sample delivered approximately 54% higher capacity than the commercial hard carbon. This improvement is primarily attributed to the porous structure of the 2.0 W sample, which provides more accessible ion diffusion pathways and enhances electrical conductivity. Therefore, CWAM irradiated with a 2.0 W laser exhibits notable long-term cycling stability and enhanced de-lithiation capacity compared to commercial hard carbon, demonstrating its potential as an anode material for lithium-ion batteries.

Conclusions

Spent coffee grounds (SCG) powders, uniformly spread on Cu substrates, were successfully converted into carbon anodes by femtosecond laser irradiation under ambient, solvent-free, and non-vacuum conditions. The systematic investigation revealed that laser power determines the microstructure, phase composition, and electrochemical behavior of the coffee-waste–derived anode materials. At 2.0 W, the products were hard-carbon–dominated with abundant microporosity, supporting gradual activation and stable long-term cycling. At 3.2 W, the materials exhibited higher graphitization and smaller interlayer spacing, together with an increased fraction of graphene-like domains. Importantly, substrate-derived Cu species were present in both conditions, with greater abundance and higher oxidation states at 3.2 W. SEM, XRD, Raman, XPS, and TEM consistently confirmed a hybrid framework of laser-induced hard carbon (LIH), laser-induced graphene (LIG), and Cu-containing species, with their relative proportions tunable by laser parameters.

Electrochemically, the graphitic-richer, Cu-heavier 3.2 W samples delivered higher initial capacities, partly due to conversion-type reactions involving Cu-containing species, but suffered from accelerated fading. By contrast, the hard-carbon–rich 2.0 W samples showed more diffusion-controlled storage and superior cycling stability. A sequential acid-etch/neutralization post-treatment removed Cu nanoparticles in both cases and simultaneously generated additional micro/mesopores. After etching, conversion-related signatures disappeared, Li-ion accessibility improved, and performance gains were most pronounced for the 2.0 W material, which reached ~ 273 mAh g^− 1 after 250 cycles at 1 C, surpassing commercial hard carbon (~ 176.9 mAh g^− 1).

These results yield clear design principles. First, there is a controllable trade-off between graphitization and pore formation: graphitic domains bolster conductivity and initial capacity, whereas hard-carbon frameworks with hierarchical porosity provide robust, diffusion-limited storage and durable cycling. Second, laser-induced, substrate-derived Cu nanoparticles act as transient templating/active species that can enhance early capacity but should be removed to unlock stable hard-carbon performance via porosity creation. Finally, from a processing perspective, femtosecond laser direct writing (FsLDW) overcomes key limitations of conventional pyrolysis/furnace routes—eliminating extreme temperatures, long dwell times, controlled atmospheres, and limited microstructural tunability—while enabling one-step, on-demand control of carbon phase composition and pore architecture directly on spread biomass powders.

This work establishes hard carbon as the primary functional phase governing long-term performance in SCG-derived electrodes and positions femtosecond laser processing as a scalable, sustainable platform for engineering high-performance hard-carbon anodes from waste resources. The combined control of graphitization level, hierarchical porosity, and formation/removal of substrate-derived metal by-products provides a practical blueprint for tailoring next-generation secondary battery anodes.

Authors’ contributions

S. E. Park and T. Jing contributed equally to this work. S. E. Park: Writing – original draft, Visualization, Resources, Investigation, Data curation, Formal analysis, Validation. T. Jing: Writing – original draft, Writing – review & editing, Resources, Validation, Data curation. M.-R. Kim: Visualization, Formal analysis. D. Yang: Visualization, Formal analysis. H. K. Nam: Conceptualization, Methodology, Validation, Supervision, Project administration, Funding acquisition. S. Park: Conceptualization, Methodology, Validation, Supervision, Project administration, Funding acquisition. Y.-J. Kim: Conceptualization, Methodology, Validation, Supervision, Project administration, Funding acquisition.

Funding

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (RS-2024-00401786, No. RS-2024-00447869) and supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government (MOTIE) (No. RS-2025-25460980). This work was also supported by the Korea Forestry Promotion Institute (Project No. RS-2023-KF002530), Korea Institute of Planning and Evaluation for Technology in Food, Agriculture and Forestry (IPET) through Technology Commercialization Support Program, and funded by Ministry of Agriculture, Food and Rural Affairs (MAFRA) (No. RS-2024-00401642).

Data availability

All data gathered or analyzed in this study are included in the article.

Declarations

Competing interest

The authors declare no competing financial interests.