Materials Science and Engineering C 40 (2014) 85–91
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Silk fibroin and sodium alginate blend: Miscibility and physical characteristics Mariana Agostini de Moraes, Mariana Ferreira Silva, Raquel Farias Weska, Marisa Masumi Beppu ⁎ School of Chemical Engineering, University of Campinas, UNICAMP, 13083-852 Campinas, SP, Brazil
a r t i c l e
i n f o
Article history: Received 4 November 2013 Received in revised form 6 March 2014 Accepted 18 March 2014 Available online 26 March 2014 Keywords: Biopolymers Films Phase separation Biomaterials
a b s t r a c t Films of silk fibroin (SF) and sodium alginate (SA) blends were prepared by solution casting technique. The miscibility of SF and SA in those blends was evaluated and scanning electron microscopy (SEM) revealed that SF/SA 25/75 wt.% blends underwent microscopic phase separation, resulting in globular structures composed mainly of SF. X-ray diffraction indicated the amorphous nature of these blends, even after a treatment with ethanol that turned them insoluble in water. Thermal analyses of blends showed the peaks of degradation of pristine SF and SA shifted to intermediate temperatures. Water vapor permeability, swelling capacity and tensile strength of SF films could be enhanced by blending with SA. Cell viability remained between 90 and 100%, as indicated by in vitro cytotoxicity test. The SF/SA blend with self-assembled SF globules can be used to modulate structural and mechanical properties of the final material and may be used in designing high performance wound dressing. © 2014 Published by Elsevier B.V.
1. Introduction Polymer blending is one of the most versatile and economical methods to produce new multiphase polymeric materials that are able to satisfy complex demands for performance. The final properties of blends are generally governed by the miscibility between the polymers and the phase equilibrium behavior has been one of the most important research subjects to produce polymer blends. It is known that the mixing composition, the component concentration, their relative proportions and the used solvent are parameters that strongly influence the phase separation in polymer blends, resulting in different morphologies [1]. Natural polymer blends have been studied in the last years because of their importance in several areas, from food and packing industry to the biomedical field. Natural polymers are, in general, biodegradable, biocompatible and can be obtained from a renewable source with low costs. However, in some cases, natural polymers present undesired properties, as high degradation rate or unsatisfactory mechanical properties. A way to improve the properties of these materials and to produce micro-structured materials with tunable properties is blending them with other biopolymers. The behavior of blends of proteins and polysaccharides is difficult to predict and can rarely result in a single phase (miscible system). In most cases, phase separation (immiscible system) occurs, resulting in one phase rich in one of the polymers while the other phase is poor in this same polymer [2].
⁎ Corresponding author. E-mail address: [email protected] (M.M. Beppu).
http://dx.doi.org/10.1016/j.msec.2014.03.047 0928-4931/© 2014 Published by Elsevier B.V.
Phase separation can be used to produce micro-structured materials, with unique structural and mechanical properties, that can be finetuned depending on polymer concentration and blending ratio. Additionally, specific compounds, such as drugs and nanoparticles, can be incorporated in the microdomains obtained from phase separation resulting in a material with specific target functions and modulated properties. Silk fibroin (SF) is a natural fiber extracted from cocoons of Bombyx mori silkworm that exhibit properties suitable for biomedical applications [3,4]. However, SF films cast from aqueous solution are soluble in water due to the dominating random coil structures, also called silk I. To induce conformational transitions from random coil to β-sheets (silk II) some methods are proposed, such as the use of organic solvents [5], high temperature [6] or shear stress [7]. The application of these methods turns SF films very brittle in dry state. To overcome this problem, blending SF with other polymers is a suitable alternative to improve its mechanical and physical properties. Sodium alginate (SA) appears as a promising second component to be used in blends with SF. SA is a polysaccharide of linear chain extracted from brown algae, composed by β-D-mannuronic and α-L-guluronic acids [11]. There are just a few papers in the literature regarding SF/SA blend films [8–10] and none of them have made a deep study on their miscibility. SF has been studied in recent years for application in wound healing and tissue engineering due to its affinity with several cell types [3,4]. Its films usually present good water vapor and oxygen permeability, and blood compatibility and they are claimed to improve collagen formation and fibroblast proliferation [12]. SA is hemostatic and keeps an adequate humidity for healing of wounds and burnings [9]. It is already found in the market as dressing for wounds, under commercial names such as
86
M.A. de Moraes et al. / Materials Science and Engineering C 40 (2014) 85–91
AlgiDERM® and Sorbsan® [13]. Blending SF and SA is a promising process for producing wound healing materials [9] and the control of the microstructure of the final blend is important to determine the properties and functionalities of the final material. We investigated the blending of SF with SA within the scope of miscibility and physical properties. The miscibility between the two natural polymers was studied in terms of morphological, structural and thermal properties, and the whole system was evaluated regarding thermodynamic aspects. In addition, the physical properties and the biocompatibility of the blend films were analyzed by water vapor permeability, swelling capacity, and mechanical and cytotoxicity tests. 2. Material and methods 2.1. Blend preparation Silk cocoons of B. mori silkworm (Bratac-Brazil) were degummed three times by soaking the cocoons in 1 g L−1 of Na2CO3 solution at 85 °C for 30 min, in order to remove the sericin of the cocoons, and then rinsing in distilled water. SF fibers were dried at room temperature and dissolved in a solution of CaCl2:CH3CH2OH:H2O (1:2:8 mole ratio) at 85 °C to a concentration of 5 wt.%. The SF salt solution was dialyzed in distilled water for three days at ca. 10 °C, to remove the salts of the solution. The final concentration of SF aqueous solution was 2.5 wt.%, and it was finally diluted in distilled water to 2 wt.%. Sodium alginate (Vetec-Brazil) with high mannuronic acid content, extracted from Macrocystis pyrifera seaweed, was dissolved in 0.1 M NaOH solution to a concentration of 2 wt.% SA. Glycerin was added in SA solution to act as plasticizer. The preparation of the SF/SA blend is briefly described below. SF and SA were blended at ratios of SF/SA 100/0, 75/25, 50/50, 25/75 and 0/100 wt.%. The blend solution was stirred for 15 min, cast in polystyrene dishes and dried for solvent evaporation. The whole process was performed at room temperature. Then, the blend films were immersed in a 0.1 M H2SO4 solution in 50 vol.% ethanol for 24 h, to stabilize the functional groups of SA and SF and turn the films insoluble in water. 2.2. Characterization 2.2.1. Miscibility evaluation The morphology of the blend films was observed by scanning electron microscopy (SEM). The samples were frozen in liquid nitrogen, freeze-fractured and then freeze-dried (Liobras, L101, Brazil) for 24 h. The samples were then coated with a gold layer and SEM observations were performed using a LEO 440i (Leica), with accelerating voltage of 10 kV. Further investigation of SF/SA blend miscibility was done by observing film morphology after SF extraction from the blend films. The SF was extracted from the blend films using the same protocol used to dissolve SF fibers when preparing SF solution. For that, pieces of 1 × 1 cm2 of the blend films were immersed in 50 mL of the solution CaCl2:CH3CH2OH: H2O, 1:2:8 mole ratio, at 85 °C, for 1 h, 1.5 h and 2 h. Time was varied in order to verify its influence in the final SF globule morphology. Subsequently, the blend films were rinsed with distilled water, frozen in liquid nitrogen, freeze-fractured and freeze-dried for 24 h prior to SEM observations. X-ray diffraction (XRD) was performed using a X'PERT PW3050 Philips equipment, with monochromatic Cu-Kα radiation, wavelength of 1.54 Å, in the 2θ range of 10° to 35° and scanning rate of 0.6° min− 1, to evaluate the crystallinity of the blend films. The chemical interaction between SA and SF and the molecular conformation of SF in the films were verified by Fourier transform infrared spectroscopy with attenuated total reflectance (FTIR-ATR), using a Bomem MB 102, equipped with a ZnSe cell crystal, in the range of
650–2000 cm−1. Each spectrum was acquired in transmittance mode through the accumulation of 256 scans. Thermogravimetric analysis (TGA) was performed in TGA-50 (Shimadzu) in a temperature range of 25–500 °C with a ramp rate of 10 °C min−1 and a N2 flow of 50 mL min−1. The data were normalized as a function of the initial mass of the sample. 2.2.2. Physical properties The thickness of the films was measured 10 times in the dry state and also in the wet state (after swelling test) for each sample using a digital micrometer (MDC-25S, Mitutoyo). Swelling degree of films was determined gravimetrically. Pieces of 2.5 cm in diameter of the films were weighed in the dry state (wi ), after reaching the equilibrium at 50% of relative humidity (48 h). Subsequently, the samples were immersed in 100 mL of distilled water (swelling medium) and were then weighed until reaching constant weight (wf). The swelling degree (%) was calculated as [(wf − wi)/wi] ⋅ 100 %. The water vapor permeability (WVP) was determined according to ASTM E 96M (2005). The samples were placed in a recipient with permeation area of 15.2 cm 2 containing anhydrous calcium chloride as desiccant and this recipient was then placed in a desiccator containing saturated aqueous NaCl solution, maintaining the ambient at 75% of relative humidity. The WVP through the films was determined gravimetrically by weighing the recipient every 12 h, for a period of 5 days. The rate of water vapor permeability was determined from the slope of the curve of mass change versus time. WVP was calculated as WVP = [(G/t) ⋅ e]/(A ⋅ Δp), where G/t is the mass variation rate (slope of the straight line), in g·day− 1; e is the film thickness, in mm; A is the test area, in m 2 ; and Δp is the vapor pressure difference, in kPa. Tensile tests were performed according to ASTM D882 (2002) using a TA.XT2 texture analyzer (Stable Microsystems SMD). The films (7 × 2.5 cm) were stored under standard conditions (25 °C, RH 50%) for 48 h before the tests. For tensile test, an initial grip separation of 50 mm and crosshead speed of 10 mm s−1 were used. The average values of tensile strength were obtained from 8 specimens. 2.2.3. Cytotoxicity test In vitro biocompatibility was performed according to ISO 10993-5 (2009) using Chinese hamster ovary cell line (CHO-k1). The cells were maintained in RPMI culture medium supplemented with antibiotics and antimicotic (100 units mL−1 of penicillin, 100 μg mL−1 of streptomycin and 0.025 μg mL−1 of amphotericin), 2 mM glutamine, and 10% calf serum, at 37 °C in a 5% CO2 humidified atmosphere until they reached confluence. For subculturing and for experiments, cells were harvested using 0.05% trypsin and 0.02% EDTA in phosphate-buffered saline solution at pH 7.4. The films were sterilized by using UV radiation for 30 min on each side of the film. The films were immersed in RPMI culture medium in a proportion of 1 cm2 mL−1, at 37 °C for 48 h. The extracts of the films were then diluted from 100% (original extract) to 6.25 vol.% in RPMI 1640. Cytotoxicity test was performed in a 96 well microplate containing 3000 cells per well and extract of films. The microplates were incubated for 72 h at 37 °C in a 5% CO2 humidified atmosphere. The cell viability was measured by adding MTS (supravital dye tetrazolium compound)/PMS (electron coupling agent) (20/1) solution and incubating for 2 h. The microplates were analyzed in a spectrophotometer ELISA at 490 nm. The test was compared with a negative control of high density polyethylene (HDPE) and a positive control of phenol 0.5 vol.% in culture medium. 3. Results In the present study, tests of SF/SA blend preparation were performed, in order to obtain a homogeneously mixed solution and a film without macroscopic phase separation. Initially, we used SA aqueous
M.A. de Moraes et al. / Materials Science and Engineering C 40 (2014) 85–91
solution; however, the blending solutions did not mix well and the final aspect of blends was heterogeneous, with formation of SF fibrils and phase separation in all blending ratios. Due to this fact, previous dissolution of SA in an alkaline solution, such as NaOH was proposed. When SA alkaline solution was used for blends, no macroscopic phase separation was detected, forming a homogeneous blend solution under the naked eye. At neutral to alkaline pHs, SF is negatively charged (pK ≅ 4.2), as well as SA (pK ≅ 4), which does not favor the electrostatic interactions. On the contrary, the negative charges on both biopolymers will increase the electrostatic repulsion between the molecules. This repulsion is probably responsible for the phase separation and fibril formation in blends with SA dissolved in water. Even though SA alkaline solution will not decrease the electrostatic repulsion between the biopolymer molecules, it will decrease SF–SF interactions, resulting in the predominance of silk I structure, that exhibits more active sites available [14]. The pH influence in SF molecule interactions can be confirmed by the fact that when lowering the pH of SF aqueous solution (pH = 7), the solution becomes gradually more viscous, until reaching the point in which the SF–SF molecule interactions are so strong that the SF solution undergoes gelation (pH ≅ 5) [15]. At this point, a solid structure with predominance of a stable silk II conformation is obtained. For SF contents higher than 25 wt.%, the formation of SF fibrils could be observed, characterizing macroscopic phase separation, even when
87
SA alkaline solution was used in the blend. SF/SA blend film without SF fibril formation could only be obtained in SF/SA 25/75 blend ratio or when the blend components were used alone (SF/SA 100/0 and 0/100). Probably, SF fibrillogenesis occurs after a critical concentration, higher than 25 wt.%, resulting in fibrils that can be observed under the naked eye. After these results, we decided to further investigate the SF/SA 25/75 blend film, in which SF fibrils were not observed. 3.1. SEM Films of pure SF and pure SA (Fig. 1-1 and -3) show typical morphology of dense films, with dense and homogeneous fracture surfaces. The SF/SA 25/75 blend film (Fig. 1-2) revealed globular structures (with ca. 1 to 10 μm of diameter) on its surface and also on its cross section. As explained before, a blend without macroscopic phase separation in the fibril form was only obtained when SA was dissolved in NaOH, but even under these conditions, microscopic phase separation in globular form was observed. Probably, the alkaline pH decreased SF–SF interactions; although it did not decrease the electrostatic repulsion between SA and SF. Thus, the final morphology is a globular phase dispersed in a continuous phase, as observed in Fig. 1. Um and Park [16] observed that the casting solvent plays a dominant role in determining the miscibility of SF blends. Studying blends of SF and poly(vinyl
Fig. 1. SEM micrographs of surface (a) and fractured cross section (b) of SF/SA blend films at 100/0 (1), 25/75 (2) and 0/100 wt.% (3).
88
M.A. de Moraes et al. / Materials Science and Engineering C 40 (2014) 85–91
alcohol) (PVA), the authors found that, when the films were cast from aqueous solution, phase separation occurred, producing a SF phase dispersed as ovals in a continuous phase of PVA. Conversely, when the films were cast from formic acid solution, a miscible blend was obtained at SEM observation (microscopic level). This might be due to the fact that SF molecules are unstable in water and exist in the colloidal state in aqueous systems due to its hydrophobic amino acids, resulting in molecular aggregation. When a better co-solvent, such as formic acid, is used, SF molecules are stabilized and their miscibility is improved. In our study, we used water as solvent for SF molecules; however, we observed the need of changing the SA solvent to form a blend without macroscopic phase separation. Furthermore, in acidic conditions, the\COO− ions of SA start to become\COOH, the electrostatic repulsion between chains is reduced and they form hydrogen bonds, resulting in increased viscosity of solution [17]. When pH value is between 3 and 4, SA undergoes gelation, which will probably be a limitation to the use of formic acid as casting solvent in SF/SA blends. Further studies in finding a suitable casting solvent for SF/SA blends are still needed in literature.
Additionally, other studies report that the addition of glycerin in blends with SF plays an important role in inducing intermolecular interactions between SF and the other blend component, decreasing the heterogeneity and the phase separation, and improving the mechanical properties of the films (increasing flexibility). In these cases, glycerin would act as a plasticizer [14,18]. This trend was also observed in our study (information observed in laboratory experiments, no data provided), where the addition of glycerin on SA solution improved the miscibility between SF and SA and also improved the flexibility of the films. To verify if the globular domains were composed of SF or SA, we proposed SF extraction using the solution CaCl2:CH3CH2OH:H2O, as described in the “Materials and methods” section. The morphology of the SF/SA 25/75 blend films after SF extraction is shown in Fig. 2. The globular structures were removed from the film surface and also from the cross-section area, indicating that the globular phase was mainly composed of SF. The solution was able to penetrate into the film, removing SF from its bulk. The SF globule extraction is more evident with increasing extraction time. Some residues of SF were observed in the fractured
Fig. 2. SEM micrographs of surface (a) and fractured cross section (b) of SF/SA 25/75 blend films after 1 h (1), 1.5 h (2) and 2 h of SF extraction (3).
M.A. de Moraes et al. / Materials Science and Engineering C 40 (2014) 85–91
cross-section micrographs, especially when the extraction time was 1.5 h, producing to some star-like structures (Fig. 2-2b). However, after high extraction time, these residual structures were not observed anymore. This fact indicates that the globules are formed mainly by a SF-rich phase. Also, by regulating the extraction time, it is possible to regulate the final morphology of SF, and to obtain micro-structured materials with different properties. Additionally, SF/SA 25/75 blend films were prepared without glycerin addition and the same SF extraction protocol was performed. As expected, the roughness observed on the surface of the blend films in Fig. 2 is attributed to glycerin leaching, once those images were not observed on the films without glycerin (data not shown). The SF/SA blend morphology studied in this paper is a typical discrete phase structure, with globules rich in SF dispersed in a matrix rich in SA. This morphology is similar to that observed in materials with phase separation mechanism of nucleation and growth [19]. SF phase separation in globular features was already observed in SF blend films with polyethylene oxide (PEO) [7,20,21], carboxymethyl chitin [22] and poly(vinyl alcohol) (PVA) [16]. In the blends with PEO and carboxymethyl chitin, SF was the major component and the second component was water-soluble; thus the authors were capable to extract the water-soluble component by immersion in water. As a result, a film consisting of SF globules is obtained, and these globules were found to be constituted by micelles (100–200 nm), with globule sizes varying from 15 to 0.8 μm [7,22]. Jin and Kaplan [7] attempted to mimic SF fiber processing by silkworm replacing sericin by PEO. The authors found that the presence of a hydrophilic material, such as sericin and PEO (as well as the SA studied in our work), alters the final stages of globule formation, accelerating their organization into silk I structure (amorphous). The SF globular structure is naturally presented in SF fibers spun from silkworm and its morphology is similar to those obtained by Jin and Kaplan [7] in SF/PEO blends and also to our SF/SA blend films. The globules obtained in our study are more spherical and uniform than those obtained in blends with PEO, PVA or carboxymethyl chitin. These self-assembled SF globules are interesting structures that can be used to design sophisticated structures with microdomains of SF. In addition, we believe that with some little variations on SF/SA ratio, it is possible to control the quantity and size of globule formation, fine tuning the structural and mechanical properties of the final material.
89
respectively [23,24]. SA typical halos were observed at 2θ = 13.7°, 16.7° and 20.7° (Fig. 3c), corresponding to 6.45, 5.30 and 4.28 Å, respectively [8,25]. On the other hand, SF/SA 25/75 blend film shows an amorphous diffractogram. This is an indication that the intermolecular interaction of SF and SA occurs in the amorphous region. Even after film stabilization with sulfuric acid and ethanol – that would increase the organizational level and the crystallinity of the biopolymers – the amorphous profile of the blend remains unaltered. This profile confirms the fact that SF globule formation occurs with stable silk I conformation [7] and for that, the presence of a hydrophilic component, such as SA, is crucial. 3.3. FTIR FTIR spectroscopy has been a useful tool to investigate the molecular conformation of SF. Fig. 4 shows the FTIR spectra for the blend films and the main absorption peaks are listed in Table 1. Pure SF film (SF/SA 100/0) revealed a FTIR spectrum with β-sheet predominance (silk II) for the amide I and II absorption bands. The amide III absorption band was located in a wavelength corresponding to silk I conformation. SA film (SF/SA 0/100) exhibited FTIR spectrum with typical absorption bands of C_O, C\O and O_H stretching; and is in accordance with the literature for SA [26]. The SF/SA 25/75 blend film presented a spectrum similar to the SF/SA 0/100 spectrum, but with the presence of SF absorption bands of amide I, II and III, slightly shifted to higher wavelengths when compared to pure SF spectrum. The bands related to amide I and amide III of the blend film overlap with SA characteristic groups, which can induce errors when analyzing the spectra at these specific wavelengths. However, the amide II band is clearly visualized in the blend spectra at 1521.7 cm− 1, related to the silk II conformation. 3.4. Thermal properties
SF/SA 0/100 and 100/0 blend films show diffraction patterns of crystalline phase (Fig. 3). Fig. 3a shows SF typical halos at 2θ = 14°, 16.8°, 18.5° and 20°, corresponding to the crystalline and organized silk II conformation, with interplanar spaces of 6.32, 5.27, 4.79 and 4.43 Å,
TGA and DTGA curves of the blend films are shown in Fig. 5. The SF/SA 25/75 blend film initiates its mass loss in a temperature between the values observed for pristine SF and SA alone. SF films exhibited the highest thermal stability. SF/SA 25/75 blend film showed similar behavior of SF/SA 0/100 film in DTGA; however a peak attributed to SF degradation is observed at 305.5 °C. This peak is observed in the SF/SA 100/0 film at 313.7 °C. The mass loss SF/SA 0/100 is observed at 222 °C, attributed to the decrosslinking of polymer network and formation of carbonaceous residues [27]. SF/SA 25/75 blend film revealed peaks related to thermal degradation of SF and SA components alone; however these peaks were shifted to intermediate temperatures, which indicates a semi-miscible blend.
Fig. 3. XRD patterns of SF/SA blend films 100/0 (a), 25/75 (b) and 0/100 wt.% (c).
Fig. 4. FTIR spectra of SF/SA blend films 100/0 (a), 25/75 (b) and 0/100 wt.% (c).
3.2. XRD
90
M.A. de Moraes et al. / Materials Science and Engineering C 40 (2014) 85–91
Table 1 FTIR absorption peaks of SF/SA blend films. SF/SA blend ratio 100/0
25/75
Table 2 Results of the physical and mechanical characterization of SF/SA blend films. Attribution of absorption peak
Properties
SF/SA blend ratio
0/100
– 1616.2
1731.9 1622.0
1731.9 1635.5
1508.2 1228.5 – – –
1521.7 1236.2 1080.0 1029.9 927.7
– 1238.2 1076.2 1027.9 927.7
100/0 C_O stretching Amide I and antisymmetric\COO\ stretching overlap Amide II Amide III and C\O stretching overlap O\H bending O\H bending C\O vibration
a
Dry thickness (μm) Wet thickness (μm)a Swelling capacity (%)b WVP (g·mm·m−2 day−1 kPa−1)b Tensile strength (MPa)c Elongation at the break (%)c a b c
24 32 59 2.7 52 1.8
± ± ± ± ± ±
25/75 2 5 5 0.2 10 0.6
33 105 220 4.8 73 3.4
± ± ± ± ± ±
0/100 1 5 13 0.1 11 0.9
31 145 310 5.2 95 7.7
± ± ± ± ± ±
4 6 6 0.4 6 2.2
Average ± standard deviation (n = 10). Average ± standard deviation (n = 3). Average ± standard deviation (n = 8).
3.5. Physical properties Table 2 depicts the results of physical and mechanical properties of SF/SA blend films. The films' thickness values in the wet state were significantly increased in the blends containing SA (SF/SA 25/75 and 0/100). SA is a highly hydrophilic biopolymer due to the presence of carboxyl and hydroxyl groups in its structure, which provides the capacity to absorb and retain large amounts of water when SA films are immersed in aqueous solutions [28]. On the other hand, SF possesses hydrophobic amino acids in its structure and does not allow the swelling and water plasticization, which could explain the low values of SF film thickness in the wet state and low swelling capacity. The same tendency is observed in the results of water vapor permeability, where SA film permeability was almost two times higher than that observed for SF
film. The SF/SA 25/75 blend film exhibited properties related to the mass content of SA and SF in the blend. From the tensile test results, it is possible to notice that the tensile strength and elongation at the break of SF/SA film is higher than the values observed for SF films. This result indicates that SF mechanical properties are improved when SF is blended with SA. In addition, the presence of glycerin in the SF/SA 25/75 blend is a key parameter, since it increases the film flexibility and improves the film handling (information observed during handling, no data provided). Multiphase blends can exhibit several morphologies and their properties are directly related to the phase's arrangement. When one phase of polymer A is dispersed in a matrix of polymer B, for example, the properties of the material are governed by the matrix properties (polymer B) [1]. This is in accordance with our results, where the structural properties of the blend films, such as swelling capacity, water vapor permeability, tensile strength and elongation at break were influenced by SA matrix. SA is known to be a good material to be used in wound dressings [13]. SA associated with the globular microdomains in the SF/SA 25/ 75 blend film and the presented physical properties indicate that the final blend is a promising material to be used in high performance microstructured dressings.
3.6. Cytotoxicity test The cell viability in the presence of the blend films remained between 90% and 100% (Table 3) and according to ISO 10993-5 (2009), these films can be considered nontoxic to cells. A cell viability of ca. 90% was observed for pure SA film (SF/SA 0/100), while the pure SF film (SF/SA 100/0) showed the highest cell viability (98%). The SF/SA 25/75 blend film showed behavior that is in between SF and SA and can be explored as potential biomaterials, mainly as devices where specific compounds or nanoparticles can be incorporated on SF globular domains and gradually released.
4. Discussion One of the main goals in polymeric blends field is to obtain compatible systems. Compatible blends are those who exhibit satisfactory properties, no matter if they are miscible (one phase), semi-miscible or immiscible (phase separation) blends.
Table 3 Viability of CHO cells in the extract of the blend films.
Fig. 5. TGA and DTGA thermograms of SF/SA blend films 100/0 (a), 25/75 (b) and 0/100 wt.% (c).
SF/SA blend ratio
Cell viability (%)a
0/100 25/75 100/0
89.1 ± 3.5 93.6 ± 1.4 98.7 ± 1.9
a
Average ± standard deviation (n = 4).
M.A. de Moraes et al. / Materials Science and Engineering C 40 (2014) 85–91
The miscibility of a mixture can be determined by the Gibbs free energy (Eq. (1)). For a miscible system, the variation in the Gibbs free energy needs to be negative. ΔGm ¼ ΔH m −TΔSm
ð1Þ
where ΔGm is the Gibbs free energy of mixture, ΔSm is the entropy of mixture, ΔHm is the enthalpy of mixture and T is the temperature. The Flory–Huggins theory was the first one to consider the high polymeric chain size and the polymer–polymer interactions (Eq. (2)). ΔGm ϕ ϕ ¼ kT 1 ln ϕ1 þ 2 ln ϕ2 þ ϕ1 ϕ2 χ 12 N0 x1 x2
ð2Þ
where x1 and x2 are the number of segments of polymers 1 and 2; ϕ1 and ϕ2 are the volumetric fractions of polymers 1 and 2 in the blend; and χ12 is the Flory–Huggins interaction parameter. The mixture of two or more polymers involves very low entropy values, once the entropic term is inversely related to the size of the polymeric chain. Thus, to obtain miscible systems it is necessary to overcome the unfavorable entropy with specific interactions between polymeric chains, responsible for a negative enthalpy. With this approach, a negative Gibbs free energy can be achieved, stabilizing thermodynamically the system. On the other hand, in a mixture of two polymers, it is necessary to overcome the energetic barrier to form a new interface to initiate the phase separation, once interfaces are unfavorable energetic surfaces. In our study, microscopic phase separation was observed in the SF/SA 25/75 blend film. As SF and SA have high molecular chains (350 kDa for SF and approximately 1000 kDa for SA), small entropy of mixture is observed when mixing them. Thus, a miscible system (ΔG b 0) is only obtained if the enthalpy of mixture is low. In other words, to obtain miscible blends from mixing polymers of high molecular chain, the polymers must have strong interactions between them. SA possesses carboxyl groups negatively charged on its surface, which allows the interaction with other functional groups of opposite charge. On the other hand, SF possesses amino groups – that can be protonated and interact with SA carboxyl – and also carboxyl groups — that can restrain the interaction of SF functional groups with SA carboxyl. Furthermore, at neutral pH SF molecules are negatively charged (pK = 4.2) and, probably, electrostatic repulsion between SF and SA takes place, resulting in a morphology consisting of a SF-rich phase (globules) dispersed in a SA continuous phase. Regarding the morphology of SF/SA blend films, we may affirm that the interaction between SF and SA chains is not strong enough to prevent phase separation. This could be predicted thermodynamically and it was also observed in our experimental study. Furthermore, the morphology of phase separation in the blend SF/SA 25/75 observed by SEM is typical of nucleation and growth, which is a mechanism of phase separation of metastable binary mixtures. Several research groups have tried to achieve self-assembled structures through chemical modifications or functionalization of surfaces. In our study, self-assembled materials were spontaneously obtained, with variable morphology, depending on the system composition (relative proportion of the blend components). Surface morphologies can be altered between globule or fibril formation, depending on the SF/SA ratio. Increasing SF proportion in the blend leads to macroscopic phase separation, with SF fibril formation; while increasing SA proportion in the blend leads to microscopic phase separation, with SF globule formation. Furthermore, we believe that the SF globule size and quantity can be regulated as a function of the SF/SA ratio, until reaching a maximum content, at which SF fibril formation occurs. Also, specific compounds can be incorporated in the SF microdomains, forming a microstructured material for controlled delivery. Considering the potential application of SF and SA in burns and wound healing, we can
91
incorporate silver, for example, in SF microdomains to prepare high performance dressings, with control over the silver release and concentration. The blend film has good swelling properties, which can absorb the wound exudate, has also good permeability, allowing fluid exchange with the environment, and still provides a mechanical barrier against infection. Additionally, the possibility to regulate the SF globule extraction opens several options to control the material morphology, from globules to star-like structures to a material with spherical void spaces. All those possibilities can be used to design sophisticated materials with SF microdomains. 5. Conclusion A blend film of SF and SA with microscopic phase separation in globular domains was obtained at SF/SA 25/75 blending ratio. The globules were mainly composed of SF, while the matrix was mainly composed of SA. The microscopic phase separation occurs due to the ΔG N 0, attributed to the high molecular chain of natural polymers and weak interactions between the chains, usually observed in biopolymeric blends. Furthermore, this self-assembled micro-structured blend film is a non-cytotoxic material, and exhibits water vapor transmission, swelling capacity, tensile strength and elongation values suitable for designing sophisticated high performance wound dressings. Acknowledgments The authors thank Andrea C. D. Rodas and Olga Z. Higa (Biotechnology Center, Energy and Nuclear Research Institute, São Paulo-SP, Brazil) for the cytotoxicity test and FAPESP and CNPq for the financial support. References [1] C. Harrats, S. Thomas, G. Groeninckx, Micro- and Nanostructured Multiphase Polymer Blend Systems: Phase Morphology and Interfaces, CRC Press, 2006. 442. [2] M. Panouille, V. Larreta-Garde, Food Hydrocoll. 23 (2009) 1074–1080. [3] G.H. Altman, F. Diaz, C. Jakuba, T. Calabro, R.L. Horan, J.S. Chen, H. Lu, J. Richmond, D. L. Kaplan, Biomaterials 24 (2003) 401–416. [4] C. Vepari, D.L. Kaplan, Prog. Polym. Sci. 32 (2007) 991–1007. [5] G.M. Nogueira, A.C.D. Rodas, C.A.P. Leite, C. Giles, O.Z. Higa, B. Polakiewicz, M.M. Beppu, Bioresour. Technol. 101 (2010) 8446–8451. [6] Y. Kawahara, K. Furukawa, T. Yamamoto, Macromol. Mater. Eng. 291 (2006) 458–462. [7] H.J. Jin, D.L. Kaplan, Nature 424 (2003) 1057–1061. [8] K.G. Lee, H.Y. Kweon, J.H. Yeo, S.O. Woo, J.H. Lee, Y.H. Park, J. Appl. Polym. Sci. 93 (2004) 2174–2179. [9] D.H. Roh, S.Y. Kang, J.Y. Kim, Y.B. Kwon, H.Y. Kweon, K.G. Lee, Y.H. Park, R.M. Baek, C. Y. Heo, J. Choe, J.H. Lee, J. Mater. Sci. Mater. Med. 17 (2006) 547–552. [10] C.X. Liang, K. Hirabayashi, J. Appl. Polym. Sci. 45 (1992) 1937–1943. [11] M. Rinaudo, Polym. Int. 57 (2008) 397–430. [12] H. Kweon, J.H. Yeo, K.G. Lee, H.C. Lee, H.S. Na, Y.H. Won, C.S. Cho, Biomed. Mater. 3 (2008) 034115 (5 pp.). [13] M. Rinaudo, Macromol. Symp. 245–246 (2006) 549–557. [14] C.L. Du, J. Jin, Y.C. Li, X.D. Kong, K.M. Wei, J.M. Yao, Mater. Sci. Eng. C Biomim. Supramol. Syst. 29 (2009) 62–68. [15] U.J. Kim, J.Y. Park, C.M. Li, H.J. Jin, R. Valluzzi, D.L. Kaplan, Biomacromolecules 5 (2004) 786–792. [16] I.C. Um, Y.H. Park, Fibers and Polym. 8 (2007) 579–585. [17] Y.M. Qin, Polym. Int. 57 (2008) 171–180. [18] L.X. Dai, J. Li, E. Yamada, J. Appl. Polym. Sci. 86 (2002) 2342–2347. [19] T. Inoue, Prog. Polym. Sci. 20 (1995) 119–153. [20] H.J. Jin, J. Park, R. Valluzzi, P. Cebe, D.L. Kaplan, Biomacromolecules 5 (2004) 711–717. [21] B.D. Lawrence, F. Omenetto, K. Chui, D.L. Kaplan, J. Mater. Sci. 43 (2008) 6967–6985. [22] P. Wongpanit, Y. Tabata, R. Rujiravanit, Macromol. Biosci. 7 (2007) 1258–1271. [23] M.Z. Li, S.Z. Lu, Z.Y. Wu, K. Tan, N. Minoura, S. Kuga, Int. J. Biol. Macromol. 30 (2002) 89–94. [24] G. Freddi, P. Monti, M. Nagura, Y. Gotoh, M. Tsukada, . Polym. Sci. B Polym. Phys. 35 (1997) 841–847. [25] C.B. Xiao, H.J. Liu, Y.S. Lu, L. Zhang, J. Macromol. Sci. Pure Appl. Chem. 38 (2001) 317–328. [26] T. Caykara, S. Demirci, M.S. Eroglu, O. Guven, Polymer 46 (2005) 10750–10757. [27] T.M.M. Swamy Siddaramaiah, B. Ramaraj, J.H. Lee, J. Appl. Polym. Sci. 109 (2008) 4075–4081. [28] S. Kalyani, B. Smitha, S. Sridhar, A. Krishnalah, Desalination 229 (2008) 68–81.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Silk fibroin and sodium alginate blend: Miscibility and physical characteristics Mariana Agostini de Moraes, Mariana Ferreira Silva, Raquel Farias Weska, Marisa Masumi Beppu ⁎ School of Chemical Engineering, University of Campinas, UNICAMP, 13083-852 Campinas, SP, Brazil
a r t i c l e
i n f o
Article history: Received 4 November 2013 Received in revised form 6 March 2014 Accepted 18 March 2014 Available online 26 March 2014 Keywords: Biopolymers Films Phase separation Biomaterials
a b s t r a c t Films of silk fibroin (SF) and sodium alginate (SA) blends were prepared by solution casting technique. The miscibility of SF and SA in those blends was evaluated and scanning electron microscopy (SEM) revealed that SF/SA 25/75 wt.% blends underwent microscopic phase separation, resulting in globular structures composed mainly of SF. X-ray diffraction indicated the amorphous nature of these blends, even after a treatment with ethanol that turned them insoluble in water. Thermal analyses of blends showed the peaks of degradation of pristine SF and SA shifted to intermediate temperatures. Water vapor permeability, swelling capacity and tensile strength of SF films could be enhanced by blending with SA. Cell viability remained between 90 and 100%, as indicated by in vitro cytotoxicity test. The SF/SA blend with self-assembled SF globules can be used to modulate structural and mechanical properties of the final material and may be used in designing high performance wound dressing. © 2014 Published by Elsevier B.V.
1. Introduction Polymer blending is one of the most versatile and economical methods to produce new multiphase polymeric materials that are able to satisfy complex demands for performance. The final properties of blends are generally governed by the miscibility between the polymers and the phase equilibrium behavior has been one of the most important research subjects to produce polymer blends. It is known that the mixing composition, the component concentration, their relative proportions and the used solvent are parameters that strongly influence the phase separation in polymer blends, resulting in different morphologies [1]. Natural polymer blends have been studied in the last years because of their importance in several areas, from food and packing industry to the biomedical field. Natural polymers are, in general, biodegradable, biocompatible and can be obtained from a renewable source with low costs. However, in some cases, natural polymers present undesired properties, as high degradation rate or unsatisfactory mechanical properties. A way to improve the properties of these materials and to produce micro-structured materials with tunable properties is blending them with other biopolymers. The behavior of blends of proteins and polysaccharides is difficult to predict and can rarely result in a single phase (miscible system). In most cases, phase separation (immiscible system) occurs, resulting in one phase rich in one of the polymers while the other phase is poor in this same polymer [2].
⁎ Corresponding author. E-mail address: [email protected] (M.M. Beppu).
http://dx.doi.org/10.1016/j.msec.2014.03.047 0928-4931/© 2014 Published by Elsevier B.V.
Phase separation can be used to produce micro-structured materials, with unique structural and mechanical properties, that can be finetuned depending on polymer concentration and blending ratio. Additionally, specific compounds, such as drugs and nanoparticles, can be incorporated in the microdomains obtained from phase separation resulting in a material with specific target functions and modulated properties. Silk fibroin (SF) is a natural fiber extracted from cocoons of Bombyx mori silkworm that exhibit properties suitable for biomedical applications [3,4]. However, SF films cast from aqueous solution are soluble in water due to the dominating random coil structures, also called silk I. To induce conformational transitions from random coil to β-sheets (silk II) some methods are proposed, such as the use of organic solvents [5], high temperature [6] or shear stress [7]. The application of these methods turns SF films very brittle in dry state. To overcome this problem, blending SF with other polymers is a suitable alternative to improve its mechanical and physical properties. Sodium alginate (SA) appears as a promising second component to be used in blends with SF. SA is a polysaccharide of linear chain extracted from brown algae, composed by β-D-mannuronic and α-L-guluronic acids [11]. There are just a few papers in the literature regarding SF/SA blend films [8–10] and none of them have made a deep study on their miscibility. SF has been studied in recent years for application in wound healing and tissue engineering due to its affinity with several cell types [3,4]. Its films usually present good water vapor and oxygen permeability, and blood compatibility and they are claimed to improve collagen formation and fibroblast proliferation [12]. SA is hemostatic and keeps an adequate humidity for healing of wounds and burnings [9]. It is already found in the market as dressing for wounds, under commercial names such as
86
M.A. de Moraes et al. / Materials Science and Engineering C 40 (2014) 85–91
AlgiDERM® and Sorbsan® [13]. Blending SF and SA is a promising process for producing wound healing materials [9] and the control of the microstructure of the final blend is important to determine the properties and functionalities of the final material. We investigated the blending of SF with SA within the scope of miscibility and physical properties. The miscibility between the two natural polymers was studied in terms of morphological, structural and thermal properties, and the whole system was evaluated regarding thermodynamic aspects. In addition, the physical properties and the biocompatibility of the blend films were analyzed by water vapor permeability, swelling capacity, and mechanical and cytotoxicity tests. 2. Material and methods 2.1. Blend preparation Silk cocoons of B. mori silkworm (Bratac-Brazil) were degummed three times by soaking the cocoons in 1 g L−1 of Na2CO3 solution at 85 °C for 30 min, in order to remove the sericin of the cocoons, and then rinsing in distilled water. SF fibers were dried at room temperature and dissolved in a solution of CaCl2:CH3CH2OH:H2O (1:2:8 mole ratio) at 85 °C to a concentration of 5 wt.%. The SF salt solution was dialyzed in distilled water for three days at ca. 10 °C, to remove the salts of the solution. The final concentration of SF aqueous solution was 2.5 wt.%, and it was finally diluted in distilled water to 2 wt.%. Sodium alginate (Vetec-Brazil) with high mannuronic acid content, extracted from Macrocystis pyrifera seaweed, was dissolved in 0.1 M NaOH solution to a concentration of 2 wt.% SA. Glycerin was added in SA solution to act as plasticizer. The preparation of the SF/SA blend is briefly described below. SF and SA were blended at ratios of SF/SA 100/0, 75/25, 50/50, 25/75 and 0/100 wt.%. The blend solution was stirred for 15 min, cast in polystyrene dishes and dried for solvent evaporation. The whole process was performed at room temperature. Then, the blend films were immersed in a 0.1 M H2SO4 solution in 50 vol.% ethanol for 24 h, to stabilize the functional groups of SA and SF and turn the films insoluble in water. 2.2. Characterization 2.2.1. Miscibility evaluation The morphology of the blend films was observed by scanning electron microscopy (SEM). The samples were frozen in liquid nitrogen, freeze-fractured and then freeze-dried (Liobras, L101, Brazil) for 24 h. The samples were then coated with a gold layer and SEM observations were performed using a LEO 440i (Leica), with accelerating voltage of 10 kV. Further investigation of SF/SA blend miscibility was done by observing film morphology after SF extraction from the blend films. The SF was extracted from the blend films using the same protocol used to dissolve SF fibers when preparing SF solution. For that, pieces of 1 × 1 cm2 of the blend films were immersed in 50 mL of the solution CaCl2:CH3CH2OH: H2O, 1:2:8 mole ratio, at 85 °C, for 1 h, 1.5 h and 2 h. Time was varied in order to verify its influence in the final SF globule morphology. Subsequently, the blend films were rinsed with distilled water, frozen in liquid nitrogen, freeze-fractured and freeze-dried for 24 h prior to SEM observations. X-ray diffraction (XRD) was performed using a X'PERT PW3050 Philips equipment, with monochromatic Cu-Kα radiation, wavelength of 1.54 Å, in the 2θ range of 10° to 35° and scanning rate of 0.6° min− 1, to evaluate the crystallinity of the blend films. The chemical interaction between SA and SF and the molecular conformation of SF in the films were verified by Fourier transform infrared spectroscopy with attenuated total reflectance (FTIR-ATR), using a Bomem MB 102, equipped with a ZnSe cell crystal, in the range of
650–2000 cm−1. Each spectrum was acquired in transmittance mode through the accumulation of 256 scans. Thermogravimetric analysis (TGA) was performed in TGA-50 (Shimadzu) in a temperature range of 25–500 °C with a ramp rate of 10 °C min−1 and a N2 flow of 50 mL min−1. The data were normalized as a function of the initial mass of the sample. 2.2.2. Physical properties The thickness of the films was measured 10 times in the dry state and also in the wet state (after swelling test) for each sample using a digital micrometer (MDC-25S, Mitutoyo). Swelling degree of films was determined gravimetrically. Pieces of 2.5 cm in diameter of the films were weighed in the dry state (wi ), after reaching the equilibrium at 50% of relative humidity (48 h). Subsequently, the samples were immersed in 100 mL of distilled water (swelling medium) and were then weighed until reaching constant weight (wf). The swelling degree (%) was calculated as [(wf − wi)/wi] ⋅ 100 %. The water vapor permeability (WVP) was determined according to ASTM E 96M (2005). The samples were placed in a recipient with permeation area of 15.2 cm 2 containing anhydrous calcium chloride as desiccant and this recipient was then placed in a desiccator containing saturated aqueous NaCl solution, maintaining the ambient at 75% of relative humidity. The WVP through the films was determined gravimetrically by weighing the recipient every 12 h, for a period of 5 days. The rate of water vapor permeability was determined from the slope of the curve of mass change versus time. WVP was calculated as WVP = [(G/t) ⋅ e]/(A ⋅ Δp), where G/t is the mass variation rate (slope of the straight line), in g·day− 1; e is the film thickness, in mm; A is the test area, in m 2 ; and Δp is the vapor pressure difference, in kPa. Tensile tests were performed according to ASTM D882 (2002) using a TA.XT2 texture analyzer (Stable Microsystems SMD). The films (7 × 2.5 cm) were stored under standard conditions (25 °C, RH 50%) for 48 h before the tests. For tensile test, an initial grip separation of 50 mm and crosshead speed of 10 mm s−1 were used. The average values of tensile strength were obtained from 8 specimens. 2.2.3. Cytotoxicity test In vitro biocompatibility was performed according to ISO 10993-5 (2009) using Chinese hamster ovary cell line (CHO-k1). The cells were maintained in RPMI culture medium supplemented with antibiotics and antimicotic (100 units mL−1 of penicillin, 100 μg mL−1 of streptomycin and 0.025 μg mL−1 of amphotericin), 2 mM glutamine, and 10% calf serum, at 37 °C in a 5% CO2 humidified atmosphere until they reached confluence. For subculturing and for experiments, cells were harvested using 0.05% trypsin and 0.02% EDTA in phosphate-buffered saline solution at pH 7.4. The films were sterilized by using UV radiation for 30 min on each side of the film. The films were immersed in RPMI culture medium in a proportion of 1 cm2 mL−1, at 37 °C for 48 h. The extracts of the films were then diluted from 100% (original extract) to 6.25 vol.% in RPMI 1640. Cytotoxicity test was performed in a 96 well microplate containing 3000 cells per well and extract of films. The microplates were incubated for 72 h at 37 °C in a 5% CO2 humidified atmosphere. The cell viability was measured by adding MTS (supravital dye tetrazolium compound)/PMS (electron coupling agent) (20/1) solution and incubating for 2 h. The microplates were analyzed in a spectrophotometer ELISA at 490 nm. The test was compared with a negative control of high density polyethylene (HDPE) and a positive control of phenol 0.5 vol.% in culture medium. 3. Results In the present study, tests of SF/SA blend preparation were performed, in order to obtain a homogeneously mixed solution and a film without macroscopic phase separation. Initially, we used SA aqueous
M.A. de Moraes et al. / Materials Science and Engineering C 40 (2014) 85–91
solution; however, the blending solutions did not mix well and the final aspect of blends was heterogeneous, with formation of SF fibrils and phase separation in all blending ratios. Due to this fact, previous dissolution of SA in an alkaline solution, such as NaOH was proposed. When SA alkaline solution was used for blends, no macroscopic phase separation was detected, forming a homogeneous blend solution under the naked eye. At neutral to alkaline pHs, SF is negatively charged (pK ≅ 4.2), as well as SA (pK ≅ 4), which does not favor the electrostatic interactions. On the contrary, the negative charges on both biopolymers will increase the electrostatic repulsion between the molecules. This repulsion is probably responsible for the phase separation and fibril formation in blends with SA dissolved in water. Even though SA alkaline solution will not decrease the electrostatic repulsion between the biopolymer molecules, it will decrease SF–SF interactions, resulting in the predominance of silk I structure, that exhibits more active sites available [14]. The pH influence in SF molecule interactions can be confirmed by the fact that when lowering the pH of SF aqueous solution (pH = 7), the solution becomes gradually more viscous, until reaching the point in which the SF–SF molecule interactions are so strong that the SF solution undergoes gelation (pH ≅ 5) [15]. At this point, a solid structure with predominance of a stable silk II conformation is obtained. For SF contents higher than 25 wt.%, the formation of SF fibrils could be observed, characterizing macroscopic phase separation, even when
87
SA alkaline solution was used in the blend. SF/SA blend film without SF fibril formation could only be obtained in SF/SA 25/75 blend ratio or when the blend components were used alone (SF/SA 100/0 and 0/100). Probably, SF fibrillogenesis occurs after a critical concentration, higher than 25 wt.%, resulting in fibrils that can be observed under the naked eye. After these results, we decided to further investigate the SF/SA 25/75 blend film, in which SF fibrils were not observed. 3.1. SEM Films of pure SF and pure SA (Fig. 1-1 and -3) show typical morphology of dense films, with dense and homogeneous fracture surfaces. The SF/SA 25/75 blend film (Fig. 1-2) revealed globular structures (with ca. 1 to 10 μm of diameter) on its surface and also on its cross section. As explained before, a blend without macroscopic phase separation in the fibril form was only obtained when SA was dissolved in NaOH, but even under these conditions, microscopic phase separation in globular form was observed. Probably, the alkaline pH decreased SF–SF interactions; although it did not decrease the electrostatic repulsion between SA and SF. Thus, the final morphology is a globular phase dispersed in a continuous phase, as observed in Fig. 1. Um and Park [16] observed that the casting solvent plays a dominant role in determining the miscibility of SF blends. Studying blends of SF and poly(vinyl
Fig. 1. SEM micrographs of surface (a) and fractured cross section (b) of SF/SA blend films at 100/0 (1), 25/75 (2) and 0/100 wt.% (3).
88
M.A. de Moraes et al. / Materials Science and Engineering C 40 (2014) 85–91
alcohol) (PVA), the authors found that, when the films were cast from aqueous solution, phase separation occurred, producing a SF phase dispersed as ovals in a continuous phase of PVA. Conversely, when the films were cast from formic acid solution, a miscible blend was obtained at SEM observation (microscopic level). This might be due to the fact that SF molecules are unstable in water and exist in the colloidal state in aqueous systems due to its hydrophobic amino acids, resulting in molecular aggregation. When a better co-solvent, such as formic acid, is used, SF molecules are stabilized and their miscibility is improved. In our study, we used water as solvent for SF molecules; however, we observed the need of changing the SA solvent to form a blend without macroscopic phase separation. Furthermore, in acidic conditions, the\COO− ions of SA start to become\COOH, the electrostatic repulsion between chains is reduced and they form hydrogen bonds, resulting in increased viscosity of solution [17]. When pH value is between 3 and 4, SA undergoes gelation, which will probably be a limitation to the use of formic acid as casting solvent in SF/SA blends. Further studies in finding a suitable casting solvent for SF/SA blends are still needed in literature.
Additionally, other studies report that the addition of glycerin in blends with SF plays an important role in inducing intermolecular interactions between SF and the other blend component, decreasing the heterogeneity and the phase separation, and improving the mechanical properties of the films (increasing flexibility). In these cases, glycerin would act as a plasticizer [14,18]. This trend was also observed in our study (information observed in laboratory experiments, no data provided), where the addition of glycerin on SA solution improved the miscibility between SF and SA and also improved the flexibility of the films. To verify if the globular domains were composed of SF or SA, we proposed SF extraction using the solution CaCl2:CH3CH2OH:H2O, as described in the “Materials and methods” section. The morphology of the SF/SA 25/75 blend films after SF extraction is shown in Fig. 2. The globular structures were removed from the film surface and also from the cross-section area, indicating that the globular phase was mainly composed of SF. The solution was able to penetrate into the film, removing SF from its bulk. The SF globule extraction is more evident with increasing extraction time. Some residues of SF were observed in the fractured
Fig. 2. SEM micrographs of surface (a) and fractured cross section (b) of SF/SA 25/75 blend films after 1 h (1), 1.5 h (2) and 2 h of SF extraction (3).
M.A. de Moraes et al. / Materials Science and Engineering C 40 (2014) 85–91
cross-section micrographs, especially when the extraction time was 1.5 h, producing to some star-like structures (Fig. 2-2b). However, after high extraction time, these residual structures were not observed anymore. This fact indicates that the globules are formed mainly by a SF-rich phase. Also, by regulating the extraction time, it is possible to regulate the final morphology of SF, and to obtain micro-structured materials with different properties. Additionally, SF/SA 25/75 blend films were prepared without glycerin addition and the same SF extraction protocol was performed. As expected, the roughness observed on the surface of the blend films in Fig. 2 is attributed to glycerin leaching, once those images were not observed on the films without glycerin (data not shown). The SF/SA blend morphology studied in this paper is a typical discrete phase structure, with globules rich in SF dispersed in a matrix rich in SA. This morphology is similar to that observed in materials with phase separation mechanism of nucleation and growth [19]. SF phase separation in globular features was already observed in SF blend films with polyethylene oxide (PEO) [7,20,21], carboxymethyl chitin [22] and poly(vinyl alcohol) (PVA) [16]. In the blends with PEO and carboxymethyl chitin, SF was the major component and the second component was water-soluble; thus the authors were capable to extract the water-soluble component by immersion in water. As a result, a film consisting of SF globules is obtained, and these globules were found to be constituted by micelles (100–200 nm), with globule sizes varying from 15 to 0.8 μm [7,22]. Jin and Kaplan [7] attempted to mimic SF fiber processing by silkworm replacing sericin by PEO. The authors found that the presence of a hydrophilic material, such as sericin and PEO (as well as the SA studied in our work), alters the final stages of globule formation, accelerating their organization into silk I structure (amorphous). The SF globular structure is naturally presented in SF fibers spun from silkworm and its morphology is similar to those obtained by Jin and Kaplan [7] in SF/PEO blends and also to our SF/SA blend films. The globules obtained in our study are more spherical and uniform than those obtained in blends with PEO, PVA or carboxymethyl chitin. These self-assembled SF globules are interesting structures that can be used to design sophisticated structures with microdomains of SF. In addition, we believe that with some little variations on SF/SA ratio, it is possible to control the quantity and size of globule formation, fine tuning the structural and mechanical properties of the final material.
89
respectively [23,24]. SA typical halos were observed at 2θ = 13.7°, 16.7° and 20.7° (Fig. 3c), corresponding to 6.45, 5.30 and 4.28 Å, respectively [8,25]. On the other hand, SF/SA 25/75 blend film shows an amorphous diffractogram. This is an indication that the intermolecular interaction of SF and SA occurs in the amorphous region. Even after film stabilization with sulfuric acid and ethanol – that would increase the organizational level and the crystallinity of the biopolymers – the amorphous profile of the blend remains unaltered. This profile confirms the fact that SF globule formation occurs with stable silk I conformation [7] and for that, the presence of a hydrophilic component, such as SA, is crucial. 3.3. FTIR FTIR spectroscopy has been a useful tool to investigate the molecular conformation of SF. Fig. 4 shows the FTIR spectra for the blend films and the main absorption peaks are listed in Table 1. Pure SF film (SF/SA 100/0) revealed a FTIR spectrum with β-sheet predominance (silk II) for the amide I and II absorption bands. The amide III absorption band was located in a wavelength corresponding to silk I conformation. SA film (SF/SA 0/100) exhibited FTIR spectrum with typical absorption bands of C_O, C\O and O_H stretching; and is in accordance with the literature for SA [26]. The SF/SA 25/75 blend film presented a spectrum similar to the SF/SA 0/100 spectrum, but with the presence of SF absorption bands of amide I, II and III, slightly shifted to higher wavelengths when compared to pure SF spectrum. The bands related to amide I and amide III of the blend film overlap with SA characteristic groups, which can induce errors when analyzing the spectra at these specific wavelengths. However, the amide II band is clearly visualized in the blend spectra at 1521.7 cm− 1, related to the silk II conformation. 3.4. Thermal properties
SF/SA 0/100 and 100/0 blend films show diffraction patterns of crystalline phase (Fig. 3). Fig. 3a shows SF typical halos at 2θ = 14°, 16.8°, 18.5° and 20°, corresponding to the crystalline and organized silk II conformation, with interplanar spaces of 6.32, 5.27, 4.79 and 4.43 Å,
TGA and DTGA curves of the blend films are shown in Fig. 5. The SF/SA 25/75 blend film initiates its mass loss in a temperature between the values observed for pristine SF and SA alone. SF films exhibited the highest thermal stability. SF/SA 25/75 blend film showed similar behavior of SF/SA 0/100 film in DTGA; however a peak attributed to SF degradation is observed at 305.5 °C. This peak is observed in the SF/SA 100/0 film at 313.7 °C. The mass loss SF/SA 0/100 is observed at 222 °C, attributed to the decrosslinking of polymer network and formation of carbonaceous residues [27]. SF/SA 25/75 blend film revealed peaks related to thermal degradation of SF and SA components alone; however these peaks were shifted to intermediate temperatures, which indicates a semi-miscible blend.
Fig. 3. XRD patterns of SF/SA blend films 100/0 (a), 25/75 (b) and 0/100 wt.% (c).
Fig. 4. FTIR spectra of SF/SA blend films 100/0 (a), 25/75 (b) and 0/100 wt.% (c).
3.2. XRD
90
M.A. de Moraes et al. / Materials Science and Engineering C 40 (2014) 85–91
Table 1 FTIR absorption peaks of SF/SA blend films. SF/SA blend ratio 100/0
25/75
Table 2 Results of the physical and mechanical characterization of SF/SA blend films. Attribution of absorption peak
Properties
SF/SA blend ratio
0/100
– 1616.2
1731.9 1622.0
1731.9 1635.5
1508.2 1228.5 – – –
1521.7 1236.2 1080.0 1029.9 927.7
– 1238.2 1076.2 1027.9 927.7
100/0 C_O stretching Amide I and antisymmetric\COO\ stretching overlap Amide II Amide III and C\O stretching overlap O\H bending O\H bending C\O vibration
a
Dry thickness (μm) Wet thickness (μm)a Swelling capacity (%)b WVP (g·mm·m−2 day−1 kPa−1)b Tensile strength (MPa)c Elongation at the break (%)c a b c
24 32 59 2.7 52 1.8
± ± ± ± ± ±
25/75 2 5 5 0.2 10 0.6
33 105 220 4.8 73 3.4
± ± ± ± ± ±
0/100 1 5 13 0.1 11 0.9
31 145 310 5.2 95 7.7
± ± ± ± ± ±
4 6 6 0.4 6 2.2
Average ± standard deviation (n = 10). Average ± standard deviation (n = 3). Average ± standard deviation (n = 8).
3.5. Physical properties Table 2 depicts the results of physical and mechanical properties of SF/SA blend films. The films' thickness values in the wet state were significantly increased in the blends containing SA (SF/SA 25/75 and 0/100). SA is a highly hydrophilic biopolymer due to the presence of carboxyl and hydroxyl groups in its structure, which provides the capacity to absorb and retain large amounts of water when SA films are immersed in aqueous solutions [28]. On the other hand, SF possesses hydrophobic amino acids in its structure and does not allow the swelling and water plasticization, which could explain the low values of SF film thickness in the wet state and low swelling capacity. The same tendency is observed in the results of water vapor permeability, where SA film permeability was almost two times higher than that observed for SF
film. The SF/SA 25/75 blend film exhibited properties related to the mass content of SA and SF in the blend. From the tensile test results, it is possible to notice that the tensile strength and elongation at the break of SF/SA film is higher than the values observed for SF films. This result indicates that SF mechanical properties are improved when SF is blended with SA. In addition, the presence of glycerin in the SF/SA 25/75 blend is a key parameter, since it increases the film flexibility and improves the film handling (information observed during handling, no data provided). Multiphase blends can exhibit several morphologies and their properties are directly related to the phase's arrangement. When one phase of polymer A is dispersed in a matrix of polymer B, for example, the properties of the material are governed by the matrix properties (polymer B) [1]. This is in accordance with our results, where the structural properties of the blend films, such as swelling capacity, water vapor permeability, tensile strength and elongation at break were influenced by SA matrix. SA is known to be a good material to be used in wound dressings [13]. SA associated with the globular microdomains in the SF/SA 25/ 75 blend film and the presented physical properties indicate that the final blend is a promising material to be used in high performance microstructured dressings.
3.6. Cytotoxicity test The cell viability in the presence of the blend films remained between 90% and 100% (Table 3) and according to ISO 10993-5 (2009), these films can be considered nontoxic to cells. A cell viability of ca. 90% was observed for pure SA film (SF/SA 0/100), while the pure SF film (SF/SA 100/0) showed the highest cell viability (98%). The SF/SA 25/75 blend film showed behavior that is in between SF and SA and can be explored as potential biomaterials, mainly as devices where specific compounds or nanoparticles can be incorporated on SF globular domains and gradually released.
4. Discussion One of the main goals in polymeric blends field is to obtain compatible systems. Compatible blends are those who exhibit satisfactory properties, no matter if they are miscible (one phase), semi-miscible or immiscible (phase separation) blends.
Table 3 Viability of CHO cells in the extract of the blend films.
Fig. 5. TGA and DTGA thermograms of SF/SA blend films 100/0 (a), 25/75 (b) and 0/100 wt.% (c).
SF/SA blend ratio
Cell viability (%)a
0/100 25/75 100/0
89.1 ± 3.5 93.6 ± 1.4 98.7 ± 1.9
a
Average ± standard deviation (n = 4).
M.A. de Moraes et al. / Materials Science and Engineering C 40 (2014) 85–91
The miscibility of a mixture can be determined by the Gibbs free energy (Eq. (1)). For a miscible system, the variation in the Gibbs free energy needs to be negative. ΔGm ¼ ΔH m −TΔSm
ð1Þ
where ΔGm is the Gibbs free energy of mixture, ΔSm is the entropy of mixture, ΔHm is the enthalpy of mixture and T is the temperature. The Flory–Huggins theory was the first one to consider the high polymeric chain size and the polymer–polymer interactions (Eq. (2)). ΔGm ϕ ϕ ¼ kT 1 ln ϕ1 þ 2 ln ϕ2 þ ϕ1 ϕ2 χ 12 N0 x1 x2
ð2Þ
where x1 and x2 are the number of segments of polymers 1 and 2; ϕ1 and ϕ2 are the volumetric fractions of polymers 1 and 2 in the blend; and χ12 is the Flory–Huggins interaction parameter. The mixture of two or more polymers involves very low entropy values, once the entropic term is inversely related to the size of the polymeric chain. Thus, to obtain miscible systems it is necessary to overcome the unfavorable entropy with specific interactions between polymeric chains, responsible for a negative enthalpy. With this approach, a negative Gibbs free energy can be achieved, stabilizing thermodynamically the system. On the other hand, in a mixture of two polymers, it is necessary to overcome the energetic barrier to form a new interface to initiate the phase separation, once interfaces are unfavorable energetic surfaces. In our study, microscopic phase separation was observed in the SF/SA 25/75 blend film. As SF and SA have high molecular chains (350 kDa for SF and approximately 1000 kDa for SA), small entropy of mixture is observed when mixing them. Thus, a miscible system (ΔG b 0) is only obtained if the enthalpy of mixture is low. In other words, to obtain miscible blends from mixing polymers of high molecular chain, the polymers must have strong interactions between them. SA possesses carboxyl groups negatively charged on its surface, which allows the interaction with other functional groups of opposite charge. On the other hand, SF possesses amino groups – that can be protonated and interact with SA carboxyl – and also carboxyl groups — that can restrain the interaction of SF functional groups with SA carboxyl. Furthermore, at neutral pH SF molecules are negatively charged (pK = 4.2) and, probably, electrostatic repulsion between SF and SA takes place, resulting in a morphology consisting of a SF-rich phase (globules) dispersed in a SA continuous phase. Regarding the morphology of SF/SA blend films, we may affirm that the interaction between SF and SA chains is not strong enough to prevent phase separation. This could be predicted thermodynamically and it was also observed in our experimental study. Furthermore, the morphology of phase separation in the blend SF/SA 25/75 observed by SEM is typical of nucleation and growth, which is a mechanism of phase separation of metastable binary mixtures. Several research groups have tried to achieve self-assembled structures through chemical modifications or functionalization of surfaces. In our study, self-assembled materials were spontaneously obtained, with variable morphology, depending on the system composition (relative proportion of the blend components). Surface morphologies can be altered between globule or fibril formation, depending on the SF/SA ratio. Increasing SF proportion in the blend leads to macroscopic phase separation, with SF fibril formation; while increasing SA proportion in the blend leads to microscopic phase separation, with SF globule formation. Furthermore, we believe that the SF globule size and quantity can be regulated as a function of the SF/SA ratio, until reaching a maximum content, at which SF fibril formation occurs. Also, specific compounds can be incorporated in the SF microdomains, forming a microstructured material for controlled delivery. Considering the potential application of SF and SA in burns and wound healing, we can
91
incorporate silver, for example, in SF microdomains to prepare high performance dressings, with control over the silver release and concentration. The blend film has good swelling properties, which can absorb the wound exudate, has also good permeability, allowing fluid exchange with the environment, and still provides a mechanical barrier against infection. Additionally, the possibility to regulate the SF globule extraction opens several options to control the material morphology, from globules to star-like structures to a material with spherical void spaces. All those possibilities can be used to design sophisticated materials with SF microdomains. 5. Conclusion A blend film of SF and SA with microscopic phase separation in globular domains was obtained at SF/SA 25/75 blending ratio. The globules were mainly composed of SF, while the matrix was mainly composed of SA. The microscopic phase separation occurs due to the ΔG N 0, attributed to the high molecular chain of natural polymers and weak interactions between the chains, usually observed in biopolymeric blends. Furthermore, this self-assembled micro-structured blend film is a non-cytotoxic material, and exhibits water vapor transmission, swelling capacity, tensile strength and elongation values suitable for designing sophisticated high performance wound dressings. Acknowledgments The authors thank Andrea C. D. Rodas and Olga Z. Higa (Biotechnology Center, Energy and Nuclear Research Institute, São Paulo-SP, Brazil) for the cytotoxicity test and FAPESP and CNPq for the financial support. References [1] C. Harrats, S. Thomas, G. Groeninckx, Micro- and Nanostructured Multiphase Polymer Blend Systems: Phase Morphology and Interfaces, CRC Press, 2006. 442. [2] M. Panouille, V. Larreta-Garde, Food Hydrocoll. 23 (2009) 1074–1080. [3] G.H. Altman, F. Diaz, C. Jakuba, T. Calabro, R.L. Horan, J.S. Chen, H. Lu, J. Richmond, D. L. Kaplan, Biomaterials 24 (2003) 401–416. [4] C. Vepari, D.L. Kaplan, Prog. Polym. Sci. 32 (2007) 991–1007. [5] G.M. Nogueira, A.C.D. Rodas, C.A.P. Leite, C. Giles, O.Z. Higa, B. Polakiewicz, M.M. Beppu, Bioresour. Technol. 101 (2010) 8446–8451. [6] Y. Kawahara, K. Furukawa, T. Yamamoto, Macromol. Mater. Eng. 291 (2006) 458–462. [7] H.J. Jin, D.L. Kaplan, Nature 424 (2003) 1057–1061. [8] K.G. Lee, H.Y. Kweon, J.H. Yeo, S.O. Woo, J.H. Lee, Y.H. Park, J. Appl. Polym. Sci. 93 (2004) 2174–2179. [9] D.H. Roh, S.Y. Kang, J.Y. Kim, Y.B. Kwon, H.Y. Kweon, K.G. Lee, Y.H. Park, R.M. Baek, C. Y. Heo, J. Choe, J.H. Lee, J. Mater. Sci. Mater. Med. 17 (2006) 547–552. [10] C.X. Liang, K. Hirabayashi, J. Appl. Polym. Sci. 45 (1992) 1937–1943. [11] M. Rinaudo, Polym. Int. 57 (2008) 397–430. [12] H. Kweon, J.H. Yeo, K.G. Lee, H.C. Lee, H.S. Na, Y.H. Won, C.S. Cho, Biomed. Mater. 3 (2008) 034115 (5 pp.). [13] M. Rinaudo, Macromol. Symp. 245–246 (2006) 549–557. [14] C.L. Du, J. Jin, Y.C. Li, X.D. Kong, K.M. Wei, J.M. Yao, Mater. Sci. Eng. C Biomim. Supramol. Syst. 29 (2009) 62–68. [15] U.J. Kim, J.Y. Park, C.M. Li, H.J. Jin, R. Valluzzi, D.L. Kaplan, Biomacromolecules 5 (2004) 786–792. [16] I.C. Um, Y.H. Park, Fibers and Polym. 8 (2007) 579–585. [17] Y.M. Qin, Polym. Int. 57 (2008) 171–180. [18] L.X. Dai, J. Li, E. Yamada, J. Appl. Polym. Sci. 86 (2002) 2342–2347. [19] T. Inoue, Prog. Polym. Sci. 20 (1995) 119–153. [20] H.J. Jin, J. Park, R. Valluzzi, P. Cebe, D.L. Kaplan, Biomacromolecules 5 (2004) 711–717. [21] B.D. Lawrence, F. Omenetto, K. Chui, D.L. Kaplan, J. Mater. Sci. 43 (2008) 6967–6985. [22] P. Wongpanit, Y. Tabata, R. Rujiravanit, Macromol. Biosci. 7 (2007) 1258–1271. [23] M.Z. Li, S.Z. Lu, Z.Y. Wu, K. Tan, N. Minoura, S. Kuga, Int. J. Biol. Macromol. 30 (2002) 89–94. [24] G. Freddi, P. Monti, M. Nagura, Y. Gotoh, M. Tsukada, . Polym. Sci. B Polym. Phys. 35 (1997) 841–847. [25] C.B. Xiao, H.J. Liu, Y.S. Lu, L. Zhang, J. Macromol. Sci. Pure Appl. Chem. 38 (2001) 317–328. [26] T. Caykara, S. Demirci, M.S. Eroglu, O. Guven, Polymer 46 (2005) 10750–10757. [27] T.M.M. Swamy Siddaramaiah, B. Ramaraj, J.H. Lee, J. Appl. Polym. Sci. 109 (2008) 4075–4081. [28] S. Kalyani, B. Smitha, S. Sridhar, A. Krishnalah, Desalination 229 (2008) 68–81.
