Introduction

Rising levels of environmental pollution, such as water, air, and soil pollution along with the emergence of several critical diseases highlight the urgency of developing highly efficient portable sensing devices. These devices are essential for frequent, rapid testing of key environmental factors and continuous monitoring of patients with critical diseases1,2,3,4,5. In recent years, the researchers have made tremendous strides in advances of sensing techniques for environmental testing, such as soil testing, air quality monitoring, and water pollution detection3,6,7. Notably, significant research efforts have been dedicated to developing biosensors, including wearable biosensors for healthcare monitoring, and their applications in point-of-care (POC) testing8,9,10,11,12. However, the advancements and large-scale production of portable devices have elevated use of conventional energy resources, for example rechargeable batteries and non-rechargeable batteries13,14,15. This heightened reliance on conventional energy resources negatively impacts the environment and human health, make sensors bulky, and brings safety issues16,17.

Triboelectric nanogenerators (TENGs) and triboelectric nanosensors (TENSs) have demonstrated great potential in harnessing the abundantly available low-frequency mechanical energy and for self-powered sensing applications18,19,20,21,22,23. Among TENGs, SL-TENGs offer advantages, such as highly efficient contact-separation because of lubricant nature of the contact liquid and wide range availability of liquid motion24. Moreover, SL-TENGs are advantageous for biosensing applications, including bacteria, drug, and chemical sensing, such as heavy metal ion sensing, pollutant sensing due to the presence of real samples in liquid form which make the SL-TENG as an ideal candidate for sensing purposes25,26,27,28,29,30. Considering above-mentioned advantages, the researchers have made significant efforts to the development of self-powered biosensors and chemical sensors based on SL-TENGs31. The developed SL-TENGs utilize different operating modes, including sea wave mode, droplet mode, flow mode, and vertical contact-separation mode32,33,34,35. To enhance the triboelectric performance of SL-TENGs, several approaches have been employed, for instance optical, electrical, physical, and chemical modification of contact solid and liquid surfaces. These modifications enhance important parameters, such as hydrophobicity, hydrophilicity, surface potential, effective contact area, and intimacy in contact-separation. These developments in SL-TENG have been employed for a wide range of sensing applications36,37,38,39,40.

Song et al. have recently introduced a wearable, wireless, and battery-free device for on-body sweat sensing that is powered by human motion. The power extraction from human motion was based on flexible printed circuit board-based freestanding TENG41. In a separate study, a self-powered active urea sensor was designed for smart agriculture using SL-TENG. The proposed sensor monitors concentration of urea during crop growth without interference from common fertilizers42. Additionally, Barman et al. have reported a self-powered SL-TENG-based heavy metal ion detection system, integrated with a robot hand. In this sensing system, tellurium nanowires (Te NWs) are grown on a copper sheet for the on-site detection of Hg2+ ions in polluted water in real environment, with the integration of a robot hand43.

In this review article, we have reviewed various operating modes of SL-TENGs along with their working mechanisms as reported by the researchers since the invention of SL-TENG. In the following section, we have presented a comprehensive analysis of the modifications made to solid contact surfaces to functionalize the detection of targeted objects. Moreover, a discussion is provided about the changes in surface properties resulting from these modifications and their corresponding characterizations. In the next section, we discuss the changes in electrical characterization-including voltage, current, and transfer charge outputs of SL-TENGs, following the modification of contact surfaces for sensing purposes. Finally, we detail the development of self-powered chemical sensors for heavy metal ion sensing, microplastic identification, and self-powered biosensors for bacteria and drug detections. The overview of this review is illustrated in Fig. 1, including represntation of the major sections. This review will greatly benefit the researchers working in field of self-powered sensors with the guidance on selecting suitable operating mode, idea of surface modification, and implementation in possible practical applications.

Fig. 1: Overview illustration of this review with sensing applications.
figure 1

Depicting TENGs mechanism, surface modification, surface phenomena, application in chemical sensing, and biosensing. [Reprinted figure with permission from refs. 43,57,58,63. Copyright (2021) Elsevier; (2023) American chemical society; (2023) Wiley-VCH; (2024) Elsevier].

Full size image

Working mechanism of different mode of SL-TENGs

Flow mode of SL-TENG

Solid-Liquid contact electrification (CE) is a process where electrical charges are generated on a liquid and solid material surface, when they come into contact. This process involves figuring out the quantity and nature of electric charges generated on the materials and how the properties of both liquid and solid materials affect this triboelectric charge generation. CE is also known as triboelectrification, a familiar phenomenon happens when two materials come into contact44. It has been showcased in various ways-including rubbing a plastic comb with a silk fabric, X-ray generation, xerography, and electrostatic separation45. Initially, it was believed, ions46 or water films acted as the electric carriers in solid-liquid CE47,48. However, recent finding suggest that water might not always be essential for this process to occur. The triboelectric microfluidic nanosensor is based on the combined effect of CE and electrostatic induction (Fig. 2a(i–iii)). In which, when the movement of the fluid takes place; it will drive electron flow between the Cu electrode and ground through external electric circuit. As shown in Fig. 2a(iv) when a water flow applied on the Kapton layer, it imparts negative charges on Kapton’s surface, while positive charges on the water layer. This positive charge generation on the water layer has created a charge imbalance in the system and an electron flow was driven from ground to the Cu electrode due to electrostatic phenomenon to maintain electroneutrality49. There will be no electron flow once the electrostatic equilibrium state is achieved (Fig. 2a(v)). Due to the continuous flow of the fluid, this mode can be named as flow mode of SL-TENG.

Fig. 2: Illustration of working mechanism for different modes of SL-TENG.
figure 2

Working mechanism of SL-TENG based on (a) flow mode of contact liquid [Reprinted figure with permission from ref. 49. Copyright (2015) American chemical society], b sea wave movement of contact liquid [Reprinted figure with permission from ref. 50. Copyright (2020) Elsevier], c vertical contact separation mode [Reprinted figure with permission from ref. 51. Copyright (2013) Wiley-VCH], d conventional droplet mode [Reprinted figure with permission from ref. 52. Copyright (2014) Wiley-VCH], and e droplet mode with switching effect [Reprinted figure with permission from ref. 55. Copyright (2022) Elsevier].

Full size image

Sea wave mode

In this mode of TENG, contact liquids’ motion is quite like motion of sea waves33. In a study of the effect of temperature, ionic concentration, and pH values on CE, sea wave mode of SL-TENG was utilized. The working mechanism of the SL-TENG, which operates based on CE at the solid-liquid interface in sea wave mode, as illustrated in Fig. 2b. Triboelectric charges were produced on the SiO2 substrate as water or oil slid over the solid surface (towards right side) (Fig. 2b(i, ii)). The SiO2 substrate was negatively charged, while the water became positively charged due to the electron transfer ability, resulted in an electron flow through the external electronic circuit. When the substrate was covered by a water film without separation, charges were shielded by the water, forming an electric double layer (EDL). Upon separation (when liquid moving towards left side), which create a opposite potential difference and resulting in an electron flow in opposite direction through the external circuit (Fig. 2b(iii, iv))50.

Vertical contact-separation mode

In this mode, one of the solid surface or contact liquid moving vertically for contact-separation to the corresponding contact surface, as shown in Fig. 2c. Initially, no charge occurred when there is no contact-separation (Fig. 2c(i)). Upon contact, PDMS gain negative charge on its surface while water gain positive charge on it (Fig. 2c(ii)). As PDMS separates from water, a potential difference occurs between them, facilitating electron flow in external circuit as shown in Fig. 2c(iii). In short-circuit case, electrons flow from Cu electrode 2 to Cu electrode 1, generating an instantaneous positive current. Upon PDMS recontacting water, electrons flow in the opposite direction generates a negative current, thus ensuring continuous output signal (Fig. 2c(v))51. Hydrophobic polymers with surface micro-pattern due to higher contact area enhanced the triboelectric effect.

Conventional droplet mode

This mode consists of electrodes only on the back side of solid triboelectric surface. As a water droplet contacts a polytetrafluoroethylene (PTFE) surface, it generates charges on water droplet and PTFE (Fig. 2d). Upon contact with a PTFE film, the ionization of surface groups of PTFE creates negatives charge on it and positive charges on surface of the water droplet to maintain electrical neutrality (Fig. 2d(iii)). When the water droplet leaves the PTFE thin film, a negative electric potential difference is established between the Cu electrode and ground, derive an instantaneous negative current (Fig. 2d(iv)). Subsequently, upon contact with following water drop, the negative charges on the PTFE attract counter ions from water droplet, forming an EDL and create a positive electric potential difference as shown in Fig. 2d(vi). This leads to an instantaneous positive current flow from ground to the Cu electrode until a new equilibrium is reached52.

Switching droplet mode

In switching droplet mode, triboelectric signal generation is based on spreading and shrinking of liquid droplet due to hydrophobicity of contact surface and change in interface area between droplet and top electrode53,54. As shown in Fig. 2(e), when a deionized water (DI water) droplet slides on a fluorinated ethylene propylene (FEP) film, water droplet and FEP film will get positive and negative charges on their surfaces, respectively. Upon droplet leaving the film, the FEP surface and bottom electrode carry equal and opposite charges. Due to continuous dripping of water droplets, FEP surface will achieve a saturation level of negative charge due to the charge accumulation. After the charge saturation, when the next droplet approaches towards FEP surface (Fig. 2e(i)), water droplets become polarized, where positive bound charges attract towards FEP side and negative bound charges towards air interface (Fig. 2e(ii)), this state known as “switch-off”. When the droplet spreads and touches the top electrode, an EDL forms between the bound negative charges on the droplet and bound positive charge on the top electrode as shown in Fig. 2e(iii) and this state is known as “switch-on”. As per the steady-state conditions, where the surface potential of top electrode and water droplet should be equal, led to the flow of induced positive charge on the top electrode and bound negative charge on droplet toward both sides of the EDL, resulting in a weaker attraction of positive charge on the bottom electrode, causing a current generation flow through the load into the ground55. Since, the contact area between the water droplet and top electrode increased firstly and then decreased. Due to which, in open circuit voltage first peak is positive and second peak is negative, which generate an alternating voltage output.

Modification of contact surface and contact liquid for sensing purposes

This section explores the modification of contact surfaces and contact liquids of SL-TENGs to functionalize the chemical sensing and biosensing. In a study, a water-soluble guanidinium- functionalized poly(norbornene) polymer (2Gdm) was utilized for bacterial detection and bacteria killing (Fig. 3a(i)). This choice capitalized on the favorable interaction between guanidinium groups (positively charged) and bacterial cell walls (negatively charged phospholipid bilayer). 2Gdm layer was further modified to X-2Gdm thin films via thermal cross-linking at 100 °C for 12 h and this cross-linking was allowed due to the presence of double bond in poly(norbornene) backbone. These self-powered micro biosensors consist of an X-2Gdm layer on ITO/PET and copper sheets within TENG and TENS setups as shown in Fig. 3a(ii). The interaction of positively charged guanidinium functional groups in the X-2Gdm film with negatively charged phosphate groups in the phospholipid bilayer of E. coli (Gram-positive, G+) and S. pneumoniae (Gram-negative, G−) leads to bacterial death and adhesion on the X-2Gdm surfaces. Figure 3a(iii) shows the correlation between surface potential of X-2Gdm and concentration of E. coli and S. pneumoniae, where a consistent decrease in surface potential can be observed when concentration increases. Immersing X-2Gdm in different concentrations of E. coli and S. pneumoniae directly influences the surface potential, subsequently impacting TENG output56.

Fig. 3: Modification of triboelectric surfaces for sensing applications.
figure 3

a A guanidinium functionalized poly(norbornene)-based polymer sheet for sensing of gram-positive and gram-negative bacteria. [Reprinted figure with permission from ref. 56. Copyright (2024) American chemical society], b Modification of Au NPs by D-mannose to functionalize the detection of E-coli based on carbohydrate-protein interactions. [Reprinted figure with permission from ref. 57. Copyright (2021) Elsevier], c Synthesized Te NWs with two kind of surface property for the detection of toxic Hg ions in the matrix of acetone and DI water. [Reprinted figure with permission from ref. 43. Copyright (2023) American chemical society], d Fabrication of a microfluidic device for the platelet level monitoring by trapping platelets on collagen layer. [Reprinted figure with permission from ref. 58. Copyright (2024) Elsevier].

Full size image

In another study, a label-free and self-powered sensing method is developed by utilizing gold nanoparticles (Au NPs) based on transcutaneous electrical nerve stimulation for E. coli detection. Leveraging the specific interaction between D-mannose (carbohydrate) and protein, this approach enables the detection of E. coli ORN178 strain. Since D-mannose-thiol does not directly bind to Cu atoms, Au NPs were initially coated onto copper wire as shown in Fig. 3b(i). Thereafter, the Au NPs are modified by D-mannose-thiol (m-Au NPs), where Au NPs enable Au-SH bond in presence of D-mannose-thiol. The targeted binding between carbohydrate and lectins makes m-Au NPs a highly effective sensing platform for detecting specific proteins and their associated biomolecules. The adherence of E. coli ORN178 on m-Au NPs is visibly evident as bright spots (indicated by arrow) in the Kelvin probe force microscopy (KPFM) image (Fig. 3b(ii)), denoting a significant shift in surface potential57. The KPFM images for 10 nM–0.1 mM concentration of concanavalin A (Con A) are distinctly illustrating that the surface potential escalates from 230 to 380 mV as depicted in Fig. 3b(iii), (iv). This elevated surface potential signifies an increased potential difference between the solid triboelectric layer and the contacting liquid, consequently leading to an elevation in triboelectric voltage output.

In another study, Berman et al. introduced a self-sustaining mercury ion sensor utilizing Te NWs as a solid contact material to functionalize the sensing43. Additionally, the sensor demonstrated selective detection of Hg2+ ions due to the robust binding affinity between Te NWs and Hg2+ ions. The impact of surface hydrophobicity on sensing performance was studied by measuring water contact angles (WCA) of hydrophobic and hydrophilic Te NWs under variation of Hg2+ ion concentration. Results showed a significant shift from hydrophobic to hydrophilic behavior with increasing Hg2+ ion concentrations for hydrophobic Te NWs (Fig. 3c(i)). This shift indicates surface hydrophobicity as a critical parameter in SL-TENS sensing, however for hydrophilic surfaces the changes in WCA are negligible as shown in Fig. 3c(ii). Figure 3c(iii) illustrates the electron transfer direction during CE between Te NWs, DI water, and acetone. CE of Te NWs with acetone and DI water led to a sudden rise in surface potential from −120 to 369 mV and −390 to −101 mV when react with Hg2+ ion concentrations of 0 nM–1 µM, respectively (Fig. 3c(iv)). However, the work function decreased to 4.35 eV as Hg2+ ions concentration increased from 0 nM–1 µM on Te NW surfaces after contact-separation with acetone (Fig. 3c(v))43. Moreover, in Fig. 3c(vi) the KPFM images are showing change in surface potential of Te NWs upon change in concentration of Hg2+ ions.

A POC device called the triboelectric microfluidic nanosensor has been developed for real-time platelet level quantification. It utilizes liquid-solid CE and charge induction to monitor the flow resistance in a microfluidic channel, primarily reflecting interactions between platelet- containing plasma and a collagen-coated surface inside microfluidic channel. Elevated flow resistance indicates platelet entrapment by the collagen layer. Figure 3d(i) illustrates the mechanism of flow resistance functionalization in the microfluidic channel due to the activation of platelets and platelet- collagen interactions. Platelets develop surface spikes upon encountering collagen due to receptors on its surface, enhancing binding and functionalizing flow resistance within channel. Platelet quantification sensing data depicted a consistent 10 mV decrease in output voltage difference at terminal electrodes with increasing platelet concentrations from 0% (platelet poor plasma, PPP) to 100% (platelet rich plasma, PRP), as shown in Fig. 3d(ii). The fluorescence images of the inner wall of the microfluidic channel with and without collagen coating, after a flow of real blood samples, are shown in Fig. 3d(iii, iv). The fluorescence image for pure PRP solution is depicted in Fig. 3d(v), where the green bright spots represent platelets. These fluorescence images are validating the trapping of platelets on the collagen layer in the microfluidic channel. Surface morphology of the inner wall of the microfluidic channel during fabrication process was examined using FESEM images. Figure 3d(vi) shows a FESEM image of the clean glass substrate, while Fig. 3d(vii) displays the fibrous structure of the collagen layer coated inside the microfluidic channel. Platelet binding onto the collagen layer is evident in the FESEM image presented in Fig. 3d(viii), following plasma solution flow through Cu/CuO NWs tubes for accessing flow resistance58.

Surface phenomena during chemical sensing and biosensing

The modified surfaces for chemical sensing and biosensing are further electrically characterized by measuring triboelectric current, voltage, and transfer charge outputs to realize the reflection of surface modification on sensing signals. In a study of wearable device for sweat analytes monitoring and human motion sensing, a fiber-based triboelectric nanogenerator (F-TENG) is fabricated59. A stretchable conductive fiber Ecoflex/multiwalled carbon nanotubes/polyaniline (PANI) was modified by glucose oxidase (GOX), creatininase/creatinase)/sarcosine oxidase (CA/CI/SO), and lactate oxidase (LOX) for the detection of glucose, creatinine, and lactate, respectively. As shown in Fig. 4a(i), when glucose, creatinine, and lactate encounter GOX, CA/CI/SO, and LOX, H2O2 production takes place. However, H2O2 is unstable which leads to production of H+ and e−60. Therefore, this charge generation resulted in elevation of triboelectric charge density, which leads to enhancement in triboelectric current output. The sensing of the above-mentioned analytes from the F-TENG is further validated by comparing its sensing data with commercial meters. Where it is claimed that the sensing data for glucose and creatinine are comparable to the data from commercial device, however, the sensing data for lactate is not following the data from commercial device which is possibly due to impurity in sweat and less sensitivity of selected materials (Fig. 4a(ii)). This sensing capability is also can be observed in selectivity data depicted in Fig. 4a(iii).

Fig. 4: Surface phenomena upon contact with target analytes and corresponding sensing signals.
figure 4

a Sensing mechanism for glucose, creatinine, and lactate upon the surface modification and its sensing performance. [Reprinted figure with permission from ref. 59. Copyright (2022) Elsevier], b A triboelectric microfluidic (TEMF) device for reduction of 4-nitrophenol (4-NP) and its monitoring. [Reprinted figure with permission from ref. 61. Copyright (2021) Elsevier], c The effect of different liquid molecules (water, ethanol, and hexane) on the FEP interface and change in charge density. [Reprinted figure with permission from ref. 62. Copyright (2019) American chemical society], d Demonstrating electric potential maps of different molecules, formation of H-bond with sucrose, and change in the EDL at DI and FEP interface with and without sucrose. [Reprinted figure with permission from ref. 63. Copyright (2023) Wiley-VCH].

Full size image

In another study, a POC device is fabricated with the combination of triboelectric effect and microfluidics. In this study, a PDMS-based microfluidic device is fabricated for monitoring the degradation of an industrial pollutant, 4-nitrophenol (4-NP), through catalytic process61. As shown in Fig. 4b(i), the catalytic reaction takes place inside the microfluidic channel, where a mixture of NaBH4, Au NPs, and 4-NP, introduced inside the microfluidic channel. In this process, NaBH4 works as reductant and Au NPs are working as catalytic agent. Initially, when the solution introduced inside the microfluidic device, the measured triboelectric voltage output is low because of presence of Na+ ions in the solution and they interact with PDMS because of its negative surface potential. By the time, when reduction of 4-NP takes place and degraded into 4-aminophenol (4-AP), it binds with PDMS stronger than Na+ ions, which leads to make more negative surface potential of PDMS and resulted in higher triboelectric voltage output as shown in Fig. 4b(ii).

In view of chemical sensing, Want et al. have proposed a ring shape device based on FEP and copper as shown in Fig. 4c(i). In this study, effect of the inherent properties of various liquids (hexane (1), isopropanol (2), ethanol (3), acetone (4), ethylene glycol (5), and water (6) on triboelectric voltage output have been analyzed62. The triboelectric voltage outputs with different liquids were analyzed by studying their macroscopic properties (interface friction/contact angle) and microscopic properties (polarity) as shown in Fig. 4c(ii). It is observed that voltage output of acetone, ethylene glycol, and water is much higher because of their high contact angles, whereas voltage output for ethanol, hexane, and isopropanol are lower because of their lower contact angles. In the study of density function theory (DFT), it found that the interface distance in equilibrium states for water, ethanol, and hexane are positively related with the polarity and the charge density at the FEP-liquid interface is also positively related with the triboelectric output as shown in Fig. 4c(iii). Liu et al. has reported a self-powered sucrose solution concentration sensing, where changes in the fluid mechanics of the solution due to changes in sucrose concentration. In Fig. 4d(i) the charge density of sucrose and water molecule is shown, calculated from DFT, where it can be observed that the sucrose molecules’ potential is higher than water molecules’ potential. This difference in potential leads to the formation of H-bond between sucrose and water molecules and destroys the H-bond network between water molecules, as shown in Fig. 4d(ii), (iii). This phenomenon will alter the charge density at the interface of FEP and sensing solution, which leads to less charge transfer as the concentration of sucrose increases, as shown in Fig. 4d(iv)63. This analysis was further validated by the measurement of triboelectric voltage output, where it is observed that the voltage output is decreases as concentration of sucrose increases, as shown in Fig. 4d(v).

SL-TENG based applications for chemical sensing

SL-TENG have shown its implications for several type of chemical sensing which includes sensing of enzymes41, pesticides, heavy metal ions43, and microplastic particles64. Huang et al. has reported a self-powered system for recognition of type of microplastic and its quantitative detection based on integration of SL-TENG and deep learning model. In this study, a FEP-based SL-TENG is designed to detect the quantity and recognition of five type of microparticles polyethylene (PE), polypropylene (PP), polyvinyl chloride (PVC), polyethylene terephthalate (PET), and polystyrene (PS)64. As shown in Fig. 5a(i)–(iii), the quantitative detection is performed in deionized water with fraction 0.025%, 0.1%, 0.175%, and 0.25% of microplastics and it is observed that voltage output increases when microplastics’ quantity increases. For the rapid and accurate recognition of microplastics based on its waveform and concentration detection, a deep learning model is constructed named as (1-dimensional convolutional neural networks) CNN-1D and for this model PET and PS are selected with fractions 0.025%, 0.1%, and 0.25% and labeled from 1-6, respectively, as shown in Fig. 5a(iv), (v). Finally, it is observed the integration of sensing data from SL-TENG and the CNN-1D model has shown the average recognition efficiency around 86.7% (for PS: 100% and PET: 82%). In another study of the development of self-powered chemical sensor for detection of different kind of solutions, along with the detection of the flow of fluid. Proposed liquid-metal-based triboelectric nanogenerator (LM-TENG), utilized Galinstan as working electrode encapsulated with PDMS, has shown superior performance than that of copper-electrode-based TENG65. The layered structure of the fabricated horizontal LM-TENG is shown in Fig. 5b(i), where the thickness of the PDMS encapsulation layer is 200 µm and optimized microfluidic channel width is 2 mm. A syringe pump is used to flow the liquid through the microfluidic channel for identification and flow sensing of the liquid as shown in Fig. 5b(ii). The fabricated LM-TENG is further tested for three different liquids (tap water, DI water, and ethanol), where it can be observed that the values of measured current and voltage are quite different as shown in Fig. 5b(iii). This difference is reflected the potential of this device for liquid recognition as well as for the flow sensing as shown in Fig. 5b(iv).

Fig. 5: Chemical sensing applications based on SL-TENG.
figure 5

a Microplastic detection and identification of type of microplastic by deep learning. [Reprinted figure with permission from ref. 64. Copyright (2023) American chemical society], b SL-TENG device for the detection of different kind of liquids with integration of microfluidics. [Reprinted figure with permission from ref. 65. Copyright (2023) Wiley-VCH], c Self-powered flow sensing device for biofluids in hospitals. [Reprinted figure with permission from ref. 66. Copyright (2020) American chemical society], d Dual signal self-powered dopamine concentration detection device based on SL-TENG. [Reprinted figure with permission from ref. 67. Copyright (2019) Wiley-VCH], e Detection of organometallic compound and different ration of organic solvents in mixture solution based on SL-TENG. [Reprinted figure with permission from ref. 68. Copyright (2021) American chemical society].

Full size image

To develop a self-powered sensor for medical applications based on SL-TENG, a tube-shaped SL-TENG with circular cross-section is designed to monitor the clinical drainage operation66. In this study, a superhydrophobic layer of silica nanoparticles has been employed as solid triboelectric layer and various kind of liquids (HCL, urine, blood, NaOH, PBS) have been used to demonstrate the potential of fabricated device for clinical IOT applications. As shown in Fig. 5c(i), the fabricated triboelectric layer along with trimmed copper electrodes is attached on the inner wall of a round silicone rubber tube. This silicone tube is further connected to another silicone tube with smaller diameter. Thereafter, this small diameter silicone tube enables the flow of testing liquid in form of droplets through the tubular drop counter. This flow of droplets through the drop counter generated pulsed current output as shown in Fig. 5c(ii), where it can be observed that current outputs are different for the different liquid droplets. These results are showing the potential of fabricated droplet counter in clinical monitoring purposes as illustrated in Fig. 5c(iii). In another study, Jiang et al. has reported a self-powered dual signal detection of dopamine based on SL-TENG which is based on a PTFE film, a copper electrode and glass substrate67. In this study, the fabricated sensor has demonstrated two kind of signals, one (ITENG) is from CE and electrostatic induction between liquid (oil/water) and PTFE film whereas other signal (Iinterface) is from electrostatic induction on copper electrode due to change in oil/water interface when the whole device introduced across the oil/water interface. As shown in Fig. 5d(i), the change in contact angle has been studied upon modification with polydopamine (PDA) of concentration range 0–500 µM L−1 for a glass substrate and a PTFE surface. From the contact angle study, it is observed that the contact angle decreases upon elevation in the PDA concentration due to elevation in adsorption of PDA on the surface of TENG. Furthermore, the relation between dual short-circuit current and concentration of PDA is studied, where it is observed that the ITENG is decreasing with the increment in PDAs’ concentration whereas Iinterface is increased (Fig. 5d(ii–iv)). The calculated limit of detection (LOD) for ITENG and Iinterface are 5.15 and 3.96 µM, respectively.

Moreover, Zhang et al. has reported a droplet triboelectric nanogenerator (droplet-TENG) with spatial configuration of electrodes and these electrodes facilitated as sensing probe for measurement of charge transfer phenomena68. In this work FEP was used as sensing surface and the transferred charges output data were studied for different organic solvents (ethanol, acetone, hexane, tetrahydrofuran, and benzene) to prove the dominance of electron transfer in comparison to ion transfer at sold-liquid interface. As shown in Fig. 5e(i, ii), during the contact- separation of organic solutions’ droplet (absence of ions) with FEP surface, the transferred charges are q1 and q2 to the electrode 1 (copper foil on back side of FEP, top side of the sensor) and electrode 2 (copper foil on back side of FEP, bottom side of the sensor), respectively. These results in Fig. 5e(ii) proved that electron transfer has a dominant role instead of ion transfer at solid-liquid interface. The different charge transfer outputs with different organic solvents are also showing the potential of droplet-TENG towards chemical sensitivity. Furthermore, effect of change in concentration of an organometallic compound (ferrocene) in ethanol on transferred charges generation has studied and it is observed that the output was decreased with the increment in concentration of ferrocene as shown in Fig. 5e(iii). Afterward, the effect of change in proportion of the solute, in organic solvent, on the transferred charges q1, q2, and its difference (Δq = q2 − q1) was studied, as shown in Fig. 5e(iv). It can be observed that the transferred charge output was increases first till 25% of ethanol because of intermolecular hydrogen-bond formation between acetone-ethanol molecules. However, beyond 25% of ethanol, the output is decreased due to hydrogen-bond formation between ethanol molecules68.

SL-TENG based applications for biosensing

SL-TENGs have also demonstrated significant progress in biosensing applications in recent years. For example, Chatterjee et al. has designed a TENS utilizing solid-liquid CE for detection of catechin and developed a method to enhance the efficiency of solid-liquid TENGs69. Specifically, TiO2 nanosheet arrays were used as the solid triboelectric material, while various solvents (water, ethanol, and acetone) act as contact liquids to demonstrate solid-liquid CE for mechanical energy harvesting and the catechin detection. As depicted in Fig. 6a(i, ii), with the increasing catechin concentration, the output voltage response of TENS gradually rises, peaking at 1.2 V for a concentration of 80 μM. The TENS showed better chemically enhanced triboelectric performance for acetone as compared to water and ethanol, as proved by highest voltage enhancement factor corresponding to acetone (Fig. 6a(iii)). The ultraviolet photoelectron spectroscopy (UPS) analysis showed that the work function of the TiO2 nanosheet arrays of the TENS decreased from 6.54 eV to 5.52 eV after modification with catechin (Fig. 6a(iv,v)). In another study, Gao et al. developed a novel TENG-based self-powered active urea sensor42. This sensor demonstrated exceptional sensitivity and specificity by leveraging a dual-electrode structure to enhance charge utilization through a volume effect and by utilizing an enzyme-catalyzed reaction to enhance the specificity of urea detection. The device has successfully monitored variations in urea concentration during crop growth, detecting concentrations as low as 4 μM, and showing resilience to common fertilizers like potassium chloride or ammonium dihydrogen phosphate. Figure 6b(i) shows stems and roots of beans cultured in different concentrations of urea solution for different time periods. The sensor was used to monitor nutrient solutions over the time during cultivation periods, while accuracy in urea concentration analysis was verified through HPLC. Figure 6b(ii, iii) demonstrates similar trends in urea concentration from HPLC and voltage variations from the sensor over incubation time. As shown in Fig. 6b(iv), the selectivity test of the device was performed by testing several salts commonly found in culture media. It was observed that urease addition in urea solution notably boosts the output voltage during device operation, a trend not observed with other solutions.

Fig. 6: Applications of biosensing based on SL-TENG.
figure 6

a Detection of catechin in real black tea samples based on conventional vertical contact-separation mode SL-TENG. [Reprinted figure with permission from ref. 69. Copyright (2020) Elsevier], b Monitoring of concentration of urea solution during the crop cultivation from day to day. [Reprinted figure with permission from ref. 42 Copyright (2024) Wiley-VCH], c Detection of anti-tumor drugs loading on protein based on SL-TENG. [Reprinted figure with permission from ref. 70 Copyright (2024) Elsevier], d Monitoring of total aerobic counts for a commercial water container on different time stages with measurement of triboelectric measurements. [Reprinted figure with permission from ref. 71 Copyright (2018) Wiley-VCH].

Full size image

In view of applications for clinical usage, an innovative label-free sensing platform was developed that can facilitate high-throughput screening for various protein targets70. The platform measured the affinity of molecule-molecule interactions by detecting voltage output variations, achieving unprecedented sensitivity and selectivity. It combined computational analysis with experimental validation using a solid-liquid TENS. Using FKBP-rapamycin as a model, the platform demonstrated high sensitivity. The study further explored drug interactions with the oncogenic protein ATG4B, identifying the binding of S130 and Tioconazole to ATG4B and showing that dexamethasone does not bind with ATG4B. As demonstrated in Fig. 6c(i)–(iii), with the increasing concentration of Tioconazole and S130, a positive correlation was obtained with the measured voltage output. However, dexamethasone exhibited no variation in voltage output with changing concentrations, indicating a lack of interaction with ATG4B. To clearly understand the voltage response observed, alterations in surface potential following the binding of various drugs to ATG4B were assessed utilizing KPFM, as depicted in Fig. 6c(iv). Further, an LC3-GST-based enzyme activity assay for purified ATG4B confirmed the inhibitory effects of S130 and Tioconazole, with 20 µM of S130 suppressing approximately 14% and 20 µM Tioconazole displaying around 31% cleavage suppression of LC3-GST, while dexamethasone shows no inhibitory effects (Fig. 6c(v)). In another study, Chen et al. has developed a microfluidic sensor based on capillary-tube triboelectric nanogenerator (ct-TENG) which comprised a highly flexible structure with an ultrafine tubular sandwich design71. This device enabled nondestructive and highly flexible microliter sampling (0.5 µL) for micro liquid sensing, a first in the field. The self-powered ct-TENG produced selectable electrical signals suitable for sensing micro liquid volumes, utilizing Maxwell’s displacement current generated from micro liquid flow. The ct-TENG was used for monitoring of total aerobic count (TAC) (cfu mL−1) in barreled water. Figure 6d(i) shows drinking scenario of barreled water, the TAC in water exhibits exponential growth as the duration of drinking time increases. After 7 days, TAC exceeds 110 cfu mL−1, surpassing water quality standards, as depicted by the varying appearance of TAC with different time in Fig. 6d(ii). Output electrical signals of the ct-TENG at different drinking time is shown in Fig. 6d(iii). The response of the ct-TENG output signals to changes in total dissolved solids and electrical conductivity (κ) in water over a period of 0 to 60 days suggested a decrease in Isc, Voc, and Qsc. This decline is attributed to the increase in TAC over prolonged drinking time. Figure 6d(iv) illustrates the optimal drinking intervals for barreled water monitored by the ct-TENG, categorized into four intervals based on two sets of output signals: Set A (191 nA, 3.1 nC, 6.3 V) and Set B (174 nA, 2.4 nC, and 6 V). These intervals include: (1) directly drinkable water, (2) water requiring heating before consumption, (3) water not recommended for drinking, and (4) water deemed completely unfit for consumption.

Conclusion and future perspectives

In summary, this review article discusses about the recent advancements in biosensing, and chemical sensing applications based on SL-TENGs. Initially, the different modes of SL-TENGs: droplet mode, sea wave motion mode, liquid flow mode, and conventional vertical contact separation mode along with their working mechanisms, have been described. Further, the surface modifications for the detection of targeted analytes are described along with the changes in surface properties upon modification. Following this section, we have described the reflection of this surface modification on the triboelectric output of the fabricated SL-TENG device. Finally, in view of applications, development of chemical and biosensors based on SL-TENG has been described like monitoring of urea concentration during crop cultivation, anti-tumor drug loading detection, catechin detection in real samples, heavy metal ion sensing, etc.

In view of future perspectives, some of the studies reported in this review for chemical sensing are showing the sensing data with different chemicals as contact liquid. However, to understand this phenomenon in a more scientific way, the researchers need to come up with appropriate material characterization and mechanism at solid-liquid interface at molecular level. Moreover, the sensing surfaces for chemical sensing can be modified for the specific detection of substances which can be utilized for specific purposes like detection of hazardous chemicals in challenging environments with the integration of robotic technology. Moreover, with the integration of multiple electrodes along with corresponding modifications for different analytes/targets can be achieved for the detection of multiple analytes with high selectivity. On the other hand, in this review one study is reported on anti-tumor drug screening, where can be concluded that the SL-TENG can be developed for the drug screening for other sever dieses like COVID19, cancer, neurological dieses. Moreover, the SL-TENG can also be developed for biosensing and chemical sensing by integrating with microfluidic technology, where the properties of the sensing surfaces like hydrophobicity, underwater oleophobicity can be a key factor of this technology. This review article will provide a pathway to the researchers’ understanding of current state of the art of SL-TENG in sensing era and significant directions for further advancement of SL-TENG based sensing devices.