Synthesis and characterization of conductive, biodegradable, elastomeric polyurethanes for biomedical applications

Abstract

Biodegradable conductive polymers are currently of significant interest in tissue repair and regeneration, drug delivery and bioelectronics. However, biodegradable materials exhibiting both conductive and elastic properties have rarely been reported to date. To that end, an electrically conductive polyurethane (CPU) was synthesized from polycaprolactone diol, hexadiisocyanate and aniline trimer, and subsequently doped with (1S)-(+)-10-camphorsulfonic acid (CSA). All CPU films showed good elasticity within a 30% strain range. The electrical conductivity of the CPU films, as enhanced with increasing amounts of CSA, ranged from 2.7±0.9 × 10⁻¹⁰ to 4.4±0.6 × 10⁻⁷ S/cm in a dry state and 4.2±0.5 × 10⁻⁸ to 7.3±1.5 × 10⁻⁵ S/cm in a wet state. The redox peaks of a CPU1.5 film (molar ratio CSA:aniline trimer = 1.5:1) in the cyclic voltammogram confirmed the desired good electroactivity. The doped CPU film exhibited good electrical stability (87% of initial conductivity after 150 h charge) as measured in a cell culture medium. The degradation rates of CPU films increased with increasing CSA content in both phosphate buffer solution (PBS) and lipase/PBS solutions. After 7 days of enzymatic degradation, the conductivity of all CSA-doped CPU films had decreased to that of the undoped CPU film. Mouse 3T3 fibroblasts proliferated and spread on all CPU films. This developed biodegradable conductive polyurethane with good elasticity, electrical stability and biocompatibility, may find potential applications in tissue engineering, smart drug release, and electronics.

1. Introduction

Conductive polymer-based biomaterials have been used in a variety of biomedical fields.¹ For example, these polymers have been employed to control or regulate cellular behavior (e.g. cell proliferation, differentiation and metabolism) with/without electrical stimulation for tissue engineering applicatons.^2–4 Additionally, biomolecules entrapped within conductive polymers can be released with electrical stimulation.^5–7 They also can be utilized as bioactuators due to their redox activity^8,9 and as biosensors by entrapping targeted molecules.^10,11 However, current conductive polymer families have limitations to meet the needs of various biomedical applications. Specifically in soft tissue engineering, a conductive material with softness, elasticity (stretchability) and full degradation is expected. One major limitation of current conductive polymers is their poor flexibility.^12–16 It creates difficulties in processing and it also results in high mechanical stiffness, which negatively influences the mechano–biological interactions between cells and polymers in soft tissue engineering.^17,18 Blending a soft elastomer with an intrinsically conductive polymer is simple and effective to improve its flexibility and stretchability,^19–22 which has also been utilized in tissue engineering application.^23–27 For example, elastic poly(glycerol sebacate) or poly(L-lactide-co-ε-caprolactone) was blended with polyaniline as a soft conductive scaffold for cardiac or skeletal tissue repair.^23,24 Polyurethane elastomer was also mixed with polypyrrole or polythiophene derivative to yield an electroactive composite with improved mechanical resilience for tissue engineering applications.^25–27 However, these non-biodegradable conductive polymers were dispersed in the insulating polymer matrices without covalent bonding, which may lead to poor controllability in mechanics and conductivity due to the possible immiscibility of two polymers. Furthermore, after degradation of the biodegradable polymer, the non-degradable conductive polymers still exist in the body, and may induce chronic inflammation and infection, then implant failure. Direct conjugation of appropriate conductive oligomers into polymer backbones may achieve a conductive elastomer with the desirable biodegradable, electrical and mechanical properties, which may address the above concerns.

In this study, we utilized polyurethane chemistry to combine biodegradable soft segments and conductive oligomers into a polymer chain using diisocyanate. Specifically, a biodegradable conductive polyurethane elastomer was synthesized from biodegradable polycaprolactone diol and conductive aniline trimer with hexadiisocyanate. Linear polyurethane was selected because of its well-known flexibility, biocompatibility, biodegradability and processability.^28,29 Aniline trimer has a well-defined electroactive structure, and it can easily be eliminated by macrophages in vivo.^30–32 A common dopant, (1S)-(+)-10-camphorsulfonic acid (CSA), was used to dope the synthesized conductive polyurethane. The conductive polyurethane films were fabricated by solvent-evaporation, and their mechanical, electrical and biodegradable properties were characterized. Electrical stability of the films was assessed with degradation and in cell culture medium with long-time charging, respectively. Cytocompatibility evaluation of the conductive polyurethane film was conducted using mouse 3T3 fibroblasts.

2. Materials and Methods

2.1. Materials

Polycaprolactone diol (PCL, number average molecular weight =2000, Sigma-Aldrich) was dried in a vacuum oven at 60 °C to remove residual water before use. Putrescine (Sigma) and 1, 6-hexamethylene diisocyanate (HDI, Sigma) were purified by distillation prior to use. CSA (Sigma), stannous octoate (Sn(Oct)₂, Sigma), p-phenylenediamine (Sigma), 4-fluoronitrobenzene (Sigma), triethylamine (TEA, Sigma), tin granular (Sigma), ammonium persulfate (Sigma), hydrochloric acid (HCl, Sigma), sodium hydroxide (NaOH, Sigma), anhydrous dimethyl sulfoxide (DMSO, Sigma), acetone (Sigma), 1,1,1,3,3,3-hexafluoroisopropanol (HFIP, Oakwood Product), dimethylformamide (DMF, Sigma), hexamethyldisilazane (HMDS, Sigma) and lipase from Thermomyces lanuginosus (≥100,000 U/g, Sigma) were used as received.

2.2. Synthesis of oxidized aniline trimer

p-Phenylenediamine (1.54 g, 14.2 mmol), 4-fluoronitrobenzene (5.06 g, 3.78 mL, 35.6 mmol), and triethylamine (2.88 g, 3.97 mL, 28.5 mmol), were initially mixed in a round-bottom flask containing DMSO at 125 °C for 3 days under an argon atmosphere, then cooled to room temperature. Concentrated HCl was then quickly added and a red precipitate was formed. This precipitate was collected and subsequently dissolved in concentrated HCl along with granulated tin prior to refluxing for 5 h. After cooling to room temperature and washing in succession by concentrated HCl and 5 M NaOH, a whitish-blue solid was collected. This solid was then completely dissolved in ethanol/acetone (1/1, v/v) and 1 M HCl. Ammonium persulfate (1.98 g, 8.67 mmol) was added and stirred in a cold bath (−17 °C) for 10 min. The resultant blue precipitate was filtered, washed by DI water and dried overnight resulting in pure oxidized aniline trimer (2.31 g, dark-blue solid).

Chemical structure characterization of the oxidized aniline trimer, possessing two NH₂ end groups, is as follows: ¹H NMR (DMSO, 500 MHz): δ = 5.43 (s, 4 H), 6.60–6.79 (m, 4 H), 6.89–7.05 (m, 4 H). ¹³C NMR (CDCl₃, 125 MHz): δ = 114.0, 123.0, 124.1, 124.3, 135.2, 136.8, 139.2, 139.3, 147.6, 147.8, 155.1. IR (neat, cm⁻¹): 3379, 3309, 3206, 1630, 1542, 1318, 1166, 984, 830, 699, 541, 506, 411. HRMS m/z (ESI) calculated for C₁₈H₁₇N₄⁺(M + H)⁺ 289.1448, found 289.1443.

2.3. Synthesis of conductive polyurethane (CPU)

The CPU polymer was synthesized using PCL, HDI and aniline trimer via a two-step process (Fig. 1).²⁸ PCL (3.1 mmol) was dissolved initially in DMSO at 70° C in a 3-neck flask under N₂ protection. HDI (6.2 mmol) was then added into the flask, followed by 3 drops of Sn(Oct)₂ catalyst. After 3 h reaction at 70 °C, the prepolymer solution was cooled to room temperature, the aniline trimer/DMSO solution was then added dropwise to the prepolymer solution in the flask. The molar ratio of PCL/HDI/aniline trimer was fixed as 1:2:1. The final polymer concentration was 4% (w/v). The reaction was carried out for 18 h at room temperature. The polymer was then precipitated in distilled water, rinsed with ethanol to remove unreacted components, and dried in a vacuum oven at 60 °C for 3 days. The yield of the conductive polyurethane was 96% of the total feeding amounts of PCL, HDI, and aniline trimer.

Figure 1.

Synthesis of a biodegradable conductive polyurethane (CPU).

2.4. Fabrication of CSA doped CPU films

The synthesized CPU polymer was dissolved in HFIP at a concentration of 2 % (w/v). The CSA dopant was mixed with the CPU polymer in HFIP at different molar ratios of 0.5/1, 1/1 and 1.5/1 (CSA:aniline trimer), which were referred as CPU0.5, CPU1 and CPU1.5, respectively (Table 1). The CPU/CSA/HFIP solution was then poured into a Teflon dish in the absence of bubble formation. After complete HFIP evaporation, the films were dried in a vacuum oven at 60°C for 3 days. The CPU film without CSA dopant was a control.

Table 1.

Polymer characterization^*

Samples	Molar ratio of CSA:aniline trimer	Water absorption (%)	Initial modulus (MPa)	Peak stress (MPa)	Breaking strain (%)	Instant recovery (%)	Conductivity (S/cm)
							Dry state	Wet state
CPU	0:1	6±2a	7±1a	17.9±2.0a	728±88a	99±1	2.7±0.9 ×10−10 a		4.2±0.5 ×10−8 a
CPU0.5	0.5:1	9±1b	13±6b	7.3±1.0b	288±37b	98±1	4.0±0.7 ×10−9 b		1.8±0.6 ×10−7 b
CPU1	1:1	8±3b	25±3c	5.0±1.4c	238±66b	98±1	5.0±1.8 ×10−8 c		5.5±1.9 ×10−6 c
CPU1.5	1.5:1	11±2c	35±11c	3.1±0.3d	75±18c	97±2	4.4±0.6 ×10−7 d		7.3±1.5 ×10−5 d

^*a, b ,c, d and e represent significantly different groups for each characteristic.

2.5. CPU film characterization

Fourier transform infrared (FTIR) spectra were recorded on a Nicolet 6700 (Thermo Scientific, Germany) spectrometer to verify the chemical structure of the CPU polymers. The UV-visible spectra of CPU in DMF were recorded on a UV-vis spectrometer (Perkin-Elmer Lambda 35). For water absorption, the weighed CPU films (W₀) were immersed in a phosphate buffer solution (PBS, Sigma) at 37°C. After 24 h, the films were weighed (W₁) after removal of the surface water by filter paper. The water absorption (n=3 for each CPU film) was calculated as (W₁-W₀)/W₀ × 100%. The surface morphologies of CPU films were observed using a scanning electronic microscope (SEM, Hitachi S-4800 HRSEM). One-dimensional X-ray diffraction (1-D XRD) measurements of CPU films were carried out using a Bruker D8 Advance X-ray diffractometer.

2.6. Electrical conductivity and electrochemical measurements

The electrical conductivities of CPU films were measured in both dry and wet (24h PBS immersion) states at room temperature using a standard four-probe technique.³³ The various CPU films were placed under a home-made four-point probe and the corresponding voltage drops across the two inner probes were obtained under a direct current through the two outer probes. A PARSTAT 2273 potentiostat was employed for the measurement. The electrical conductivities (σ) of the samples were calculated using the equation: σ (S/cm) = (ln 2/π)(I/V)(1/t),^23,34 where I is the current through the outer probes in ampere, V is the voltage drop across the inner probes in volt, and t is the sample thickness in cm. Four measurements were taken for each sample group.

The cyclic voltammogram (CV) of the CPU polymer was recorded using the potentiostat instrument (PARSTAT 2273) to characterize their electrochemical properties.³⁴ The CSA-doped CPU polymer was coated on a platinum sheet as the working electrode. A platinum mesh was employed as the auxiliary electrode. The reference electrode was Ag/AgCl in 1M H₂SO₄ solution. The scanning potential ranged from −0.1 to 1.3 V with a scan rate of 50 mV/s.

2.7. Mechanical testing

The uniaxial tensile mechanical properties of the strips (2×20 mm strips; n=6) cut from the CPU films were measured on a MTS Insight Testing System with a 500 N load cell and a cross head rate of 10 mm/min following ASTM D638–03 standard.³⁵ The initial modulus (E) was determined by the slope (strain < 10%) of the stress-strain curve.³⁶ Instant recovery of CPU films was measured under the same conditions as described above. The strips (2 × 20 mm strips; n = 4) marked with two distal ends were stretched to 10% strain, held for 1 min, then released. The procedure was repeated 3 times. The instant strain recovery was calculated as (1-(L₁-L₀)/L₀) × 100%. For cyclic stretch testing, the samples (2 × 20 mm; n = 3) were stretched to a maximum strain of 30%, which was set because of the deformations (< 30%) of most tissues (e.g., cardiac muscle, bladder, and blood vessel) during normal activities,^37–39 then retracted back to the original length repeatedly for 10 cycles at a constant rate of 10 mm/min. The test was conducted on a uniaxial cyclic tensile test system as previously described with a 500 N load cell.^29,40

2.8. In vitro hydrolytic and enzymatic degradation

The in vitro degradation studies of the CPU films were carried out in 10 mL PBS, and in 2 mL PBS containing 100 U/mL lipase at 37°C, respectively.²⁹ The samples (n = 3) were cut from the CPU films and weighed (W₀), then immersed in PBS or lipase/PBS (refreshed every 3 days) at 37 °C. At each predetermined time point, the samples were picked and rinsed with deionized water, dried in a freeze-dryer for 3 days, then weighed (W₁). The mass remaining was calculated as W₁/W₀ × 100%.

The electrical conductivity changes of CPU films (n = 4) were measured after 3, 7, and 14 days degradation in 100 U/mL lipase/PBS solution at 37 °C. The degraded CPU films were rinsed with PBS solution, then their electrical conductivities in the wet state were measured using the four-probe technique as described above.

2.9. Electrical stability

To ascertain electrical stability under physiological conditions, the CPU1.5 film (n=3) was connected to an external power source (PARSTAT 2273) by NEM tape (Nisshin EM Co., Ltd) and immersed in Dulbecco’s Modified Eagle Medium (DMEM, Sigma) containing 0.05% sodium azide (Sigma) to prevent bacterial growth.¹⁴ Sample incubation was carried out for 150 h at 37 °C. A constant DC voltage of 100±2 mV was applied to the CPU1.5 film by a PARSTAT 2273 potentiostat, and the current-potential curve was recorded electronically. The measurement was undertaken in triplicate.

2.10. In vitro cellular growth on the CPU films

Mouse 3T3 fibroblasts (ATCC, Manassas, VA) were used to evaluate the cell compatibility of CPU films. The 6 mm CPU disks were punched from the CPU films using standard biopsy punches (6 mm, Miltex), and then sterilized by UV radiation for 1 h (two sides). The disks were rinsed 3 times with PBS prior to immersion in cell culture medium (DMEM supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin) overnight. The 3T3 fibroblasts (3 × 10³ cells per sample) were seeded on the disk surface in 96 well plates, then incubated at 37 °C for 1, 3, and 5 days. The cell culture medium was exchanged every 3 days. The cellular metabolic activity was measured using a mitochondrial activity assay (MTT, Sigma) at each time point.⁴¹ Five samples were used for each polymer group. Tissue culture polystyrene (TCPS) was employed as a positive control. To verify the results qualitatively based on the MTT assay and observe the cell morphology on CPU films, the cell-seeded CPU disks were fixed in 4% paraformaldehyde, dehydrated by an ascending gradient of ethanol (from 30% to 100%), and finally dried with HMDS. The cell morphology was observed using scanning electronic microscopy.

2.11. Statistical analysis

All data are presented as mean ± standard deviation (SD). All data were analyzed by one-way ANOVA followed by a post-hoc Tukey-Kramer test. Repeated measures ANOVA was used for hydrolytic and enzymatic degradation of CPU films using the Statistics Analysis System (SAS). P < 0.05 was considered statistically significant.

3. Results

3.1. CPU characterization

The chemical structure of the CPU polymer was verified using ATR-FTIR (Fig. 2). The urethane and urea groups were confirmed by specific peaks at 3320 cm⁻¹ (N-H stretching of urethane and urea groups), 1720 cm⁻¹ (C=O stretching of urethane and urea groups), and 2940 cm⁻¹ and 2860 cm⁻¹ (symmetric and asymmetric C-H stretching).⁴² The specific peaks for aniline trimer were located at 1590 cm⁻¹ and 1510 cm⁻¹ (C=C stretching of quinoid and benzenoid rings), 1300 cm⁻¹ (C-N stretching of aromatic amine), and 860 cm⁻¹ (C-H bending in benzenoid rings).⁴³ For CSA-doped CPU films, there was an additional peak at 1040 cm⁻¹ attributed to the asymmetrical stretching of the sulfonyl group interacting with aniline trimer in the polyurethane backbone, which is a characteristic of the doped form of polyaniline and its derivatives.⁴⁴

Figure 2.

ATR-FTIR spectra of a) CPU, b) CPU0.5, c) CPU1, and d) CPU1.5.

The bulk hydrophilicity of CPU films was characterized by water absorption (Table 1). The water absorption of CPU films increased with increasing CSA content. The CPU film without CSA dopant had the lowest water absorption at 6 ± 2%, while the CPU1.5 film had the highest water absorption at 11 ± 2% (p < 0.05). The undoped CPU film had a smooth surface and the CPU films with CSA dopant showed increased surface roughness (Supporting Information Fig. S1). The XRD spectrum of CPU1.5 showed two characteristic peaks at 21.9 2θ and 24.3 2θ corresponding to the diffraction of the 110 and 200 lattice planes of the crystalline PCL (Supporting Information Fig. S2).²⁹ However, the undoped CPU showed weaker and broader crystalline peaks for PCL, indicating lower crystallinity of undoped CPU compared with CPU1.5. In the UV-vis spectra of CPU and CPU1.5 [Fig. 3(A)], the undoped CPU polymer had absorption peaks at 329 nm and 528 nm, resulting from the π-π* transition in benzene rings and the π_b- π_q transition from the benzene ring to the quinoid ring in the aniline trimer segment, respectively.³² When CSA was added to the CPU polymer, the absorption peak at 528 nm disappeared and the absorption peak at 329 nm was blue-shifted to 303 nm. A shoulder band with a maximum at 492 nm from the polaron- π* transition appeared, and a broad peak at 827 nm was observed due to the localization of radical polaron along the doped CPU backbone.^30,45

Figure 3.

(A) UV-vis spectra of a) undoped CPU and b) CPU1.5 in DMF. (B) Cyclic voltammogram of CPU1.5 polymer on Pt electrode in 1.0 M H₂SO₄ using Ag/AgCl as a reference at a scan rate of 50 mV/s.

3.2. Electrical and electrochemical properties

The conductivity of CPU films in the dry state ranged from 2.7 ± 0.9×10⁻¹⁰ to 4.4 ± 0.6×10⁻⁷ S/cm (Table 1). There was a clear trend toward increased conductivity of CPU films with increasing CSA content. Compared with the dry state, the wet CPU films (PBS immersion) showed markedly increased conductivities, ranging from 4.2 ± 0.5×10⁻⁸ to 7.3 ± 1.5×10⁻⁵ S/cm.

The cyclic voltammogram of the CPU1.5 polymer is shown in Figure 3(B). The initial oxidation peak at 0.17 V corresponded to transition from the leucoemeraldine state to the emeraldine state, and the second oxidation peak at 0.82 V was attributed to the transition from the emeraldine state to the pernigraniline.³⁰ The well-defined redox peaks, corresponding to the transitions of three oxidation/reduction forms in CPU1.5, confirmed the good electroactivity of this composite polymer.

3.3. Mechanical properties

The stress-strain curves of the CPU films are presented in Figure 4(A) and their uniaxial mechanical properties are summarized in Table 1. The tensile strength of CPU films decreased from 17.9 ± 2.0 MPa to 3.1 ± 0.3 MPa with increased CSA content (p < 0.05). The same trend was seen for the ultimate elongation. The CPU had the highest ultimate elongation (728 ± 88%) while the CPU1.5 had the lowest (75 ± 18%) (p < 0.05). The initial moduli of the CPU films increased with increasing CSA content. The CPU1.5 film had the highest initial modulus (35 ± 11 MPa), and the CPU had the lowest initial modulus, at 7 ± 1 MPa (p < 0.05). The instant recovery of all CPU films was ≥ 97% after 3 cycles of stretching at 10% strain.

Figure 4.

(A) Stress-strain curves of a) CPU, b) CPU0.5, c) CPU1, and d) CPU1.5 films. (B) Cyclic stretching curves of CPU films at 30% deformation.

Cyclic stretching of CPU films at a maximum strain of 30% was performed to detect their resiliency [Fig. 4(B)]. A large hysteresis loop was observed in the first cycle for all the CPU films, followed by smaller hysteresis loops in the next nine cycles. All of the CPU films showed small irreversible deformations (~10%) at a maximum strain of 30%, except for the CPU1.5 film (~15%).

3.4. In vitro degradation

In vitro hydrolytic and enzymatic degradation of the CPU films were carried out in PBS and lipase/PBS solutions at 37 °C, respectively [Fig. 5(A, B)]. The degradation rate of the CPU films is relevant to the CSA amount. After 8 weeks degradation in PBS [Fig. 5(A)], the undoped CPU film maintained 98.6±0.4% of its initial weight, and the mass remaining of CPU0.5, CPU1 and CPU1.5 decreased from 97.8±0.2% to 96.6±0.4% (p < 0.05). In lipase/PBS solution [Fig. 5(B)], the CPU films showed faster degradation than in PBS solution. Within 14 days, the undoped CPU film had the lowest degradation rate (93.3 ± 0.3% mass remaining), while the CPU1.5 film had the highest degradation rate (80 ± 1.8% mass remaining) (p < 0.05).

Figure 5.

CPU film degradation. (A) Mass remaining of CPU films in PBS solution at 37 °C. (B) Mass remaining of CPU films in100 U/mL lipase/PBS solution at 37 °C. (C) Changes in electrical conductivities of CPU films during degradation in lipase/PBS solution for 14 days. * represents significant different groups (p < 0.05).

The electrical conductivity changes of CPU films with enzymatic degradation at days 3, 7 and 14 were shown in Figure 5(C). The conductivity of the undoped CPU film did not significantly change within 14 days degradation in lipase/PBS solution (4.2 ± 0.5×10⁻⁸ S/cm at day 0, 4.9 ± 0.8×10⁻⁸ S/cm at day 3, 4.0 ± 0.3×10⁻⁸ S/cm at day 7 and 3.6 ± 0.8×10⁻⁸ S/cm at day 14) (p > 0.05). However, the electrical conductivity of CSA-doped CPU films markedly decreased with enzymatic degradation. After 3 days, the electrical conductivities of the CPU0.5 (1.8 ± 0.6×10⁻⁷ S/cm at day 0), CPU1 (5.5 ± 1.9×10⁻⁶ S/cm at day 0) and CPU1.5 films (7.3 ± 1.5×10⁻⁵ S/cm at day 0) were 5.2 ± 1.1×10⁻⁸, 1.2 ± 0.3×10⁻⁶ and 5.9±2.1×10⁻⁶ S/cm, respectively. After 7 days, the conductivities of the CSA-doped CPU films had decreased to the same level as that of the undoped CPU film (4.5 ± 0.7×10⁻⁸, 3.5 ± 0.6×10⁻⁸, and 4.2 ± 1.1×10⁻⁸ S/cm for the CPU0.5, CPU1 and CPU1.5 films, respectively) (p > 0.05). After 14 days, the electrical conductivities of CSA-doped CPU films showed no significant difference from those after 7 days degradation (p > 0.05).

3.5. Electrical stability

The current conductivity of the CPU1.5 film increased from 15.7 nA to 21.6 nA in the first 14.6 h in DMEM, followed by a relatively quick decrease to 15.1 nA at 67 h, representing 96% of the initial current (Fig. 6). After that, the conductivity slowly decreased to 13.6 nA at 150 h, which was 87% of the initial value.

Figure 6.

Electrical stability. Relationship between electrical current and incubation time of a CPU1.5 film in cell culture medium charged with a fixed voltage.

3.6. In vitro cellular growth on the CPU films

The cell viability of the 3T3 fibroblasts seeded on the CPU films and the control TCPS increased from day 1 to day 5 (p < 0.05) [Fig. 7(A)]. There were no significant differences in cell proliferation between the CPU films and TCPS within 5 days of incubation (p > 0.05), except for the CPU1.5 film. The CPU1.5 film showed less cell viability than other samples and TCPS at day 3 and day 5 (p < 0.05). The SEM images of 3T3 fibroblasts cultured on CPU films at day 5 are shown in Figure 7(B). The cells with clear pseudopods spread on the CPU films. However, the cells on the CPU1.5 film had relatively lower density than those on the other CPU films, which was consistent with the cell viability results.

Figure 7.

Cytocompatibility of CPU films. (A) Metabolic index of mouse 3T3 fibroblasts seeded on polymer films (TCPS as a control) at days 1, 3 and 5. +, #: p < 0.05, CPU1.5 compared with other groups at day 3 and 5, respectively. (B) SEM images of mouse 3T3 fibroblasts cultured on the polymer films at day 5.

4. Discussion

Biodegradable elastomeric polyurethane has been employed for biomedical applications due to its tunable mechanical properties, processability, biodegradability, and good biocompatibility.^28,29 It has been processed into films,^46,47 electrospun fibers,^48,49 and porous scaffold,^28,50 for tissue engineering application as well as nano/micro particulates,^51–53 membranes,^54,55 and matrices⁵⁶ for controlled drug release. The biodegradable polyurethane has also been combined with traditional conductive polymers or organic additives to form PU-based conductive composites for tissue engineering.^{25–27,57,58} However, few studies were reported to combine conductive segments with a PU backbone to form a PU-based conductive polymer. In one relatively complicate synthesis process, a conductive polyurethane was synthesized based on poly(ethylene glycol) (PEG), PCL, isophorone diisocyanate (IPDI) and aniline pentamer via 3 steps (NCO-terminated prepolymer, aniline-dimer-ended polyurethane, then polyurethane containing aniline pentamer).^59,60 The PU containing the pentamer could not be directly processed into a film due to its poor solubility,¹² and it had to be blended with other polymers, as a dispersed additive, in the film or scaffold for further use.¹² In contrast to these prior studies, a typical two-step PU synthesis process was used to synthesize an elastomeric biodegradable polyurethane containing aniline trimer. Importantly, the resulting films exhibited good mechanical (soft and elastic) and electrical properties, and did not require a secondary polymer.

The conductivity of CPU films increased exponentially with increased amounts of CSA dopant, which was consistent with previous studies.^61,62 It can be attributed to the increased concentration of carriers hopping between polymer chains available for electrical conduction.⁶³ The conductivities of the synthesized CPU films in this study were comparable with polyurethane-siloxane-aniline tetramer (6.5 × 10⁻¹¹-1.3 × 10⁻⁵ S/cm),³² and had relatively lower conductivities than those of the polyurethane-aniline pentamer (on the order of 10⁻⁵ S/cm).⁵⁹ The relatively lower conductivity was attributed to a small number of aniline repeat units and the low content of the aniline trimer segment in the CPU backbone. The electrical conductivity of oligomers with 7 or 8 aniline repeat units can be equal to that of pure polyaniline.^64,65 The higher content of aniline oligomer in polymer can also enhance its conductivity due to better π-π stacking of the conductive moieties.^34,66 However, high molecular weight and content of aniline oligomer would make the polymer mechanically rigid, brittle and insoluble,^12,23,24,65 which would be hard to be processed into implants with desirable morphology to meet the needs for biomedical applications. The conductivities of wet CPU films were in the range of semiconductor materials (~1–10⁻⁸ S/cm)⁶⁷, that are similar to those of human physiological environments. The semiconductor range of conductivity is sufficient for tissue engineering and regeneration use because of the low micro-current intensity present in human bodies.^68,69 For example, a porous conductive scaffold based on polyurethane and PCL with conductivity at around 10⁻⁵ S/cm has shown the ability to improve the cardiomyocyte adhesion, growth and cardiac gene expression without external electrical stimulation.⁶⁰

The CPU film possessed good elasticity and robust mechanical properties. The elasticity of polyurethane is related to its chemical structure, including a hard segment, soft segment, and chain extender. Linear aliphatic HDI was employed as the hard segment to impart more flexibility to the CPU polymer than aromatic diisocyanates [e.g., methylene diphenyl diisocyanate and alicyclic diisocyanates (e.g. IPDI)], due to the inherent rigidness of the latters.^70,71 Linear PCL is semicrystalline and has often been used as a soft segment in polyurethane synthesis.^28,29,46,49 The current initial moduli of the polymer solid films are higher than those of soft tissues [e.g. human heart (0.01–0.5 MPa) and skin (0.1–2 MPa)],^18,72 When the solid polymer is processed into a porous scaffold, the initial modulus of the polymer would be significantly reduced,⁴⁹ and may match with the mechanics of the native tissue. Furthermore, the soft segment can also be replaced by softer polymer diols, such as poly(trimethylene carbonate) (PTMC)⁷³ and poly(δ-valerolactone-co-ε-caprolactone) (PVLCL),²⁹ which can reduce the initial modulus of the CPU . The chain extender aniline trimer, which has the minimum number of aniline repeat units, has fewer negative impacts on polymer flexibility than aniline oligomers with higher molecular weights (e.g. tetramer and pentamer).⁶⁵ However, the CPU conductivity is reduced because of the small number of aniline repeat units. Thus, it is necessary to consider the elasticity and conductivity of the polyurethane comprehensively. Additionally, the dopant involvement greatly affects the material mechanical properties. The incorporation of CSA dopant led to an increased initial modulus with decreased tensile strength and ultimate elongation of the CPU films, which might result from the interaction between CSA and the aniline oligomer segment. As the charge donors, the dopants can introduce charge carrier between polymer chains,^12,63 thus, there is an electrical interaction between polymer chains and dopants.⁷⁴ This limits the mobility of CPU polymer chains, which reduces the elasticity of doped CPU films while increasing their initial moduli. Furthermore, the doped polyaniline showed higher crystalline at higher doping levels,^75–76 which was confirmed by the XRD spectra of CPU and CPU1.5. Thus, it also reduced the elasticity of the doped CPU films compared with the undoped CPU film.

The conductivities of CPU films decreased with degradation. During the degradation of CPU polymers in lipase/PBS solution, the dopant CSA gradually leached out within 7 days, which resulted in an obvious reduction in conductivity.⁷⁷ A similar test conducted on a poly(glycerol-sebacate)/polyaniline composite in PBS solution showed that within 4 days, the conductivities of the composites decreased by around an order of magnitude.²³ Importantly, the enzymatic degradation of the polymer is much faster than its hydrolytic degradation in PBS. Thus, the conductivity of the CPU films may persist for a long term in a physiological condition, which also was evidenced in the conductivity stability testing of the CPU in cell culture medium. The CPU1.5 film retained 87% of its initial conductivity in cell culture medium after 150 h of immersion. The change of the current is directly responsible for the change of film conductivity under a fixed voltage. The slightly reduced conductivity of the CPU1.5 film is primarily attributed to dedoping and deprotonation under the synergic action of the cell medium and the current.⁷⁸ Because of the wet environment in biomedical applications, long-term electrical stability of the conductive polymers is required for practical purposes.¹⁴ The good electrical stability of the CPU1.5 film exhibited the potential to be appropriate for biomedical applications.

The CPU film has good cell compatibility, which was evidenced by cell proliferation on the film surface. The traditional conductive polymers (e.g., polyaniline) have been proved to support the growth of various cell types.¹² But dopant toxicity and polymer surface morphology have a significant impact on cell adhesion, proliferation and differentiation.⁷⁷ Further, the diffusion of the dopant into the cell culture medium may induce cytotoxicity.⁷⁹ When CSA was incubated with rat thymocytes, it resulted in significant toxicity at a high concentration (50 μg/mL).⁸⁰ Thus, at high dosages. CSA may exhibit toxicity to the cells during the culture. In addition, the leaching out of CSA can cause a pH value drop in the cell culture medium, which also might influence cell growth.²³ These factors may result in CPU1.5 having less cell proliferation than other samples. In future studies, a dopant with less toxicity may be employed to substitute CSA to achieve better cytocompatibility.

The CPU may have broad applications in the biomedical field, not limited to the tissue engineering. The repair and regeneration of some tissues, such as myocardium, nerve, muscle, skin and bone, respond positively to the presence of electrical fields, which makes conductive polymers attractive as tissue-engineered scaffolds.¹ The CPU exhibits good elasticity and electrical conductivity with biodegradable ability. It can be processed into tissue engineering scaffolds because the developed CPU can be dissolved in organic solvent, which is convenient for various scaffold processing approaches. In addition, drug release may be precisely controlled under applied electrical current or potential stimulus on drug-loaded conductive polyurethane through a de-doping procedure. For example, the conductive composites, such as polypyrrole/poly[(D,L-lactic acid)-co-(glycolic acid)]-b-poly(ethylene oxide)-b-poly[(D,L-lactic acid)-co-(glycolic acid)] (PLGA-PEG-PLGA) composite,⁸¹ PEDOT/poly(vinyl alcohol) (PVA) composite,⁸² and polyaniline/PVA composite,⁸³ carried drugs during the doping process, and then released them with electrical stimuli, suggesting the potential application of the CPU as a smart drug carrier. Furthermore, the synthesized CPU may also find opportunities for biodegradable, soft/wearable, and stretchable electronics use because of its biodegradability, conductivity, flexibility, and elasticity.

5. Conclusion

A biodegradable conductive polyurethane containing aniline trimer has been synthesized. These CPU films with CSA dopant exhibited good elasticity and increased initial moduli with increasing amounts of CSA. The electrical conductivities of wet CPU films are in the range of semiconductive materials. The CPU films exhibited good electroactivity and electrical stability under a physiological condition. All CPU films had good cytocompatibility to support cell growth on their surfaces. These results show that the conductive polyurethanes offer opportunities to be applied for tissue engineering, smart drug delivery, and electronics.