Influence of Carbon Nanotube Hydrodynamic Radius on Electrical Conductivity in Photoresists ^†

Abstract

Tuning the electrical properties of biocompatible materials with minimal amounts of nanofiller presents a significant challenge in neuroimplant development. This study investigates the influence of single-walled carbon nanotube (SWCNT) agglomerate size on the electrical conductivity of a bioinspired biopolymer composite. The composites were fabricated via photolithography. We analyzed the effect of ultrasonic homogenization on the size distribution of SWCNT bundles. Our results demonstrate that the degree of nanotube dispersion is critical for determining electrical conductivity. The highest conductivity was achieved with an average bundle size of 95 µm and a defect level of no more than 0.143, as measured by the I_D/I_G+ band ratio using Raman spectroscopy. We attribute this to the formation of an interconnected percolation network within the biopolymer matrix. These findings demonstrate a viable approach for controlling the conductive properties of bioinspired composites.

1. Introduction

The development of neuroimplantable devices is a key frontier in modern bioelectronics. Neuroimplants are devices integrated into the nervous system that enable stable, bidirectional communication between biological tissues and electronic systems. They primarily expand therapeutic possibilities by regenerating damaged nerves, restoring lost bodily functions, and blocking chronic pain signals [1].

Neuroimplants must meet biosafety requirements and ensure the long-term functionality of the device within the body [2]. Traditional electrode materials, such as platinum, gold, and titanium, are perceived by the body as foreign substances. Consequently, glial scarring occurs, greatly reducing electrical conductivity due to mechanical incompatibility [3,4,5]. Composite hydrogels mimic the mechanical and adhesive properties of the extracellular matrix [6]. Several options provide the hydrogel’s electrical conductivity: conductive polymers, nanoparticles, or ionogels [7].

This paper proposes a bio-inspired electrically conductive composite hydrogel based on bovine serum albumin (BSA), chitosan, eosin Y, and modified single-walled carbon nanotubes (SWCNTs). In contrast to the previously developed material [8], this work also includes type II collagen in the biopolymer matrix, which, when used with chitosan, allows control of another important property of the hydrogel, such as the degree of swelling [9], and also contributes to improved compatibility with native tissue [10,11].

The primary objective of this study is to investigate the influence of SWCNT bundle dimensions on the electrical conductivity of the material. The morphology of the SWCNTs network within the biopolymer matrix is the key parameter governing the composite’s electrical properties. Strong van der Waals interactions between individual nanotubes promote the formation of bundles and agglomerates of varying sizes and connectivity that can be regulated by ultrasonic processing [12]. During photopolymerization under laser irradiation, SWCNTs can weld together at defect sites, forming a percolating conductive network [13] throughout the polymer matrix. In this study, we regulate the hydrodynamic radius of SWCNT bundles to optimize the material’s electrical conductivity without increasing the nanofiller concentration.

2. Materials and Methods

2.1. Preparation of the Photocurable Hydrogel

Photocurable hydrogel composites were produced according to the method described in [8]. In this work, in addition to bovine serum albumin (BSA) (BioClot, Aidenbach, Germany), chitosan (Bioprogress LLC, Losino-Petrovsky, Russia), eosin Y (Agat-Med, Moscow, Russia), and single-walled carbon nanotubes (SWCNTs) (TUBALL^TM, OCSiAl, Moscow, Russia), type II collagen (MacMedi LLC, Moscow, Russia) was also used.

SWCNTs were used, as in [8]. In this work, their bundle size was controlled, and aqueous nanotube dispersions were subjected to ultrasonic homogenization using a Q700 sonicator (Qsonica, Newtown, CT, USA). During sample processing, the duration and power of exposure were varied. As a result, four distinct dispersions were prepared with predominant hydrodynamic bundle radii of approximately 250 μm (5 min, 50 W), 95 μm (15 min, 110 W, filter with a pore size of 100 μm (Minimed, LTD, Bryansk, Russia)), 28 μm (30 min, 500 W, filter with a pore size of 100 μm), and 0.4 μm (45 min, 500 W, filter with a pore size of 10 μm (Minimed, LTD, Bryansk, Russia)). The sonication process was performed with cooling to prevent overheating. To evaluate the influence of SWCNT bundle size on electrical conductivity properties after filtration, all samples were aligned by optical density at a wavelength of 550 nm using a GENESYS UV-Vis-Nir spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). The hydrodynamic radius of SWCNTs was characterized by Dynamic Light Scattering (DLS) using a Photocor Complex particle size analyzer (Photocor, Moscow, Russia). The final composition of all hydrogels was identical: 0.03 wt.% SWCNTs, 5 wt.% BSA, 10 wt.% chitosan, 2.5 wt.% collagen, and 0.1 wt.% eosin Y.

2.2. Electrical Conductivity and Swelling Rate of Composite Hydrogels Investigation

The specific electrical conductivity of the fabricated composite material structures was evaluated using the four-point probe method [8]. For hydrogels, a key property for ensuring tight contact between the electrode and nerve is the degree of swelling, given the nerve’s mobility. Vibra AJ-420CE scales (Shinko Denshi Co., Ltd., Tokyo, Japan) were used to determine mass. The degree of swelling was determined by the ratio S = m_w/m₀ × 100%, with m_w wet sample weight and m₀ dry sample weight.

3. Results and Discussion

3.1. Investigation of Hydrodynamic Radius

The change in the hydrodynamic radius distribution of SWCNT bundles as a function of ultrasonic homogenization time was analyzed using the DLS method (Figure 1). Increasing the processing time was found to reduce the size of the SWCNT bundles (

Table 1. Distribution analysis DLS for SWCNTs.

Ultrasonic Homogenization Time t, min	Sonication Power P, W	Sonication Energy E, kJ	SWCNTs	Type of Particles	Hydrodynamic Radius R, µm	Standard Deviation, µm	Contribution, %
5	50	15	SWCNTs (250 µm)	Bundles	2.5	1.9	15
				Bundles	250.0	53.0	85
15	110	100	SWCNTs (95 µm)	Separated nanoparticles	0.3	0.1	3
				Bundles	95.0	34.0	97
30	500	900	SWCNTs (28 µm)	Separated nanoparticles	0.3	0.1	1
				Bundles	28.0	23.0	99
45	500	1350	SWCNTs (0.4 µm)	Separated nanoparticles	0.1	0.1	3
				Separated nanoparticles	0.4	0.1	73
				Bundles	8.6	2.8	24

). After 5 min of treatment, the dispersion contained large SWCNT agglomerates with a predominant radius of 250 µm (85%), while the smallest agglomerates were 2.5 µm. No individual SWCNTs were detected. This sample is characterized by an extremely high degree of nanotube agglomeration.

Increasing the ultrasonic treatment time to 15 min significantly reduced the size of the primary bundles to 95.0 µm, confirming the effectiveness of ultrasound in breaking down large structures. A weak signal (3%) from particles with a radius of 0.3 µm was also present, indicating the presence of individual single-walled carbon nanotubes (SWCNTs). After 30 min of treatment, the bundle size continued to decrease; however, the fraction of small particles (0.3 µm) remained insignificant.

The most substantial changes were observed after 45 min of ultrasonic treatment. The dispersion was dominated by particles with radii in the 0.1–0.4 µm range, comparable to the expected hydrodynamic radius of individual SWCNTs. The proportion of large bundles measuring 8.6 µm decreased to 24%.

Thus, the results clearly demonstrate that prolonged ultrasonic treatment is essential for achieving a high degree of SWCNT dispersion. It has been established that the hydrodynamic radius (R) decreases according to a power function of the processing time (t). R(t)~E^−m, with a degree index equal to 0.5. This indicates a rupture caused by stress along the SWCNTs [12]. Sonication energy is defined by the expression E = t × P, with P the sonication power. The sample of SWCNTs (0.4 µm) deviates from this dependence because it is strongly influenced by filtering. In our previous work, we evaluated the defectiveness of SWCNTs using a sonication energy of 720 kJ. The I_D/I_G+ band ratio in Raman spectroscopy was 0.143, indicating the presence of defects, which may be due to carboxyl groups or arise from ultrasonic treatment. To further evaluate the influence of SWCNT bundle dimensions and defectiveness on electrical conductivity, studies were conducted on photocured composite hydrogels using the obtained nanotubes.

3.2. Electrical Conductivity and Swelling Rate of the Bioinspired Material

The electrical conductivity of the bioinspired material was investigated using the four-point probe method (

Table 2. Electrical conductivity of samples.

Photoresist	Hydrodynamic Radius, µm	Specific Conductivity, mS∙cm−1
1	250.0	10
2	95.0	54
3	28.0	27
4	0.4	14

).

Sample 1, containing the largest bundles of SWCNTs (250 µm), exhibited the lowest conductivity. The SWCNTs are grouped into dense, isolated clumps, which explains this effect. A connected network of SWCNTs does not form within the dielectric biopolymer matrix. The maximum conductivity was achieved with SWCNTs (95 µm) that were subjected to 15 min of ultrasonic treatment. In this sample, large aggregates were broken down and individual SWCNTs were present, which promoted particle distribution within the matrix and the formation of multiple contacts. This leads to the formation of a branched percolation network and increased conductivity [13]. The defectiveness of these two samples’ SWCNTs was estimated in relation to the I_D/I_G+ bands at a level below 0.143 [8]. With continued ultrasonic treatment, the number of defects increases [12]. The defectiveness of the SWCNTs in the other two samples was already estimated to be above 0.143 in relation to the I_D/I_G+ bands. With further dispersion, the electrical conductivity of the samples decreases. This effect indicates that the short length and high degree of isolation of individual SWCNTs within the biopolymer matrix hinder the formation of stable contacts between them. Another possible reason is an increase in defects above 0.143 in relation to the strips [12]. Controlling such key factors as defectiveness and hydrodynamic radius simultaneously allows for adjustment of the composite’s electrical conductivity properties. All samples demonstrated sufficient swelling, with an average of 170 ± 20%. The impact of the ratio between collagen and chitosan was evident in this result, as previously reported in [9]. Notably, the variation in the hydrodynamic radii of the utilized SWCNTs did not influence this value.

4. Conclusions

This study demonstrates that the electrical conductivity of the bio-inspired composite is strongly governed by the size of SWCNT agglomerates and degree of defectiveness of their structure, which in turn is determined by the ultrasonic homogenization regime. Optimal performance is achieved at an average bundle radius of approximately 95 µm and defectiveness not exceeding 0.143 in relation to the strips I_D/I_G+, enabling the formation of a stable percolating network [13] and enhancing the material’s specific conductivity. In contrast, excessive dispersion leads to the presence of numerous isolated SWCNTs and an increase in their defect rate [12] that cannot form a continuous conductive pathway. Therefore, controlling the degree of SWCNT agglomeration and defectiveness represents an effective strategy for tailoring the electrical properties of biopolymer composites. These findings hold significant practical relevance for the design and optimization of neuroimplantable devices. At the same time, the use of a combination of collagen and chitosan made it possible to achieve a swelling rate of approximately 170%, which is important for ensuring close contact between the mobile nerve and the electrode.