Article pubs.acs.org/Biomac
Dielectric Breakdown Strength of Regenerated Silk Fibroin Films as a Function of Protein Conformation Matthew B. Dickerson,† Scott P. Fillery,† Hilmar Koerner,† Kristi M. Singh,† Katie Martinick,‡ Lawrence F. Drummy,† Michael F. Durstock,† Richard A. Vaia,† Fiorenzo G. Omenetto,‡ David L. Kaplan,‡ and Rajesh R. Naik*,† †
Materials and Manufacturing Directorate, Air Force Research Laboratory, Wright-Patterson Air Force Base, Dayton, Ohio 45433 Biomedical Engineering Department, Tufts University, Medford, Massachusetts 02155
‡
S Supporting Information *
ABSTRACT: Derived from Bombyx mori cocoons, regenerated silk fibroin (RSF) exhibits excellent biocompatibility, high toughness, and tailorable biodegradability. Additionally, RSF materials are flexible, optically clear, easily patterned with nanoscale features, and may be doped with a variety bioactive species. This unique combination of properties has led to increased interest in the use of RSF in sustainable and biocompatible electronic devices. In order to explore the applicability of this biopolymer to the development of future bioelectronics, the dielectric breakdown strength (Ebd) of RSF thin films was quantified as a function of protein conformation. The application of processing conditions that increased β-sheet content (as determined by FTIR analysis) and produced films in the silk II structure resulted in RSF materials with improved Ebd with values reaching up to 400 V/μm.
■
these applications, and the interest in utilizing flexible RSF materials as insulating substrates, offers immense potential for the development of biocompatible and sustainable bioelectronic devices.4,7 However, there is a critical need to understand and optimize the dielectric properties of fibroin in order to explore and maximize the potential of RSF films to operate as insulating materials in high electric fields. The high electric field insulation properties of dielectric materials are defined by their dielectric breakdown field strength, which is determined by dividing the voltage at insulation breakdown (failure) by the material thickness. Amorphous polymers (e.g., polystyrene (PS) and PMMA) and semicrystalline polymers (e.g., biaxially oriented polypropylene (BOPP), polyethylene (PE) and poly(vinylidene fluoride) (PVDF)) typically exhibit breakdown strengths of 200−400 and 600−800 V/μm, respectively.11−14 The breakdown strength of semicrystalline polymers is affected by the crystalline content, morphology, habit, and orientation within the sample.14−17 RSF materials are semicrystalline and, thus, offer a biologically derived and biocompatible analog to the synthetic polymer dielectrics traditionally used in high electric field environments.18 Significantly, processing conditions controlling the crystalline structure and content of RSF are well established.19−21 Given the range of possible protein conformations and crystallite fractions (up to 60%) possible for
INTRODUCTION Regenerated silk fibroin (RSF) is a mechanically robust biopolymer that is extracted and purified from silkworm (Bombyx mori) cocoons.1−3 This biologically derived material is sustainable, nontoxic, biodegradable, bioresorbable, biocompatible, and induces minimal immune and inflammatory responses when implanted in the body.1−4 The aqueous-based processing of fibroin facilitates the incorporation of cells, biomolecules, and small molecules into RSF materials where these labile elements remain active and are protected from harsh environments.4 Additionally, fibroin-based materials are highly amenable to modification (i.e., both genetic and chemical), may be patterned by soft-lithography or nanoimprinting techniques, are visibly transparent, and possess tailorable/programmable environmental stability and mechanical characteristics.4−6 This unique combination of attractive properties has positioned RSF as a leading material for the generation of bioresponsive, bioactive, and biocompatible electronic devices.4,7 For example, RSF has been utilized as a biocompatible, surface-conforming substrate that facilitated the interfacing of electrode arrays with brain tissue.8 Organic field effect and light emitting transistors (OFETs and OLETs) featuring RSF gate dielectrics have been observed to possess charge mobilities, on/off ratios and optoelectronic characteristics similar to devices featuring polymethyl-methacrylate (PMMA) or SiO2 gates.9 RSF has also been utilized as a gate dielectric in an organic thin film transistor (OTFT), where its presence promoted the orthorhombic phase of pentacene, resulting in high charge mobility values.10 The success of RSF as a dielectric media in © 2013 American Chemical Society
Received: June 10, 2013 Revised: August 26, 2013 Published: August 29, 2013 3509
dx.doi.org/10.1021/bm4008452 | Biomacromolecules 2013, 14, 3509−3514
Biomacromolecules
Article
fibroin, there is substantial opportunity and need to explore structure−dielectric property relationships for RSF materials.19 Herein, we demonstrate the dependence of the dielectric breakdown strength of RSF films on crystalline structure and βsheet content of this protein. Utilizing water- and methanolbased processing treatments to facilitate a broad range of structural motifs, the breakdown strength of RSF was explored and a structure−property relationship based on β-sheet content (as determined by FTIR analysis) and protein conformation was defined. The resultant relationship between RSF structure and breakdown strength offers a pathway to the optimization of film processing conditions and dielectric properties. We anticipate that establishing this structure−property relationship will facilitate the further development of RSF-based electronic and bioelectronic devices.
■
mode using Cu Kα, generated by a Rigaku Ultrax 18 system. Silk fibroin films were delaminated from the Si(100) substrate and transmission X-ray of the samples was conducted in-plane. 2D image reduction/analysis was carried out using the software package Fit2D. FTIR analysis was achieved using a FTIR-ATR attachment (Bruker αP) with silk fibroin films deposited on Si (100) substrates. Five (5) spectra per treatment were captured, Fourier self-deconvolution (FSD) of the infrared spectra in the amide I region was performed according to previously published procedures.19 Grazing incidence small angle scattering experiments were carried out at the SAXS/WAXS beamline 7.3.3 of the Advanced Light Source at Lawrence Berkeley National Laboratory at 10 keV (1.24 Å) from a bend magnet and focused via a Mo/B4C double multilayer monochromator. A Dectris Pilatus 1 M detector was used to collect 2D SAX patterns at a detector to sample distance of 4m. 2D images were reduced using Nika 1 macros for Igor Pro.22 Images were corrected for transmission, background, dark current, and initial beam intensity fluctuations. Dielectric characterization, including impedance spectroscopy, dielectric breakdown strength and displacement-energy characterization measurements were performed on silk fibroin films deposited on ITO/glass substrates, using a metal−insulator−metal (MIM) geometry. The top aluminum electrodes were deposited through thermal evaporation (Edwards Auto 306) with an approximate thickness of 200 nm, as determined by profilometry. A total of ten (10) top finger-shaped electrodes were deposited per sample. Individual electrodes featured an area of 16.45 mm2, with a 6 mm2 contact pad at the substrate edge, overlapping the glass portion of the substrate. Electrical contact was facilitated using micromanipulator probes to a colloidal Ag dot on the Al contact pad. A flowing dry N2 gas environment was maintained for all dielectric measurements. Impedance measurements were made on silk fibroin films using a Novocontrol Alpha Analyzer using a frequency sweep of 0.1 Hz to 1 MHz, at an AC driving voltage of 1 V. The dielectric constant and loss were determined from phase-sensitive measurements of current and voltage. Dielectric breakdown strength was measured using a Keithley 6517B 1 kV high voltage source (Keithley Instruments), under computer control to enable current/voltage logging during the test procedure. Voltage was applied to silk fibroin films under DC conditions at 25 V/sec up to the point of catastrophic failure, as evidenced from large changes to the logged current (>1 μA) and visual sparking.
MATERIALS AND METHODS
Fibroin solutions were prepared from Bombyx mori cocoons (Mulberry Farms, Fallbrook, CA) following Na2CO3-based degumming, LiBrbased dissolution, purification, and concentration processes previously detailed in the literature.3 An important modification from this standard process was the extensive dialysis (10 water changes, dialysis bath to sample ratio of ∼200:1, over a period of 7−10 days) conducted at 4 °C against 18.2 MΩ water, utilizing dialysis tubing that had been previously prepared according to the manufacturer’s instructions. Following dialysis, insoluble materials were removed from the fibroin solution by repeated centrifugation at 5000 rpm and filtration through a 5 μm syringe filter. The silk fibroin solution was concentrated by reverse dialysis at 4 °C, against an aqueous, 20 wt % PEG (8000 average MW) solution utilizing 3,500 kDa MWCO tubing (Fisher Scientific Inc.) prepared according to the manufacturer’s instructions. Regenerated fibroin dopes containing 10−14 wt % protein were spread onto substrates by spin coating, using a 500 rpm/15 s spread step followed by a 2000 rpm/60 s thinning and solidification step. Films were subjected to various post spin coating treatment steps: a noncrystallization procedure (films isolated under vacuum); water vapor exposure; water vapor exposure with post-thermal dehydration at 180 °C; methanol vapor exposure; methanol vapor exposure with post-thermal dehydration at 180 °C; a 5 min 50 vol % methanol/5 min 90 vol % methanol sequential immersion; and a single step 5 min 90 vol % methanol immersion. Water vapor annealing was conducted by placing films within a glass desiccator vessel (5.7 L volume) containing 100 mL of 18.2 MΩ water bath separated from the samples by a raised platform.20 A 100% RH atmosphere was created by evacuating the air from the glass vessel and sealing the vessel to the ambient atmosphere and vacuum line. Samples were exposed to water vapor for 16 h at ambient laboratory temperature (20−25 °C). Methanol vapor annealing was conducted by sealing samples within a 100 mL polypropylene jar containing Kimwipes saturated with 7 mL of methanol that were separated from the samples by a raised platform. The samples were exposed to methanol vapor at 37 °C for 24 h. Sequential methanol bath treatments were conducted by submerging substrate-supported regenerated silk fibroin films in a 50 vol % methanol/18.2 MΩ water bath for 5 min, the films were then transferred to a 90 vol % methanol/18.2 MΩ water bath for 5 min, removed and blown dry with N2 gas. Select silk fibroin films were exposed to a single step 90 vol % methanol/18.2 MΩ water bath for 5 min, removed and blown dry with N2 gas. Where appropriate, silk fibroin films were thermally annealed for 1 h at 180 °C under a 25 mbar pressure within a vacuum oven (Memmert GmbH). Cross-sectional images of the silk fibroin films were obtained from scanning electron microscopy (FEI Quanta II ESEM), utilizing a Au conductive coating. Silk fibroin film thickness was measured using contact profilometry (P-15, KLA-Tencor). Residual H2O weight was determined through a Q500 Thermogravimetric Analyzer (TGA; TA Instruments), carried out under a dry flowing N2 gas environment. Wide angle X-ray experiments were carried out on a Statton Box camera at a 53 mm sample to image plate distance in transmission
■
RESULTS AND DISCUSSION High purity, aqueous-based fibroin solutions (prepared according to standard methods, slightly modified to eliminate ionic impurities) were utilized to produce RSF films for this study.3 The careful removal of ionic impurities from fibroin solutions was accomplished via the use of pretreated dialysis tubing and extensive dialysis. This high-purity, aqueous solution of fibroin was spin-coated onto substrates to create RSF thin films. Following spin-casting, RSF materials were exposed to a variety of water- and methanol-based treatment regimes in order to develop samples with a broad range of protein conformations and β-sheet contents.19,20 The primary structure of B. mori fibroin is composed of repetitive amino acid blocks that form crystalline domains (consensus repeat Gly-Ala-GlySer-Gly-Ala) and less-crystalline/amorphous regions (consensus repeat Gly-Ala-Gly-(Tyr/Val)-Gly-Ala) within the solidified material.23 Depending on the processing of RSF films (e.g., exposure to mechanical stress, organic solvents or thermal dehydration) the fibroin crystalline blocks may take on one of three structure types.19−21 These three distinct forms of fibroin are: amorphous/random coil (water-soluble, disordered structure), silk I (water-soluble, type II β-turns), and silk II (water insoluble, antiparallel β-sheet structure).21 Seven treatment procedures utilizing water, methanol and thermal 3510
dx.doi.org/10.1021/bm4008452 | Biomacromolecules 2013, 14, 3509−3514
Biomacromolecules
Article
solutions were deposited on the patterned ITO/glass substrates by spin-casting. Finger-shaped aluminum pads, produced by physical vapor deposition, served as the top electrode, providing 10 testable devices per substrate with individual contact areas of 16.45 mm2. Cross-sectional scanning electron microscopy (SEM) of RSF films exposed to water vapor and methanol are presented in Figure 1b and c, respectively. These films exhibit homogeneous microstructures and are largely free of microscopic voids that would impart lower breakdown strength values through field intensification and partial discharge mechanisms. The surface morphology and average roughness of the RSF films, as characterized by atomic force microscopy (AFM) is presented in Figure 1d,e, Supporting Information, Figure S1 and Table S1. The spin-casting of RSF produced flat, smooth films with roughness values
