1 Introduction

The growing demand for green technology has subsequently resulted in an increased need for environmentally-friendly materials and technologies. A hurdle hindering this quest for green technology is the task of scaling up the utilization of biodegradable and sustainable materials, which is a critical step toward achieving a more sustainable and environmentally responsible future. The substitution of environmentally harmful materials with green alternatives offers a promising avenue for substantially mitigating environmental pollution associated with petroleum-based polymers or plastics (Ramesh et al. 2017). Using these natural green materials can help diminish environmental pollution, lower production expenses, and provide a safe solution to current issues (Ehsani and Parsimehr 2020a, b). Lignocellulosic materials, such as wood, leaves, and plants, are the most representative green materials. They have many advantages, including being renewable, abundant, biodegradable, cheap, carbon neutral, and capable of reducing pollution (Mehta et al. 2020). Lignocellulosic materials are being used in various ways to implement green technology, including the development of bio-refinery technology to replace petroleum-based refineries (Arevalo-Gallegos et al. 2017), engineered woods that enhance the performance of materials to overcome the limitations of natural wood (Chen et al. 2020), the production of bioplastics (Thakur et al. 2018), the fabrication of fibrillated cellulose soft gels (Li et al. 2021), and the use of green electronic materials. Moreover, subjecting lignocellulosic materials to high-temperature carbonization via thermal decomposition processes broadens their potential applications to encompass electronics and energy harvesting/storage (Parsimehr et al. 2021). This technology enables the recycling and utilization of all lignocellulosic biomass, encompassing fruits, seaweed, and wood wastes, as green electronics (Ehsani and Parsimehr 2020a; Parsimehr and Ehsani 2020a, b). Hence, ongoing advancements in technologies harnessing lignocellulosic materials, including biomass, are fostering the application of green technology across diverse fields.

The generation of electronic waste, also known as e-waste, is becoming more prevalent with the development of the electronic market, and the shortened replacement cycle (Kang et al. 2004). The electronics industry is known for its high energy intensity and carbon emissions (Chen et al. 2021; Chih-Lin et al. 2020; Martins 2021); therefore, it is essential to replace current electronics with sustainable and green alternatives in the long run. To address this, Irimia-Vladu (2014) has identified four requirements for green electronics: (i) precursor materials should be abundant, cheap, and organic, (ii) synthetic routes need to be economically viable and have high production rates, (iii) synthesized electronic materials must have a low-cost fabrication process, and (iv) the device should be biodegradable and biocompatible. The ongoing development of electronics technologies utilizing lignocellulosic materials with various metal coating methodologies enhances the utilization of green materials in electronics (Guo et al. 2018; Wan et al. 2017; Zhu et al. 2018). However, these methods commonly require additional materials to be added through a coating or doping process, which results in technical limitations involving production time and cost. Moreover, these additional materials may not be biodegradable or biocompatible, leading to difficulties in waste disposal (Li et al. 2020). To overcome these limitations, Ruquan Ye et al. (2017) demonstrated the first LIG (Laser-Induced Graphene) formation on wood. This novel graphene patterning technology can pattern arbitrary-shaped carbon electrodes without pre-produced masks or additional conductive materials through direct carbonization and graphitization of residual carbon. Synthesizing LIG on wood, however, presents unique challenges since wood can easily combust and be ablated. Researchers have tried to resolve this issue through several methods, including non-oxygen atmosphere condition fabrication (Ye et al. 2017), fire-retardant treatment and multiple lasing technologies (Chyan et al. 2018), and ultrafast laser-based fabrication that suppresses thermal damage and combustion for direct LIG formation in ambient air, without additional treatment (Nam et al. 2023). Since the introduction of LIG formation on wood, various lignocellulosic materials such as potato skins, coconut shells (Chyan et al. 2018), fallen leaves (Le et al. 2019), corks (Carvalho et al. 2020), paper (Jung et al. 2022), and cellulose paper (Kulyk et al. 2021) have been used as the precursors to create LIG electrodes, which have been applied to various electronics such as sensors, heaters, and batteries. (Marques et al. 2020) To increase the utilization and performance of LIG, researchers have also developed several innovative methods, including transferring LIG on wood into a polymer to create flexible electrodes (Kim et al. 2021), creating metal-embedded graphene by soaking wood in metal nanoparticles and subsequent laser patterning (Han et al. 2018), and fabricating iron-embedded LIG formation onto wood by applying iron paint and subsequent laser patterning (Dreimol et al. 2022). These methods show promise in creating sustainable and green electronic devices that meet the requirements for biodegradability and biocompatibility while also being economically viable with high production rates. However, LIG's conductivity and chemical performance fall short compared to single-layer or few-layer graphene, indicating a need for technology to enhance LIG's performance through environmentally-friendly chemical doping.

The emerging trend of green LIG electrodes has also led researchers to the development of green energy storage devices to operate green electronics in an environmentally viable manner. Supercapacitors are becoming increasingly popular as a form of next-generation energy storage owing to their capability to provide high-power density, fast charge/discharge rates, and nearly infinite cycle life (Simon and Gogotsi 2010). However, supercapacitors suffer from low energy densities, as their performance is highly influenced by the electrochemical properties of electrode materials (Raza et al. 2018). Carbon-based materials containing pores, with a high specific surface area, such as activated carbon, are commonly used for efficient electric double-layer capacitors (EDLCs) (Frackowiak and Béguin 2001; Kötz and Carlen 2000). A series of technologies have been developed to transform eco-friendly materials, such as coconut shells and corn biomass, into activated carbon for green supercapacitor fabrication (Parsimehr and Ehsani 2020a, b). However, this requires carbonization at high temperatures above several hundred degrees, or attachment to alkaline solutions like sodium hydroxide (Laine and Yunes 1992; Phiri et al. 2019; Wang et al. 2013). Consequently, LIG supercapacitors based on various lignocellulose materials including wood (Le et al. 2019; Ye et al. 2017), coconut (Chyan et al. 2018), fallen leaves (Le et al. 2022), and paper (Lu et al. 2022a, b) have been fabricated because they can easily and quickly make 3D porous graphene structures of the desired design in an eco-friendly and inexpensive manner. The use of lignocellulosic material-derived LIG in supercapacitors has garnered significant attention due to its potential to address several limitations such as production speed and cost associated with current porous carbon electrode manufacturing. In addition to LIG itself, previous studies have reported several methods to embed transition metal oxides onto carbon, which promotes surface redox reactions as pseudocapacitors (Zhu et al. 2020; Xu et al. 2022). Even though these previous studies have succeeded in ameliorating capacitive performance, they commonly require copious amounts of chemical precursors using commercially synthesized substrates over a prolonged production duration, rendering them unsuitable for future green mass production of MSCs. Despite these pioneering works, there remains the need for further development of green supercapacitors in terms of the synthetic production speed, biodegradability limitations, complexity of the pre-treatment process, residual chemical waste discarded during the production process, as well as low energy density of carbon materials based on EDLC mechanism. By addressing these challenges, LIG supercapacitors based on lignocellulosic materials could become a promising alternative to current carbon-based electrode materials for supercapacitors, offering a sustainable and eco-friendly solution for energy storage.

Herein, we demonstrated a LIG-based supercapacitor on wood via femtosecond laser direct writing (FsLDW). Our novel supercapacitor satisfies the standards for green electronics by utilizing wood, an abundant, inexpensive, and natural material. Moreover, our process is simple, fast, and inexpensive compared to past microsupercapacitors, and finally, our eco-friendly and low-cost synthetic process results in a biodegradable and biocompatible device. To create this green supercapacitor, we employed FsLDW technology to fabricate high-quality LIG. Specifically, we utilized a two-step process of drop-casting potassium permanganate (KMnO4) onto prefabricated LIG and laser patterning for MnO/LIG hetero-nanostructure formation. Our optimized MnO/LIG micro-supercapacitors demonstrated superior performance compared to prior LIG micro-supercapacitors, specifically in terms of areal capacitance, power density, and energy density. The optimal laser condition for producing high-quality prefabricated LIG was a scanning speed of 20 mm s−1 at 1 W power, while the optimal laser condition for producing MnO/LIG after drop-casting was a scanning speed of 50 mm s−1 at 1 W power. By using 0.66 mg cm−2 of Mn source in drop-casting, we efficiently absorbed only the necessary capacity, thereby removing chemical waste. Our supercapacitor exhibited exceptional performance compared to previous LIG supercapacitors, with an areal capacitance of 35.54 mF cm−2 at 10 mV s−1, a power density of 80 µW cm−2, and an energy density of 0.67 µWh cm−2 at 200 µA cm−2. Moreover, the device's durability was impressive, as it maintained a performance of 82.31% after 10,000-cycle cyclic voltammetry (CV) tests for the MnO/LIG MSC. Our MnO/LIG micro-supercapacitors leverage wood, a prominent green material, employ a precise chemical drop-casting method that produces no chemical waste during synthesis and undergoes laser patterning within 3 min after chemical treatment, obviating the necessity for further steps. This straightforward process enables cost-effective, easy, and rapid production while fully meeting the criteria of green electronics. This development represents significant progress in sustainable battery manufacturing technology, as our green supercapacitor not only exhibits outstanding performance but also addresses limitations inherent in prior technologies, such as e-waste and chemical waste, along with manufacturing time and costs.

2 Materials and methods

2.1 LIG and MnO/LIG synthesis

LIG was synthesized by a Yb-doped fiber femtosecond laser (FsLDW, Lasernics, FUPL-250-6). The wavelength of the laser was 1030 nm, the pulse duration was 255 fs, and the pulse repetition rate was 201.5 kHz. The average power was modulated from 0.8 to 1.5 W. The scanning speed of the laser beam was controlled from 10 to 80 mm s−1. Notably, the near-IR wavelength was selected because it minimizes unwanted ablation as compared to ultraviolet (Pawlak et al. 2016). For the fabrication of LIG electrodes, both laser power and scanning speed were optimized for the minimization of the electrical resistance of LIG; accordingly, the laser power with 1.0 W, scanning speed with 20 mm s−1, and hatch distance of 20 µm were selected. The raster scanning of the delivered beam via the Galvano scanner enabled the pattern-customized design of LIG. The electrical resistance of LIG was measured by both four-point probes (MS tech, M4P-302) and two-point probes (Saehan, SH-852TR). The KMnO4 (Sigma-Aldrich, 223468) aqueous solution with a concentration of 76 mg mL−1 was synthesized to disperse permanganate ions homogeneously in a porous LIG structure. Subsequently, a minute quantity of the precursor (25 µL cm−2) was dropped on the LIG surface. The infiltration of KMnO4 solution into LIG took less than 3 min. Finally, the secondary laser was irradiated with a laser power of 1.0 W, a scanning speed of 50 mm s−1, and a hatch distance of 75 µm to synthesize MnO/LIG.

2.2 Characterization

Surface morphology was observed by FE-SEM (Hitachi, SU5000) using a 10 kV acceleration voltage. HRTEM (Talos F200X) and SAED (Talos F200X) images were obtained with a 200 kV acceleration voltage. To prepare the specimens for imaging, powders of LIG and MnO/LIG were delicately detached from the surface of the materials, and subsequently dispersed in isopropyl alcohol over 1 h of sonication with 40 kHz of ultrasonic frequency (Seong dong, SD-301H). The crystal structure was determined using an X-ray diffractometer (Rigaku, SmartLab) with a Cu K-α1 incident radiation wavelength of 1.541 Å. XPS (Thermo Scientific, Nexsa G2) was employed to investigate the surface chemistry using an Al K-α radiation source. Raman spectra of prepared materials were acquired by the HORIBA, LabRAM HR Evolution Raman spectrometer with a laser wavelength of 514 nm. TGA (Mettler, TGA2) was performed in the air condition with a temperature range between 0 and 1000 ℃ and a temperature step size of 10 ℃ min−1. FT-IR (Thermo Fisher, Nicolet iS50) was employed on dispersed powders in KBr pellets. The specific pore area and volume of prepared materials are compared based on BET theory (MicrotracBEL, BELSORP-miniX).

2.3 Electrochemical measurements

The electrochemical measurements were performed by a potentiostat system (HS technologies, Vertex) to scrutinize the performance of prepared materials. CV, GCD, and EIS were performed at ambient temperature. The CV measurements were executed at varying scan rates ranging from 10 to 100 mV s−1 in − 0.4 to 0.5 V for LIG and 0–0.8 V for MnO/LIG electrodes, respectively. These different potential windows are adopted for each electrode to operate in an electrochemically stable state versus Ag/AglCl (Guo et al. 2014; Piñeiro-Prado et al. 2016). The GCD was carried out with current densities ranging from 1000 to 5000 µA cm−2. The potential window for LIG and MnO/LIG MSC is from 0 to 0.8 V with a current density from 200 to 700 µA cm−2. EIS was obtained in the frequency range of 10 mHz to 1 MHz with an AC voltage amplitude of 10 mV at the open circuit condition. A three-electrode system was adopted to analyze the electrochemical properties of cell-type electrodes made of the prepared materials (LIG electrodes and MnO/LIG electrodes). These electrode materials have a square pattern of 1 cm × 1 cm which is employed directly as a working electrode. The Ag/AgCl electrode was used as a reference electrode, the platinum electrode as a counter electrode, and 1 M Na2SO4 aqueous solution (Sigma-Aldrich, 403008) as an electrolyte. To assess the microsupercapacitor properties, a two-electrode system was also implemented (LIG MSC and MnO/LIG MSC). As-fabricated MSC acted as both the working and counter electrodes in 1 M Na2SO4 electrolyte. The solid gel electrolyte for the practical demonstration was PVA/Na2SO4, which is a mixture of 4 g of PVA and 0.852 g of Na2SO4 in 40 mL of distilled water.

The specific areal capacitance of a device (Cdevice, F cm−2) can be calculated from curves of CV, and GCD, respectively, via the following equations:

$${{\text{C}}}_{{\text{device}}}=\frac{{\int }_{{V}_{i}}^{{V}_{f}}idV}{2\nu A\Delta V}$$
(1)
$${{\text{C}}}_{{\text{device}}}=\frac{It}{A\Delta V}$$
(2)

where i is the current (A), Vf and Vi are potential limits of the CVs (V), ν is the scan rate (mV s−1), A is the area of active electrodes (cm2), ΔV is the potential window (V), I is the areal charge and discharge current density (A cm−2) and t is the discharge time (s). The specific areal energy and power density of MSC (E and P, in unit Wh cm−2 and W cm−2) can be calculated from the galvanostatic charge and discharge curves using the following equations:

$${\text{E}}=\frac{1}{2}\times \frac{1}{3600}\times C\times {\left(\Delta V\right)}^{2}$$
(3)
$${\text{P}}=\frac{{\text{E}}}{{\text{t}}}$$
(4)

3 Results and discussion

3.1 Fabrication of MnO/LIG MSCs via FsLDW

Figure 1a illustrates the overall diagram of the process employed in synthesizing wood-derived LIG and MnO/LIG hetero-nanostructure via FsLDW. As illustrated in Additional file 1: Fig. S1, the intense ultrafast pulse energy photothermally converted the excess carbon of the organic compounds into 3D mesoporous LIG in ambient conditions (Le et al. 2019). The ultrafast pulse locally induces high temperature and pressure on the surface of the wood. This leads to breaking the bonds of the lignocellulose structure and rearranging them into sp2-hybridized carbon with rapid liberation of gaseous products (Ye et al. 2017, 2018). This photoconversion of wood into LIG can be described as successive carbonization and graphitization. The ensuing 3D mesoporous structure can facilitate easy access for electrolyte ions to establish a Helmholtz layer, which is advantageous for EDLC applications (Lin et al. 2014). However, owing to the intrinsic limitations of EDLCs in energy density, it is essential to enhance their capacitive performance. To address this concern, we synthesized MnO/LIG hetero-nanostructures via two-step femtosecond laser direct writing.

Fig. 1
figure 1

Schematics of the fabrication process for MnO/LIG hetero-nanostructure on wood via ultrafast pulses. a Green, facile, and fast femtosecond laser direct writing enables to convert wood to LIG, and Mn precursor to MnO which is in the limelight for supercapacitors. b Applications of supercapacitors and eco-friendly operation of electric devices. c XRD patterns of LIG and MnO/LIG. d HRTEM images of MnO/LIG, revealing two distinctive morphologies: graphitic layer and crystalline MnO nanoparticle. e Raman spectra of wood, LIG, and MnO/LIG

Full size image

We adopted the FsLDW by two-times to homogeneously embed MnO nanoparticles in the LIG matrix to form MnO/LIG hetero-nanostructures using only a drop of precursor over a short time duration. While the first FsLDW was able to synthesize wood-derived LIG, the following second FsLDW converted the dispersed potassium permanganate precursor into MnO nanoparticles. In this process, laser processing parameters played a critical role in synthesizing highly conductive LIG and crystalline MnO/LIG. Therefore, we have conducted a comprehensive investigation of the effects of laser fluence by measuring the electric conductivity of LIG and ex-situ X-ray diffraction (XRD) of MnO/LIG (Additional file 1: Figs. S2–S4). In addition, the XRD pattern, when replacing the secondary femtosecond laser with the CO2 laser under identical laser processing parameters, represents amorphous potassium permanganate in Additional file 1: Fig. S4. These infer that both insufficient laser fluence and low peak power of CO2 laser incompletely convert the targeted materials into LIG, while excessive laser fluence leads to thermal ablation and unwanted heat-affected zones. This principle also applies to the synthesis of MnO/LIG. Not only does the insufficient laser fluence fail to synthesize high-crystalline MnO/LIG, but excessive laser fluence seems to deteriorate the crystallinity. Therefore, it is imperative to synthesize LIG with optimized laser fluence. Regarding this optimized LIG, writing speed, and hatch size for secondary FsLDW, concentration and dropped amount of potassium permanganate for MnO/LIG are optimized respectively in Additional file 1: Figs. S5–S6. The as-fabricated pseudocapacitor with MnO/LIG electrodes demonstrated an increased areal capacitance of 35.54 mF cm−2 at 10 mV s−1, compared to the traditional supercapacitors that have an areal capacitance of 22.37 mF cm−2 at 10 mV s−1. Moreover, this MnO/LIG MSC consumes only 0.66 mg cm−2 of Mn source and takes less than 10 min for all processes from wood to MnO/LIG MSC, including preceding FsLDW, solution infiltration, and following FsLDW, which is suitable for mass production. On the contrary, previous energy storage production methods take over 24 h for a single-cell test including both EDLCs and lithium-ion batteries (Ma et al. 2016; Murray et al. 2019). This MnO/LIG MSC can operate electronic devices such as LEDs and digital clocks, as depicted in Fig. 1b. Therefore, we anticipate that this green, facile, and fast method for the fabrication of the MnO/LIG MSC could be promising for sustainable future energy storage.

3.2 Characterization of LIG and MnO/LIG

The XRD patterns of LIG and MnO/LIG are depicted in Fig. 1c. The crystal structure of LIG is a turbostratic structure that consists of both sp2 and sp3 hybridized orbitals. Accordingly, the d-spacing between (002) planes in the hexagonal close-packed (hcp) structure is 0.66 nm at 2θ = 25.84°. In contrast, MnO/LIG exhibits the following peaks of the octahedron structure of MnO: (111) plane at 2θ = 35.03°, which gives 0.22 nm of d-spacing, (200) plane at 2θ = 40.68°, (220) plane at 2θ = 58.85°, (311) plane at 2θ = 70.44°, and (211) plane at 2θ = 74.12°. The reference peaks of graphene and MnO were obtained from JCPDS #75-16212, and JCPDS #78-0424, respectively. The depicted high-resolution transmission electron microscopy (HRTEM) images of MnO/LIG in Fig. 1d exhibited a clear distinction between MnO and LIG. The MnO nanoparticles are supported by a LIG matrix, a propitious medium for heightened electric and ionic conductivity. The wrinkled pattern of LIG (left) exhibits a d-spacing of 0.348 nm that corresponds to the (002) planes of the graphitic layer. Conversely, the lattice fringes of MnO (right) are separated by 0.224 nm, representing the characteristic peak of the (200) planes of MnO. The insets in Fig. 1d show fast Fourier-transformed images of each region, which infer the crystallinity of LIG and MnO. Meanwhile, Fig. 1e portrays the Raman spectrum of wood, LIG, and MnO/LIG. The Raman spectrum is used to discern the molecular in-plane vibration of multilayer graphene and its heterostructures. The D peak Raman shift of 1350 cm−1 originates from the breathing modes of the six-atom rings (A1g) and requires defects for activation. The G peak Raman shift of 1580 cm−1 is attributed to one degenerate in-plane optical mode (E2g). The 2D peak Raman shift of 2700 cm−1 is the overtone of the D peak, while the Mn–O peak Raman shift of 650 cm−1 is assigned to the symmetric vibration of Mn–O (Ferrari and Basko 2013; Lin et al. 2020). The intensity ratio of ID/IG, and I2D/IG for LIG is 0.458, and 0.340, respectively, whereas those for MnO/LIG is 0.293, and 0.334, respectively, which means the enhancement of structure ordering by metal nanoparticles.

The arbitrary patterned MnO/LIG in Fig. 2a inferred the user-customized flexible design of as-synthesized MnO/LIG and LIG. The microstructure at the interface between MnO/LIG and wood is observed by field emission scanning electron microscopy (FE-SEM) in Fig. 2b, c. The interface between wood and MnO/LIG was intentionally ablated by high laser fluence to separate electrodes from wood to minimize the heat-affected zone. The ablated interface acts as a thermal insulator so that the irradiated pulse energy is solely used to convert the wood into LIG and MnO/LIG consecutively. We can also identify the porous structure of MnO/LIG derived from the degassing process when irradiated with ultrafast pulse energy. From the energy dispersive spectrum (EDS) images of MnO/LIG, it is clear that the manganese elements are selectively detected where LIG is already synthesized. During the process of infiltration, the mesoporous LIG structure more favorably absorbs manganese ion precursors. Conversely, wood comprises both oxygen and carbon in polysaccharide molecules, whereas, during the LIG synthesis, degassing liberates the oxygen as a gaseous form during the photoconversion process. The uniform thickness of MnO/LIG, detected solely through selective manganese element detection, is observed to be around 200 µm from cross-sectional SEM imaging (Fig. 2d). This implies that MnO nanoparticles are fully embedded in the porous structure of LIG. Furthermore, discerning the morphological difference between MnO/LIG and LIG in Additional file 1: Fig. S7 is elusive because the MnO nanoparticles are embedded homogeneously in the LIG matrix structure.

Fig. 2
figure 2

Schematic illustration of FsLDW for LIG and MnO/LIG synthesis on wood. a Photo of arbitrary patterned MnO/LIG on wood. b Surface SEM images at the interface between wood and MnO/LIG, and its corresponding EDS images. c Highly magnified SEM images at the surface of MnO/LIG. d Cross-sectional SEM images of MnO/LIG and its corresponding EDS image. e TEM images of LIG and MnO nanoparticles. e-I HRTEM image of MnO nanoparticles and its fast Fourier transform (FFT) image. e-II HRTEM images of LIG. f EDS mapping, and g SAED image of the MnO nanoparticle/LIG on wood. h Thermogravimetric analysis for wood, LIG, and MnO/LIG. i Fourier transform infrared spectroscopy spectrum for wood, LIG, and MnO/LIG. j X-ray photoelectron spectroscopy of wood, LIG, and MnO/LIG

Full size image

The TEM images of MnO/LIG were presented in Fig. 2e, revealing a MnO single grain with a diameter of approximately 50 nm distributed uniformly in the LIG matrix. Upon magnifying the MnO grain image in Fig. 2e-I, we can observe the distinctive fringe patterns with a d-spacing of 0.223 nm, corresponding to the (200) plane of MnO. The inset FFT image of the HRTEM image of MnO suggests the high crystallinity of as-synthesized MnO single grain since the FFT image shows distinct spots at different orientations, as shown in Additional file 1: Fig. S12 (Liang et al. 2020; Sopha et al. 2020). The wrinkled pattern of LIG, surrounding the MnO grain, was observed in Fig. 2e-II with a d-spacing of 0.663 nm, which matches the (002) plane of graphitic carbon. These ripples are developed from the thermal expansion by ultrafast pulse and are exposed at the edge of the graphene layer (Lin et al. 2014). The EDS image of MnO single grain was shown in Fig. 2f, where manganese and oxygen elements were detected in a spherical shape within a diameter of 50 nm, surrounded by carbon. The selected area diffraction (SAED) image in Fig. 2g obtained from a MnO particle was inset with a zone axis [001]. This suggests that polycrystalline MnO nanoparticles are synthesized with a high degree of crystallinity, which is favorable to unimpeded ion diffusion for surface redox reactions.

The thermogravimetric analysis (TGA) for wood, LIG, and MnO/LIG was illustrated in Fig. 2h. The weight loss up until 150 ℃ is due to the evaporation of moisture. The decomposition of wood begins at 250 ℃. Among the three main components of wood, lignin is thermally the most stable. Therefore, lignin decomposes in a wide range of temperatures while cellulose and hemicellulose decompose in a small interval between 300 and 400 ℃, which makes lignin the critical factor in pyrolysis. In contrast, hemicellulose is thermally the most unstable, which decomposes faster and at lower temperatures than cellulose and lignin(Burhenne et al. 2013). The first derivative of the TG (DT) in Additional file 1: Fig. S8 reveals that the thermal decomposition of wood can be sorted into three phases: 225–300 ℃, 300–400 ℃, and 400–500 ℃. Each phase represents the decomposition of hemicellulose, cellulose, and lignin or subsidiary carbon reactions, respectively (Xu et al. 2017). On the other hand, LIG gradually degrades from 250 to 600 ℃ to form volatile gaseous products such as CO and CO2. The residual 3.38 wt% of LIG may be attributed to pre-existing impurities. Meanwhile, MnO/LIG has a distinct weight decrease from 250 to 550 ℃, which also comes from the pyrolysis of LIG. Consecutively, above 500 ℃, MnO oxidizes to Mn2O3 (Deljoo et al. 2020; Zaki et al. 1997), resulting in the remaining 17.34 wt% of MnO/LIG, which is ascribed to the integrated effect of weight increase by oxidation of MnO and weight loss by gasification of organic components (Di Blasi 2009). As a result, 15.58 wt% of MnO is embedded in the MnO/LIG hetero-nanostructure.

The Fourier transform infrared spectroscopy (FT-IR) spectrum portraying the absorption bands of each material for wood, LIG, and MnO/LIG was unveiled in Fig. 2i. The absorption intensities at 3440–3340 cm−1 correspond to hydroxyl groups (O–H) being absorbed by the surface of LIG, hemicellulose, and lignin molecules for wood, while the peaks near 1730 cm−1 for the carbonyl groups (C=O) involve the absorption of esters, aldehydes, and ketones (Md Salim et al. 2021). The peak near 1630 cm−1 represents aromatic groups (C=C) of LIG and lignin (Emmanuel et al. 2015). The peaks near 1230, 1106, and 1035 cm−1 indicated aliphatic groups (C–O), while the peak at 1150 cm−1 corresponds to glycosidic bond (C–O–C) vibration. The photothermal energy results in the highly localized heating on the wood surface, which consecutively leads to bond dissociation of C–O, C–H, and C=O bonds (Vashisth et al. 2020). Notably, the two peaks at 630 and 525 cm−1 arose from the manganese oxide (Mn–O, Mn–O–Mn) vibrations, indicating the embedment of MnO nanoparticles in LIG (Zheng et al. 2013). The diminished amplitude of two peaks observed in the FT-IR spectrum suggests a significant presence of a carbonaceous matrix (Qiao et al. 2022). Correspondingly, the X-ray photoelectron spectroscopy (XPS) of wood, LIG, and MnO/LIG depicted in Fig. 2j showed two peaks of 637 and 650 eV belonging to Mn 2p3/2 and Mn 2p1/2, respectively, in agreement with those of MnO (Lu et al. 2022a, b). Another peak near 528 eV belongs to O 1s, and the other peak near 280 eV belongs to C 1s. The O 1s peak represents the single and double bonds of oxygen to carbon atoms (–C=O, C–O–C). Likewise, the C 1s peak indicates carbon–oxygen, and carbon-hydrogen atoms (C=O, O–C–O, C–OH) (Azadfar & Wolcott 2020). Compared with wood and LIG, the carbon-to-oxygen ratio of LIG increased more than twice from 1.60 to 3.88. This confirms the efficient removal of oxygen with successful carbonization and graphitization (Le et al. 2022).

3.3 Electrochemical properties

Figure 3 shows the comparative electrochemical performance for LIG and MnO/LIG through three-electrode configurations (Fig. 3a–c) and MSC applications (Fig. 3d–i). In Fig. 3a, the quasi-symmetric triangular shape in the galvanostatic charge–discharge (GCD) curve represents the electrochemical capacitive behavior of LIG and MnO/LIG with a current density of 1000 µA cm−2 within the potential range of − 0.4 to 0.5 V for LIG, and 0–0.8 V for MnO/LIG, respectively. Herein, the additional surface Faradaic reaction by MnO nanoparticles contributes to increasing areal capacitance so that the discharge time increased from 36.7 to 87.3 s. The CV curves of LIG and MnO/LIG electrodes in Fig. 3b presented quasi-rectangular shapes, which is a typical characteristic of EDLC. The embedment of MnO nanoparticles increased The areal capacitances of MnO/LIG electrodes (and that of LIG) are 41.82 (26.11), 30.42 (20.55), 25.69 (17.85), 22.55 (15.80), 20.36 (14.34), and 14.26 (10.21) mF cm−2 at scan rate from 10, 20, 30, 40, 50, and 100 mV s−1, respectively as demonstrated in Fig. 3c. The CV curves under different scan rates and GCD curves under different current densities were illustrated in Additional file 1: Fig. S9–S11. The improvement of capacitance can also be observed in the MSCs which are composed of three interdigitated pins at both electrodes of which the design is presented in Additional file 1: Fig. S12. Figure 3f compares the areal capacitance of MnO/LIG MSC (and that of LIG), which is 35.54 (22.37), 27.33 (17.58), 23.46 (15.31), 20.99 (13.87), 19.15 (12.84), and 13.73 (9.82) mF cm−2 at scan rate from 10 to 100 mV s−1, respectively. The CV and GCD curves of MSCs were also presented in Additional file 1: Figs. S13–S15. The reason why areal capacitance decreases according to scan rate is that at a low scan rate (~ 10 mV s−1), it gives sufficient time for the ion to diffuse into the porous MnO electrodes so that maximization of the activation site is possible. On the other hand, at a fast scan rate (~ 100 mV s−1), it only allows surface ions to participate in the redox reaction (Tholkappiyan et al. 2018).

Fig. 3
figure 3

Electrochemical properties of LIG and MnO/LIG with 1 M Na2SO4 electrolyte. a GCD curve of LIG and MnO/LIG electrodes in three-electrode configuration at a current density of 1000 µA cm−2. b CV curves of LIG and MnO/LIG electrodes in three-electrode configuration at a scan rate of 10 mV s−1. c Areal capacitances of LIG and MnO/LIG electrodes in three-electrode configuration at a scan rate from 10 to 100 mV s−1. d GCD curve of LIG and MnO/LIG MSC at a current density of 200 µA cm−2. e CV curves of LIG and MnO/LIG MSC at a scan rate of 10 mV s−1. f Areal capacitances of LIG and MnO/LIG MSC at a scan rate from 10 to 100 mV s−1. g Nyquist plots of LIG and MnO/LIG MSC. h Cycling stability of LIG and MnO/LIG MSC at a scan rate of 100 mV s−1. The inset indicates the CV curves of each MSC from the 1st, 2500th, 5000th, 7500th, and 10000th cycle. i Ragone plots of LIG and MnO/LIG MSC to compare the power density and energy density with previously reported natural source-based MSCs

Full size image

Compared with previous studies that fabricated LIG on a green substrate in Additional file 1: Table S1, MnO/LIG MSC has superior areal capacitance (35.54 mF cm−2 at 10 mV s−1, and 30.30 mF cm−2 at 200 µA cm−2). Previous literature reported LIG on pine wood (4 mF cm−2 at 1 mA cm−2) (Ye et al. 2017), LIG MSC on leaves (26.44 mF cm−2 at 10 mV s−1) (Le et al. 2022), boron-doped LIG MSC on cork (18.68 mF cm−2 at 0.1 mA cm−2) (Imbrogno et al. 2022), mushroom derived LIG MSC (36.20 mF cm−2 at 0.05 mA cm−2) (Yadav et al. 2016), sandwich-LIG on Kraft lignin (3.52 mF cm−2 at 10 mV s−1) (Mahmood et al. 2020), LIG MSC on biodegradable parylene-C (6.64 mF cm−2 at 0.01 mA cm−2) (Correia et al. 2022), LIG MSC on cork (5.72 mF cm−2 at 0.1 mA cm−2) (Silvestre et al. 2022), and LIG MSC on the phenolic resin (1.60 mF cm−2 at 0.1 mA cm−2) (Zhang et al. 2018). This demonstrates that our green pseudocapacitor has a higher power and energy density compared with other natural source-derived LIGs.

The Nyquist plot obtained from electrochemical impedance spectroscopy (EIS) of LIG MSC and MnO/LIG MSC in the range between 10 mHz and 1 MHz was illustrated in Fig. 3g. Both MSCs exhibited the linear capacitive behavior of EDLCs. Considering the high-frequency region of the Nyquist plot, we can apply Randles cell for MnO/LIG and coated metal cell for LIG, respectively. For both cases, the intercept at the x-axis signifies the ideal equivalent circuit resistance (ESR, Rs) for supercapacitors, which is the summation of the bulk electrolyte, electrode, and contact resistance between the electrode and the current collector. For LIG, the diameter of the first semi-circle represents pore resistance Rpo, which is the resistance of ion conducting paths in the coating. From this circle, the capacitance of intact coating Cc can be inferred as well. For the second semi-circle of LIG, and that of MnO/LIG, the diameter of the semi-circle indicates charge transfer resistance Rct, which is the resistance at the interface between the electrode and electrolyte (Mei et al. 2018). In the same way, double-layer capacitance Cdl can be inferred. The ESR and Rct for MnO/LIG have been reduced compared to LIG from 106.6 Ω and 58.6 Ω to 61.24 Ω and 43.52 Ω, respectively. On the other hand, double-layer capacitance Cdl has been increased from 37.63 µF to 12.43 mF. In addition, the capacitance of intact coating Cc, and pore resistance Rpo for LIG is 3.62 nF and 64.68 Ω, respectively. This implies that the rate of electrostatic adsorption and desorption between electrodes and electrolytes is enhanced due to the MnO nanoparticle embedment. This can be attributed to not only the surface redox reaction but also to the effective surface area expansion, owing to the MnO nanoparticles. The expanded effective surface area acts as the active site for electrolyte ions that can adsorb at the interface of electrodes. The specific area and volume comparison between LIG and MnO/LIG in the range of pore size 10–200 nm based on Brunauer–Emmett–Teller (BET) theory in Additional file 1: Fig. S17 confirms this expanded surface area. The specific surface area of LIG is 3.19 m2 g−1, whereas that of MnO/LIG is 7.26 m2 g−1, which is more than twice that of LIG. Moreover, MnO nanoparticle embedment significantly improves the hydrophilicity, thus increasing the rate of adsorption. The rate of water droplet infiltration was compared in Additional file 1: Fig. S18. The video recording of a water droplet infiltration at the interface of LIG and MnO/LIG is provided in Additional files 2, 3: Video S1 and S2. This increased surface area and hydrophilic nature of MnO/LIG effectively lower the ESR, thus ameliorating the areal capacitance of MnO/LIG. Additionally, the stability of MnO/LIG MSC was measured at 100 mV s−1 for 10,000 cycles, as demonstrated in Fig. 3h. The capacitance of LIG MSC was retained at 88.69%, and that of MnO/LIG MSC was retained at 82.31% after 10,000 cycles. Figure 3i shows the Ragone plot of as-fabricated MnO/LIG MSC, which exhibits an energy density (0.67 µWh cm−2) at a specific power density (80 µW cm−2) when compared to other natural source-derived LIG MSCs.

3.4 Electronics demonstration

Green MnO/LIG MSCs are promising candidates for next-generation sustainable energy storage devices. To power electronic devices, it is often necessary to connect multiple MSCs in parallel and series to attain higher voltage and current. The convenient and adaptable design of MSCs via FsLDW has paved the way for producing pre-determined arrays of MSCs that cater to the specific needs of various electronics. As shown in Fig. 4a, a single MSC was composed of three interdigitated pins in both working and counter electrodes (1 × 1). By connecting multiple MSCs in parallel and series, we can amplify their charge and discharge duration and regulate their operating voltages to requisite specifications. This can be confirmed in Fig. 4b which exhibits GCD curves for various MSC arrays. A series connection augments the charge and discharge duration, whereas a parallel connection elevates the operating voltage. By means of these MSC arrays, we can successfully power commercial electronic devices, such as LEDs, digital clocks, and electronic paper as exemplified in Fig. 4c–e and the accompanying Additional file 4: Video S3.

Fig. 4
figure 4

Green wood-derived MSCs for next-generation energy storage. All MnO/LIG MSC arrays are operated with solid gel electrolytes of PVA/Na2SO4. a Single MSC design with its microstructure. b GCD curve of MSC arrays at a current density of 100 µA cm−2. c Turning on an LED with the switch of copper tape. d Powering a digital clock for over 30 s. e Electronic paper represents ‘F’ in FsLDW by MnO/LIG MSC

Full size image

4 Conclusions

In summary, we presented a facile, fast, and green fabrication of wood-derived MnO/LIG MSC via FsLDW technology. Ultrafast fs laser pulses enable the creation of LIG electrodes on wood through carbonization and graphitization, allowing for high-resolution microelectrode patterning while minimizing thermal ablation, without additional chemical treatments in the atmosphere. Ultrafast laser pulses not only transform minimal precursor material into MnO/LIG through metal ion synthesis but also establish FsLDW as a fast and simple fabrication method, capable of fabricating graphene patterns and embedding metal oxide nanoparticles within the LIG matrix. Compared to wood-derived LIG, small amounts of Mn source (0.66 mg cm−2) induce the formation of MnO nanoparticles within LIG, facilitating surface redox reactions. This change significantly boosts the energy density by 2.1 times (from 0.32 to 0.67 µWh cm−2) and increases the areal capacitance by 1.6 times (from 22.37 to 35.54 mF cm−2 at 10 mV s−1), improving the performance of MSC. Our MnO/LIG MSCs employ precise chemical drop casting, generating minimal chemical waste in small-scale synthesis. After only 3 min of drop casting, laser patterning can proceed, facilitating a straightforward and rapid production process that is cost-effective and green-compliant. This MSC demonstrates an 82.31% capacity retention at 10,000 cycles and requires only 10 min cm−2 for production, ultimately embodying performance stability, cost-effectiveness, ease of manufacturing, and rapidity while meeting the stringent criteria of green electronic standards. Furthermore, these easily fabricated, eco-friendly MnO/LIG MSCs can power an LED, digital clock, and electronic paper. We believe that these MSCs will be a promising, and excellent candidate for next-generation energy storage.