Research Progresses and Application of Biofuel Cells Based on Immobilized Enzymes

: Enzymatic biofuel cells (EBFCs) are devices that use natural enzymes as catalysts to convert chemical energy from bio-sourced fuels into electrical energy. In this review, we summarize recent research progress and applications in the ﬁeld of biofuel cells based on immobilized enzymes. Specif-ically, we discuss how to optimize and improve the electrochemical performance and operational stability of enzymatic biofuel cells through enzyme immobilization materials, enzyme immobilization methods, electron transfer improvement on enzyme electrodes, and cell construction methods. We also cover current and future practical applications of biofuel cells based on immobilized enzymes, including implantable enzymatic biofuel cells and wearable enzymatic biofuel cells. Additionally, we present some of the issues that still need to be addressed in the ﬁeld of biofuel cells based on immobilized enzymes to ensure their


Introduction
Enzymatic biofuel cells (EBFCs) are electrochemical devices that use natural enzymes as catalysts to convert chemical energy in some bio-sourced fuels into electrical energy under mild conditions. EBFCs have gained significant attentions due to their potential applications in various fields, including biomedical devices, environmental monitoring, consumer electronics, military and defense, and among others, providing a sustainable and environmentally friendly alternative to traditional power sources. This technology has opened a new way for the utilization of clean energy and renewable resources, as EBFCs generally do not require expensive precious metal catalysts [1,2]. The working principle of biofuel cells is depicted in Figure 1. In the anode region, EBFCs utilize biological enzymes such as glucose oxidase (GOx), glucose dehydrogenase (GDH), fructose dehydrogenase (FDH), lactate oxidase (LOx), alcohol dehydrogenase (ADH), and alcohol oxidase (AOx), including ethanol oxidase (EOx), as catalysts to catalyze the oxidation of biofuels such as glucose, fructose, and ethanol to lose electrons [3]. The electrons then reach the cathode through the external circuit of the cell. Sometimes, metallic platinum is used as the cathode material to receive the electrons. Additionally, some biological enzymes, bilirubin oxidase (BOD), laccase (Lac) and maybe some other enzymes, are used to catalyze the reduction of oxygen [4], so that the chemical energy is converted into electrical energy. EBFCs are promising because of their fuel abundance, biocompatibility, mild reaction conditions (physiological pH, temperature, etc.), and simplicity of operation, and they can find a wide range of applications including powering biomedical and wearable devices [5][6][7].
However, EBFCs also face some significant challenges such as low output power, low open-circuit voltage, poor power stability, electron transfer barriers, easy leakage of electron transfer mediators and immobilized enzymes, and poor stability of enzymes [8]. One of the prominent problems is the activity stability of the immobilized enzymes on the electrodes [9]. Therefore, researchers use different immobilization methods, different We will briefly review and discuss the research progresses of enzymatic biofuel cells and their application prospects in terms of the selection of enzyme immobilization methods and materials related to electrode modification, the improvement of electron transfer between enzymes and electrodes, and the optimization of cell construction methods [1,2]. Finally, we will explore current and future practical applications of EBFCs, including implantable and wearable enzymatic biofuel cells, and the issues that still need to be addressed to ensure the technical and commercial viability and sustainability of EBFCs [5][6][7].

Some Nanomaterials and Nanocomposites for Enzyme Immobilization in EBFCs
EBFCs require enzyme immobilization materials with high surface area, high enzyme loading density, high adsorption durability and recyclability. In the technology of EBFCs, the enzyme immobilization material, that is, the carrier material for enzyme immobilization, is also an electrode material or electrode modification material; thus, its conductivity is also a key factor [1,2]. Enzyme immobilization materials include those carbon-based nanomaterials with good electrical conductivity, large specific surface area and good biocompatibility, allowing large enzyme loading densities and comfortable microenvironments for the enzymatic reaction. Some flexible carbon materials such as buckypaper (BP), carbon paper (CP), graphene paper (GP), carbon cloth (CC), due to their good mechanical maneuverability and portability, good electrical conductivity, and good biocompatibility, are also widely used in EBFCs. Moreover, some conductive polymers We will briefly review and discuss the research progresses of enzymatic biofuel cells and their application prospects in terms of the selection of enzyme immobilization methods and materials related to electrode modification, the improvement of electron transfer between enzymes and electrodes, and the optimization of cell construction methods [1,2]. Finally, we will explore current and future practical applications of EBFCs, including implantable and wearable enzymatic biofuel cells, and the issues that still need to be addressed to ensure the technical and commercial viability and sustainability of EBFCs [5][6][7].

Some Nanomaterials and Nanocomposites for Enzyme Immobilization in EBFCs
EBFCs require enzyme immobilization materials with high surface area, high enzyme loading density, high adsorption durability and recyclability. In the technology of EBFCs, the enzyme immobilization material, that is, the carrier material for enzyme immobilization, is also an electrode material or electrode modification material; thus, its conductivity is also a key factor [1,2]. Enzyme immobilization materials include those carbon-based nanomaterials with good electrical conductivity, large specific surface area and good biocompatibility, allowing large enzyme loading densities and comfortable microenvironments for the enzymatic reaction. Some flexible carbon materials such as buckypaper (BP), carbon paper (CP), graphene paper (GP), carbon cloth (CC), due to their good mechanical maneuverability and portability, good electrical conductivity, and good biocompatibility, are also widely used in EBFCs. Moreover, some conductive polymers such as polyaniline (PANI), polypyrrole (Ppy), polyindole (PIn), and various metal nanoparticles and inorganic nanoparticles can also improve the efficiency of EBFCs.
Carbon nanotubes (CNTs), graphene oxidase (GO), and other carbon-based nanomaterials are generally hydrophobic, while enzymes require a hydrophilic microenvironment to remain active. Functionalization and modification using water-soluble polymers, pro-teins, and hydrophilic inorganic materials can improve the hydrophilicity of carbon-based nanomaterials. Kim et al. functionalized single-walled carbon nanotubes (SWCNTs) using H 2 N-DNA to obtain DNA@SWCNTs/GOx anodes and cathodes of SWCNTs/Lac, forming single-chambered diaphragmless EBFC [11]. The EBFC had a stable output of 730-760 µW cm −2 over a week [11]. Ferritin (Frt), a widely available iron storage protein, comprises a hydrated iron oxide core with a nanometer size and a protein shell with a cage-like structure. Jeon and Kim et al. functionalized SWCNTs with Frt to obtain SWCNTs/Frt/GOx bioanodes [12]. The resulting EBFC achieved a current density of 4.6 mA cm −2 in 20 mM glucose solution. Several other research groups also used Frt as a biocompatible material for the electrode construction of EBFCs [13][14][15][16][17].
Cosnier et al. developed a bioenzymatic electrode using GOx and Catalase (Cat) as dual anodic enzymes and Lac as a cathodic enzyme, which were mechanically compressed with multi-walled carbon nanotubes (MWCNTs) to make a disc electrode [18]. The inclusion of Cat can eliminate the anodic by-product hydrogen peroxide and can resupply the anodic region with oxygen. The resulting bioenzymatic electrode of the cell showed high current densities for both glucose oxidation at the anode and oxygen reduction at the cathode, with an open-circuit voltage of 0.95 V and a maximum power density of 1.3 mW cm −2 . Under simulated physiological conditions (with 5 mM of glucose), the cell can operate stably for 1 month at 1 mW cm −2 . This simple method of compressing MWCNTs can be fully extended to other types of enzymes, which can guide the construction of electrodes for other types of biofuel cells subsequently [18]. Based on this method of making electrodes by mechanical compression, Cosnier et al. also designed a cell that utilizes both indirect and direct electron transfer, using naphthoquinone (NQ)-mediated GOx oxidation of glucose at the anode for indirect electron transfer and Lac-catalyzed direct electron transfer at the cathode for direct reduction of oxygen [19]. The battery had an open-circuit voltage of 0.76 V, a current density of 4.47 mA cm −2 , and a maximum power density of 1.54 mW cm −2 . At 0.5 V discharge, the constant output power of the cell was 0.56 mW h cm −2 . With an external charge pump, the cell successfully powered a light-emitting diode (LED), where the power could be scaled up from µW to mW, showing the potential to power electronic devices [19].
Özacar and Holzinger et al. developed a glucose-fueled photobiofuel cell based on two-dimensional graphite-phase carbon nitride (g-C 3 N 4 ) nanosheets, which are widely used in photocatalysis [20]. They first prepared different ruthenium complexes and then incorporated the ruthenium complexes on a substrate composed of MWCNTs and g-C 3 N 4 to obtain MWCNTs/g-C 3 N 4 @Ru anode. The anode was loaded with GDH as a photobioanode. BOD was used at the cathode. The resulting cell produced a maximum power density of 26.7 ± 0.4 µW cm −2 under normal conditions (no light) and 28.5 ± 0.10 µW cm −2 under light. The current density increases by 6.2% under light conditions, probably due to the use of ruthenium complexes that promoted anode electron transfer and thus increase the current density [20]. Blay and Pinyou et al. successfully immobilized GDH by electropolymerizing toluidine blue (Poly (toluidine blue), PTB) on rGO to create a GDH/PTB/rGO bioanode [21]. In addition, they prepared a GOx@HRP/MWCNTs bio-cathode by cross-linking GOx and HRP on MWCNTs. The resulting cell had an open-circuit voltage of 0.65 V in 40 mM glucose solution and a maximum power density of 31.3 µW cm −2 at 0.3 V [21]. Similarly, EBFCs with excellent performance using peroxidase elimination of by-products in the anode region were also reported [22][23][24].
Kanchi and Inamuddin et al. demonstrated the synthesis of ZnO nanoparticles by mixing neem leaf extracts with zinc nitrate solution [16]. Then, PIn was used to functionalize MWCNTs for preparing ZnO/PIn@MWCNTs/Frt/GOx electrode. The maximum current density of this electrode was 4.9 mA cm −2 [16]. Inamuddin and Ahmed et al. put SWCNTs in a mixed acid to obtain hydrophilic HOOC-SWCNTs [25]. Then, a surface defective two-dimensional nanomaterial NiMoSe 2 was prepared by hydrothermal synthesis. HOOC-SWCNTs were then mixed with NiMoSe 2 in a ferric chloride solution, thereby the pyrrole monomer was polymerized onto the mixture to obtain HOOC-SWCNTs@Ppy@NiMoSe 2 .
Similarly, the resulting HOOC-SWCNTs@Ppy@NiMoSe 2 /Frt/GOx electrode was obtained. The open-circuit potential of the resulting cell was 0.35 V and the maximum current density of 9.01 mA cm −2 was obtained in 50 mM glucose solution [25]. Hu et al. demonstrated a method for preparing a 3D carbon composite of PANI1600@CNTs by in situ polymerizing aniline monomer in MWCNTs and carbonizing it at 1600 • C. The resulting composite exhibited a rhizobia-like structure that interconnected adjacent carbon tubes, leading to a high-performance bioenzyme electrode when used for GOx and Lac [26]. The assembled cells had an open-circuit voltage of 0.78 V and a maximum power density of 1.12 mW cm −2 at 0.45 V. The three prepared cells were connected in series to successfully light up a yellow light-emitting diode with a voltage requirement of about 1.8 V. The power density was still maintained at 82.85% after two weeks of smooth operation. Composites of conductive polymers with nanomaterials are efficient loading materials for improving material conductivity for the immobilization of enzymes in the field of EBFCs or biosensors [15][16][17][25][26][27].
Kwon and Park et al. have made modifications to the reduced graphene oxide (rGO) structure to increase enzyme immobilization sites, leading to improved performance of EBFCs [28]. They used rGO with three-dimensional porous structure, prepared by spray freeze-drying method, as the substrate material for immobilized GOx. Based on this material, the bioenzyme electrode made with glutaraldehyde as a cross-linking agent had twice the loading capacity of GOx than when GO was used. The assembled cell has an open-circuit voltage of 0.85 V, and the maximum power density is 380 µW cm −2 , which is 1.4 times higher than that when using GO. Similarly, Zhou and Bai et al. prepared three-dimensional coralloid nitrogen-doped hierarchical micro-and mesoporous carbons aerogels (3D-NHCAs) as the electrode substrate for immobilization of AOx and BOD for wearable EBFCs [29].
Chen et al. developed a 3D foam material through vapor deposition using nickel foam and ethanol as the carbon source [30]. Then, the SWCNTs were directly deposited on this foam to obtain CNT@Gr material, with which GOx and Lac were immobilized, respectively. The open-circuit voltage of this cell reached about 1.2 V with a power density of 2.27 ± 0.11 mW cm −2 [30]. Ma and Zuo et al. designed a spider nest-shaped 3D substrate composed of rGO and nickel foam (Ni foam) with excellent electrical conductivity and large electroactive surface area [31]. GOx and Lac were encapsulated thereby into bioanode and biocathode, respectively ( Figure 2). The resulting cell had an open-circuit voltage of 0.7 V and a maximum power density of 7.05 ± 0.05 mW cm −2 . After 60 days of cyclic storage experiments, the performance of the cell remained 84.2% [31]. Hu and Li et al. synthesized a bamboo-shaped hollow composite of N-doped carbon nanotubes (N-CNTs) and g-C 3 N 4 nanosheets (N-CNTs@g-C 3 N 4 ) by a simple synthetic method [32]. The combination of one-dimensional N-CNTs with an open structure and two-dimensional g-C 3 N 4 nanosheets formed a three-dimensional cross-linked network on which GOx was well immobilized to produce a high open-circuit voltage of 0.93 V with a maximum power density of 0.57 mW cm −2 and the ability to harvest energy directly from various soft drinks. This composite also demonstrated the ability to harvest energy directly from various soft drinks, suggesting the promising application of g-C 3 N 4 -based nanocomposites as electrode materials [32]. These results demonstrate that the use of 3D substrates with large surface areas, special morphology and good electrical conductivity can significantly improve the performance and stability of bioelectrochemical cells. Table 1 provides an overview of several other nanomaterials and nanocomposites used in EBFCs. The use of some carbon-based nanomaterials as electrode substrates have solved more or less some bottlenecks of EBFCs: (1) the high specific surface area of nanomaterials and nanocomposites increases the enzyme attachment sites and ensures efficient consumption of biofuels; (2) the stability and biocompatibility of EBFCs can be enhanced with incorporation of carbon-based nanomaterials; (3) the unique 3D structure can shorten the distance between the enzyme active site and the electrode, which increases the electron transfer efficiency, high conductivity and thus enhances the performance of the cell; and (4) the conductive polymers can help improve material conductivity for enzymatic electrodes.  Table 1 provides an overview of several other nanomaterials and nanocomposites used in EBFCs. The use of some carbon-based nanomaterials as electrode substrates have solved more or less some bottlenecks of EBFCs: (1) the high specific surface area of nanomaterials and nanocomposites increases the enzyme attachment sites and ensures efficient consumption of biofuels; (2) the stability and biocompatibility of EBFCs can be enhanced with incorporation of carbon-based nanomaterials; (3) the unique 3D structure can shorten the distance between the enzyme active site and the electrode, which increases the electron transfer efficiency, high conductivity and thus enhances the performance of the cell; and (4) the conductive polymers can help improve material conductivity for enzymatic electrodes.

Anode
Cathode Open-Circuit Voltage

Some Paper-Based and Other Flexible Materials for Enzyme Immobilization in EBFCs
Some paper-based and other flexible electrodes have been used more and more widely due to miniaturization, low cost, mechanical flexibility and flexible processing. In order to broaden the application of enzymatic biofuel cells, Dong et al. printed MWCNTs-ionic liquid (IL) nanocomplexes on paper instead of conventional electrodes to immobilize GDH and BOD [42]. The open-circuit voltage of this cell was 0.56 V with a maximum power output of 13.5 µW cm −2 . After one week of placement, the open-circuit voltage was 0.55 V with a maximum power output of 12.2 µW cm −2 with good stability and successfully obtained energy using some commercially available soft drinks as fuel [42]. Additionally, they developed another miniature origami battery made of a GDH bioanode and a consumable MnO 2 -graphite sheet cathode, which differs from the conventional EBFCs in that it uses MnO 2 as the cathode material to be consumed to accept external electrons, rather than conventionally exhibiting an oxygen reduction reaction [43]. This work is the first attempt to introduce consumable sheet or paper materials as cathodes for EBFCs. It will provide a disposable, cost-effective and clean solution for practical green energy systems.
Buckypaper (BP), which is made by dispersing carbon nanotubes and graphene into a slurry and drying them, creates a densely packed carbon material that facilitates efficient electron transfer from the enzyme, resulting in improved cell performance. Slaughter et al. proposed a hybrid biofuel cell (HBFC) where the anode used metal to catalyze oxidation of glucose and oxygen was reduced at the cathode with immobilized BOD [44]. Platinum nanoparticles were loaded on braided gold wire at the anode, and then the BOD was cross-linked on BP at the cathode. The resulting cell had an open-circuit voltage of 0.735 V and a power density of 46.31 µW cm −2 in 3 mM of glucose. This cell can supply continuous and stable power to small electronic devices [44].
Goel et al. developed a MWCNTs paper-based enzyme cell with glutaraldehyde cross-linked electron transfer mediators Tetrathiofulvalene (TTF) and GOx at the anode and Lac at the cathode [45]. The resulting single-chamber MWCNT-based enzymatic biofuel cells with four Y-shaped paper microchannels cascaded in series had a maximum stable open-circuit potential of 1.65 V and a maximum power density of 46.4 µW cm −2 was obtained at 0.8 V (Figure 3). Comparing to cross-shaped paper microchannels, Y-shaped paper microchannels could allow more sustainable and stable microfluidics of biofuel and oxidants for various series-parallel arrangements of enzymatic anodes and cathodes in a shelf-stack configuration with maximum and stable power output [45]. The output power generated was sufficient to power small portable devices [45]. Appl  Ahn et al. developed a disposable paper-based membrane-free microfluidic EBFCs by preparing a composite slurry of cellulose ionic liquids and MWCNTs as carriers for GOX at the anode and Lac at the cathode [46]. The resulting cell achieved a better power density of 141.2 ± 3.35 µW cm −2 and current density of 615.6 ± 3.14 µA cm −2 in a Y-type paper substrate structure and it was more stable than the cross-type structure. They also discussed the effect of glucose concentration with the best cell performance at 100 mM for the Y-type and 200 mM for the cross-type. The better cell performance was obtained when the electrode was prepared with multi-walled carbon nanotubes instead of single-walled carbon nanotubes, and the enzyme was immobilized on the electrode instead of dissolved in the electrolyte [46]. Ahn et al. developed a disposable paper-based membrane-free microfluidic EBFCs by preparing a composite slurry of cellulose ionic liquids and MWCNTs as carriers for GOX at the anode and Lac at the cathode [46]. The resulting cell achieved a better power density of 141.2 ± 3.35 µW cm −2 and current density of 615.6 ± 3.14 µA cm −2 in a Y-type paper substrate structure and it was more stable than the cross-type structure. They also discussed the effect of glucose concentration with the best cell performance at 100 mM for the Y-type and 200 mM for the cross-type. The better cell performance was obtained when the electrode was prepared with multi-walled carbon nanotubes instead of single-walled carbon nanotubes, and the enzyme was immobilized on the electrode instead of dissolved in the electrolyte [46]. Shitanda et al. prepared bioelectrodes to construct EBFCs by immobilizing GOx and BOD, respectively, on screen-printed paper and successfully obtained energy from glucose [47]. The LEDs were successfully lit without the use of a boost circuit, and the output voltage of the series battery pack could be adjusted as needed. In addition, they developed another battery pack by matrix arrangement to acquire 0.97 ± 0.02 mW at 1.4 V in a 4-series/4-parallel arrangement toward wearable devices [48]. Moreover, Wang and Yang et al. successfully expanded the application of CC for EBFCs by modifying the CC surface to immobilize GOx and supply fuel to the battery in a self-pump manner [49,50].

Implantable EBFCs
Many implantable devices have been developed for medical monitoring so that vital life parameters can be acquired to alert medical professionals or device bearers for appropriate care. Implantable EBFCs can help power those implantable devices as well as the devices used outside the bodies. Flexible and biocompatible electrode materials are often required as enzyme carriers in EBFCs for implantable biomedical devices. With this in mind, the carbon materials, such as buckypaper (BP), carbon paper (CP), graphene paper (GP) and carbon cloth (CC), are often selected for EBFCs due to their good mechanical maneuverability and portability, good electrical conductivity and good biocompatibility. For example, Chi et al. employed two-dimensional graphene paper (2D-GP) as the enzyme carrier for fabricating EBFCs toward wearable and implantable devices [53]. EBFCs can convert chemical energy from molecules in living organisms into electrical energy. Glucose is the main energy supplying substance in living organisms and is widely distributed in living organisms, such as blood, tear fluid and sweat. Thanks to innovative advances in the design of nanostructured materials, implantable EBFCs based on molecular energy supply in living organisms have been developed rapidly.
Mano and Heller implanted their biofuel cell into a grape [60]. It was the first diaphragmless single-chamber biofuel cell operating in an organism to oxidize glucose with GOx and reduce oxygen with BOD, yielding an open-circuit cell voltage of 0.8 V and producing a maximum power density of 240 µW cm −2 at 0.52 V [60]. Cinquin and Cosnier et al. implanted the first glucose biofuel cell based on an independent enzyme electrode in rats [61]. The cell used insoluble ubiquinone (UQ) as an electron transfer mediator bound to GOx for the anode. Polyphenol oxidase (PPO) was used for the cathode with hydroquinone (HQ) as an electron transfer mediator. It was implanted into the retroperitoneal space of rats. The open-circuit voltage of the cell was 0.275 V with a maximum output power of 6.5 µW, and the open-circuit voltage only dropped to 0.25 V after 40 days of operation, indicating good operational stability [61].
To ensure the biocompatibility and minimal impact on living organisms, implantable EBFCs need to be carefully designed. Katz et al. used GDH as the anode enzyme and Lac as the cathode enzyme [62]. The enzymes were immobilized onto the BP electrode by cross-linking. The resulting cell was implanted into a living snail, obtaining an opencircuit voltage of 0.53 V and a maximum output power of 30 µW cm −2 . The battery was still functional after two weeks of operation in the snail under the condition that the snail survived normal feeding, demonstrating that the metabolically regenerated glucose can continuously produce electricity for the battery implanted in the living body. They also implanted the same EBFCs into clams and lobsters [63,64]. The EBFCs implanted in clams obtained an open-circuit voltage of 0.3 V~0.4 V with a maximum power density of 40 µW cm −2 . The two lobster "batteries" implanted with EBFCs were connected in series to obtain an open-circuit voltage of 1.2 V, which successfully enabled the operation of an electronic meter. Connecting five lobster "batteries" in series also enabled the activation of a medical pacemaker requiring an operating voltage of 2.8 V and a power consumption of 90 µW [63,64].
In the development of implantable biofuel cells (EBFCs), biocompatibility and minimal impact on the living organism are crucial considerations. Scherson et al. grafted alglucosidase (Trehalase, Tre) and GOx dual enzyme anodes and BOD cathodes onto a polymer backbone with Os complexes and implanted them through an abdominal incision into female Blaberus discoidalis (insect name) [65]. The resulting cell had a maximum power density of about 55 µW cm −2 at 0.2 V and only a 5% decrease after 2.5 h of operation. Morishima et al. also made EBFCs using Tre and GOx dual enzyme anodes and BOD cathodes. The electrodes were protected with dialysis membranes and implanted into cockroaches [66]. The resulting cell obtained a maximum power density of 6.07 µW cm −2 using alginose in the cockroaches. In addition, using air diffusion cathodes, the maximum power density could reach 10.5 µW cm −2 . They also connected five cockroach EBFCs in series to successfully operate a piezoelectric speaker [66]. This result suggests that insect EBFCs are a promising biobattery that can be used as a power source for environmental monitoring micro-tools.
Using KB (Ketjen black, conductive carbon black) as the base electrode [34]. Nishizawa et al. developed a needle bioelectrode using FDH and GDH as dual enzyme anodes and BOD as cathode, which produced a power density of 115 µW cm −2 at 0.34 V after implantation into grapes [67]. The needle electrode was inserted into a rabbit ear vein vessel and the open-circuit voltage of the battery could reach 0.81 V [67]. Using CP (carbon paper) as the base electrode, Miyake et al. integrated hydrophilically treated with GOx and BOD on carbon fibers to form a bioanode and a biocathode onto a metal needle to form a needle cell [68]. The resulting cell obtained 10 mA cm −2 in simulated artificial glucose containing 5 mm of glucose. When the tip of the anode needle was inserted into the natural samples of grapes, kiwi and apples, the power generated by the assembled battery from the glucose was 55 µW, 44 µW and 33 µW, respectively. In addition, the needle cell was inserted into the heart of a mouse and produced 16.3 µW at 0.29 V. Additionally, they coated the anode to be implanted into the organism with an antifouling polymer and sealed the cathode with medical tape to minimize the evaporation of water without affecting the oxygen permeability, thus improving the battery life towards wearable electronics [68].
The activity of human dermal fibroblasts (Human Dermal Fibroblasts-adult, HDF) is a key biological factor affecting post-transplant tissue or organ wound healing. Jeon et al. performed a basal cell culture using GDH as an anodic enzyme and BOD as a cathodic enzyme to study the cytocompatibility of EBFCs using HDF as a model cell [69]. The cellular inflammatory response of HDF was low, while EBFCs showed extreme biotoxicity to HDF after replacing GDH with Gox. The power density of EBFCs prepared using GDH ranged from 15.26 to 38.33 nW cm −2 , indicating that GDH is more suitable for implantable EBFCs than GOx and is a promising implantable power source for biomedical applications [69]. Shleev et al. were able to generate electrical energy from human tears for the first time [70]. One-hundred µm diameter gold wires were covered with 17 nm Au nanoparticles to form a three-dimensional nanostructured microelectrode. Anodic enzyme electrode and cathodic enzyme electrode were prepared with CDH (Cellobiose Dehydrogenase) and BOD, respectively. Tested in the tear environment of a healthy human body, the battery had an open-circuit voltage of 0.57 V, and the maximum power density at 0.5 V is 1 µW cm −2 . After 20 h of continuous operation at 0.51 V, the output current of the cell was still 68% of the initial current with no significant change in the open-circuit voltage. The electrodes were stably stored in buffer at room temperature [70]. These findings suggest that tears could potentially serve as a sustainable and renewable source of energy for various implantable devices.
After Cinquin and Cosnier et al. developed an EBFC implanted in rats [61]. Martin and Cinquin et al. implanted a wireless-controlled EBFC in a rabbit [71]. The EBFC functioned well in vivo for a period of 2 months, although at the end of the implantation, the power output diminished, most likely due to an inflammatory process [61,71]. After developed the implantable EBFC in a living snail [62]. Katz et al. prepared an EBFC based on BP electrodes with GOx and BOD [72]. The resulting biofuel cell was implanted in a living gray garden slug which produced electrical power in the range of 2-10 µW and it was successfully used to power one microelectronic device. The whole system was able to be operated autonomously by extracting electrical energy from the glucose-containing hemolymph (blood substituting biofluid) in the slug and readout data wirelessly [72].
Kim et al. implanted carbon nanotube-based EBFC in the abdominal cavity of mice [73]. The carbon nanotube electrodes were coated with Nafion and twisted into a micro-sized, two-ply, one-body system with improved mechanical properties for convenient implantation, providing high power density of 0.3 mW cm −2 . The two-ply EBFC system exhibited good biocompatibility in vivo [73]. Lee et al. developed an EBFC using carbon felt electrodes loaded with immobilized Gox and BOD via MWCNTs as bioanode and biocathode, respectively [74]. The EBFC was combined with a brain stimulator and implanted in a pigeon. The power output from the EBFC reached 0.12 mW in vitro and 0.08 mW in vivo using only the natural glucose and oxygen in the pigeon's body [74].
Although implantable EBFCs have been comprehensively studied, in order to achieve sustainable operation, the following issues should be addressed: (1) the enzyme should be simply and firmly anchored to the electrode and no substances with poor biocompatibility should be arbitrarily introduced; (2) it should be ensured that oxygen does not interfere with the catalytic oxidation of the anode biofuel; and (3) the bioelectrode should be able to operate at low concentrations of biofuel and oxygen, especially in the case of slow diffusion and high viscosity of biological tissues, such as in human blood.

Wearable EBFCs
Wearable devices have been changing how we live and work when tracking our life physically and physiologically. Flexible and biocompatible electrode materials are often required as enzyme carriers in EBFCs for wearable devices. EBFCs in wearable devices can convert molecular chemical energy inside body fluids such as sweat, tears and blood into electrical energy to produce reasonable power output [7]. Shitanda et al. prepared paper-based EBFCs based on a screen-printing array structure toward wearable energy storage devices [48]. Liu et al. developed the wearable self-powered biosensor system integrated with diaper for detecting the urine glucose of diabetic patients (Table 2) [51]. The 2D-GP nanosheets were assembled in a paper-like architecture with good mechanical strength and high conductivity for potential applications of EBFCs in wearable and implantable devices [53]. The implantable needle-type biofuel cell using enzyme/mediator/carbon nanotube composite fibers has been developed towards wearable electronics [68].
Magner et al. prepared a nanoporous gold-based EBFC for wearable microelectronic device applied on contact lenses (Figure 4) [75]. Nanoporous gold (NPG) electrodes were mechanically stable and flexible. The resulting flexible EBFC loaded with LOx and BOD was placed between two commercially available contact lenses, exhibiting a maximum power density of 1.7 ± 0.1 µW cm −2 and an open-circuit voltage of 380 ± 28 mV when tested in air-equilibrated artificial tear solutions [75]. Niiyama et al. reported a high-performance EBFC based on flexible carbon cloth loading with MgO-templated porous carbon [76]. The open-circuit potential was 0.75 V and the maximum output power density was 2 mW cm −2 . The resulting flexible composite carbon cloth exhibited very promising for the development of wearable EBFCs [76]. Li et al. developed an integrated EBFC based on laser-scribed N-doped graphene (LSNG) [77]. The integrated EBFC produced a maximum power density of 27 ± 1.7 µW cm −2 at open-circuit voltages of 0.45 ± 0.03 V. LSNG displayed excellent mechanical robustness, conductivity and electrocatalytic performance for the development of wearable device [77].
Appl. Sci. 2023, 13, x FOR PEER REVIEW 11 of 17 between the two electrodes as the diaphragm of the battery, allowing the anode to be in contact with sweat and the cathode to be exposed to air. The battery was attached to the skin surface with a maximum output power of 938 µW and successfully provided electrical power to the wearable sensor and was able to operate stably for two weeks [79]. Bandodkar et al. prepared an Au/CNTs/NQ/LOx/CTS biological anode. The anode was based on CTS-coated carbon nanotubes with NQ as the electron transfer mediator. The Au/CNTs/Ag2O was used as the cathode. The resulting cell was assembled into a wearable battery with an open-circuit voltage of 0.5 V, producing a power density of 1.2 mW cm −2 at 0.2 V [80]. Zhu and Liao et al. also used NQ as an electron transfer mediator and integrated MWCNTs onto CC to produce a CC/MWCNTs/NQ/LOx anode [81]. Pt was mixed with conductive carbon black on CC to produce a cathode. The resulting EBFCs produced a power density of 62.2 ± 2.4 µW cm −2 . Reid et al. integrated poly (methylene green) (PMG), LOx and electron mediator NAD+ on BP paper to make an anode [82]. The quaternary ammonium cation-modified Perfluorosulfonic acid film was used to protect and immobilize the BOD as the cathode. The cathode and anode were immobilized on the surface of silicone rubber elastomer and combined with contact lenses to assemble EBFCs that could generate energy from lactic acid in tears [82]. The cell had an open-circuit voltage of 0.413 ± 0.06 V and a maximum output power of 8.01 ± 1.4 µW cm −2 and could provide micro wearable devices such as contact lenses with electrical power. However, PMG leaching from the anode easily degraded the performance of the cell and the problem of poor biocompatibility still needs to be solved [82]. These are several examples of wearable biofuel cells that use different approaches to generate electrical power from human sweat or tears. Mercier and Wang et al. fabricated printed electrodes on textiles, using tetrathiofulvlene-7,7,8,8-tetracyanoquinodimethane (TTF-TCNQ) as the electron transfer mediator for LOx to obtain the anode and platinum black as the cathode, producing a power density of 100 µW cm −2 at 0.34 V [83]. The battery pack was mounted into a textile headband and wristband, and a stable voltage and current were obtained from the sweating human body with the help of a DC/DC converter to directly power LEDs and digital watches [83]. Miyake et al. also developed a high-power biofuel cell bracelet using lactic acid production capacity from human sweat [84]. This battery used LOx/Os-based complexes/CNTs/carbon fibers as the anode for lactic acid oxidation and BOD/CNTs/carbon fibers as the cathode for oxygen reduction and is woven into a hydrophilic support fabric for storing sweat. The cell reached a maximum opencircuit voltage of 0.6 V and obtained a maximum power of 74 µW at 0.39 V in artificial sweat containing 20 mm of lactic acid. The performance remained above 80% after 12 h of operation. Six cells were connected in series and bundled and wrapped around the wrist, and the open-circuit voltage reached 2.0 V, successfully powering an electronic watch [84]. In another work, they coated carbon fibers with MWCNTSs, and integrated a GDH Wang et al. prepared a CTS-protected, TTF-mediated, CNTs-based lactate oxidase anode electrode (CNTs/TTF/LOx/CTS) with Pt as cathode [78]. The resulting wearable battery placed on a human arm produced a power density of 5-70 µW cm −2 under exercise sweating conditions [78]; Baikun et al. assembled a battery with a CC electrode loaded with LOx as the anode and Pt as the cathode [79]. A filter membrane was sandwiched between the two electrodes as the diaphragm of the battery, allowing the anode to be in contact with sweat and the cathode to be exposed to air. The battery was attached to the skin surface with a maximum output power of 938 µW and successfully provided electrical power to the wearable sensor and was able to operate stably for two weeks [79]. Bandodkar et al. prepared an Au/CNTs/NQ/LOx/CTS biological anode. The anode was based on CTScoated carbon nanotubes with NQ as the electron transfer mediator. The Au/CNTs/Ag 2 O was used as the cathode. The resulting cell was assembled into a wearable battery with an open-circuit voltage of 0.5 V, producing a power density of 1.2 mW cm −2 at 0.2 V [80]. Zhu and Liao et al. also used NQ as an electron transfer mediator and integrated MWCNTs onto CC to produce a CC/MWCNTs/NQ/LOx anode [81]. Pt was mixed with conductive carbon black on CC to produce a cathode. The resulting EBFCs produced a power density of 62.2 ± 2.4 µW cm −2 . Reid et al. integrated poly (methylene green) (PMG), LOx and electron mediator NAD+ on BP paper to make an anode [82]. The quaternary ammonium cation-modified Perfluorosulfonic acid film was used to protect and immobilize the BOD as the cathode. The cathode and anode were immobilized on the surface of silicone rubber elastomer and combined with contact lenses to assemble EBFCs that could generate energy from lactic acid in tears [82]. The cell had an open-circuit voltage of 0.413 ± 0.06 V and a maximum output power of 8.01 ± 1.4 µW cm −2 and could provide micro wearable devices such as contact lenses with electrical power. However, PMG leaching from the anode easily degraded the performance of the cell and the problem of poor biocompatibility still needs to be solved [82].
These are several examples of wearable biofuel cells that use different approaches to generate electrical power from human sweat or tears. Mercier and Wang et al. fabricated printed electrodes on textiles, using tetrathiofulvlene-7,7,8,8-tetracyanoquinodimethane (TTF-TCNQ) as the electron transfer mediator for LOx to obtain the anode and platinum black as the cathode, producing a power density of 100 µW cm −2 at 0.34 V [83]. The battery pack was mounted into a textile headband and wristband, and a stable voltage and current were obtained from the sweating human body with the help of a DC/DC converter to directly power LEDs and digital watches [83]. Miyake et al. also developed a high-power biofuel cell bracelet using lactic acid production capacity from human sweat [84]. This battery used LOx/Os-based complexes/CNTs/carbon fibers as the anode for lactic acid oxidation and BOD/CNTs/carbon fibers as the cathode for oxygen reduction and is woven into a hydrophilic support fabric for storing sweat. The cell reached a maximum opencircuit voltage of 0.6 V and obtained a maximum power of 74 µW at 0.39 V in artificial sweat containing 20 mm of lactic acid. The performance remained above 80% after 12 h of operation. Six cells were connected in series and bundled and wrapped around the wrist, and the open-circuit voltage reached 2.0 V, successfully powering an electronic watch [84]. In another work, they coated carbon fibers with MWCNTSs, and integrated a GDH bioanode mediated by MG and a BOD biocathode for oxygen reduction on a cotton textile fabric to make a battery with an open-circuit voltage of 0.51 V [85]. The power density generated from 0.1 mM of glucose (simulating the glucose concentration in human sweat) at 0.24 V was 48 µW cm −2 . When used in a hydrogel containing 200 mM of glucose, a power density of 216 µW was generated at 0.36 V. They also transformed the cell into an S-shape with no significant loss of performance. In addition, they bundled four cells on clothing to generate 1.9 V and successfully lit LEDs on clothing [85]. Wang and Cosnier et al. immobilized NQ-mediated LOx as anode and BOD as cathode on a BP paper substrate [86]. The two bioelectrodes were integrated onto a stretchable screen-printed circuit substrate. The battery open-circuit voltage of 0.74 V and the maximum power density of 520 µW cm −2 were obtained with in vitro-simulated sweat containing 0.5 M lactic acid [86]. The battery is immobilized in the human arm and can generate a maximum power of 450 µW during exercise.
Zhou and Bai et al. developed a microfluidic wearable battery consisting of a skininterfaced microfluidic module and a cell module based on an ethanol/oxygen biofuel cell [29]. Using 3D-NHCAs as carriers of AOx and BOD to form electrodes, the battery system was placed on different skin regions of the forearm, the back of the neck, and the forehead of the drinker. The battery exhibited a maximum output of 1.01 µW cm −2 for the battery system worn on the forearm and a minimum output of 0.65 µW cm −2 on the forehead [29]. They also considered and studied some practical conditions related to human drinking and realized the collection of sweat bioenergy in a controlled environment. The battery system fully proves the feasibility of replacing exogenous substances in human sweat with human body-derived substances (such as ethanol) as a new biofuel.
Over time, although EBFCs have made significant improvements in power output and stability, in paving the way for the use of EBFCs to power portable electronic devices, there are considerable challenges in developing these into cost-effective, mass-produced power supplies.

Discussion
Enzyme-based biofuel cells (EBFCs) have attracted significant attention in recent years as a promising technology for sustainable and eco-friendly power generation. The use of immobilized enzymes provides many advantages, including high stability, substrate specificity, and mild reaction conditions, making EBFCs suitable for various applications. However, there are still some challenges that need to be addressed to further improve the performance and practicality of EBFCs.
One major challenge is the limited power output and energy density of EBFCs, which may not be sufficient for some applications. That said, the problems of low power density and short service life have not been well solved, and the existing EBFCs are far inferior to chemical fuel cells in terms of lifetime and operational stability. To overcome these limitations, researchers have been exploring novel enzyme immobilization techniques to optimize the electrode designs and materials and employing new types of biofuels that can generate more energy. In order to maintain the high performance of EBFCs, it is necessary to maintain the intact structure of the enzyme and the catalytic active site and retain the unique and efficient catalytic activity of the enzyme. To improve the cell performance, people would take many other factors into account to select a suitable immobilization method for the enzyme, to promote the electron transfer between the electrode surface and the enzyme, and to choose suitable electrode materials to maintain and enhance enzymatic activities on the electrode, all of which are important. For example, the use of nanomaterials such as carbon nanotubes, graphene, and metal nanoparticles as enzyme immobilization matrices has been shown to enhance enzyme activity and improve power output. While enzyme immobilization can improve enzyme stability, the enzymes still undergo gradual deactivation over time. To address this issue, researchers have been exploring ways to improve enzyme stability and extend their lifespan. This includes the use of enzyme engineering techniques to enhance enzyme stability.
Another challenge is the stability and durability of EBFCs, especially under physiological conditions. The enzymes used in EBFCs can be sensitive to various factors such as temperature, pH, and reactive species, which may affect their catalytic activity and stability. Therefore, it is important to develop more robust and stable enzymes, as well as improve the design and materials of the immobilization matrix to extend the lifespan of the immobilized enzymes.
Despite the challenges, the potential applications of EBFCs are vast, particularly in the fields of implantable and wearable devices. Implantable devices, such as pacemakers and insulin pumps, require a reliable and long-lasting power source, which can be provided by EBFCs. Wearable devices, such as smartwatches and fitness trackers, can benefit from the use of EBFCs as a sustainable and eco-friendly power source that can be recharged using the wearer's own biofluids. Although EBFCs have been successful in obtaining energy from body fluids in vitro and have been successfully tested on animals, it is inevitable that when they come into direct contact with tissues or organs, the implanted EBFCs will produce by-products or electrical signals that will damage or irritate the implanted body and the cell itself. Therefore, there is still a long way to go in order to achieve a good and stable application of EBFCs in living organisms.

Conclusions and Outlook
In conclusion, enzyme-based biofuel cells represent a promising technology for sustainable and eco-friendly power generation. The use of immobilized enzymes provides many advantages, including high stability, substrate specificity, and high efficiency, making EBFCs suitable for various applications.
Despite significant progress in the development of EBFCs, several challenges still need to be addressed in order to fully realize their potential. One of the most pressing issues is the need for improved stability and lifetime of the enzymes used in EFBCs. Enzymes are typically susceptible to denaturation and deactivation under conditions such as high temperature, extreme pH, and exposure to organic solvents. Thus, finding strategies to enhance the stability of enzymes under such conditions is crucial for the long-term operation of EBFCs. Another important issue is the need for improved efficiency and power output of EBFCs. Although enzymes have high catalytic activity, the power densities achieved by EBFCs are still relatively low compared to conventional fuel cells. Thus, developing new enzyme-based systems or optimizing existing ones to improve power output is a critical area of research. Finally, the integration of EBFCs into practical devices and systems requires the development of suitable electrode materials and designs, as well as the development of cost-effective and scalable production methods. Overall, addressing these urgent research issues is key to realizing the full potential of EBCs for a range of applications. In this review, although we have summarized the recent literatures' work using a variety of immobilization and modification strategies for enzymatic bioanodes and biocathodes, especially using those carbon-based nanomaterials which can be further improved by incorporating conductive polymers and other nanomaterials, there have been not many direct applications of these new strategies on the wearable and implantable devices based on EBFCs. There are still some technical gaps to be filled between the research of EBFCs and their applications in practice, and so far, there have been very few commercially EBFC available in the market, if any.
While the low power output and limited enzyme lifespan are still major challenges, ongoing research in novel enzyme immobilization techniques, hybrid EBFC systems, and enzyme stabilization strategies offer great potential for improving the performance and lifespan of EBFCs. Although there are many types of natural enzymes, almost all of them more or less have a stability issue and require mild reaction conditions and the cost of biological enzymes is not low, which limits the expanded application of EBFCs and makes it difficult to commercialize them. So, it is also important to reduce the cost of manufacturing EBFCs and apply them to complex environments. It is expected to develop suitable and efficient artificial or mimetic enzymes to replace natural enzymes for catalytic reactions.
The future of EBFC technology looks promising, particularly in the fields of implantable and wearable devices. In the coming years, we expect to see further progress in enzyme engineering and immobilization techniques, as well as the integration of EBFCs with other devices and systems. With ongoing advancements in enzyme engineering, enzyme immobilization strategies, and hybrid energy conversion systems, the power output and efficiency of EBFCs are expected to continue to improve. Moreover, the use of sustainable and eco-friendly biofluids as fuel sources makes EBFCs an attractive option for sustainable energy production. In the future, we can expect to see more practical applications of EBFCs, particularly in the medical and wearable device industries. With the increasing demand for sustainable and renewable energy sources, the potential for EBFCs to revolutionize the energy industry is enormous.

Conflicts of Interest:
The authors declare no conflict of interest.