Bioadhesive polymer semiconductors and transistors for intimate biointerfaces

Abstract

The use of bioelectronic devices relies on direct contact with soft biotissues. For transistor-type bioelectronic devices, the semiconductors that need to have direct interfacing with biotissues for effective signal transduction do not adhere well with wet tissues, thereby limiting the stability and conformability at the interface. We report a bioadhesive polymer semiconductor through a double-network structure formed by a bioadhesive brush polymer and a redox-active semiconducting polymer. The resulting semiconducting film can form rapid and strong adhesion with wet tissue surfaces together with high charge-carrier mobility of ~1 square centimeter per volt per second, high stretchability, and good biocompatibility. Further fabrication of a fully bioadhesive transistor sensor enabled us to produce high-quality and stable electrophysiological recordings on an isolated rat heart and in vivo rat muscles.

Fig. 1.. Bioadhesive polymer semiconductors for electrochemical-transistor-based tissue interfacing.

(A) Use of electrochemical transistors at tissue interfaces for biosensing with built-in amplification, for which bio-signals couple into the polymer semiconductor channels through direct tissue contact. (B to C) Conventional device attachment methods on tissue surfaces, such as peripheral suturing and applying a non-electrical adhesive layer, and their corresponding limitations. (D) Direct adhesive attachment achieved by a bioadhesive polymer semiconducting (BASC) channel and a wet tissue surface. The double-network design of the BASC contains a semiconducting polymer and an adhesive polymer achieving covalent bonding with tissue surfaces. (E) Chemical structures of the adhesive monomers with even longer linear side chains terminated with NHS ester and COOH groups, and the schematic of formed brush-architectured bioadhesive polymer (BAP). (F) Chemical structure and a schematic of the utilized polymer semiconductor p(g2T-T) with long linear side chains. (G) Photograph showing a fully-bioadhesive OECT with BASC channel adhered to a rat heart for ECG recording, which can stand for mechanical agitations.

Fig. 2.. Adhesive-relevant properties of the BASC films.

(A to B) Water content measured through gravimetric analysis (A) and dimensional swelling (B) of the BAPs with three types of side-chain designs when soaked in PBS solution over time. (C) SEM images showing the microscale features of a BAP-COOH film and a PAAc film in the dry state. (D) AFM phase images showing the top and bottom surfaces of a BASC film. (E) 3D schematic morphology of a BASC film. (F) XPS-measured ratios between the S element and the N element from the top to the bottom surface of a BASC film. (G) Rheological measurement of the BASC polymer in the dry state. (H) 180-degree peel test (ASTM D3330) for interfacial toughness measurement on rigid substrates. (I) Interfacial toughness of the adhesion between a BASC film and an amine-functionalized, dry glass substrate, in comparison to a neat p(g2T-T) film, and a BAP film. (J) On amine-functionalized glass substrates, interfacial toughness achieved by BASC films and BAP (including PAAc) films with different types of side chains. The two dashed lines mark the levels of interfacial toughness for the BASC and BAP films from (I). (K) 180-degree peel test (ASTM F2256) for interfacial toughness measurement on bio-tissues. (L) Interfacial toughness, shear strength, and tensile strength of the adhesion between wet porcine muscle tissues and a BASC film, a BASC-COOH film, a BASC-NHS film, and a neat p(g2T-T) film, respectively. (M) Interfacial toughness, shear strength, and tensile strength achieved by BASC films on various wet tissue surfaces. Values in I-J, L-M represent the mean and the standard deviation (n = 3). Statistical significance and P values are determined by two-sided Student’s t-test: *P<0.05; **P<0.01.

Fig. 3.. Electrical and structural characterizations of the BASC film.

(A) Setup for the OECT-based characterizations. (B) Transfer curves from a BASC film and a p(g2T-T) film serving as the OECT channel (V_g: gate voltage, I_d: drain current, V_d: drain voltage, g_m: transconductance). (C) Charge-carrier mobility and g_m for the BASC and p(g2T-T) films. Values represent the mean and the standard deviation (n = 5). (D) Response speed measurement with applied gate voltage pulse and drain current response. (E) 2D GIXD patterns of the BASC and p(g2T-T) films. (F) 1D linecuts in the out-of-plane direction (top) and in-plane direction (bottom) of a BASC and a p(g2T-T) film. (G) Normalized UV-vis absorption of a BASC and a p(g2T-T) film. (H) Schematic diagram showing the setup for characterizing the influence of the device-tissue distance caused by the use of a separate bioadhesive layer. (I) AC signal input (grey) and acquired signals between the p(g2T-T) electrodes that are covered with a layer of BAP with two different thicknesses. The thin adhesive layer (d = 3.8 μm) decreases the signal amplitude by 15 % while the thick adhesive layer (d = 300 μm) decreases the amplitude by 50 %.

Fig. 4.. Abrasion resistance, stretchability, and biocompatibility of BASC films.

(A) Schematic diagram illustrating physical abrasions that can happen on the surfaces of implantable devices. (B) Photographs showing a BASC film and a p(g2T-T) film before and after abrasion by a PTFE-covered glass plate under 1 kPa for 500 cycles. The arrows indicate the direction of the abrasion. (C) Changes of OECT on-current from the two films after the abrasion cycles along the charge transport direction. Values represent the mean and the standard deviation (n = 4-5). (D) Schematic illustrating a BASC film under stretching. The bottom photograph shows a stretched BASC film on a PDMS substrate at 100 % strain. (E) Optical microscopy and AFM images showing a BASC film stretched to 100 % strain without forming cracks. (F) Transfer curves of BASC films in the pristine state, and stretched to 100 % strain for 1 and 100 cycles, which were measured with V_d = −0.6 V. (G to H) Masson’s trichrome staining of surrounded tissues of a subcutaneously implanted BASC film (G) and a SEBS film (control, H) after one month in mice. (I) Calculated fibrotic capsule thickness. (J to M) Immunofluorescence staining of α-SMA for fibroblasts (yellow, J and L) and CD68 for macrophages (red, K and M). Statistical significance and P values are determined by two-sided Student’s t-test: ns, not significant; **P<0.01.

Fig. 5.. Fully-bioadhesive OECT sensor and the use for ex vivo and in vivo electrophysiological recording.

(A) Device structure and picture. Scale bar: 5 mm. (B) Transfer curves for a fully-bioadhesive OECT under 0 % and 50 % strains. (C) Shear strength of the adhesion between a fully-bioadhesive OECT on the porcine muscle tissue, in comparison to an OECT with non-bioadhesive surface. (D) Schematic showing the use of the OECT sensor for ECG recording on a heart surface, and the circuit diagram. (E) Photographs showing a fully-bioadhesive OECT attached to an isolated rat heart surface maintaining stable contact during mechanical agitation. (F) Comparison of a non-bioadhesive OECT, for which capillary-based attachment cannot maintain conformable and stable contact. (G) ECG signals recorded by the fully-bioadhesive OECT on the LV. (H) ECG recording by the non-bioadhesive OECT, which ceased after the device detachment from the heart. (I) Schematic showing the use of the OECT sensor for EMG recording on the GM muscles upon stimulation of the sciatic nerve. (J) Photographs showing a fully-bioadhesive OECT attached to the GM muscle maintaining stable contact during mechanical agitation. (K) Comparison of a non-bioadhesive OECT, for which capillary-based attachment cannot maintain conformable and stable contact. (L) EMG signals recorded by the fully-bioadhesive OECT on the GM muscle. (M) EMG signals recorded by the non-bioadhesive OECT on the GM muscle, which ceased after the device detached from the muscles.