Anti-fatigue adhesive non-swelling hydrogel constructed by covalent topological structure and micro-nano gel for stretchable bioelectronics

Abstract

Hydrogel adhesives are rapidly emerging as a promising candidate toward flexible bioelectronics due to their adhesive characteristics and tissue-like mechanical properties. However, current hydrogel adhesives manifest weak anti-fatigue adhesion and an inability to ensure long-term integration of bioelectrodes on wet and dynamic tissue surfaces because they are constrained by their high swelling ratio and exclusive formation of covalent bonds at the tissue interface and its own weak cohesion. Here, we for the first time develop covalent bond topological adhesion paired with double covalent bond cross-linking in hydrogel to enhance cohesive force and adhesive force, achieving excellent anti-fatigue tissue adhesion and adhesive's capacity to follow significant tissue deformation. The adhesive strength of our hydrogel (Sodium alginate-polyacrylamide-acrylic acid N-hydroxysuccinimide ester hydrogel (SPAN) as the substrate and liquid adhesive containing chitosan (LC) as the adhesive layer) reaches impressive 290 kPa, surpassing that of the reported hydrogels (∼130 kPa). Additionally, fatigue threshold of SPAN/LC adhesion (240 J m⁻²) far exceeds SPAN (48.6 J m⁻²) and SPAN/LC (without NHS ester) (71.6 J m⁻²). Simultaneously, micro-nano gel and pre-swelling strategy enhance the elongation at break (1330 %) and limit swelling of SPAN in vivo (V/V₀ = 1) by storing SPAN chains and acting as physical cross-linking points, thereby increasing adhesion stability and biocompatibility. The adhesion strength of SPAN/LC to the tissue consistently remains above 125 kPa after 70 days of immersion in a buffer solution. Employing the hydrogel as the soft interfacing material, we further demonstrate stretchable micro-electrode arrays (MEAs) for long-term electrophysiological recording and stimulation in rat models. Thanks to the superior anti-fatigue performance of the hydrogel adhesives, this MEAs adheres tightly to the wet and continuously moving subcutaneous muscle of a living rat, enabling the stable collection of electrophysiological signals with high signal-to-noise ratios for 35 days. These excellent performances pave the way for establishing a new paradigm in long-term stable and highly efficient signal transmission at the dynamic electrodes–tissue interface.

Graphical abstract

The as prepared SPAN/LC with micro-nano gel form double covalent bond crosslinking and topological adhesion with the tissue, achieving non-swelling and generating tough and excellent anti-fatigue adhesion. Subsequently, by using SPAN/LC, we fabricated stretchable micro-electrode arrays with stable and tough adhesion. This microelectrode array can collect stable electrophysiological signals on the subcutaneous muscle of rats for 35 days, and exhibits excellent stimulating ability.

Highlights

• •A new adhesive strategy was demonstrated by introducing covalent bond topological adhesion and double covalent bonds.
• •Introduction of micro-nano gel and pre-swelling strategy enhance the elongation at break and limit swelling of SPAN in vivo.
• •The addition of NHS ester in SPAN enhances the cohesion between chitosan and SPAN.
• •Tissue interface bioelectronics can record electrophysiological signals with high SNR for an extended duration of 35 days.

1. Introduction

Reliable integration of bioelectronics with tissues is crucial to ensure accurate signal exchange, efficient stimulation of tissues and long-term stable recording of electrophysiological signal [[1], [2], [3], [4], [5]]. Signal exchange is critical for the diagnosis and treatment of neurological diseases, including deafness, Parkinson's disease, epilepsy, and so on [[6], [7], [8], [9]]. Due to the huge mechanical mismatch between rigid bioelectrodes and tissues, the ideal seamless integration can only be achieved with flexible and stretchable materials [10,11]. Flexible and stretchable bioelectrodes offer improved conformal contact with curved tissues [[12], [13], [14], [15]], yet weak adhesion to tissues poses a significant threat to stable and long-term integration on wet and dynamic tissue surfaces, impairing efficient stimulation of tissues and long-term stable recording of electrophysiological signal (Fig. S2, Supporting Information).

In recent years, liquid adhesives (Precise definition of “liquid adhesive” is provided in the Supporting Information) [[16], [17], [18], [19], [20]], and hydrogel adhesives [[21], [22], [23], [24], [25], [26], [27], [28], [29]], have been explored for the integration of bioelectronic devices with wet and dynamic tissues [17,22,23,[30], [31], [32], [33]]. Despite these advancements, several challenges persist. For instance, liquid adhesives rely on the diffusion of molecules, a process that takes a considerable amount of time (>5 min) to form interfacial interlocking adhesion (Physical or chemical adhesion (amide covalent bonds and Schiff base bonds with tissue constituents (-COOH, -NH₂ and -SH groups))) (Fig. 1a) [20,[34], [35], [36]]. This extended duration means an inevitable increase in the anesthesia time during electrode implantation procedure, leading to possible neuronal apoptosis and brain damage [18,37]. Also, the unstable and poor anti-fatigue adhesion of liquid adhesives and hydrogels to tissues poses an obstacle to ensuring the long-term stable integration of bioelectronics on wet and continuously moving tissue surfaces (Fig. 1b). This is attributed to the fact that the interfacial interlocking adhesion and covalent cross-linking of the liquid adhesive/hydrogel-tissue interface result in a poor ability to dissipate stress, making the interface prone to fracture and adhesion failure [19,20,34,36,[38], [39], [40], [41]]. Recently, topological adhesion contributed by chitosan as bridging polymer and sodium alginate-polyacrylamide hydrogel as substrate has been proposed to overcome the above problems. This type of adhesive exhibits great application prospects in the field of biomedicine [[42], [43], [44], [45]]. However, these adhesives only focus on the adhesion between the bridging polymer and the tissue, overlooking crucial factors such as the weak cohesive force between the bridging polymer and hydrogel substrate, as well as interface adhesion between hydrogel and tissue. The inadequate cohesive strength and interface adhesion contribute to frequent fractures and failures in adhesion between the chitosan-hydrogel and hydrogel-tissue interfaces, significantly compromising the adhesive's anti-fatigue stability. This weak cohesion is attributed to the carboxyl group of sodium alginate in hydrogel being occupied and cross-linked by calcium ions, preventing chitosan from forming stable covalent topological adhesion even if it diffuses into the interior of hydrogel (weak cohesive force). Strong and stable anti-fatigue adhesion requires both strong cohesive force and adhesive force [46,47]. Simultaneously, topological adhesion is easily compromised with the dissolution of chitosan under acidic conditions (for example in the stomach and bladder) (Fig. S3, Supporting Information) [[42], [43], [44],[48], [49], [50]]. Therefore, tissue adhesion formed solely through covalent bonds or topological network is not robust enough to ensure the enduring adhesion of bioelectrodes on wet tissue for long-term applications because they have weak anti-fatigue adhesion. Meanwhile, high swelling ratio of the hydrogel can further reduce adhesion stability, mechanical performance and increase rejection reaction of tissue [[51], [52], [53], [54]].

Fig. 1.

a-b) Schematic diagram of the adhesion mechanism of the current liquid adhesives and hydrogel. c) Schematic illustration of the adhesion mechanism of our hydrogel-tissue interface formation. d) Our hydrogel adhesive exhibits excellent adhesion and stability to previously reported literature. e) Digital photos of the stomach adhered by SPAN/LC under different strains. f) Adhesion strength of SPAN/LC, SPAN and SPAN/LC without NHS ester to stomach after the stomach adhered by they were applied different strains. Values in figure f show the mean and the standard deviation (n = 4).

Herein, adhesion strength, stability, anti-fatigue ability and mechanical performance of adhesive are significantly enhanced by synergistic effects of NHS ester-enhanced cohesive strength, EDC/NHS-activated dual-covalent bonding, fatigue-resistant topological networks, micro-nano gel and pre-swelling strategy. Subsequently, we introduce, for the first time, NHS ester groups into polyacrylamide by copolymerization to enhance cohesive force between bridging polymer chitosan and hydrogel, simultaneously, developing an adhesive capable of forming double-covalent bonds on tissue surface and covalent bond topological adhesion in tissue interior by introducing EDC/NHS into chitosan (activating the carboxyl groups). In addition, the micro-nano gel and pre-swelling strategy (in vitro) were employed to enhance the elongation at break (after swelling) and limit swelling of SPAN in vivo (V/V₀ = 1). These strategies significantly enhance both the cohesive force between hydrogel and chitosan, the adhesive strength between the adhesive and tissue, as well as the adhesion stability and biocompatibility of the hydrogel. Consequently, the robust anti-fatigue tissue adhesion was achieved. This adhesive is composed of sodium alginate-polyacrylamide-acrylic acid N-hydroxysuccinimide ester hydrogel (SPAN) as the substrate and a liquid adhesive containing chitosan (LC) as the adhesive layer (acrylic acid N-hydroxysuccinimide ester is named AAc-NHS ester) (Fig. S4, Supporting Information). The chitosan within LC exhibits the ability to diffuse into both SPAN and tissue, forming covalent bonds with the inside NHS ester of SPAN (enhancing cohesive force) and carboxyl groups of tissue (activating carboxyl groups through the use of a coupling agent, carbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS)) (Fig. 1c). This unique mechanism is defined as “covalent bond topological adhesion” in this work (specifically, the formation of covalent bonds inside the SPAN and tissue). The adhesion generated by covalent bond topology is not affected by pH change in tissues. Simultaneously, the NHS ester groups and amino groups of SPAN form double-covalent adhesion with amino groups and carboxyl groups of tissue surface (EDC and NHS activating carboxyl groups of tissue), respectively, enhancing adhesive force with wet tissues. The synergistic effect of topological network and double-covalent bonds effectively prevents stress concentration on the covalent bonds or topological network, yielding anti-fatigue and tough tissue adhesion. surpassing that reported in the literatures and reaching 290 kPa (Fig. 1d–Table S1, Fig. S5, Supporting Information). The SPAN/LC toughly adhere to tissue surface even after the tissue, adhered by SPAN/LC, undergoes stretching/releasing cycles 3000 times under a strain of 30 %. In contrast, SPAN without LC, which exclusively forms covalent bonds with the tissues surface, loses adhesion after tissue adhered by SPAN undergoes 700 stretching/releasing cycles under a strain of 30 %. Meanwhile, fatigue threshold of SPAN/LC adhesion (240 J m⁻²) far exceeds SPAN (48.6 J m⁻²) and SPAN/LC (without NHS ester) (71.6 J m⁻²). In addition, the adhesion strength of SPAN/LC to tissue remains 154.5 kPa even after an extended period of 70 days. This tough, stable, and anti-fatigue adhesion ensures the long-term stable integration of electrodes onto wet and dynamic tissue surfaces. Simultaneously, SPAN and LC demonstrate excellent biocompatibility, with the viability of rat cardiomyocytes exceeding 96 % and without significant inflammatory response after 24 h cultivation (in vitro) and long-term implantation (in vivo) for 14 days. Extensive literature supports the use of related polymers in vivo, demonstrating favorable biocompatibility [45,[55], [56], [57]] (Figs. S6–S7, Supporting Information). As a proof of concept, by employing SPAN as the stretchable substrate and LC as the adhesive layer, we fabricate a highly stretchable microelectrode array (SPAN/LC-PPy MEAs). This electrode, with its superior anti-fatigue adhesion (polypyrrole (PPy) as the electrode material) properties, remains stably adhered to wet and dynamic tissue surfaces in rats for an impressive duration of 35 days. Furthermore, it accurately monitors electrophysiological signals with high signal-to-noise ratios (SNRs) in rats. The success of these adhesive and bioelectrode holds significance promise for the diagnosis of electrophysiological diseases and the precise manipulation of prostheses.

2. Results and discussion

2.1. Adhesion stability and mechanism of SPAN/LC

The adhesive stability of hydrogel is a crucial performance factor, it allows bioelectrodes to adhere stably on wet tissue surface for an extended period. Tissues undergo large deformation and remain in a prolonged state of motion; for example, the strain of the stomach and bladder reaches 70 % and 250 %, respectively (Fig. S8, Supporting Information). This requires that the adhesion formed by hydrogel-tissue interface can drive the bioelectrode to follow the tissue's significant deformation. However, currently reported adhesive bioelectrodes only form covalent bond cross-linking with the tissue surface to generate adhesion [[22], [23], [24],30]. Stress concentrated at the bioelectrode-tissue interface easily results in covalent bond fracture and adhesive failure when the tissue undergoes large strains repeatedly. To address this problem, we designed a hydrogel adhesive (SPAN/LC) that can form both double covalent bond and topological adhesion with tissues. Simultaneously, the amide bonds formed by the NHS ester of SPAN and the NH₂ groups of chitosan significantly enhance cohesion between SPAN and chitosan. The covalent bonds and topological networks can effectively dissipate stress (Fig. 1e), thereby generating tough and anti-fatigue adhesion. Despite subjecting the stomach adhered by SPAN/LC was applied to strain of 70 %, the adhesion strength of SPAN/LC to tissue remains stable at 92 kPa (Fig. 1f). This excellent adhesion strength is attributed to the transformation of the amino groups on the chitosan chains to -NH₃⁺ in the LC solution (pH = 6) [42], causing chitosan to dissolve in the solution and diffuses into the tissue and the SPAN. As shown in Fig. 2a, FITC-labeled chitosan continuously diffused into the tissue with the increase of time, resulting in an increase in the thickness of the fluorescent layer. Simultaneously, the same phenomenon is observed with SPAN hydrogel (Fig. 2b), effectively verifying that the chitosan diffuses into the tissue and the hydrogel. Subsequently, the high pH (∼7.35) at tissues leads to an increase of pH around chitosan, causing the transformation of -NH₃⁺ of chitosan to -NH₂ [36,43,58]. The -NH₂ groups of chitosan then form amide bonds with carboxyl groups of the tissue (with coupling agent) and NHS ester of the hydrogel, which promotes the chitosan chains to establish covalent bond topological adhesion within the tissue and hydrogel (enhancing cohesion), respectively. Therefore, the applied stress is dispersed to the entire topological polymer networks, thereby decreasing the stress concentration [16]. Meanwhile, dual-covalent bonding between tissue and SPAN/LC surface further dissipates stress concentration, enhancing the adhesion strength and stability.

Fig. 2.

a) Fluorescent images of FITC-labeled chitosan diffusing into the muscle. b) Fluorescent images of FITC-labeled chitosan diffusing into the hydrogel. c) Adhesion strength of SPA, SPAN/LC and SPAN/LC without NHS ester to the tissue. d) Schematic illustration of covalent crosslinking between amine-coupled fluorescent microbeads with PAM with EDC/NHS (the SPAN without AAc-NHS ester and sodium alginate was named PAM). Values in figure c show the mean and the standard deviation (n = 4).

Contrastingly, the adhesion strength of SPAN and SPAN/LC without NHS ester drops sharply under varying strains (Fig. 1f, Fig. S9, Supporting Information), indicating that the interfacial adhesion (formed by SPAN hydrogel-tissue) and weak cohesion (formed by SPAN-chitosan) possess weak ability to dissipate stress. This is because the tissue needs to drive the hydrogel through covalent bonds and physical topology (weak cohesion) at the interface to move in coordination with the tissue. Consequently, the stress easily concentrates on the adhesive interface and SPAN-chitosan interface (weak cohesion) during stretching, causing interfacial adhesion fracture [3]. On the other hand, the NHS ester enhancing cohesion, dual-covalent bonding (on tissue) and topological network (in tissue) formed between SPAN/LC and tissue effectively dissipate the stress, preventing interfacial adhesive fracture, and improving the stability of adhesion.

NHS ester groups and -NH₂ groups of SPAN also affect the stability of adhesion because they can form double covalent bonds with the amino groups and carboxyl groups on the tissue surface, respectively (Figs. S10–S11. Supporting Information). Meanwhile, the amide bond formed by NHS ester of SPAN and amino groups of chitosan (in LC) greatly increased the cohesion between hydrogel and chitosan (Fig. S12. Supporting Information). This interaction in tough adhesion greatly enhancing the adhesion stability of SPAN/LC. The combination of toughness and stability in adhesion plays a crucial role in promoting seamless integration of bioelectrodes on wet dynamic tissues. Consequently, this facilitates bioelectrodes in accurately collecting physiological signals from these moving tissues. SPAN/LC demonstrates a robust adhesion strength of 120 kPa to the tissue (Fig. 2c, Video S1, Fig. S13, Supporting Information). On the contrary, SPA (The hydrogel without AAc-NHS ester groups and LC was named SPA) and SPAN/LC without NHS ester only achieve an adhesion strength of 2.6 kPa and 62.2 kPa. This further indicates that the NHS ester group increases the cohesion of the chitosan-SPAN, enhancing the adhesive adhesion. Subsequently, after using FITC-labeled cysteine covalent crosslinking with polyacrylamide (PAM) in coupling agent (EDC/NHS) (Fig. 2d), we observe that the PAM surface possesses large number of fluorescent microspheres (Fig. 3a). Meanwhile, FTIR spectrum of PAM with EDC/NHS reveals characteristic peaks of amide bonds at 1525 cm⁻¹ and 1409 cm⁻¹ (Fig. 3b), indicating that -NH₂ groups of SPAN are capable of forming amide bonds with the carboxyl group of the tissue. Employing FITC-labeled chitosan covalent crosslinking with SPAN (with NHS ester) (Fig. S14, Supporting Information), SPAN surface also shows a large number of fluorescent microspheres (Fig. 3c), and its FTIR spectrum displays characteristic peaks of amide bonds (Fig. 3d). These characterizations effectively validate that NHS ester groups and -NH2 groups of SPAN/LC generate double covalent bond with tissue, thereby determining the adhesion strength of the adhesive to the tissue.

Fig. 3.

a) The fluorescence microscopy photos of covalently crosslinked microbeads with SPAN with coupling agent (EDC and NHS) (left) and SPAN without coupling agent (right). b) FTIR spectrum of the cysteine covalently crosslinking with PAM with EDC/NHS. c) The fluorescence microscopy photos of covalently cross-linked microbeads with SPAN with NHS ester (left) and SPA without NHS ester (right). d) FTIR spectrum of the chitosan covalently crosslinking with SPAN with NHS ester. e) The adhesion strength of SPAN/LC to tissue after the tissue adhered by SPAN/LC was applied 3000 stretching/relaxing cycles under 30 % strain (The stretching rate is 50 mm/s. The length and width of muscle are 3 cm and 1.5 cm, respectively. The length and width of hydrogel are 1.5 cm and 1 cm, respectively). f) Images of SPAN/LC peeling from the tissue surface after the tissue adhered by SPAN/LC was applied 3000 stretching/relaxing cycles under 30 % strain. g) Interfacial fatigue threshold of SPAN, SPAN/LC (without NHS and coupling agent), SPAN/LC (without NHS ester) and SPAN/LC. Values in figure e and g show the mean and the standard deviation (n = 4).

2.2. Adhesion anti-fatigue properties of SPAN/LC

In addition to enduring large strain, biotissues are constantly in motion. Therefore, the adhesion strength generated by hydrogels to tissues is required to possess anti-fatigue properties to ensure the long-term stable integration of bioelectrodes on the wet dynamic tissue surface. The adhesion strength of SPAN/LC to muscle still maintains stable even after the tissue undergoes 3000 cycles of stretching and releasing under a 30 % strain (Fig. 3e, Fig. S15a, Supporting Information). Moreover, SPAN/LC exhibits robust adhesion to the muscle surface (Fig. 3f, Video S2-S3, Supporting Information). In contrast, both the SPAN and SPAN/LC (without NHS ester and coupling agent) lose adhesion undergoing 3000 cycles of stretching and releasing under 30 % strain (Fig. S15b, Supporting Information). Simultaneously, the interface fatigue threshold and anti-fatigue adhesion strength of SPAN/LC are significantly superior to those of SPAN, SPAN/LC without NHS ester and SPAN/LC without NHS ester and coupling agent (Fig. 3g, Figs. S16–S20, Supporting Information), indicating that interfacial covalent bonds, weak cohesive force and physical topological adhesion possess weak anti-fatigue abilities.

2.3. Non-swelling properties of SPAN/LC

The stable adhesion strength is critical for long-term implantation. However, the SPAN of high swelling ratio will seriously weaken the its mechanical properties [[51], [52], [53], [54]], adhesion strength stability and increase the rejection reaction of the tissue. Although the swelling of the hydrogel can be limited by introducing hydrophobic groups or hydrophobic polymers [24,59,60], these methods will enhance hydrophobic properties of hydrogel and limit the diffusion of hydrophilic chitosan solution into the hydrogel, thus weakening the adhesion. We introduce micro-nano gel (micro-nano gel is prepared by ball milling SPAN hydrogel, and named mn-SPAN (Figure S21a-b, Figure S22, Supporting Information)) into SPAN to reduce its swelling ratio and simultaneously enhance its mechanical properties. The swelling ratio was reduced to 4 (Fig. 4a, Fig. S21c–f, Supporting Information) and the elongation at break was increased to 1330 % (before swelling) and 570 % (after swelling) (Fig. 4b–c, Fig. S23a, Supporting Information).

Fig. 4.

a) Swelling ratio of SPAN with micro-nano gel and without micro-nano gel. b) Stress-strain curves of SPAN with micro-nano gel and without micro-nano gel before swelling. c) Stress-strain curves of SPAN with micro-nano gel and without micro-nano gel after swelling. d) Swelling ratio of pre-swelling SPAN and non pre-swelling SPAN in vivo for 14 days, the illustration is that digital photos of SPAN/LC toughly adhering on muscle after pre-swelling SPAN were implanted for 14 days. e) Digital photos of pre-swelling SPAN in vivo for 14 days. f) Images of non pre-swelling SPAN in vivo for 14 days. g) Adhesive stability of SPAN/LC to the tissue, the illustration is that the photograph of SPAN/LC tough adhering on muscle surface after the muscle adhered by SPAN/LC was placed in wet environment for 70 days. h) Microscope images of tissues surrounding pre-swelling SPAN implanted for 14 days. i) Microscope images of tissues surrounding non pre-swelling SPAN implanted for 14 days. j) Microscope images of rat subcutaneous muscle without hydrogel. Values in figure a, d and g show the mean and the standard deviation (n = 4).

This can be attributed to the large number of coiled SPAN chains stored in the micro-nano gel, which allows the elongation at break to be greatly increased (Fig. S23b Supporting Information) [52]. Simultaneously, the polymer chain of SPAN penetrates into the interior of the micro-nano gel and forms a chain entanglement with topological constraints, thus reducing swelling ratio of SPAN. Subsequently, SPAN was pre swollen to equilibrium in vitro, as a result, the swelling ratio of the SPAN in vivo was further limited to 1 (Fig. 4d), exhibiting strong adhesion in vivo. Simultaneously, no significant swelling phenomenon was observed after the pre-swelling SPAN was implanted in vivo for 14 days (Fig. 4e–f). Meanwhile, the embedded micro-nano gel acts as a multi-functional physical cross-linking point and stores coiled molecular chains, endowing SPAN with excellent mechanical properties even after swelling [52]. This performance will not affect its adhesion strength to the tissue. The adhesion strength to the tissue remains consistently above 125 kPa even after placing the tissue in sealed wet bag with phosphate-buffered saline solution (PBS) (pH = 6.8) for 70 days (Fig. 4g, Video S4, Fig. S23c–d Supporting Information). Additionally, pre-swelling SPAN showed lower immune response than non pre-swelling SPAN after the pre-swelling SPAN and non pre-swelling SPAN were implanted into subcutaneous of rats for 14 days, respectively (Fig. 4h–j, Figure S24, Supporting Information). The above data indicate that the synergistic effects of NHS ester-enhanced cohesion, dual covalent bonding, fatigue-resistant topological networks, micro-nano gel and pre-swelling strategy enhance adhesion stability and anti-fatigue ability, while reducing the rejection reaction of the tissue. These excellent performances can completely ensure the long-term stable adhesion of bioelectrodes to the wet dynamic tissues and the collection of accurate electrophysiological signals.

2.4. Instant adhesion of SPAN/LC

Instant adhesion ensures that anesthesia time is reduced during electrode implantation, avoiding neuronal apoptosis and brain damage [18,37]. Employing high concentration of chitosan (10 %) enables rapid formation of physical cross-links with tissues and hydrogels within 10 seconds, such as through hydrogen bonding and electrostatic interaction. In contrary, when a 2 % concentration of chitosan was employed, no obvious adhesion formed (Video S5-S6, Supporting Information), which further proved that the high concentration of chitosan is necessary for the formation of quick adhesion. it's important to note that the adhesive formed within 10 seconds represents an initial adhesion that is instant but not stable and tough. Notably, SPAN/LC produces an adhesion strength exceeding 60 kPa within 1 min (Fig. 5a, Fig. S25, Supporting Information). After 10 min, the adhesion strength surpasses 120 kPa. Additionally, in practical applications, it desires that the bioelectrodes possess tough adhesion to diverse wet dynamic tissues to monitor their electrophysiological signals. The SPAN/LC can form tough adhesion with different tissues, including skin, fat, stomach, heart, spinal cord and muscle. The adhesion strength of SPAN/LC to these tissues reaches 195.5 kPa, 203.4 kPa, 156.9 kPa, 139.3 kPa, 136.5 kPa, and 123.2 kPa, respectively (Fig. 5b, Figure S26-S29, Supporting Information). However, the adhesion strength of SPAN/LC (without NHS esters) and SPAN/LC (without NHS esters and coupling agents) to different tissues decreased sharply, indicating that weak cohesion and physical topology are difficult to generate tough adhesion strength. Subsequently, we further investigate the effect of the initially applied pressure and the degree of deacetylation of chitosan on the adhesion strength. When the initially applied pressure and the degree of deacetylation are 1.5 kPa and >95 %, respectively, an adhesion strength of 143 kPa can be produced (Fig. 5c, Fig. S28, Supporting Information). These adhesive properties of SPAN/LC are sufficient to ensure long-term stable integration of bioelectrodes on wet dynamic tissue surfaces.

Fig. 5.

a) The adhesion strength of SPAN/LC to the porcine muscle after different times. b) The adhesion strength of SPAN/LC, SPAN/LC (without NHS ester) and SPAN/LC (without NHS ester and coupling agent) to different tissues. c) The effect of applied pressure to adhesion strength (porcine muscle). d) Comparison of the resistance of the SPAN/LC-PPy under different strains. e) Comparison of the resistance of SPAN/LC-PPy under different cyclic stretching/releasing. f) Electrical impedance spectra of the PPy and Au (Area: 0.2826 cm²). Values in figure a, b, c and d show the mean and the standard deviation (n = 4).

2.5. The electrical stability of SPAN/LC-PPy MEAs

Tissue adhesive microelectrode arrays was are then fabricated by employing the SPAN hydrogel and LC as the adhesive material. The detailed preparation process of SPAN/LC-PPy MEAs is shown in supporting information (Fig. S29, Supporting Information). Good tensile stability is important for the electrode to realize synchronous movement with the wet dynamic tissue and maintain stable electrical performance. No obvious change in resistance is observed when SPAN/LC-PPy MEAs are subject to different strains (20 %, 40 %, 60 %, 80 % and 100 %) (Fig. 5d, Fig. S30, Supporting Information), demonstrating the good stretchability of the SPAN/LC-PPy MEAs. Here, the SPAN/LC-PPy MEAs are prepared by a pre-stretching method (More information can be found in supporting information). When this structure is pulled, the wavy structure of PPy changes its amplitude accordingly to dissipate the applied stress without causing stress concentration on PPy (Fig. S31, Supporting Information), thereby ensuring that SPAN/LC-PPy MEAs can maintain stable electrical properties under high strain. Simultaneously, the SPAN/LC-PPy MEAs also demonstrate excellent mechanical cycling stability, as there is no noticeable change in resistance after the electrode undergoes 1000 cycles of stretching and releasing under a 30 % strain (tensile speed of 500 mm min⁻¹) (Fig. 5e). Such a superior stretching stability assures that the electrodes can accurately record electrophysiological signals when the electrodes move together with the tissue.

2.6. The electrochemical performance of SPAN/LC-PPy MEAs

The electrochemical performance of neural electrodes plays an important role in the transfer efficiency of electrophysiological signals at the electrode-tissue interface [[61], [62], [63]]. Thus, the electrochemical properties of the neural electrode are characterized in PBS (pH = 6.8). As shown in Fig. 5f, the interface impedance of the PPy is much lower than that of the widely used Au electrode (We selected an Au electrode of the same size as the control samples). The low interfacial impedance allows the PPy electrode to accurately record electrophysiological signals [62,64]. Moreover, the electrode can also transmit electrical signals from external devices to the biological nervous system. These electrical signals can be used to stimulate biological systems to treat nervous diseases. The charge storage capacity is often used to predict the stimulation capacity of the electrode [65,66]. PPy possesses a charge storage capability of ≈69.9 mC cm⁻² and 3.4 mC cm⁻² at scan rates of 50 mV s⁻¹ and 1 V s⁻¹ (Fig. S32, Supporting Information), respectively. In comparison, the Au electrode only exhibits a charge storage capability of ≈5 mC cm⁻² and 0.7 mC cm⁻² at the same scan rates. As mentioned above, the stable tough adhesion and excellent electrical properties indicate that SPAN/LC-PPy MEAs are potential candidates for bioelectronics applications.

2.7. Bioelectronic functionalities of SPAN/LC-PPy MEAs

Finally, to verify the bioelectronic function of the adhesive electrode, the SPAN/LC-PPy MEAs are adhered to the rat's subcutaneous muscle to record the rat's electromyography (EMG) signals (Fig. S33a–b, Supporting Information). This SPAN/LC-PPy MEAs can accurately record the EMG signals of the rat following electrical stimulating of the ischiadic nerve (Electrical stimulation intensity is 300 μA, and electrical stimulating pulse width is 20 μs) (Fig. S33c, Supporting Information). It demonstrates that the SPAN/LC-PPy MEAs stably adhere to the wet moving tissue surface and accurately record action potentials. Subsequently, to verify the long-term adhesive stability of the electrode, the adhesive electrodes are implanted into the subcutaneous muscle of a rat (Fig. 6a), and demonstrate stable adhesion for 35 days (Fig. 6b, Fig. S34, Supporting Information). Simultaneously, high-amplitude EMG signals can still be recorded after 35 days (Fig. 6c–d), and the obtained EMG signal exhibits high SNRs (Fig. 6e). This is significance for monitoring and diagnosis of muscle-related diseases and precise manipulation of prostheses. In addition, we verify the in-situ stimulating capability of the SPAN/LC-PPy and PDMS/Au electrodes by stimulating the sciatic nerve of the rat to induce the flexion of left hind leg (Fig. 6f–g). It can be found that the left hind leg bending angle of the rat induced by the adhesive electrode is larger than that of the rat induced by the Au electrode under the same stimulating conditions (Fig. 6h). The main reason is that the electrode can conformally adhere to the soft sciatic nerve, possess low interface impedance and high charge storage capacity. These excellent performances endow adhesive electrode with good stimulating ability. This facilitates efficient and stable signal transmission between the electrode and tissue interface.

Fig. 6.

a) Schematic diagram of adhesive electrodes adhered to subcutaneous muscles. b) Photograph of the adhesive electrode implanted in the rat subcutaneous muscle for 35 days. c) Schematic photo of long-term EMG recording of adhesive electrodes. d) EMG signals collected by the adhesive electrodes implanted in the rat at different times. e) Signal-to-noise ratios of EMG signal after implantation at different times. f) Photograph of the rat left hind leg flexed after the ischiadic nerve was stimulated by SPAN/LC-PPy. g) Photograph of the rat left hind leg flexed after the ischiadic nerve is stimulated by the PDMS/Au. h) The bending angle of the rat left hind leg after the ischiadic nerve was stimulated by SPAN/LC-PPy and PDMS/Au, respectively. Values in figure e and h show the mean and the standard deviation (n = 4).

Concurrently, this electrode is further adhered to the heart of a living rat to monitor its electrocardiography (ECG) signals (Fig. S35a, Supporting Information). SPAN/LC-PPy MEAs can record stable and high-quality ECG signals once it adhered to the rat's heart (Fig. S35b, Supporting Information). The ECG signals have no observable baseline beating or high-amplitude noise. This result further validates that the electrode's ability to form conformal adhesion to the wet moving tissues even under significant strains. The rapid and tough adhesion, as well as good stretchability, position these electrodes as promising candidates for in vivo electrode-neural interface.

3. Conclusions

In summary, we develop an innovative hydrogel with both double-covalent bond and topological adhesion (Forming covalent bond inside tissue) by using SPAN hydrogel as substrate and the LC as the adhesive layer. Employing SPAN/LC as adhesive material, tissue adhesive SPAN/LC-PPy MEAs have been fabricated. The adhesion strength of SPAN/LC to tissues surpasses that of reported hydrogel adhesives (130 kPa) and reaches 290 kPa. This enhanced performance is attributed to the diffusion of LC into SPAN and tissue, bonding them together by topological adhesion. Simultaneously, the NHS ester groups and amino groups of SPAN form double-covalent adhesion with tissues. Double-covalent adhesion and topological adhesion endow SPAN/LC with excellent ability of anti-fatigue adhesion. SPAN/LC stably adheres on the tissue surface and exhibits adhesion strength of 92 kPa and 32 kPa after being subjected to a 70 % strain and cyclic stretching of 3000 times at 30 % strain, respectively. Furthermore, micro-nano gel and pre-swelling enhance the elongation at break (1330 %) and limit swelling of SPAN in vivo (V/V₀ = 1), further increasing adhesion stability and biocompatibility. These excellent properties ensure that the bioelectrode can be stably and long-term integrated on the wet dynamic tissue surface. Simultaneously, the high stretchability of the obtained MEAs enables the electrode-tissue interface to stably transmit signal even when the devices move with the tissue (with no electrical conductivity change observed under 100 % stretch). Finally, the adhesive stability and bioelectronic functionality of the SPAN/LC-PPy MEAs in living tissues are systematically validated. This electrode can be implanted in rats for 35 days and obtain EMG signals with high SNRs. Meanwhile, the device possesses good stimulating ability, and stably recorded ECG signals on the beating heart. These abilities are great significance for the precise control of neural prostheses, diagnosis, and treatment of neurological diseases. Moreover, the success of this adhesive microelectrode offers a promising solution to establishing long-term stable and conformal electrode-tissue interface.

CRediT authorship contribution statement

Gongwei Tian: Writing – review & editing, Writing – original draft, Software, Investigation, Formal analysis, Data curation, Conceptualization. Ming Zhu: Writing – review & editing. Jianhui Chen: Software, Formal analysis. Cuiyuan Liang: Data curation. Qinyi Zhao: Data curation, Conceptualization. Dan Yang: Investigation. Yan Liu: Data curation. Shuanglong Tang: Formal analysis. Jianping Huang: Data curation. Zhiyuan Liu: Supervision. Weihong Lu: Writing – original draft, Validation. Meifang Zhu: Writing – review & editing, Writing – original draft. Wei Yan: Writing – review & editing, Writing – original draft. Dianpeng Qi: Writing – review & editing, Writing – original draft, Supervision, Resources, Funding acquisition.

Data availability statement

The data presented in the study are included in the article and Supporting Information, further inquiries can be directed to the corresponding author.

Ethics approval and consent to participate

All animal experimental protocols were performed in accordance with the guidelines of the Harbin Institute of Technology ethics review board (Approval Code: IACUC-2024022).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: