Electrospinning of food-grade nanofibers from cellulose acetate and egg albumen blends

Journal of Food Engineering 98 (2010) 370–376

Contents lists available at ScienceDirect

Journal of Food Engineering journal homepage: www.elsevier.com/locate/jfoodeng

Electrospinning of food-grade nanofibers from cellulose acetate and egg albumen blends Saowakon Wongsasulak a,*, Manashuen Patapeejumruswong a, Jochen Weiss b, Pitt Supaphol c, Tipaporn Yoovidhya a a b c

Department of Food Engineering, Faculty of Engineering, King Mongkut’s University of Technology Thonburi, Bangkok 10140, Thailand Department of Food Science and Biotechnology, University of Hohenheim, Garbenstraße 21, 70599 Stuttgart, Germany The Petroleum and Petrochemical College, The Center for Petroleum, Petrochemical and Advanced Materials, Chulalongkorn University, Pathumwan, Bangkok 10330, Thailand

a r t i c l e

i n f o

Article history: Received 24 April 2009 Received in revised form 7 January 2010 Accepted 9 January 2010 Available online 18 January 2010 Keywords: Biomaterial Cellulose acetate Electrospinning Egg albumen Nanofibers

a b s t r a c t Edible nanofibrous thin films were fabricated for the first time from blend solutions of cellulose acetate (CA) in 85% acetic acid and egg albumen (EA) in 50% formic acid by electrospinning. The mass percentage ratios of CA–EA in the mixed solvents varied from 100:0 to 91:9, 77:23, 66:34 and 0:100. Effects of the blend ratios on the solution properties and morphology of the resulting electrospun products were studied. The results showed that EA lacked sufficient entanglement and also possessed very high surface tension, thereby being unable to form nanofibers. The addition of CA and surfactant (Tween40Ò) decreased both the electrical conductivity and the surface tension of the blends (p < 0.05), which facilitated the formation of CA–EA blend nanofibers. Scanning electron microscopic images showed that the continuity of the blend fibers was improved with an increase in the EA ratio. Fourier-transformed infrared spectroscopy and thermo-gravimetric analysis results indicated that the obtained fibers were composed of both CA and EA constituents. This study demonstrated a potential to fabricate edible nanofibers from natural food biopolymers using the electrospinning technique. Due to the properties of EA, these nanofibers could provide new functionalities with respect to in vivo-controlled release of nutraceuticals and drugs. Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction Over the past two decades, the use of novel biodegradable polymers as drug delivery vehicles in the human body has drawn increased attention (Karp and Langer, 2007; Ranganath and Wang, 2008). This is attributed to the expectancy of non-existent or minimum harm to the treated human biological system as well as to the non-requirement for post-surgical removal. Other important properties of the polymers or materials used for drug or nutraceutical carriers are that they can improve delivery efficiency by responding directly to physiological or external stimuli, can improve patient compliance, are non-toxic, and are also cost effective (Gunasekaran et al., 2007; Karp and Langer, 2007; Uchegbu and Schätzlein, 2006). The problems of conventional drug/nutraceutical delivery are low loading efficiency and the initial burst-release of the encapsulated component. Recently, release kinetics of many encapsulated active ingredients has been improved to have a closeto-zero order kinetics and a low overall degradation of the encapsulating materials. Zeng et al. (2003) reported that drugs could be encapsulated directly into electrospun fibers, whose fibrous structure had a close-to-zero order kinetics of drug release when com* Corresponding author. Tel.: +66 2 470 9341; fax: +66 2 470 9240. E-mail address: [email protected] (S. Wongsasulak). 0260-8774/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jfoodeng.2010.01.014

pared with a nanospherical structure. The other advantages of electrospun fibrous film over conventional solution-cast films is the highly porous nature of the fibrous films; these exhibit a much greater surface area, thus allowing active molecules to diffuse out from the matrix much more freely (Agarwal et al., 2008; Kenawy et al., 2002, 2009; Zong et al., 2002). Electrospun biodegradable nanofibers used as drug carriers (Agarwal et al., 2008; Kenawy et al., 2002, 2009; Chew et al., 2005; Taepaiboon et al., 2007) and used for post-operative local chemotherapy (Ranganath and Wang, 2008; Xu et al., 2005) have thus become promising new developments. Potential applications of nanofibrous films in food sectors are, for instances, delivery vehicles for nutrients and other active compounds, protective entities for encapsulated active compounds during food processing, and bioseparation media to enhance food safety (Kriegel et al., 2008; Li et al., 2009; Sozer and Kokini, 2009). Electrospinning is a simple technique used to form long, ultrafine fibers in the form of a non-woven web. The diameters of the electrospun fibers vary from sub-micrometers down to nanometers, depending on the polymer solution properties and processing conditions (Deitzel et al., 2001; Sill and von Recum, 2008). In the fiber-formation process, the hemi-spherical shape of an electrostatically-charged polymer solution droplet at the tip of a capillary is distorted to finally assume the shape of a cone, known as the Taylor cone, upon the application of the electric field across the

371

S. Wongsasulak et al. / Journal of Food Engineering 98 (2010) 370–376

Nomenclature c* K n

critical chain overlap concentration flow behavior consistency flow behavior index

capillary tip and a conductive collecting target. Beyond a critical value of the applied electric field, a stream of electrostaticallycharged polymer solution (i.e., charged jet) is ejected from the apex of the cone (Deitzel et al., 2001; Sill and von Recum, 2008; Yarin et al., 2001; Reneker and Chun, 1996). During the transportation of the charged jet to the collecting target, the jet elongates (due to the electrical forces associated with the interaction of the excess charges within the jet to the electric field and the Coulombic repulsion of the mutual charges within the jet) and, at the same time, the solvent evaporates, leaving fibers in the form of a non-woven web on the collecting target. Egg albumen (EA) is an excellent natural source of protein. In this research, EA was chosen as the material for further fabrication into a fibrous form by electrospinning because it is a low cost material, compared with other proteins, and it is readily available in Thailand as a by-product of Thai dessert manufacturing. Moreover, EA exhibits several interesting functionalities that make it interesting for various industries. It has the ability to stabilize emulsions and a high potential to be used as carriers for controlled release of active compounds (due to its pH- and temperature-sensitive properties). In the latter, the release pattern of the active compounds can be easily manipulated through regulation of the structure-forming condition (Wongsasulak et al., 2007a). Some of us have demonstrated previously that egg albumen-based ultrafine fibers are superior nutraceutical carriers, in that they can withstand the stomach’s severe acidic condition and release the bioactive material at the alkaline pH of the intestine (Wongsasulak et al., 2007b). However, the globular molecular structure of EA causes low polymeric solution entanglement, leading to failure in nanofiber formation by electrospinning. This research work was, therefore, aimed at fabricating edible EA-based nanofibers by means of blending EA with cellulose acetate (CA), an edible polymer, which is an electrospinnable material (Taepaiboon et al., 2007; Ma et al., 2005; Son et al., 2006; Tungprapa et al., 2007; Suwantong, et al., 2007; Han et al., 2008).

r c_ ga

shear stress (Pa) shear rate (s1) apparent viscosity (Pa s)

acid (50% w/v) and acetic acid (85% w/v) were respectively chosen as the solvent for egg albumen and cellulose acetate with respect to their ability to dissolve the polymers. Furthermore, formic acid was previously found to be a good solvent to initiate electrospinning of proteinaceous polymers (Martin et al., 1997; Bucho et al., 1999; Ki et al., 2005; Park et al., 2004; Leeden et al., 2000; Um et al., 2001; George and Sander, 2004). Table 1 summarizes the schemes for the preparation of the stock and the blend solutions. Briefly, CA stock solution with a concentration of 20% (w/w) in 85% acetic acid and EA stock solution with a concentration of 12% (w/w) in 50% formic acid were prepared. Each stock solution was gently stirred to ensure a complete dissolution prior to blending together. The mass compositions of CA to EA in the blend solution were varied as follows: 91:9, 77:23 and 66:34. The resulting blends were further stirred to obtain a homogeneous mixed solution. In each respective CA–EA blend, Tween40Ò was added (i.e., 10%, 18% and 23% w/w) corresponding to the CA:EA blend ratio of 91:9, 77:23 and 66:34, respectively. These blend solutions, referred to as blend A, blend B and blend C, were then subjected to the electrospinning. All solutions were prepared, characterized and electrospun at room condition (i.e., 25 ± 1 °C and 50 ± 5% RH). 2.3. Electrospinning A 30-mL syringe (TOP, Japan) containing 10 mL of the blend solution, attached with a stainless steel needle with an inner diameter of 0.69 mm (Hamilton No. 91019), was mounted on a syringe pump (New Era Pump Systems; NE1000, USA). The needle was connected to a high-voltage generator (Model ES30P-5W; Gamma High Voltage, FL, USA), operating in the positive DC mode at 15 kV. A tip-to-target distance of 15 cm was used. A grounded and cleaned copper plate was used to collect the electrospun products. The solution flow rate was controlled and kept constant at 2.2 mL/ h by the syringe pump.

2. Materials and methods 2.1. Materials Egg albumen (EA) powder (Igreca, France) was purchased from Winner Group Enterprise (Thailand). Egg albumen is a commercial product consisting of a mixture of different proteins with molecular weights ranging from 14 to 76 kDa (Li-Chan et al., 1995). Cellulose acetate (CA; white powder; Mw  30 kDa; acetyl content = 39.7 wt.%; degree of acetyl substitution  2.4) was purchased from Sigma–Aldrich (Switzerland). Formic acid (commercial grade; Samsung Fine Chemicals, Korea) and glacial acetic acid (analytical reagent grade; Carlo Erba, Italy) were used as solvents of the polymer. Tween40Ò (analytical reagent grade; Fluka, Germany) was used as a non-ionic surfactant. 2.2. Preparation of polymer solutions for electrospinning CA–EA blend solutions were prepared by blending two primary solutions: 20% (w/w) of CA dissolved in 85% acetic acid and 12% (w/w) of EA dissolved in 50% formic acid, at various ratios. Formic

Table 1 Preparation scheme of the primary CA and EA solutions and their resulting blend solutions (i.e., blend A, blend B and blend C) with the addition of Tween40Ò, a nonionic surfactant, for electrospinning. Preparation step of

Components

Stock solution of CA; stock CA

CA powder 85% acetic acid

20.0 g 80.0 g

Stock solution of EA; stock EA

EA powder 50% formic acid

12.0 g 88.0 g

CA–EA blending solutions

Stock solution blending volume ratio CA:EA

Samples

Polymer weight ratio (CA:EA)

Total polymer conc. (% w/w)

Stock CA (g)/stock EA (g)/Tween40Ò (g)

Pure CA Blend A Blend B Blend C Pure EA

100:0 91:9 77:23 66:34 0:100

20.0 19.25 18.25 17.25 12.00

20.00/–/– 3.64/0.36/0.40 3.08/0.92/0.72 2.64/1.36/0.92 –/20.00/–

(The concentrations of the stock solutions were prepared according to the results reported in Section 3.1).

372

S. Wongsasulak et al. / Journal of Food Engineering 98 (2010) 370–376

2.4. Rheology The apparent viscosity of the polymer solution was determined by using a rotational rheometer (Haake VT500, Germany), equipped with a double coaxial stainless steel bob and cup (length = 61.4 mm, diameter = 10.1 mm, gap width = 1.45 mm). Shear stress (r, Pa) of the blend solution was recorded as a function of shear rate (c_ , s1), ranging between 0 and 400 s1. The reported results were averaged from triplicate measurements. The recorded data of shear stress versus shear rate were fitted to the power law model (Eq. (1)). Values of the flow behavior consistency K and the flow behavior index n were then calculated, as described by Lapasin and Pricl (1995):

r ¼ K c_ n ;

ð1Þ

where r is the shear stress and c_ is the shear rate. The flow behavior index n reflects the type of the solution: i.e., n = 1, if the solution behaves as a Newtonian liquid and n – 1, if the solution behaves as a non-Newtonian liquid (e.g., shear thinning or shear thickening). 2.5. Tensiometry Surface tension of the polymer solution was determined by the Du Noüy ring method, using a digital tensiometer (Model DCT11; Dataphysics, Germany). Ring diameter was 18.7 mm. Twenty milliliters of each sample solution was poured into a clean glass container, having a diameter of 6.7 cm. The De Noüy ring was first submerged into the solution below the solution meniscus and then vertically pulled out of the solution. The surface tension (mN/m) was recorded at the maximum pulling force prior to the ring breaking through the solution meniscus. 2.6. Electrical conductivity Electrical conductivity of the polymer solution was determined by a digital conductivity meter (Model CG855; Schott, Germany). Calibration was done using NaCl standard solution (1000 lS/cm). Results are averaged from triplicate measurements and two sets of sample solutions. 2.7. Scanning electron microscopy The morphology of the electrospun products (fibers and/or beads) was evaluated by scanning electron microscope (SEM) (Model JSM 5800; JEOL, USA). The electrospun products were collected directly on ground SEM brass stubs that were mounted on the collector plate. After deposition of the samples, the stubs were removed and sputter-coated with gold using a coating condition of 2 kV, 5 mA for 2 min, with an argon backfill at 8 Pa. The average diameters of the obtained fibrous products were determined by an image-analytical software (Image J; NIMH, Maryland, USA). 2.8. Fourier-transformed infrared spectroscopy To determine composition and chemical characteristics of the electrospun fibers, infrared spectra of the fibers were recorded using a Fourier-transformed infrared spectrophotometer (Model Spectrum GX; Perkin–Elmer, USA), operated at a wave number range of 4000–400 cm1 and a resolution of 4 cm1. Each measurement was carried out from an average of 16 scans. 2.9. Thermal gravimetric analysis Thermal gravimetric analysis in the form of the first derivative mass loss curves (dm/dT) was performed on a PerkinElmer Pyris I TGA HT instrument. The samples (7–8 mg) were heated on plati-

num pans at a heating rate of 20 °C/min, and under a nitrogen atmosphere at a controlled flow rate of 20 mL/min. TGA and differential thermal gravimetric (DTG) curves were recorded from 50 to 700 °C. The reported data were averages of three scans. 2.10. Statistical analysis All the measured data were ranked and evaluated for their statistical differences, using one-way analysis of variance (ANOVA) and turkey’s multiples range test. A value of p < 0.05 was considered significantly different. The data were expressed as mean ± SD. 3. Results and discussion 3.1. Optimization on concentrations of primary stock solutions of EA and CA In a set of preliminary studies, CA solutions of varying concentrations of 8%, 15% and 17% (w/w) in 85% acetic acid were prepared and tested for their ability to produce electrospun nanofibers (data not shown). SEM micrographs indicated that the minimum concentration of CA that resulted in the formation of smooth fibers was 17% (w/v) in 85% acetic acid. This was in good agreement with Han et al. (2008), who reported that pure CA fibers were successfully produced from 17% (w/v) of CA in 85% acetic acid. However, electrospinning of the CA solution in its blends with the primary solution of EA (i.e., 5% (w/w) in 85% formic acid) failed to form fibers, as only discrete droplets were obtained at all of the mass ratios of CA:EA investigated (i.e., 91:9, 77:23 and 66:34). Analysis of the shear flow behavior of the blends revealed that the zero-shear viscosity values of the blends were too low (data not shown here), which could lead to inadequate molecular entanglements necessary for the formation of stable continuous jets, hence the failure to finally form the fibers. In a more technical context, the failure to form the fibers could be explained in terms of the critical chain overlap concentration (c*) of the polymer solution, i.e., the cross-over between the dilute and the semi-dilute concentration regimes, which is normally used to determine the degree of polymer entanglement that is an important key factor for fiber formation (Gupta et al., 2005). In a recent work by some of us (Wongsasulak et al., 2007b), the concentration exponent for EA solution in formic acid was estimated to be 1.32 (g  c1.32). The value of 1.32 indicates the semi-dilute un-entangled solutions of polymers with respect to the determined values of Gupta et al. (2005). Accordingly, the concentration of the primary EA solution was increased from 5% to 12% (w/w), whilst that of the primary CA solution was increased from 17% to 20% (w/w). Nevertheless, because of the higher EA concentration, the formic acid concentration had to be reduced from 85% to 50%, in order to avoid gelation of the EA solution. Preliminary results indicated the successful production of fibers from the blends of 12% (w/w) EA solution in 50% formic acid and 20% (w/w) CA solution in 85% acetic acid. These stock solutions were then used to prepare the blend solutions for all of the subsequent studies. 3.2. Certain properties of CA–EA blend solutions Viscosity, surface tension and electrical conductivity of the blend solutions were measured and the results are shown in Table 2. The apparent viscosity of the pure CA solution at a shear rate of 100 s1 (ga,100) was 3.10 ± 0.07 Pa s, while that of the EA counterpart was 0.37 ± 0.04 Pa s. As the ratio of EA increased, the viscosity of the blend solutions decreased significantly (p < 0.05). The surface tension of the primary EA solution was slightly greater than that of the primary CA solution (i.e., 34.28 ± 0.06 versus

373

S. Wongsasulak et al. / Journal of Food Engineering 98 (2010) 370–376 Table 2 Solution properties of the polymer blends with no Tween40Ò added for electrospinning; measured at 25 ± 0.5 °C. Percentage ratio (CA:EA)

Apparent viscosity ga,100 (Pa s)

Power law consistency coefficient (K)

Power law flow behavior index (n)

Surface tension (mN/m)

Electrical conductivity (mS/cm)

100:0 91:9 77:23 66:34 0:100

3.10 ± 0.07e 2.80 ± 0.02d 1.48 ± 0.04c 1.36 ± 0.04b 0.37 ± 0.04a

4.33 ± 0.20a 5.43 ± 0.12b 5.63 ± 0.04c 5.42 ± 0.11b 6.85 ± 0.37d

0.93 ± 0.01e 0.92 ± 0.01d 0.83 ± 0.02c 0.79 ± 0.01b 0.37 ± 0.02a

29.68 ± 0.06a 30.44 ± 0.03b 31.25 ± 0.10c 32.03 ± 0.10d 34.28 ± 0.06e

0.02 ± 0.00a 0.17 ± 0.00b 0.69 ± 0.00c 1.05 ± 0.00d 5.75 ± 0.01e

The superscripts a–e indicate statistically significant differences among the values of each column.

29.68 ± 0.06 mN/m). This results in the observed increase in the surface tension of the blend solutions, with an increase in the EA composition (p < 0.05). It should be noted again that the particular set the blend solutions was prepared without the addition of Tween40Ò. In a similar manner, the electrical conductivity of the primary EA solution was very high (5.75 ± 0.01 mS/cm), while that of the primary CA solution was much lower (0.02 ± 0.00 mS/cm). As a result, the electrical conductivity of the blend solutions increased from that of the primary CA solution, with an increase in the EA composition. With respect to the fiber-forming ability of the blend solutions, SEM micrographs revealed that only the primary CA solution yielded beaded fibers or fibers with some bead defects. The primary EA and the resulting blend solutions failed to produce fibers (data not shown here). This could be attributed to the fact that the solution of EA in 50% formic acid possesses a high surface tension and a high electrical conductivity (Table 1). The high surface tension might counteract the electrical forces, thus preventing the successful ejection of a steady polymer jet from the tip of syringe. In order to solve the problem associated with the high surface tension of the blend solutions, a food-grade non-ionic surfactant (Tween40Ò) was added to produce blend A, blend B and blend C at 10%, 18% and 23% (w/w), respectively. The use of surfactant to aid the electrospinning of a polymer solution has previously been reported (Kriegel et al., 2009; Peng et al., 2008). The effect of Tween40Ò on viscosity, surface tension and electrical conductivity of the resulting blend solutions is shown in Table 3. The apparent shear viscosity values of the CA–EA blend solutions containing Tween40Ò, at a shear rate of 100 s1 (ga,100), decreased slightly from those of the blend solutions without the surfactant. Evidently, the addition of Tween40Ò decreased both the surface tension and the electrical conductivity values of the blend solutions significantly (p < 0.05) (Table 4). As previously noted, the reduction in surface tension of the blend solutions should be a key factor determining their fiber formation (Fig. 1). Specifically, to form fibers, the applied voltage should be sufficient to overcome the surface tension at the tip of the Taylor cone, in order to initiate the ejection of the charged jet. The entanglement of the polymeric molecules in the jet segments should thus be high enough to prevent the total break-up of the jets from the stretching forces imposed by both the electric field and the repulsion of mutual charges (Reneker et al., 2000). However, the shear viscosity of EA in the blend solutions was low, owing to its globular nature, and Tween40Ò was added to reduce the surface tension of

the blend solutions, thereby improving the electrospinnability. This result agreed well with the study of Peng et al. (2008), who reported the important role of surfactants in regulating morphology of the electrospun fibers. Specifically, they reported that electrospun poly (L-lactic acid) fibers with smaller and more uniform diameters were obtained when certain amounts of surfactant were added to the PLLA solutions. Jung et al. (2005) also reported that morphologies and diameters of electrospun poly (vinyl alcohol) fibers were influenced by the type and the amount of surfactant that was added into the polymer solutions. Here, the reduction in the electrical conductivity upon the addition of the non-ionic surfactant (Tween40Ò) to the CA–EA blend solutions could be explained by the binding of surfactant monomers to the backbone of EA, thereby reducing its polyelectrolytic property. This possibility was recently discussed by Kuhn et al. (1998), Klinkesorn and Namatsila (2009), Mun et al., (2006), and Tadros (2009). As mentioned previously, that the shear viscosity of the blend solutions was not significantly affected by the addition of the surfactant suggests that the presence of Tween40Ò should not interfere with the dynamic properties of the polymers, but it instead facilitates the dispersion of EA molecules throughout the mass of the solutions (Klinkesorn and Namatsila, 2009; Tadros, 2009). A similar result was reported by Kriegel et al. (2009), who showed that the addition of non-ionic surfactants had little influence on the shear viscosity of the chitosan–PEO blend solutions. In contrast, the addition of an ionic surfactant, due to the much greater extent of polymer–surfactant interactions, may lead to substantial changes in the flow behavior of the blend solutions (Kriegel et al., 2009; Jung et al., 2005). 3.3. Electrospinning of CA–EA blend solutions containing Tween40Ò and morphology of the obtained fibers In this study, the primary solutions of CA and EA and their corresponding blend solutions containing Tween40Ò were electrospun under the applied electrical potential of 15 kV over a tip-totarget distance of 15 cm and a flow rate of 2.2 mL/h (tip’s inner diameter 0.69 mm). Fig. 1 shows SEM micrographs (5000) illustrating morphologies of the electrospun products. Electrospinning of the primary EA solution resulted in the deposition of submicron particles, while electrospinning of the primary CA counterpart yielded nanofibers with an average diameter of 265 ± 48 nm (Fig. 1a). For the CA–EA blend solutions containing Tween40Ò (i.e., blend A, blend B and blend C), the average fiber diameters

Table 3 Solution properties of the polymer blends with Tween40Ò added for electrospinning; measured at 25 ± 0.5 °C. Percentage ratio (CA:EA)

Apparent viscosity, ga,100 (Pa s)

Power law consistency coefficient (K)

Power law flow behavior index (n)

Surface tension (mN/m)

Electrical conductivity (mS/cm)

91:9 77:23 66:34

2.73 ± 0.06c 1.32 ± 0.01b 1.19 ± 0.09a

5.42 ± 0.23b 5.81 ± 0.17b 3.51 ± 0.26a

0.85 ± 0.01c 0.68 ± 0.01a 0.77 ± 0.01b

28.69 ± 0.02a 29.16 ± 0.05b 29.28 ± 0.05c

0.14 ± 0.01a 0.42 ± 0.00b 0.73 ± 0.01c

The superscripts a–c indicate statistically significant differences among the values of each column.

374

S. Wongsasulak et al. / Journal of Food Engineering 98 (2010) 370–376

Table 4 Solution properties of the polymer blends without/with Tween40Ò added for electrospinning; measured at 25 ± 0.5 °C. Percentage ratio (CA:EA) Samples 91:9 77:23 66:34

Apparent viscosity; ga,100 (Pa s) Without Tween40 a

2.80 ± 0.02 1.48 ± 0.04b 1.36 ± 0.04b

Ò

Surface tension (mN/m)

With Tween40 a

2.73 ± 0.06 1.32 ± 0.01a 1.19 ± 0.09a

Ò

Without Tween40 b

30.44 ± 0.03 31.25 ± 0.10b 32.03 ± 0.10b

Ò

Electrical conductivity (mS/cm) With Tween40

Ò

a

28.69 ± 0.02 29.16 ± 0.05a 29.28 ± 0.05a

Without Tween40Ò b

0.17 ± 0.00 0.69 ± 0.00b 1.05 ± 0.00b

With Tween40Ò 0.14 ± 0.01a 0.42 ± 0.00a 0.73 ± 0.01a

The superscripts a, b indicate statistically significant differences between the values of each parameter for each sample.

Fig. 1. SEM micrographs of the electrospun products from (a) the primary CA solution (20 wt.% in 85% acetic acid), (b) blend A, (c) blend B, (d) blend C and (e) the primary EA solution (12 wt.% in 50% formic acid). Please see Table 1 for the definitions of blend A, blend B and blend C. The bar scales are 5 lm.

were 242 ± 32, 384 ± 54 and 410 ± 38 nm (Fig. 1b–d), respectively. Clearly, the addition of EA increased the average fiber diameters and also improved the continuity of the blend fibers. The observed increase in the diameters of the blend fibers with an increase in the EA content should be due to the increase in the electrical conductivity of the blend solutions, leading to an increase in the electrical forces exerting on the ejected jet, which, in turn, increased the mass throughput (in addition to that driven by the syringe pump). A closer examination of the blend fibers revealed some cracks on the fiber surfaces. These cracks were not observed on the surfaces of the CA fibers. Nevertheless, the surface of the blend fibers became smoother with an increase in the EA content (Fig. 1b–d). This could be a result of the diffusion of EA molecules to locate close to the surface of the fibers. To verify such a hypothesis, additional experiments using a more sophisticate tool like X-ray photoelectron spectroscope are necessary to gain an insight into the chemical composition of the fiber surfaces.

3.4. Compositional analysis of electrospun nanofibers from the CA–EA blend solutions containing Tween40Ò FTIR spectra of the electrospun products from the primary solutions of CA and EA indicated that electrospun fibers were composed of both CA and EA (Fig. 2). The FTIR spectra of electrospun particles of EA had characteristic peaks at about 3400 and 1600 cm1 indicating amine and carboxylic acid groups, respectively (Stuart, 2004). On the other hand, the characteristic peaks of pure CA fibers were located at about 3500, 1750, 1250 and 1050 cm1. The absorption peak at 3500 cm1 could be attributed to the presence of a hydroxyl group, while the peaks at 1750, 1250 and 1050 cm1 indicated the presence of carbonyl group, ether group, and –C–O– bond of the –CH2–OH group, respectively (Brandenburg and Seydel, 1996). Thermal gravimetric results in the form of the first derivative mass loss curves (dm/dT) of electrospun fibers or particles sub-

375

S. Wongsasulak et al. / Journal of Food Engineering 98 (2010) 370–376

CA

EA

Blend C

Blend A

Blend B

Blend B

Blend C

Blend A

EA

CA

100

200

300

400

500

600

700

Temperature (oC) Fig. 3. First derivatives dm/dT in mg/°C of TGA results as a function of the furnace temperature of the electrospun products from the primary CA solution (20 wt.% in 85% acetic acid), the CA–EA blend solutions (i.e., blend A, blend B and blend C) and the primary EA solution (12 wt.% in 50% formic acid).

4000

3000

2000

1000

Wavenumber (cm -1 )

jected to a temperature ramp in a range of 50–700 °C, are shown in Fig. 3. The dTGA data confirmed that electrospun blend fibers were composed of both CA and EA. Thermal degradation of pure CA electrospun fibers was observed at 370 °C, while degradation of pure EA electrospray occurred in the temperature ranges of 170–250 °C and 270–350 °C. The char content at 700 °C of the electrospun pure CA fibers was about 5%, while that of pure EA electrospray was about 15%. The characteristic weight loss revealed an improvement in the thermal stability of the electrospun blend fibers. Fig. 4 shows the compositions of the composite electrospun fibers as well as those of polymer solution obtained from deconvolution of the dTGA spectra of Fig. 3 followed by integration of the peak areas. Electrospun composite fibers produced from the solutions of blend A, blend B and blend C contained 21.3%, 48.3% and 59.1% of EA, respectively; the levels that were noticeably higher than the concentration of EA in the electrospinning solution. It is interesting to note that EA was preferentially electrospun since the protein

100

Blend solutions

Blend electrospun fibers

CA EA

80 Composition ratio(%)

Fig. 2. FT-IR spectra of the electrospun products from the primary CA solution (20 wt.% in 85% acetic acid), the CA–EA blend solutions (i.e., blend A, blend B and blend C) and the primary EA solution (12 wt.% in 50% formic acid) over the wave number range of 4000–400 cm1.

60

40

20

0

A pure C Blend A Blend B Blend C

A pure C Blend A Blend B Blend C

Fig. 4. Comparison of cellulose acetate (CA):egg albumen (EA) compositional ratios in the blend solutions and those in the resulting blend fibers. These were estimated from thermo-gravimetric analysis.

has a globular structure and thus did not readily entangle to form a continuous fiber, as evidenced by the fact that pure EA could not be electrospun into fibers. EA, similar to many proteins, possesses

376

S. Wongsasulak et al. / Journal of Food Engineering 98 (2010) 370–376

surface active property and thus readily accumulates at surfaces and interfaces of the constituent polymer molecules (Mun et al., 2006; Tadros, 2009) of the constituent polymers’ molecules. The preferential adsorption of EA at the surface of the electrospinning blend solution could thus be a contributing factor to the observed increased concentrations of EA.

4. Conclusion This study demonstrated the successful fabrication of edible electrospun fibrous thin film from blend solutions of CA and EA. The study illustrated that electrospun fibers from edible biopolymers could be fabricated through regulation of solution composition as well as addition of surfactant to modulate the solution properties. The fabricated edible polysaccharide–protein composite nanofibers are expected to find applications in food packaging materials due to their potential in increasing safety and quality of packed foods. The fibers could also be potentially used for controlled release of nutraceuticals or pharmaceuticals in gastrointestinal tract. Acknowledgement This research project was financially supported by a grant from the Thailand Research Fund (TRF, Grant No. TRG 5080008). References Agarwal, S., Wendorff, J.H., Greiner, A., 2008. Use of electrospinning technique for biomedical applications. Polymer 49 (26), 5603–5621. Brandenburg, K., Seydel, U., 1996. Fourier transform infrared spectroscopy of cell surface polysaccharides. In: Mantsch, H.H., Chapman, D. (Eds.), Infrared Spectroscopy of Biomolecules. John Wiley & Sons, New York, pp. 207–216. Bucho, C.J., Chen, L.C., Shen, Y., Martin, D.C., 1999. Processing and microstructural characterization of porous biocompatible protein polymer thin films. Polymer 40 (26), 7397–7407. Chew, S.Y., Wen, J., Yim, E.K., Leong, K.W., 2005. Sustained release of proteins from electrospun biodegradable fibers. Biomacromolecules 6 (4), 2017–2024. Deitzel, J.M., Kleinmeyer, J.D., Hirvonen, J.K., Beck Tan, N.C., 2001. Controlled deposition of electrospun poly(ethylene oxide) fibers. Polymer 42 (19), 8163– 8170. George, L., Sander, W., 2004. Matrix isolation infrared and ab initio study of the hydrogen bonding between formic acid and water. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 60 (13), 3225–3232. Gunasekaran, S., Ko, S., Xiao, L., 2007. Use of whey proteins for encapsulation and controlled delivery applications. Journal of Food Engineering 83 (1), 31–40. Gupta, P., Elkins, C., Long, T.E., Wilkes, G.L., 2005. Electrospinning of linear homopolymers of poly(methyl methacrylate): exploring relationships between fiber formation, viscosity, molecular weight and concentration in a good solvent. Polymer 17 (13), 4799–4810. Han, S.O., Youk, J.H., Min, K.D., Kang, Y.O., Park, W.H., 2008. Electrospinning of cellulose acetate nanofibers using a mixed solvent of acetic acid/water: effects of solvent composition on the fiber diameter. Materials Letters 62 (4–5), 759– 762. Jung, Y.H., Kim, H.Y., Lee, D.R., Park, S.Y., 2005. Characterization of PVOH nonwoven mats prepared from surfactant–polymer system via electrospinning. Macromolecular Research 13 (5), 385–390. Karp, J.M., Langer, R., 2007. Development and therapeutic applications of advanced biomaterials. Current Opinion in Biotechnology 18 (5), 454–459. Kenawy, E.R., Bowlin, G.L., Mansfield, K., Layman, J., Simpson, D.G., Sanders, E.H., Wnek, G.E., 2002. Release of tetracycline hydrochloride from electrospun poly(ethylene-co-vinylacetate), poly(lactic acid), and a blend. Journal of Controlled Release 81 (1–2), 57–64. Kenawy, E.R., Abdel-Hay, F.I., El-Newehy, M.H., Wnek, G.E., 2009. Processing of polymer nanofibers through electrospinning as drug delivery systems. Materials Chemistry and Physics 113 (1), 296–302. Ki, C.H., Baek, D.H., Gang, K.D., Lee, K.H., Um, I.C., Park, Y.H., 2005. Characterization of gelatin nanofiber prepared from gelatin–formic acid solution. Polymer 46 (14), 5094–5102. Klinkesorn, U., Namatsila, Y., 2009. Influence of chitosan and NaCl on physicochemical properties of low-acid tuna oil-in-water emulsions stabilized by non-ionic surfactant. Food Hydrocolloid 23 (5), 1374–1380.

Kriegel, C., Arrechi, A., Kit, K.M., McClements, D.J., Weiss, J., 2008. Fabrication, functionalization, and application of electrospun biopolymer nanofibers. Critical Reviews in Food Science and Nutrition 48 (8), 775–797. Kriegel, C., Kit, K.M., McClements, D.J., Weiss, J., 2009. Electrospinning of chitosan– poly(ethylene oxide) blend nanofibers in the presence of micellar surfactant solutions. Polymer 50 (1), 189–200. Kuhn, P.S., Levin, Y., Barbosa, M.C., 1998. Complex formation between polyelectrolytes and ionic surfactants. Chemical Physics Letters 298 (1–3), 51–56. Lapasin, R., Pricl, S., 1995. Secondary and Tertiary Structure of Polysaccharides in Solutions and Gels: Rheology of Industrial Polysaccharides – Theory and Applications. Blackie Academic and Professional Publishers, Melbourne. Leeden, V.D.M.C., Rutten, A.A.C.M., Frens, G.J., 2000. How to develop globular proteins into adhesives. Journal of Biotechnology 79 (3), 211–221. Li-Chan, E.C., Powrie, W.D., Nakai, S., 1995. The chemistry of eggs and egg products. In: Stadelman, W.J., Cotterill, O.J. (Eds.), Egg Science and Technology, fourth ed. Food Product Press, New York, pp. 105–175. Li, Y., Lim, L.T., Kakuda, Y., 2009. Electrospun zein fibers as carriers to stabilize ()epigallocatechin gallate. Journal of Food Science 74 (3), C233–C240. Ma, Z., Kotaki, M., Ramakrishna, S., 2005. Electrospun cellulose nanofiber as affinity membrane. Journal of Membrane Science 265 (1–2), 115–123. Martin, D.C., Jiang, T., Buchko, J., 1997. Protein-Based Materials. Birkhauser, Boston. Mun, S., Decker, E.A., McClements, D.J., 2006. Effect of molecular weight and degree of deacetylation of chitosan on the formation of oil-in-water emulsions stabilized by surfactant–chitosan membranes. Journal of Colloid and Interface Science 296 (2), 581–590. Park, W.H., Jeong, L., Yoo, D., Hudson, S., 2004. Effect of chitosan on morphology and conformation of electrospun silk fibroin nanofibers. Polymer 45 (21), 7151– 7157. Peng, H., Zhou, S., Guo, T., Li, Y., Li, X., Wang, J., Weng, J., 2008. In vitro degradation and release profiles for electrospun polymeric fibers containing paracetanol. Colloids and Surfaces B: Biointerfaces 66 (2), 206–212. Ranganath, S.H., Wang, C.H., 2008. Biodegradable microfiber implants delivering paclitaxel for post-surgical chemotherapy against malignant glioma. Biomaterials 29 (20), 2996–3003. Reneker, D.H., Chun, I., 1996. Nanometre diameter fibres of polymer, produced by electrospinning. Nanotechnology 7 (3), 216–223. Reneker, D.H., Yarin, A.L., Fong, H., Koombhongse, S., 2000. Bending instability of electrically charged liquid jets of polymer solutions in electrospinning. Journal of Applied Physics 87 (9), 4531–4547. Sill, T.J., von Recum, A., 2008. Electrospinning: applications in drug delivery and tissue engineering. Biomaterials 29 (13), 1989–2006. Sozer, N., Kokini, J.L., 2009. Nanotechnology and its applications in the food sector. Trends in Biotechnology 27 (2), 82–89. Son, W.K., Youk, J.H., Park, W.H., 2006. Antimicrobial cellulose acetate nanofibers containing silver nanoparticles. Carbohydrate Polymers 65 (4), 430–434. Stuart, B.H., 2004. Organic Molecules: Infrared Spectroscopy: Fundamentals and Applications. John Wiley & Sons, New Jersey (pp. 7193). Suwantong, O., Opanasopit, P., Ruktanonchai, U., Supaphol, P., 2007. Electrospun cellulose acetate fiber mats containing curcumin and release characteristic of the herbal substance. Polymer 48 (26), 7546–7557. Tadros, T., 2009. Polymeric surfactants in disperse systems. Advances in Colloid and Interface Science 147–148(Special issue), 281–299. Taepaiboon, P., Rungsardthong, U., Supaphol, P., 2007. Vitamin-loaded electrospun cellulose acetate nanofiber mats as transdermal and dermal therapeutic agents of vitamin A acid and vitamin E. European Journal of Pharmaceutics and Biopharmaceutics 67 (2), 387–397. Tungprapa, S., Jangchud, I., Supaphol, P., 2007. Release characteristics of four model drugs from drug-loaded electrospun cellulose acetate fiber mats. Polymer 48 (17), 5030–5041. Uchegbu, F.I., Schätzlein, G.A., 2006. Polymer in Drug Delivery. CRC Press, Taylor & Francis Group, FL. Um, I.C., Kweon, H.Y., Park, Y.H., Hudson, S., 2001. Structural characteristics and properties of the regenerated silk fibroin prepared from formic acid. International Journal of Biological Macromolecules 33 (2), 91–97. Wongsasulak, S., Yoovidhya, T., Bhumiratana, S., Hongsprabhas, P., 2007a. Physical properties of egg albumen and cassava starch composite network formed by a salt-induced gelation method. Food Research International 40 (2), 249–256. Wongsasulak, S., Kit, K.M., McClements, D.J., Yoovidhya, T., Weiss, J., 2007b. The effect of solution properties on the morphology of ultrafine electrospun egg albumen–PEO composite fibers. Polymer 48 (2), 448–457. Xu, X., Yang, L., Xu, X., Wang, X., Chen, X., Liang, Q., 2005. Ultrafine medicated fibers electrospun from W/O emulsions. Journal Controlled Release 108 (1), 33–42. Yarin, A.L., Koombhongse, S., Reneker, D.H., 2001. Taylor cone and jetting from liquid droplets in electrospinning of nanofibers. Journal of Applied Physics 90 (9), 4836–4846. Zeng, J., Xu, X., Chen, X., Liang, Q., Bian, X., Yang, L., Jing, X., 2003. Biodegradable electrospun fibers for drug delivery. Journal of Controlled Release 92 (3), 227– 231. Zong, X.H., Kim, K., Fang, D.F., Ran, S.F., Hsiao, B.S., Chu, B., 2002. Structure and process relationship of electrospun bioabsorbable nanofiber membranes. Polymer 43 (16), 4403–4412.