A comprehensive review of laser-induced-graphene for sensor applications: fabrication, properties, and performance evaluation

Abstract

Graphene, renowned for its excellent mechanical and electrical properties, can be fabricated through various methods such as mechanical exfoliation, chemical vapor deposition (CVD), and epitaxial growth, each facing challenges that hinder commercial applications. However, laser-induced graphene (LIG), a three-dimensional porous material synthesized by laser technology from sustainable sources, could reduce e-waste while supporting circular economy strategies. In addition to its superior properties, including high thermal conductivity and a high carrier mobility rate, LIG has found extensive applications across diverse fields. Notably, the fabrication process obviates the need for harsh environments and photolithography processes, reducing costs. This review introduces the manufacturing processes of some common sensors and categorizes discussions based on different working mechanisms. LIG-driven devices are emphasized in biology and chemistry, particularly in energy storage devices. Finally, some perspectives on potential development are presented.

---

1. Introduction

As Moore's Law nears its limits, researchers are relentlessly developing new materials with exceptional properties.^1,2 Graphene, a two-dimensional material consisting of a single layer of carbon atoms arranged in a hexagonal lattice, has emerged as a leading candidate due to its superior electrical, mechanical, and thermal properties. Its high carrier mobility and thermal conductivity make it an ideal material for various applications, from energy storage systems to advanced sensing technologies.^3–5 Despite its potential, the widespread utilization of graphene in commercial applications has been hindered by the complexities associated with its production.

In 2004, the discovery of graphene by Geim et al. through mechanical exfoliation marked a significant milestone in materials science.³ However, this method cannot produce large-size graphene sheets, and alternative techniques, such as liquid-phase stripping, present similar issues, such as low productivity and scalability.^6,7 Therefore, CVD emerged which can be used to synthesize graphene on a large scale. It typically requires high-temperature annealing which causes issues such as substrate lattice mismatch leading to wrinkles and defects.^8,9 Additionally, different metal substrates will significantly affect the number of graphene layers and lattice defects, as illustrated in Fig. 1(a).^10,11 The idea of growing high-quality graphene on single crystal copper (111) substrates was proposed a decade ago,^12,13 which requires a vacuum or inert gas environment, but it is always difficult to remove oxygen during the evacuation process.¹⁴ The mechanical properties are insufficient due to residual oxygen groups and reduced electron mobility.

Fig. 1 (a) Internal folding of the four layers.¹⁰ (b) The curve of sheet resistance with different temperatures forms various graphene layers.¹⁵ All the images enlisted here were reproduced with the copyright permission of Elsevier.

In terms of other synthesis methods, epitaxial growth on silicon carbide (SiC) leads to the formation of a graphene/SiC layer which has excellent sheet quality and electrical conductivity, and the sheet resistance can reach as low as ∼0.43 Ω sq⁻¹, as shown in Fig. 1(b).¹⁵ However, challenges such as lattice matching will cause uncontrollable defects, and the presence of polycrystalline domain structures makes it challenging to control the number of layers.¹⁶ Additionally, there are effective roll-to-roll techniques for mass production of graphene that meet the scale and quality for various applications.^17,18 The issues are also apparent: shear stress at the substrate edges can result in wasted yield and mechanical damage to the graphene films.

In 2014, Tour et al. proposed a new one-step synthesis method to fabricate porous graphene film using a CO₂ laser to directly irradiate the polyimide (PI) substrate. This porous graphene film is often called LIG.¹⁹

The production technology for LIG utilizes efficient laser scanning, significantly increasing the yield. This approach also addresses the issue of smaller flake sizes associated with traditional exfoliation methods, outperforming mechanical and liquid-phase exfoliation techniques. The production technology of LIG effectively circumvents the lattice-matching issues associated with epitaxial growth and CVD methods. First, the formation of LIG involves direct laser irradiation to decompose the precursor into graphene without relying on the substrate's lattice structure. Secondly, the precursors are often polymers, which are inherently polycrystalline or amorphous, lacking a fixed lattice, thus eliminating the problem. Furthermore, this method can produce graphene at room temperature via localized laser heating.

Compared to the most commonly used Hummers’ method, Hummers’ multi-step synthesis is complex, while the LIG synthesis process is quick and straightforward. Economically, the substantial cost of the chemicals needed for the Hummers’ method is a significant consideration. In contrast, LIG stands out because of its low-cost and sustainable substrates. In industrial production, stability and uniformity are of the utmost importance. The Hummers’ method, however, performs poorly in this aspect due to uncertainties in its multi-step process. On the other hand, LIG can easily achieve these industrial requirements by simply maintaining the stability of the laser system. In terms of material properties, LIG exhibits a highly porous structure with a large specific surface area, which provides significant advantages in applications such as supercapacitors and sensors. In contrast, graphene produced via the Hummers’ method displays limited porosity, and its specific surface area primarily depends on the number of layers and the degree of exfoliation. In addition, recognizing the escalating challenges posed by environmental degradation, there is an increasing understanding of the critical importance of sustainable economic development of LIG, synthesized from renewable carbon sources, which plays a vital role in reducing electronic waste and plastic pollution while supporting the advancement of circular economy models and green technologies.^20,21

Overall these advantages make LIG commercially significant for future applications in electronic devices including energy storage devices, sensors, bioelectronics, and biomedical devices.²²

2. Energy storage devices

Nowadays, sensors must be equipped with compatible, high-performance energy storage systems. LIG, with its unique structural characteristics, can be a promising candidate for supercapacitors, and these devices demonstrate remarkable long-term stability and high capacitance storage capabilities under both low and high-voltage conditions.²³ The porous nature of LIG significantly enhances the specific surface area of these devices, thereby enhancing their overall energy storage capacity.

The fabrication process of LIG can precisely control the pattern of graphene which enables direct doping without the need for photolithography. Typically, LIG needs further doping to improve carrier concentration and modify the electronic band structure, thereby boosting conductivity.^24,25

What is even more worthy of attention is that a study revealed the areal capacitance of LIG fabricated from F-doped and undoped PI substrates is enhanced 27 times.²⁶ Therefore, it is worth confirming that doping truly is an effective method to enhance energy storage performance. Hence, researchers have conducted doping experiments on a series of low-cost polymers, to delve into and enhance the performance of these polymers following the doping process.

Poly furfuryl alcohol (PFA), an economical and sustainable material with a highly crosslinked structure that is similar to PI, is effective for LIG formation under direct laser writing. Salt doping can be performed by spray-drying the solution and doping through laser localization heating. A study has shown that salt doping in PFA enhances the performance of LIG,²⁷ and the salt iron increases the surface area, contributing to several thousand-fold increases in capacitance. Composites of PFA/20% Na₂SO₄ exhibit average areal capacitances up to about 80 mF cm⁻² at 0.05 mA cm⁻², however, the areal capacitance of undoped LIG is on the order of μF cm⁻². The average areal capacitance of LIG from non-salt-doped PFA is three orders of magnitude lower than 20% Na₂SO₄-doped PFA, demonstrating the effectiveness of doping in enhancing capacitance.

Phenolic resin (PR) is doped with various metal ions. In this process, PR powder is added to ethanol for sonication, the PR solution and metal ions are mixed, the solution is controlled using adhesive tape, the solution is cleaned by sonication using water and ethanol and dried, the solution is spin-coated on the polyethylene terephthalate (PET) substrate to form a uniform film, and finally, the residual compound is cleaned using alcohol. The whole process is shown in Fig. 2(a). In the doped PR layer, the layer with zinc doping exhibits the weakest absorption peak at a wavelength of 405 nm in Fig. 2(b), resulting in the highest sheet resistance in Fig. 2(c).²⁸ This phenomenon demonstrates the complexation behavior of metal ions and polymers. The key lies in the varying viscosities of different metal salts, leading to inconsistent penetration rates and diverse catalytic effects on accelerating polymerization reactions. These effects affect the rate of graphitization and decrease sheet resistance. In this study, we can consider employing various surface treatment techniques or altering the laser absorption rate of the substrate to achieve precise control of LIG. This strategy offers vital guidance and direction for the development of future related research.

Fig. 2 (a) Schematic procedures for the fabrication of LIG patterns; (b) UV-vis spectra of PR coatings doped with 2 mg mL⁻¹ different metal salts. (c) The sheet resistance with PR doping the metal ion.²⁸ All the images enlisted here were reproduced with the copyright permission of Elsevier.

Additionally, monolayer graphene's surface area can reach up to 2630 m² g⁻¹, which makes it an ideal material for energy storage devices such as supercapacitors. However, its potential application in energy storage is limited by factors including the potential window and chemical properties. Unlike monolayer graphene, the 3D structure of graphene not only retains the intrinsic properties of graphene but also enhances electron transfer, and adsorption–desorption processes, and facilitates material modifications through methods such as metal doping.²⁹ Studies have shown that metal doping can significantly increase the capacitance of graphene.³⁰ The large potential window of LIG will result in increasing the energy density of the device,³¹ because the wrinkle structure provides more ion channels. Moreover, the process based on LIG is time-efficient, requiring only the spin-coating of a metal film on the substrate, followed by laser beam irradiation at high temperatures to generate metal droplets that fall back and form nanoparticles, as shown in Fig. 3(a).³² The diameter of the metal nanoparticles is about 0.9–1.2 μm. Fig. 3(b)–(d) show that the surface folds of LIG and Ag–LIG increase the contact area. Before doping, the interlayer distance of LIG is 0.34 nanometers, and it possesses a significant number of layers. Fig. 3(c) illustrates the uniformity before doping, and Fig. 3(e) shows that the interlayer distance of the LIG increases by 0.04 nm after Ag particle doping, proving the lattice expansion of LIG due to high temperature. These observations suggest a good match and can be explained as doping causing a change in the Fermi energy level, increasing the storage capacity of capacitors.³³

Fig. 3 (a) Fabrication and morphological characterization of metal–LIG (M–LIGs). (b) and (c) Transmission electron microscope (TEM) images of LIG. (d) and (e) TEM images of doped LIG.³² All the images shown here were reproduced with the copyright permission of Elsevier.

The metal-solution mixed spin-coating method allows for control of capacitance through solution concentration, but suffers from poor film uniformity and limited metal element range. Vapor deposition, while simpler, faces challenges in controlling doping concentrations.

Certainly, beyond improving performance through doping, performance can also be enhanced by optimizing the device structure. Traditional sandwich-type supercapacitors face challenges in meeting the volume requirements for wearable devices. Hence, designing novel electrode architectures is a crucial task. The LIG can be made into electrodes with arbitrary shapes, thus eliminating the need for photolithographic processes. This approach can easily allow the manufacture of unique electrodes.

Interdigital structured graphene electrodes represent a significant advancement in electrode technology, offering high-density, compact sizes that facilitate easier integration compared to traditional designs. These electrodes mitigate the risk of short circuits inherent in conventional sandwich structures.^34,35 Their compact size eliminates the need for mechanisms such as rotation and the utilization of redox intermediates’ concentration gradients for diffusion. For example, LIG maintains more stable capacitance through numerous cycles compared to coin cell batteries, highlighting its durability and efficiency. Fig. 4(a) and (b) illustrate the electrode structure, where the electrode width (W_e), channel spacing (W_g), and thickness (t) are key parameters.^36,37 The energy density and power of the device can be increased by controlling the radio of W_e/W_g. This unique configuration also reduces the impact of electrode thickness in ion diffusion, enhancing ion transport laterally between electrodes while allowing thickness modifications to optimize power density. Researchers have engineered specific structures within LIG to enhance ion adsorption, such as the foam-like structures shown in Fig. 4(c), which expand the contact area between the electrolyte and the electrode. Additionally, integrating materials like molybdenum disulfide (MoS₂) with graphene on polyimide foils is shown to optimize electrochemical properties significantly. Fig. 4(d) shows that the modified electrode has lower transferred resistance than a single LIG electrode. This feature helps reduce energy losses, which is crucial for high-power applications. The cyclic voltammetry (CV) curve illustrated in Fig. 4(e)³⁸ exhibits a square shape indicative of pseudo-capacitive behavior.^39,40 This behavior associated with the charging and discharging cycles of the double-layer capacitor, produces very high area capacitance, leading to significant enhancements in storage capacity as shown by the curve's rectangular shape.⁴¹ The electrochemical behavior and morphology image fit together.

Fig. 4 (a) Schematic sketch of the conversion of PI-foil into porous LIG and a sketch of an in-plane micro-supercapacitor (IMSC).³⁶ (b) Scheme of sandwich-style supercapacitors and micro-supercapacitors featuring in-plane interdigitated electrode designs.³⁷ (c) Schematic diagram of laser processed MoS₂-decorated LIG and field emission scanning electron microscope (FESEM) images. (d) Nyquist plot of MoS₂–LIG. (e) The capacitance of an MoS₂–LIG device at a rate of 10 mV s⁻¹.³⁸ All the images enlisted here were reproduced with the copyright permission of MDPI, Wiley and ACS.

Additionally, Martins et al. fabricated LIG-based micro-supercapacitors (MSCs) on paper substrates.⁴² This research indicates the possibility of reducing costs and mitigating the environment. The MSCs, with a similar structural design, exhibit not only excellent capacitive performance but also outstanding electrochemical stability. After 10 000 charge–discharge cycles, the initial specific capacitance was retained at over 190%. This significant performance enhancement is likely attributed to the activation of electrochemical sites on the electrode surface. So combining LIG and interdigital structured graphene electrodes improved energy storage problems from some aspects.⁴³ These findings imply that exploring various materials combinations could be crucial for future developments in energy storage technologies.

There have been many research groups focused on elucidating the relationship between the parameters of the laser and the structure of LIG. Fig. 5(a) and (b) present synergistic control of laser power and raster speed, highlighting their impact on LIG properties.²⁴ The findings reveal that when the engraving is conducted in raster mode with a power exceeding 18 W, it causes damage to the PI substrate. In contrast, laser-induced graphene foam (LIGF) is formed within the power range of 13.5 to 18 W. Furthermore, the production of LIG is sensitive to variations in both laser power and engraving speed. Specifically, laser power settings between 4.5 and 6.5 W combined with engraving speeds of up to 24 cm s⁻¹ can form LIG. As the engraving speed increases to between 24 and 35 cm s⁻¹, the required laser power adjusts to a range of 6.5 to 10 W. For even higher speeds, from 33 to 35 cm s⁻¹, the necessary laser power further increases to between 10 and 15 W. In this process, raster speed controls the thickness of the burned LIG, and conductivity.

Fig. 5 (a) Influence of power and raster speed on morphology and scanning electron microscope (SEM) images (scale = 200 μm). (b) Spot size and raster speed utilizing various laser powers.²⁴ All the images shown here were reproduced with the copyright permission of Springer Nature.

Fig. 6(a) and (b) show the morphologies of fibers after laser ablation on PI substrates under different power levels. When the laser output power increases to 0.4 W, dendritic fibers appear due to the thermal effect of the laser. Fig. 6(c) is an HSEM image showing the morphological evolution of the material under different power conditions. From top to bottom, the structures correspond to a layered porous morphology at power levels below 0.35 W, a loose porous morphology at 0.35–0.4 W, and a fiber-like morphology at power levels above 0.4 W.⁴⁴ This phenomenon implies that the local thermal effects produced by laser power significantly impact the morphology of LIG, which may be related to the process of forming amorphous carbon. This one-dimensional configuration can be transformed into a tunable one-dimensional nano-sensor under varying parameters, exhibiting unique properties under specific conditions, such as surface effects caused by periodic interruptions. As shown in Fig. 6(d), an increase in laser power correlates with enhanced conductivity, reaching approximately 600 S m⁻¹, while the tensile strength decreases simultaneously. These changes indicate an increase in the graphitization of the PI film. Fig. 6(e) and (f) show the correlation between increasing laser power and the ratio of I_G/I_D. Studies have shown that the ratio of I_G/I_D is an essential criterion for the quality of graphene, layers, and the degree of graphitization.^45,46 The graphitization process of some sulfur polymers has also been studied.⁴⁷ Resistance is also an essential factor to consider in the material application of sensors. Resistance affects electrical power consumption, and power consumption has always been a concern in electronic devices.

Fig. 6 (a) Optical images of LIG fibers (scale = 100 μm). (b) Low magnification SEM image of LIG fibers (scale = 50 μm). (c) High magnification SEM (HSEM) images of the LIG fiber electronics (LIGEF) at different graphitization stages (scale = 100 nm). (d) The relationship between the tensile strength and electrical conductivity of LIGEF and the laser power. (e) Raman spectra of LIGFE using different laser powers. (f) Correlation between laser power and the intensity ratio of Raman G peak to D peak (I_G/I_D).⁴⁴ All the images shown here were reproduced with the copyright permission of Elsevier.

Moreover, the laser wavelength might also be an important influencing factor, but detailed studies focusing on this aspect remain sparse. It is evident that the absorption rate corresponding to the laser wavelength dramatically affects the quality of LIG.⁴⁸ When the laser wavelength is relatively long, thermal effects predominate, resulting in the disruption of chemical bonds via thermochemical reactions. Conversely, when the laser wavelength is shorter, high-energy photons facilitate photochemical reactions that lead to the cleavage of covalent bonds and the formation of LIG. Overall, this process is primarily governed by the laser fluence per second. A reduced scanning speed results in a smaller energy radiation area, thereby enhancing energy density. In scenarios where laser power falls within the low graphitization range, decreasing scanning speed can augment light fluence per unit time.²⁰ Moreover, there exists an inverse relationship between laser wavelength and light fluence per unit time. In other words, the impact of laser scanning can be measured not only by laser energy density, but also by other parameters such as power, which can be evaluated in a similar manner.

3. Mechanical sensors

With the increasing demand for smaller sensors, automation and intelligent systems require easier operation methods, real-time operation, and control of various devices and equipment. However, there are two main challenges. Firstly, small sized devices require low operation voltages or other driving methods. Secondly, the materials used must not only be small but also flexible and durable. LIG, known for its outstanding properties, can also be synthesized using sustainable substrates, thereby reducing pollution and enhancing sustainability. Thus, LIG is frequently utilized in these applications. Here it is mainly introduced that common mechanical sensors of LIG incorporate resistive, capacitive, and triboelectric sensors.

3.1 Resistive sensors

Resistive sensors are designed to measure physical quantities by detecting changes in resistance by strain, humidity, etc. In industry, LIG is typically produced on a PI substrate due to the flexibility of PI. However, the poor stretchability, robustness, and biochemical properties of PI limit applications in wearable sensors. Furthermore, the mismatch of Young's modulus between PI and LIG can lead to the formation of cracks in LIG.⁴⁹ Additionally, due to the lack of biocompatibility of PI substrate materials and the need for additional processing steps, their application in the biomedical field is hindered. These limitations must be considered when designing wearable sensors. To address these issues, transferring the PI film onto a flexible substrate is a decent choice. Researchers also used CO₂ laser irradiation to create LIG patterns on the PI substrate, followed by heating and curing. Since the Young's modulus of polydimethylsiloxane (PDMS) is closer to that of LIG compared to PI, in order to reduce the occurrence of cracks, researchers use PDMS to peel off LIG from the PI substrate and transfer it onto a flexible PDMS substrate. Based on this transfer method, the PI transfer to PDMS (PTP) transfer technique has been developed as shown in Fig. 7(a),⁵⁰ although it poses certain difficulties in controlling surface wrinkles and ensuring the integrity of LIG transfer. PDMS's breathability is weak, necessitating exploration into alternative materials for wearables, specifically porous polymers with strong mechanical attributes. For example, it is remarkable that LIG can also be formed on common materials like paper; however, this requires specific laser absorption rates and the application of flame retardants to prevent combustion.

Fig. 7 (a) The process of peeling LIG from the PI film using PDMS. (b) Human–machine interaction system utilizing LIG-based sensors and actuators. (c) The relationship between the resistance of LIG sensors and changes in bending angle and temperature.⁵⁰ (d) The reversible behavior of resistance changes at a 25% safe strain limit.⁴⁹ All the images enlisted here were reproduced with the copyright permission of ACS.

The LIG sensor is integrated with a microcontroller, transmitting signals via resistance changes. This setup enables the microcontroller to regulate the connected actuator via voltage changes at the output port, allowing for direct human interaction with the sensor and subsequently with the actuator. This approach eliminates the requirement for a computer host and introduces a novel and straightforward method of remote control between humans and robots, as shown in Fig. 7(b). With temperature changes, the actuator undergoes deformation, which in turn causes a change in resistance. The functional relationship is shown in Fig. 7(c).⁵⁰ Such an approach could potentially offer commercial opportunities for remote control, thereby minimizing human involvement in hazardous operations. Research has shown that some heterostructures exhibit excellent properties, avoiding Schottky barrier contacts and demonstrating higher electron mobility and improved noise performance.

Park et al. has developed a LIG/MoS₂ strain sensor, which protects the graphene from fracture, and uniform fracture in a certain range prompted the sensor to produce a reversible strain response, as shown in Fig. 7(d).⁴⁹ Although the difference in Young's modulus between LIG and MoS₂ is higher than that between LIG and PDMS, LIG/MoS₂ has a lubricating effect. This is because van der Waals forces have contacted the layers of MoS₂ and been contacted by van der Waals forces. The layered structure of the heterostructure formed with graphene exhibits excellent friction and lubrication characteristics,^51,52 adjusting its interlayer spacing under strain, demonstrating certain flexibility and compressibility. Sensors with this heterogeneous structure have an operating strain limit of 37.5% and offer a higher gauge factor (GF) over the strain range. GF is defined as in eqn (1).

R₀ represents the initial resistance value when there is no strain, R is the resistance of the material after experiencing strain, and ε represents the relative change in length of the material, that is, the ratio of the changed length to the initial length. These sensors demonstrate low initial resistance, which minimizes static power consumption and has contributed to their widespread applications. In addition, the MoS₂/graphene configuration results in a reduced Schottky barrier, facilitating ohmic contacts without the need for high-temperature annealing.⁵³ However, piezoresistive sensors despite their advances still struggle with high power consumption and only moderate response speed during operation. Conversely, piezoelectric sensors offer distinct advantages such as self-powering capabilities but are hindered by the toxicity and brittleness of the materials used, limiting their application. Despite these challenges, similar technologies have been utilized in the medical field, particularly in wearable health monitoring devices, where they can leverage their dynamic response characteristics for effective patient monitoring.

The mismatch in Young's modulus between MPU and LIG falls between that of PDMS/LIG and MoS₂/LIG, while being smaller than that of LIG/PI. Due to the stringent requirements for biocompatibility and durability in biomedical applications, MCU has been introduced as an innovative substrate for sensors. Thus, MPU/LIG has been developed for medical devices. The device initiates with LIG synthesis on a PI substrate and then transitions to medical-grade polyurethane (MPU). This configuration includes a copper connector paired with the LIG electrode using conductive silver paste, forming a typical sandwich structure. A zero insertion force (ZIF) connector completes the assembly, as illustrated in Fig. 8(a). This pressure sensor can detect various human tissue surfaces, such as collecting signals by monitoring muscle movements and throat vibrations, as shown in Fig. 8(b).⁵⁴ In medical applications, it avoids the noise generated by wire displacement disturbances of traditional signal detection devices. Additionally, due to the VIAS structure, it increases the density of transferred electrons, further favoring the integration of electronic devices. In the medical field, LIG synthesized from wood-derived precursors has been used for gait detection, demonstrating excellent piezoresistive response characteristics. As a green product that combines both high performance and environmental sustainability, LIG exhibits distinct advantages over other graphene-based materials in this regard.⁵⁵

Fig. 8 (a) Scheme with ZIF-connector and using VIAs. (b) Skin sensor applications for the detection of various tissues.⁵⁴ All the images enlisted here were reproduced with the copyright permission of ACS.

However, the GF of LIG-based devices is not exceptionally high. Challenges such as cracking and wrinkling in some carbon-based strain sensors reveal reversible behaviors and demonstrate potential for achieving ultra-high GF properties.⁵⁶ However, optimizing performance, particularly under stress levels that could induce cracks leading to line defects, is an ongoing issue.⁴⁹ Strategies for enhancing durability and performance, including the integration of an intermediate layer such as hexagonal boron nitride (hBN), are currently under research. These methods aim to eliminate structural weaknesses and improve the overall functionality of the device in medical applications, ensuring reliable diagnostics in dynamic medical environments.

3.2 Capacitive sensors

The stability of resistive sensors limits their application prospects, particularly in a good linearity range. Under large strains they display significant nonlinearity and diminished output signals. In contrast, capacitive sensors, which operate based on the principle of charge storage, measure physical quantities through changes in capacitance. Capacitive pressure sensors use a polymer as an insulating layer. On the PI film, a 405 nm diode laser is used for multiple laser irradiations, forming LIG electrodes on both sides of the film, which constitute a capacitor, as shown in Fig. 9(a). The LIG electrodes use conductive copper tape to transmit signals, as shown in Fig. 9(b). The electrodes are encapsulated by heat-transferring thermoplastic polyurethane (TPU) film, and the lower LIG electrode is grounded. The sensitivity of the capacitive sensor is defined as the ratio slope of capacitance change to pressure. In the study, the sensor implies a low-pressure sensitivity of up to 7.74% in the range of 0 to 2.865 kPa.⁵⁷ For similar resistive pressure sensors, a good linearity range is maintained between 0 and 0.2 kPa.⁵⁸ Although the sensitivity of the resistance sensor is better than that of the capacitive sensor, capacitive sensors offer a broader linear range, which indicates greater stability and consistency under varying pressures. This is attributed to the limited compression of the three-dimensional structure of LIG. Efforts to enhance the sensor's high-sensitivity range include adjusting the pore size of the LIG. Despite sacrificing the sensitivity for reduced power consumption, capacitive sensors offer greater stability during operations. The response of loading and unloading time is milliseconds with capacitance changes and Fig. 9(c) demonstrates the response of the sensor during loading and unloading, showing good repeatability of voltage over 1000 cycles, as shown in Fig. 9(d). In comparison, the relative capacitance change rate of capacitive sensors is lower than that of resistive sensors, but exhibits better stability. The size of the dielectric layer can be controlled by adjusting its thickness, which aids in the storage of charge, according to Fig. 9(e). Similarly, research into fabricating electrodes at both ends of precursors like PI is also gradually becoming a popular topic.

Fig. 9 (a) Schematic diagram of the capacitive pressure sensor. (b) The sensor performance under pressure. (c) Pressure pulses and (d) the response under pressure. (e) A schematic diagram of the working principle of the sensor.⁵⁷ All the images shown here were reproduced with the copyright permission of MDPI and RSC.

There is a general consensus that commonly used metal electrodes are more susceptible to polarization, affecting the stability of the electrodes, for example, TENG studies also show that it can lead to polarization fatigue and internal electric field instability.^59–61 Additionally, metal interconnect electromigration presents a significant challenge, not only generating electromigration noise but also creating excessive local currents that can burn out devices. Forming a heterostructure passivation layer is necessary to reduce overcurrent.²⁵ K. Ghosh et al. has already replaced metal interconnects with CNT/graphene, and the result demonstrates good contact properties.⁶² New materials have emerged as a potential alternative to metal electrodes. Therefore, LIG may potentially provide a breakthrough in solving these issues in the future.

3.3 Triboelectric sensors

Unlike capacitive and resistive sensors that depend on an external power source, triboelectric sensors offer a self-powered alternative, addressing environmental concerns over long-term energy use. A notable advancement in this area is triboelectric nanogenerator (TENG) technology, developed by Wan et al. This technology utilizes the minimal drive force required to harvest random and small-scale energy efficiently. Triboelectric sensors, such as TENG, operate by using conductors to extract charges from opposite ends of the electrodes and store them in capacitors. Fig. 10(a) presents a schematic of interfacial laser-induced fluorinated graphene (F-LIG). In this setup, a solution of fluorinated ethylene propylene (FEP) is spin-coated onto a PI substrate. The laser treatment causes localized pyrolysis, decomposing F atoms that combine with LIG to form F-LIG, as detailed in Fig. 10(b). This innovative doping method signifies a breakthrough in material engineering.⁶³ Additionally, TENG devices can generate electricity through mechanisms like a droplet-based electricity generator (DEG) in Fig. 10(c), where different substrates impact the current response as shown in Fig. 10(d).⁶³ FEP, when used as a triboelectric interface material, effectively transfers electrons, creating a positive current from FEP to F-LIG. Conversely, PI and PDMS substrates generate only minimal current. The key innovation here involves employing LIG as an interfacial material to increase surface roughness and contact area, significantly enhancing the triboelectric effect between liquids and solids. Advantages of friction-based sensors include their transient characteristics and self-powered nature. By adjusting parameters such as laser scanning speed and focal length, the hydrophobicity of LIG can be fine-tuned, thus controlling the width of LIG filaments. Introducing metal complexes further creates a superhydrophobic surface on the LIG electrode, enhancing its chemical stability and maintaining the pH window even after 10 000 cycles. Compared to traditional metal electrodes, which are susceptible to corrosion and flaking under long-term exposure to liquids, LIG offers superior chemical resistance and stability. Wang et al. has innovatively combined slippery liquid-infused porous surfaces (SLIPS) with TENG technology, significantly improving efficiency under low-temperature conditions compared to earlier models.⁶⁴ This advancement underscores the potential of TENG systems, particularly in the context of wearable devices requiring low-voltage energy storage solutions.

Fig. 10 (a) A schematic diagram of interfacial F-LIG. (b) The fabrication process of F-LIG. (c) The working mechanism of an F-LIG-based droplet-based electricity generator (DEG). (d) Current response of devices with different triboelectric interface materials.⁶³ All the images shown here were reproduced with the copyright permission of Wiley.

Despite these advancements, triboelectric sensors still face challenges in achieving consistent linearity and stability, as their self-powered systems primarily produce transient pulse signals. Developing a stable power supply for electronic equipment remains a significant technical challenge that needs addressing to fully harness the benefits of this promising technology.

4. Gas sensors

Gas sensors are widely used in civil, industrial, environmental monitoring, health diagnostics, and other fields. For example, with the development of new energy, the instability of traditional lithium batteries is caused by parasitic reactions and dendrite growth between metallic lithium and the electrolyte.^65,66 Gas fuel cells and related research directions are becoming a new trend, and simultaneously, there is a huge demand for microelectronic devices. The detection of hydrogen, a key reactive gas, has become particularly important because hydrogen is applied to petrochemicals, semiconductor production, and so on. However, hydrogen poses significant risks, including the potential for explosions when mixed with oxygen, which amplifies the need for highly sensitive and reliable gas sensing technologies. Environmental factors heavily influence the sensitivity of gas sensors. Therefore, adjustable sensitivity is required for different application scenarios. LIG offers a promising solution to this challenge. By precisely controlling the laser parameters during the production process, LIG allows for the customization of nanopore sizes within the graphene structure, which in turn can fine-tune the sensitivity of the sensors. This level of control makes LIG an excellent candidate for developing sensors that need to perform reliably across different environmental conditions.¹⁴

4.1 Thermistor gas sensor

This section discusses a method for identifying gas compositions through the measurement of thermal conductivity, utilizing the unique properties of LIG. LIG serves as an embeddable structure that facilitates gas detection, as shown in Fig. 11(a) and (c), and the contact pads leave an electrical path through the LIG filament connections synthesized on the flexible substrate.⁶⁷

Fig. 11 (a) Optical image of the LIG-based gas sensor. (b) and (c) SEM image of the conductivity channel. (d) Response of different gases by sensors.⁶⁷ All the images enlisted here were reproduced with the copyright permission of ACS.

Unlike other materials where resistance increases with temperature, LIG exhibits a decrease in resistance as temperatures rise. This phenomenon can be explained by two main factors: firstly, LIG maintains high charge carrier mobility at elevated temperatures with minimal charge carrier-phonon scattering, which moderates the increase in resistance. Secondly, graphene's distinctive zero band gap allows electrons to move freely from the conduction band to the valence band without the need to overcome an energy gap. Methods such as surface modification or doping can be employed to control the electrical properties of graphene. By applying bias, the LIG filament undergoes Joule heating to alter its temperature. When exposed to air, the LIG experiences temperature reduction and an increase in resistance due to convective effects. The properties of LIG vary in different gas environments, leading to differences in thermal conductivity. Consequently, the resistance is altered by different types of gasses. The device's response in various environments is shown in Fig. 11(d),⁶⁷ with a strong response to helium gas and the weakest response to oxygen gas. This is attributed to helium having the highest thermal conductivity while oxygen has the lowest. Researchers use oxygen plasma to control the string's size, which in turn determines the device's sensitivity.

Moreover, the thermal conductivity of porous materials like LIG can be finely controlled by adjusting the pore size, which directly affects heat transfer pathways and channels. This property is particularly useful in applications like thermal insulation in aerogels.⁶⁸ Despite their simple design, exceptional sensitivity, and rapid response, these sensors exhibit limited gas selection and high power requirements.

4.2 Semiconductor gas sensors

Thermal conductivity gas sensors have a large detection range but their detection sensitivity is not very high. Research indicates that ZnO exhibits high sensitivity in gas sensors while its stability is limited. LIG, with its intrinsic porous structure, significantly enhances the stability and sensitivity of these sensors. The porosity of LIG improves the interaction between the sensor and combustible gasses, optimizing the utilization rate of the gas-sensitive layer and thereby enhancing sensor response.⁶⁹ Therefore, a layer of ZnO nanorods was coated on the LIG's surface layer to increase the gas adsorption. During the experimental process, conductive silver paste is heated for one hour, and ZnO nanorods are synthesized using the hydrothermal method as shown in Fig. 12(a).⁷⁰ Temperature control is crucial during this process as it directly influences the length of the nanorods due to the temperature-dependent corrosive behavior occurring during the reaction.⁷¹ The solution is dissolved in alcohol, dispersed using ultrasound, and dropped into the middle of the silver paste. The corrosion behavior of the ZnO bar is unfavorable to gas adsorption in the experiment. Consider changing the method for the deviation of the metal oxide layer ff the LIG. Because of the rapid degradation of ZnO's high sensitivity films such as indium oxide as a substitute. In addition, changing the shape of the electrode layer can also be used as a feasible method.

Fig. 12 (a) Schematic diagram of the manufacturing process of the ZnO sensor.⁷⁰ (b) Schematic of the fabrication process of the electrochemical methane sensor. (c) and (d) SPE morphology. (e) Sensor's current responses at a constant methane concentration (50 ppm) under varying applied potentials. (f) Current measurements for various gases at a potential of 0.6 V.⁷² All the images enlisted here were reproduced with the copyright permission of ACS and Elsevier.

A lot of reports have mentioned various structural forms, such as network-like networks,⁷³ bridge-shaped designs,⁷⁴ and wrinkles.^75,76 These structures exhibit high sensitivity in the detection of multiple gasses and are advantageous for sensor arrays to distinguish different gasses in multi-gas environments. However, current research still needs to be improved, especially in mixed gas detection and the integration of multifunctional sensors.

4.3 Electrochemical gas sensors

Electrochemical gas sensors represent a sophisticated approach to detecting gas compositions and concentrations, leveraging specific gas reactions and electrochemical processes. Although semiconductor gas sensors are capable of detecting gas mixtures and exhibit sensitivity at the parts-per-million (ppm) level with a precision of 1 part-per-billion (ppb), they are limited by a narrow detection range and significantly influenced by environmental factors such as temperature and humidity, which can affect their stability. Electrochemical sensors often utilize liquid electrolytes as the medium for gas detection. However, these liquid electrolytes can block pores and inhibit gas diffusion within the electrodes, posing a challenge for effective gas sensing. Polyvinylidene difluoride (PVDF) is employed as a porous solid polymer–electrolyte (SPE) to overcome this. This viscous mixture is applied to the LIG interdigitated electrodes using a glass rod and is subsequently solidified by cooling at room temperature to form a pseudo-solid state. This setup is depicted in the sensor's response to O₂ and CH₄ interactions within the electrolyte, as illustrated in Fig. 12(b), and the morphology of the SPE is showcased in Fig. 12(c) and (d). At room temperature, the Pd metal layer acts as a catalyst, partially embedded into the upper layer of LIG electrodes, and absorbs methane to generate water and carbon dioxide. The chemical reaction of interference acids enables the gas chemical sensors to function normally, meeting practical life requirements. However, due to the specificity of chemical reactions, this type of sensor is limited by its single-functionality. Fig. 12(e) and (f) show low voltage operation, with air in the reaction to form a current, and most notably low detection limits and high sensitivity, down four orders of magnitude. 0.55 μA ppm⁻¹ cm⁻² with a response time of 40 s and an experimental detection limit of 9 ppm are the advantages of this sensor.⁷² Sensors’ gas speciation composition detection presents challenges: detection is specific, reducing the array of gases detectable. The detection of a multi-component mixed system is inaccuracy. Its potential as a solution remains uncertain.

5. Biosensors

In recent years, agricultural development has been increasingly emphasized, playing a crucial role in food safety and the detection of substance residues. Additionally, the global pandemic has highlighted the importance of public health, significantly elevating the relevance and application of biosensors in medical diagnostics. Traditional diagnostic methods include magnetic resonance imaging (MRI), chromatography, and in vivo testing. MRI is a static measurement method with strict sample requirements, is generally too time-consuming and resource-intensive for routine use, and often relegated to laboratory research. Chromatography involves a complex preprocessing and analysis process. In vivo testing requires sampling from an organism and minor surgery, which is invasive. These methods do not meet the needs for real-time detection and are often costly, hence not ideal for biological detection. In contrast, biosensors with point-of-care testing (POCT) reduce the time between sample collection and analysis, providing point-to-point treatment,⁷⁷ real-time tracking of patient physiological conditions, reducing direct contact with patients and lowering the risk of infection in medical diagnostics. LIG masks have almost 100% antibacterial efficiency under certain lighting conditions, further optimizing this function.⁷⁸ Additionally, biosensors should be cost-effective, presenting advantages over other detection methods.

5.1 Enzyme sensors

Biosensors often operate in various complex environments, such as body fluids containing a variety of organic interferents. Patients often need an accurate test for a particular substance, but these chemicals often have similar concentrations and redox potentials, making the specificity of recognition very important. They typically analyze signals that are easy to distinguish. For biomolecules, enzymes are commonly chosen as specific substances. When using enzymes as recognition elements in biosensors, special attention should be paid to the immobilization of enzymes. Maintaining the stability and activity of these immobilized enzymes is an important factor to consider.

The biochemical reagents play a crucial role in anchoring enzymes on the electrode surface, and redox reactions are employed to detect specific molecules in the preparation of functional enzymatic biosensors. This type of biosensor is widely used to detect uric acid, glucose, and dopamine. These sensors are highly valued in the fields of biological detection and medical diagnosis due to their exceptional selectivity and sensitivity. Ping et al. and colleagues adopted this approach, modifying electrodes with organophosphorus hydrolase (OPH) to hydrolyze methyl parathion and release reactive para-nitrophenol (PNP).⁷⁹ They used the linear relationship between the current peak generated by the applied voltage and the concentration of PNP to determine the residual concentration of the chemical substance. However, it should be noted that excessive enzyme modification might decrease electron transfer efficiency due to the high concentration of Nafion solution.⁸⁰ As previously discussed, LIG has a three-dimensional porous structure that allows semi-solid gelatin to be well embedded in electrodes. Thus, in terms of characterization, the outstanding mechanical properties of LIG, along with its efficient electron transfer capability and the ability to maintain almost identical electrochemical performance to that before bending, plays an essential role in enhancing sensor performance. Common LIG/PDMS encapsulated electrodes exhibit good mechanical properties and show consistent electrochemical responses at different bending angles, as shown in Fig. 13(a).⁷⁹ They can closely adhere to curved surfaces, providing a flexible and reliable electrode choice for long-term detection. However, it should be noticed that even a small range of stress can change the resistance to some extent, and the issue of low sensitivity in small strain ranges remains to be resolved, as shown in Fig. 13(b).⁷⁹ Piezoelectric sensors equipped with charge amplifiers perform well under low stress conditions and can detect small deformations, which might be a new route to be explored. For PDMS/LIG sensors, it is important to ensure that no interference signals are observed in their response curves or that they can effectively distinguish the oxidation or reduction peaks of interfering substances, as shown in Fig. 13(c).⁸¹ Taking the example of Gao et al. of wearable devices for detecting uric acid in sweat,⁸² researchers induced the oxidation reaction of uric acid by applying a voltage. In such biosensors that selectively redox processes, it is necessary to eliminate the influence of product solutions on the CV response. This issue will hinder the development of related sensors.

Fig. 13 (a) CV curves during bending. (b) The resistance–strain curve of the PDMS/LIG electrode during stretching.⁷⁹ (c)DPV curves of different biomolecules.⁸¹ All the images shown here were reproduced with the copyright permission of ACS and Elsevier.

Additionally, by combining biomolecules and enzymes, where enzymes act as inhibitors of redox reactions, their response can be used to detect biomolecules.⁸³ Laser-scribed graphene (LSG) electrodes are prepared using laser scanning technology, and the electrode surface is modified with 1-pyrenebutyric acid (PBA) as an intermediary to covalently bond with anticoagulants, thereby inhibiting redox reactions, as shown in Fig. 14(a). The detection accuracy is at the pM level, and as the scanning rate increases, the oxidation peak potential rises, and a pair of redox current peaks and potentials show a high degree of symmetry in Fig. 14(b),⁸³ indicating that the electrode reactions on the LIG film are quasi-reversible.⁸⁴ Upon testing, this electrode also exhibits a higher electron transfer rate compared to other carbon materials. The electron transfer rate is k₀ = 0.0044 ± 0.0003 cm s⁻¹. Furthermore, the direct reaction between the electrode surface modifiers and the analytes promotes the electron transfer rate. This reaction leads to a change in charge transfer and distribution, thereby affecting the electrode's current response. By monitoring changes in potential or current, it can detect the presence and concentration of target biomolecules.

Fig. 14 (a) Schematic of the thrombin detection mechanism. (b) CV curves of electrodes at different scan rates.⁸³ All the images shown here were reproduced with the copyright permission of ACS.

5.2 Non-enzyme sensors

The sensitivity, stability, and reproducibility of enzyme-based biosensors are commonly compromised by environmental factors during their fabrication, storage, and operation, which restricts their applications. In 2007, a novel finding was reported that Fe₃O₄ exhibits enzyme-like activity. The discovery opened up new avenues for the development of non-biological nano enzyme sensors, which have since shown better stable chemical properties.⁸⁵ Furthermore, non-biological enzymes are user-friendly for controlling aperture and surface modifications which enhances their versatility. Their design allows for ease of use and customization, making them an attractive option for widespread implementation. These non-enzyme sensors are not only user-friendly in terms of control but also adaptable for large-scale production and highly feasible for industry use, although issues regarding biocompatibility remain. As these sensors overcome the limitations, these advancements could expand their field of application in health care, environmental monitoring, and biotechnology, all sectors that require long-lasting sensor tools. These advances tackle the shortcoming of conventional enzyme sensors, paving the path for further exploration into creating more effective and robust bio-sensing systems.

In related research, various materials, including metal nanoparticles,^86,87 two-dimensional nanomaterials,^88,89 nanoporous materials, organic frameworks, and oxides⁹⁰ have been developed through methods such as self-assembly of nano-frameworks and enhanced coordination. These materials have achieved stable activity in sensor applications. In the aspect of blood glucose monitoring, an innovative approach has been proposed: transforming the traditional early-stage diabetes blood sugar testing method into a more convenient urine sugar testing technique. Since the concentration of sugar in urine is usually lower than in blood, this requires the sensors to have high sensitivity. Zhou et al. proposes an innovative method, using a two-dimensional material similar to graphene, MXene, as a connecting medium. By utilizing a π-conjugated system similar to LIG, they combine a metal nanoparticle material with LIG to successfully create non-enzymatic biosensors. As shown in Fig. 15(a)–(c),⁹¹ the sensitivity of these sensors is calibrated using the CV curve to establish a linear relationship between I_pa and v_1/2 according to the Randles–Sevcik equation. The results show that the electrochemically active area of the Au@CuO/LIG electrode is 4.4 times that of the Au@CuO/graphene electrode, which is attributed to the porous structure of LIG, and the high specific surface area can affect the rate of recombination. The sensitivity of this sensor is 1.124 × 10⁶ μA mM⁻¹ cm⁻², and within its detection range, the limit of detection (LOD) is as low as 1.8 μM.⁹¹ The tunability of nanostructuring, thickness, and morphology of LIG can be easily achieved through computer-controlled laser technology, such as enabling pore sizes to reach 10 nm. Using this method, gold nanoparticles (AuNPs) and Cu₂O can be embedded into the porous framework of LIG, thereby providing a more effective way for controlled nanostructured catalysis.⁹² It is well known that catalytic sites are typically located on the surface or edges of the material. The structure is similar to a core–shell structure. The fabrication process of the composite framework is shown in Fig. 15(d), and it presents honeycomb and cubic nanostructures, as shown in Fig. 15(e). Using an app to display the response, the single LIG current response is subtle, as shown in Fig. 15(f). In contrast, after embedding pristine Cu₂O nanoparticles (NPS) into the electrode surface, the peak current increased by about 10 times, as shown in Fig. 15(g).⁹² This response also suggests the redox of the solution. Moreover, Cu₂O has the advantages of being low-cost and nontoxic, which are also utilized in the study. Similar hybrid nanosensors exhibit outstanding sensitivity.⁹³ Besides performance metrics, the economic impact is essential. The existing research has successfully synthesized copper nanoparticles on paper-based precursors and fabricated LIG, which provides a convenient and efficient electrode fabrication approach for non-enzymatic glucose detection.⁹⁴ Of course, this similar approach also accomplishes bimolecular recognition, and with a LOD as low as 11.98 nM, this paper-based platform also shows advantages in terms of cost control.⁹⁵

Fig. 15 (a) Catalytic current of Au@CuO/LIG and Au@CuO/GE. (b) The relationship between peak current and v_1/2 about Au@CuO/GE. (c) The relationship between peak current and v_1/2 about Au@CuO/LIG.⁹¹ (d) and (e) The fabrication process of the electrochemical chip. (f) POCT system. (g) The current responses of the bare LIG, LIG–Cu₂O, and LIG–Cu₂O–Au in 0.1 M NaOH with the addition of 7 mM glucose.⁹² All the images enlisted here were reproduced with the copyright permission of Elsevier.

5.3 DNA sensors

Although enzyme sensors and non-enzyme sensors offer various applications and flexibility by utilizing biorecognition elements such as antibodies and receptors, they primarily rely on redox reactions for detecting biomolecules. This mechanism often lacks the specificity required for detecting nucleic acid sequences. In contrast, DNA sensors are unparalleled in their specificity and sensitivity due to their reliance on complementary base pairing. This principle enables precise recognition of target nucleic acid sequences and the ability to distinguish even single-base differences. These make DNA sensors crucial for disease diagnosis, genetic screening, pathogen detection, and essential tools in modern molecular diagnostics.

By utilizing the binding of biomolecules to analytes, the electron affinity or electron transfer properties of the electrode surface are altered. This type of binding, unlike traditional covalent bonding, is reversible. Such binding leads to a change in charge transfer or charge distribution, thereby altering the electrode's potential and current response. By measuring changes in potential or current, the presence and concentration of target biomolecules can be determined. DNA sensors, as a typical representative of biosensors, have broad application prospects in fields such as medicine, biological research, disease diagnosis, and environmental monitoring. Unlike biosensors that form strong and irreversible bonds through covalent attachment, this method directly utilizes the anchoring effect of poly-cytosine (poly-C) DNA to form a DNA probe (pDNA), which pairs with the complementary target DNA (cDNA) dropped onto the LIG surface. By hybridizing ferrocene-labeled cDNA to LIG, the single-stranded DNA probe (ssDNA) binds to graphene, and after hybridization, the double-stranded (ds-)DNAs separate due to strong repulsion from graphene, leading to different behaviors of ferrocene. This significant distinguishability can be used to detect DNA concentration in blood, as shown in Fig. 16(a), and the sensor's LOD is 57 fM. Compared to bare LIG, the Raman spectrum D peak of DNA probe functionalized LIG (LIG-pDNA) is higher, as shown in Fig. 16(b), indicating that more defects were introduced due to the adsorption functionalization of LIG-pDNA. Furthermore, the sensor's specificity is also quite significant. Compared to non-complementary DNA (ncDNA), the current change caused by the specific binding of complementary target DNA (cDNA) is about 5 times higher, as shown in Fig. 16(c),⁹⁶ demonstrating the high specificity of the sensor. In this research, the use of the anchoring block of poly-cytosine DNA can be extended to nano-materials like ZnO, GO, and carbon nanotubes, providing insights for reversible binding.

Fig. 16 (a) Schematic diagram of the detection process of a DNA biosensor. (b) The Raman spectrum of LIG-bare and LIG-pDNA. (c) The current response of hybridization experiments in different solutions.⁹⁶ (d) A schematic of the sensor. All the images shown here were reproduced with the copyright permission of Elsevier.

The biological environment within an organism is highly complex, and the precision of many nucleic acid signal detections is challenging. Subtle changes in certain DNA sequences can significantly impact the processes of gene transcription, translation, and expression within an organism. A DNA sensor that enhances the detection signal has been proposed and this sensor ingeniously utilizes a LIG electrode and carefully constructs a sensitive layer of AuNPs on its surface. The target molecules, 5-methylcytosine-single strand DNA (5mC-ssDNA) and N6-methyladenine RNA (m6A-RNA), are stably anchored onto the AuNPs via the introduction of thiol (–SH) groups. Subsequently, antibodies specific to these particular nucleic acid sequences bind to the corresponding nucleic acid chains, forming specific immune complexes.⁹⁷ At the same time, 1-hexanethiol (HT) serves to seal the nucleic acid chain sites that are not involved in the reaction, thus significantly enhancing the sensor's selectivity. The structure is shown in Fig. 16(d).

It is worth emphasizing that the sensor utilizes an additional antigen–antibody specific binding mechanism involving streptavidin-modified horseradish peroxidase (SA-HRP). In the detection solution containing hydrogen peroxide hydroquinone (H₂O₂-HQ), SA-HRP initiates its catalytic action through differential pulse voltammetry (DPV), generating redox currents to achieve signal amplification. The resulting current signal strength is directly proportional to the number of methylation nucleic acid chains captured on the electrode, effectively enhancing signal detection.

The sensor greatly enhances detection sensitivity by utilizing biological enzymes to catalyze specific biochemical reactions. This innovative design concept not only greatly expands the horizon of sensor research but also opens up new possibilities for designing and applying future biosensors.

6 Conclusions and prospects

This review has extensively discussed the application and advantages of LIG in various electronic devices and explores the challenges faced, including structural optimization and material doping. Structural optimization involves wrinkles, meshes, bridging, heterostructures, and metal–organic frameworks, while material optimization includes doping and surface modification. Doping and surface modification of LIG are common methods to enhance its functionality. Given graphene's inherent zero bandgap property, bandgap tuning through doping is necessary. Most of the gas sensors, mechanical sensors, and other devices discussed in this review have employed this technique. The production parameter operability of LIG opens possibilities for the development of electronic devices. The article thoroughly discusses the effects of laser power, speed, focal length, and wavelength on the patterning and quality of LIG, as well as how to detect signals through physicochemical principles. PI is often used as a precursor substance in commercial applications to suit sensor scenarios, combining the hardness of graphene with the compatibility of flexible substrates to fill gaps in wearable sensors. However, with increasing demands for quality of life, we still face many challenges. Broad commercialization requires faster device manufacturing speeds, yet the production speed of LIG is limited by the scanning speed of the light source. Balancing the precision control of high-quality film production with speed remains a challenge. Additionally, the quality of LIG varies significantly across different substrates. For example, under the same conditions, the graphitization degree of the PR substrate is higher than that of the PI substrate. Existing polymer organic precursors are insufficient, and deeper research is needed for other precursors to generate LIG, including polystyrene (PS) and others. Research on substrates will be a way to address the issue of LIG's quality. When selecting substrates, economic and environmental factors are considered. The use of environmentally friendly degradable substrates can reduce electronic waste and plastic pollution, offering advantages in terms of cost and post-treatment. However, current studies indicate that the performance of these sustainable substrates still lags behind that of traditional PI.

Currently a CO₂ laser source is the most common choice in commercialization, but only UV and infrared lasers have been comparatively studied, leaving gaps in our understanding of optimal light source conditions. Exploring a parallel laser beam system could potentially enhance LIG production rates by treating multiple areas simultaneously, thus improving efficiency.

Additionally, issues such as the incomplete transfer of LIG and defects in the stripping process during sensor transfer are areas needing improvement. The choice of substrate post-transfer significantly influences device performance and longevity. Although PDMS is widely used, alternatives like hydrogels, despite their high sensitivity and good biocompatibility, suffer from limitations like drying out and insufficient wetting properties, which restrict their practical use. Looking forward, enhancing mechanical sensor performance could involve adjustments in LIG's thickness, patterning, and doping. PVDF may present new opportunities for specific sensor applications. Current gas sensors face challenges with stability and differentiation in gas composition, particularly electrochemical sensors limited to detecting single gas types.

The singular functionalization of sensors and the emerging trend of substance cross-linking are paving the way for innovative solutions, with 3D printing technology opening avenues for intelligent structural designs.^98,99 In the realm of biosensors, as medical health issues become more prominent, rapid and accurate detection becomes a primary concern. Non-enzymatic sensors, due to their more relaxed preservation environment, will also become more popular. In the future, the integration of multiple sensors will become a major trend, LIG has already integrated multifunction sensors that detect temperature, gas, humidity, and pressure, which are already feasible.¹⁰⁰ As we enter the age of intelligence, ensuring compatibility with internet-based systems will be crucial for facilitating control and integration, representing the next significant challenge in sensor technology.

Author contributions

X. H contributed to writing the original article and revision. J. W. contributed to the editing. C. F., H. W., J. Y., and Q. W. contributed to supervising the final manuscript. All authors have read and agreed to the published version of the manuscript.

Data availability

Data availability is not applicable to this article as no new data were created or analyzed in this study.

Conflicts of interest

The authors declare no conflicts of interest.

Acknowledgements

This work is partially supported by the Major Program of Hubei Province (grant no. 2023BAA009).

Notes and references

1. A. K. Geim, Science, 2009, 324, 1530–1534.
2. H.-M. Dong, Y.-F. Duan, F. Huang and J.-L. Liu, Front. Phys., 2018, 13, 137203.
3. K. S. Novoselov, D. Jiang, F. Schedin, T. J. Booth, V. V. Khotkevich, S. V. Morozov and A. K. Geim, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 10451–10453.
4. J. Zhao, P. Ji, Y. Li, R. Li, K. Zhang, H. Tian, K. Yu, B. Bian, L. Hao, X. Xiao, W. Griffin, N. Dudeck, R. Moro, L. Ma and W. A. de Heer, Nature, 2024, 625, 60–65.
5. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191.
6. S. W. Mushfiq and R. Afzalzadeh, Sci. Rep., 2022, 12, 9872.
7. X. Gu, Y. Zhao, K. Sun, C. L. Z. Vieira, Z. Jia, C. Cui, Z. Wang, A. Walsh and S. Huang, Ultrason. Sonochem., 2019, 58, 104630.
8. F. Liu, P. Li, H. An, P. Peng, B. McLean and F. Ding, Adv. Funct. Mater., 2022, 32, 2203191.
9. L. Wang, R. Lai, L. Zhang, M. Zeng and L. Fu, ACS Mater. Lett., 2022, 4, 528–540.
10. H. Zhang and P. X. Feng, Carbon, 2010, 48, 359–364.
11. H. Wordenweber, S. Karthauser, A. Grundmann, Z. Wang, S. Aussen, H. Kalisch, A. Vescan, M. Heuken, R. Waser and S. Hoffmann-Eifert, Sci. Rep., 2022, 12, 18743.
12. D. Luo, M. Choe, R. A. Bizao, M. Wang, H. Su, M. Huang, S. Jin, Y. Li, M. Kim, N. M. Pugno, B. Ren, Z. Lee and R. S. Ruoff, Adv. Mater., 2022, 34, 2110509.
13. Y. Hou, B. Wang, L. Zhan, F. Qing, X. Wang, X. Niu and X. Li, Mater. Today, 2020, 36, 10–17.
14. J. Zhu, M. Cho, Y. Li, I. Cho, J. H. Suh, D. D. Orbe, Y. Jeong, T. L. Ren and I. Park, ACS Appl. Mater. Interfaces, 2019, 11, 24386–24394.
15. Z. Liu, Q. Xu, C. Zhang, Q. Sun, C. Wang, M. Dong, Z. Wang, H. Ohmori, M. Kosinova, T. Goto, R. Tu and S. Zhang, Appl. Surf. Sci., 2020, 514, 145938.
16. L. Sun, B. Chen, W. Wang, Y. Li, X. Zeng, H. Liu, Y. Liang, Z. Zhao, A. Cai, R. Zhang, Y. Zhu, Y. Wang, Y. Song, Q. Ding, X. Gao, H. Peng, Z. Li, L. Lin and Z. Liu, ACS Nano, 2022, 16, 285–294.
17. J. T. Li, M. G. Stanford, W. Chen, S. E. Presutti and J. M. Tour, ACS Nano, 2020, 14, 7911–7919.
18. S. Bae, H. Kim, Y. Lee, X. Xu, J. S. Park, Y. Zheng, J. Balakrishnan, T. Lei, H. R. Kim, Y. I. Song, Y. J. Kim, K. S. Kim, B. Ozyilmaz, J. H. Ahn, B. H. Hong and S. Iijima, Nat. Nanotechnol., 2010, 5, 574–578.
19. J. Lin, Z. Peng, Y. Liu, F. Ruiz-Zepeda, R. Ye, E. L. Samuel, M. J. Yacaman, B. I. Yakobson and J. M. Tour, Nat. Commun., 2014, 5, 5714.
20. P. I. C. Claro, T. Pinheiro, S. L. Silvestre, A. C. Marques, J. Coelho, J. M. Marconcini, E. Fortunato, L. H. C. Mattoso and R. Martins, Appl. Phys. Rev., 2022, 9, 041305.
21. S. Nandy, E. Fortunato and R. Martins, Prog. Nat. Sci.: Mater. Int., 2022, 32, 1–9.
22. T. Pinheiro, M. Morais, S. Silvestre, E. Carlos, J. Coelho, H. V. Almeida, P. Barquinha, E. Fortunato and R. Martins, Adv. Mater., 2024, 36, 2402014.
23. X. Li, W. Cai, K. S. Teh, M. Qi, X. Zang, X. Ding, Y. Cui, Y. Xie, Y. Wu, H. Ma, Z. Zhou, Q.-A. Huang, J. Ye and L. Lin, ACS Appl. Mater. Interfaces, 2018, 10, 26357–26364.
24. J. de la Roche, I. López-Cifuentes and A. Jaramillo-Botero, Carbon Lett., 2022, 33, 587–595.
25. Y. Jeong, O. Douglas, U. Misra, M. R. E. Tanjil, K. Watanabe, T. Taniguchi and M. C. Wang, Adv. Electron. Mater., 2021, 7, 2100002.
26. M. Kim, M. G. Gu, H. Jeong, E. Song, J. W. Jeon, K.-M. Huh, P. Kang, S.-K. Kim and B. G. Kim, ACS Appl. Energy Mater., 2021, 4, 208–214.
27. G. F. Hawes, P. Verma, M. Uceda, G. Karimi, B. S. Noremberg and M. A. Pope, ACS Appl. Mater. Interfaces, 2023, 15, 10570–10584.
28. Z. Zhang, M. Song, J. Hao, K. Wu, C. Li and C. Hu, Carbon, 2018, 127, 287–296.
29. M. F. El-Kady, Y. Shao and R. B. Kaner, Nat. Rev. Mater., 2016, 1, 16033.
30. L. Kong, S. Dang, B. Yao, Y. Li, B. Chen, X. Hua and G. Tian, J. Colloid Interface Sci., 2022, 628, 670–681.
31. Y. W. Chi, C. C. Hu, H. H. Shen and K. P. Huang, Nano Lett., 2016, 16, 5719–5727.
32. J. Yao, L. Liu, S. Zhang, L. Wu, J. Tang, Y. Qiu, S. Huang, H. Wu and L. Fan, Electrochim. Acta, 2023, 441, 141719.
33. Z. Peng, R. Ye, J. A. Mann, D. Zakhidov, Y. Li, P. R. Smalley, J. Lin and J. M. Tour, ACS Nano, 2015, 9, 5868–5875.
34. W. Zhao, M. Jiang, W. Wang, S. Liu, W. Huang and Q. Zhao, Adv. Funct. Mater., 2020, 31, 2009136.
35. Z. Chen, Y. Chen, Y. Zhao, F. Qiu, K. Jiang, S. Huang, C. Ke, J. Zhu, D. Tranca and X. Zhuang, Langmuir, 2021, 37, 2523–2531.
36. A. Ray, J. Roth and B. Saruhan, Molecules, 2022, 27, 329.
37. N. Liu and Y. Gao, Small, 2017, 13, 1701989.
38. F. Clerici, M. Fontana, S. Bianco, M. Serrapede, F. Perrucci, S. Ferrero, E. Tresso and A. Lamberti, ACS Appl. Mater. Interfaces, 2016, 8, 10459–10465.
39. C. Choi, D. S. Ashby, D. M. Butts, R. H. DeBlock, Q. Wei, J. Lau and B. Dunn, Nat. Rev. Mater., 2019, 5, 5–19.
40. H. W. Park and K. C. Roh, J. Power Sources, 2023, 557, 232558.
41. C. Costentin, T. R. Porter and J. M. Saveant, ACS Appl. Mater. Interfaces, 2017, 9, 8649–8658.
42. R. Abreu, M. dos Santos Klem, T. Pinheiro, J. Vaz Pinto, N. Alves, R. Martins, E. Carlos and J. Coelho, FlatChem, 2024, 46, 100672.
43. L. Li, J. Zhang, Z. Peng, Y. Li, C. Gao, Y. Ji, R. Ye, N. D. Kim, Q. Zhong, Y. Yang, H. Fei, G. Ruan and J. M. Tour, Adv. Mater., 2016, 28, 838–845.
44. M. He, Y. Wang, S. Wang and S. Luo, Carbon, 2020, 168, 308–318.
45. Y. Wang, H. Wang, J. Ye, L. Shi and X. Feng, Chem. Eng. J., 2020, 383, 123096.
46. R. Ye, Y. Chyan, J. Zhang, Y. Li, X. Han, C. Kittrell and J. M. Tour, Adv. Mater., 2017, 29, 1702211.
47. A. Lamberti, M. Serrapede, G. Ferraro, M. Fontana, F. Perrucci, S. Bianco, A. Chiolerio and S. Bocchini, 2D Mater., 2017, 4, 035012.
48. L. Wang, Z. Wang, A. N. Bakhtiyari and H. Zheng, Micromachines, 2020, 11, 1094.
49. A. Chhetry, M. Sharifuzzaman, H. Yoon, S. Sharma, X. Xuan and J. Y. Park, ACS Appl. Mater. Interfaces, 2019, 11, 22531–22542.
50. H. Wang, Z. Zhao, P. Liu, Y. Pan and X. Guo, ACS Appl. Mater. Interfaces, 2022, 14, 41283–41295.
51. C. Lee, Q. Li, W. Kalb, X.-Z. Liu, H. Berger, R. W. Carpick and J. Hone, Science, 2010, 328, 76–80.
52. K. H. Hu, X. G. Hu, Y. F. Xu, F. Huang and J. S. Liu, Tribol. Lett., 2010, 40, 155–165.
53. A. Behranginia, P. Yasaei, A. K. Majee, V. K. Sangwan, F. Long, C. J. Foss, T. Foroozan, S. Fuladi, M. R. Hantehzadeh, R. Shahbazian-Yassar, M. C. Hersam, Z. Aksamija and A. Salehi-Khojin, Small, 2017, 13, 1604301.
54. A. Dallinger, K. Keller, H. Fitzek and F. Greco, ACS Appl. Mater. Interfaces, 2020, 12, 19855–19865.
55. A. F. Carvalho, A. J. S. Fernandes, R. Martins, E. Fortunato and F. M. Costa, Adv. Mater. Technol., 2020, 5, 2000630.
56. J. Zhou, P. Guo, L. Cui, C. Yan, D. Xu, F. Li, C. Zhang and A. Wang, Diamond Relat. Mater., 2023, 132, 109619.
57. T. Zhao, H. Zhu and H. Zhang, Biosensors, 2023, 13, 546.
58. Y. Zhu, J. Li, H. Cai, Y. Wu, H. Ding, N. Pan and X. Wang, Sens. Actuators, B, 2018, 255, 1262–1267.
59. J. Jeong, B. Yoo, E. Jang, I. Choi and J. Lee, Nanoscale Res. Lett., 2022, 17, 42.
60. A. Datta, J. Fiala, P. Becla and S. Motakef, APL Mater., 2017, 5, 106109.
61. C. Chen, J. Wang, C. Li, Z. Wen, Q. Xu and J. Du, AIP Adv., 2017, 7, 055829.
62. K. Ghosh, N. Ranjan, Y. K. Verma and C. S. Tan, RSC Adv., 2016, 6, 53054–53061.
63. Y. Chen, B. Xie, J. Long, Y. Kuang, X. Chen, M. Hou, J. Gao, S. Zhou, B. Fan, Y. He, Y. T. Zhang, C. P. Wong, Z. Wang and N. Zhao, Adv. Mater., 2021, 33, 2104290.
64. W. Xu, X. Zhou, C. Hao, H. Zheng, Y. Liu, X. Yan, Z. Yang, M. Leung, X. C. Zeng, R. X. Xu and Z. Wang, Natl. Sci. Rev., 2019, 6, 540–550.
65. A. Manthiram, X. Yu and S. Wang, Nat. Rev. Mater., 2017, 2, 16103.
66. L. Zhang, T. Yang, C. Du, Q. Liu, Y. Tang, J. Zhao, B. Wang, T. Chen, Y. Sun, P. Jia, H. Li, L. Geng, J. Chen, H. Ye, Z. Wang, Y. Li, H. Sun, X. Li, Q. Dai, Y. Tang, Q. Peng, T. Shen, S. Zhang, T. Zhu and J. Huang, Nat. Nanotechnol., 2020, 15, 94–98.
67. M. G. Stanford, K. Yang, Y. Chyan, C. Kittrell and J. M. Tour, ACS Nano, 2019, 13, 3474–3482.
68. J.-J. Zhao, Y.-Y. Duan, X.-D. Wang and B.-X. Wang, J. Nanopart. Res., 2012, 14, 1024.
69. K. Suematsu, K. Watanabe, A. Tou, Y. Sun and K. Shimanoe, Anal. Chem., 2018, 90, 1959–1966.
70. S.-F. Tseng, P.-S. Chen, S.-H. Hsu, W.-T. Hsiao and W.-J. Peng, Appl. Surf. Sci., 2023, 620, 156847.
71. W. D. Hsu, J. K. Tsai, J. Y. Tang, T. H. Meen and T. C. Wu, Microsyst. Technol., 2017, 24, 3957–3963.
72. M. Dosi, I. Lau, Y. Zhuang, D. S. A. Simakov, M. W. Fowler and M. A. Pope, ACS Appl. Mater. Interfaces, 2019, 11, 6166–6173.
73. J. Han, J. Y. Lee, J. Lee and J. S. Yeo, Adv. Mater., 2018, 30, 1704626.
74. X. Wang, Y. Gu, Z. Xiong, Z. Cui and T. Zhang, Adv. Mater., 2014, 26, 1336–1342.
75. L. Liu, X. Yang, L. Zhao, W. Xu, J. Wang, Q. Yang and Q. Tang, Nano Energy, 2020, 73, 104797.
76. Z. Niu, H. Dong, B. Zhu, J. Li, H. H. Hng, W. Zhou, X. Chen and S. Xie, Adv. Mater., 2013, 25, 1058–1064.
77. P. B. Luppa, C. Muller, A. Schlichtiger and H. Schlebusch, TrAC, Trends Anal. Chem., 2011, 30, 887–898.
78. L. Huang, S. Xu, Z. Wang, K. Xue, J. Su, Y. Song, S. Chen, C. Zhu, B. Z. Tang and R. Ye, ACS Nano, 2020, 14, 12045–12053.
79. F. Zhao, J. He, X. Li, Y. Bai, Y. Ying and J. Ping, Biosens. Bioelectron., 2020, 170, 112636.
80. A. A. Karyakin, E. E. Karyakina, L. Gorton, O. A. Bobrova, L. V. Lukachova, A. K. Gladilin and A. V. Levashov, Anal. Chem., 1996, 68, 4335–4341.
81. C. Li, Y. Cai, J. Hu, J. Liu, H. Dai, Q. Xu, C. Zhang, X. Zhang, K. Liu, M. L. Kosinova, T. Goto, R. Tu and S. Zhang, ACS Appl. Mater. Interfaces, 2023, 15, 27399–27410.
82. Y. Yang, Y. Song, X. Bo, J. Min, O. S. Pak, L. Zhu, M. Wang, J. Tu, A. Kogan, H. Zhang, T. K. Hsiai, Z. Li and W. Gao, Nat. Biotechnol., 2020, 38, 217–224.
83. C. Fenzl, P. Nayak, T. Hirsch, O. S. Wolfbeis, H. N. Alshareef and A. J. Baeumner, ACS Sens., 2017, 2, 616–620.
84. X. Cui, X. Fang, H. Zhao, Z. Li and H. Ren, Anal. Methods, 2017, 9, 5322–5332.
85. L. Gao, J. Zhuang, L. Nie, J. Zhang, Y. Zhang, N. Gu, T. Wang, J. Feng, D. Yang, S. Perrett and X. Yan, Nat. Nanotechnol., 2007, 2, 577–583.
86. W. Peng, Y. Qin, W. Li, M. Chen, D. Zhou, H. Li, J. Cui, J. Chang, S. Xie, X. Gong and B. Tang, ACS Sens., 2020, 5, 1912–1920.
87. T. Dayakar, K. V. Rao, K. Bikshalu, V. Rajendar and S. H. Park, J. Mater. Sci.: Mater. Med., 2017, 28, 109.
88. M. Sivakumar, A. K. Vivekanandan, G. Panomsuwan, V. Veeramani, S. H. Chen, Z. Jiang and T. Maiyalagan, Mater. Today Chem., 2022, 26, 101156.
89. C. Zhu, Y. Bing, Q. Chen, B. Pang, J. Li and T. Zhang, ACS Appl. Mater. Interfaces, 2023, 15, 19384–19392.
90. Y. Tutel, S. Koylan, S. Tunca and H. E. Unalan, ACS Appl. Nano Mater., 2021, 4, 13871–13883.
91. F. Cui, H. Sun, X. Yang, H. Zhou, Y. Wu, J. Li, H. Li, J. Liu, C. Zeng, B. Qu, J. Zhang and Q. Zhou, Chem. Eng. J., 2023, 457, 141303.
92. Y. Huang, Y. Han, J. Sun, Y. Zhang and L. Han, Mater. Today Chem., 2022, 24, 100895.
93. X. Hui, M. Sharifuzzaman, S. Sharma, C. I. Park, S. Yoon, D. H. Kim and J. Y. Park, Anal. Chim. Acta, 2022, 1209, 339872.
94. T. Pinheiro, J. Caetano, E. Fortunato, M. G. F. Sales, H. Almeida and R. Martins, Adv. Sens. Res., 2024, 3, 2400052.
95. A. C. Marques, A. R. Cardoso, R. Martins, M. G. F. Sales and E. Fortunato, ACS Appl. Nano Mater., 2020, 3, 2795–2803.
96. M. Bahri, M. Amin Elaguech, S. Nasraoui, K. Djebbi, O. Kanoun, P. Qin, C. Tlili and D. Wang, Microchem. J., 2023, 185, 108208.
97. J. Guo, M. Zhao, C. Chen, F. Wang and Z. Chen, Analyst, 2023, 149, 137–147.
98. D. X. Luong, A. K. Subramanian, G. A. L. Silva, J. Yoon, S. Cofer, K. Yang, P. S. Owuor, T. Wang, Z. Wang, J. Lou, P. M. Ajayan and J. M. Tour, Adv. Mater., 2018, 30, 1707416.
99. Y. Chang, Q. Cao and B. J. Venton, Curr. Opin. Electrochem., 2023, 38, 101228.
100. S. Duan, B. Wang, Y. Lin, Y. Li, D. Zhu, J. Wu, J. Xia, W. Lei and B. Wang, Adv. Intell. Syst., 2021, 3, 2100056.

---