Sage Journals: Discover world-class research

Abstract

Pullulan nanofibers with long-term stabilization were fabricated by using glutaraldehyde as the cross-linking agent and sulfuric acid as the catalyzer via in situ cross-linking electrospinning. The structural, chemical, and thermal analyses have been conducted by field emission scanning electron microscopy, Fourier transform infrared spectroscopy, X-ray photoelectron spectroscopy, and differential scanning calorimetry. Cross-linked pullulan nanofibers were found thermally more stable. More importantly, the water absorption and water solubility of the cross-linked pullulan nanofiber membranes were evaluated. The results showed that the water absorption rate increased significantly (from 327% to 651%) and the weight loss rate decreased (from 19.0% to 4.1%) with the increase of glutaraldehyde content. Overall, this in situ cross-linking electrospinning technique provides a method to cross-link pullulan during electrospinning. The cross-linked pullulan nanofibers could be employed as promising materials for a variety of potential applications in adsorption and separation fields.

Keywords

Electrospinning pullulan nanofibers in situ cross-linking separation

Introduction

Biopolymers have been emerging as a new class of materials with promising applications as scaffolds for tissue engineering, implants, wound dressings, and delivery vehicles for drugs and cell therapies (Dang and Leong, 2006). Pullulan is a linear polysaccharide extracellularly produced by strains of fungus Aureobasidium pullulans and consists of α-(1→6)-linked maltotriose units, which is α-(1→4)-linked three glucose units (Leathers, 2003). The regular alternation of (1→4) and (1→6) bonds results in two distinctive properties of structural flexibility and enhanced solubility (Leathers, 2003). Pullulan is regarded as the one of the biodegradable and renewable materials obtained from the nature (Enomoto-Rogers et al., 2015). Due to its excellent properties such as non-toxic, edible, and film-forming abilities, pullulan is widely used in food or biomedical field (Karim et al., 2009).

As an increasingly popular nanofabrication technique, electrospinning has emerged as a versatile and effective method for manufacturing nanofibers with small and uniform diameters, large specific surface area, and interconnected porous structure (Kuo et al., 2014; Lin et al., 2016; Xie et al., 2015). Benefitting from the fascinating features, electrospinning pullulan nanofibers provide excellent prospects for food packaging, biomedical, adsorption, and separation applications due to the bio-functionality and bio-compatibility of the pullulan (Bonino et al., 2011; Lee et al., 2011). However, pullulan is water-soluble which will damage its structure and reduce its mechanical strength, limiting its practical applications as structural materials (Kishan et al., 2015a). Therefore, cross-linking of pullulan is necessary to improve the degradation resistance and mechanical properties for tissue engineering, adsorption, and separation applications.

The most commonly implemented technique for cross-linking natural polymers is post-cross-linking (Taepaiboon et al., 2007; Tang et al., 2010). The post-cross-linking methods include physical cross-linking such as UV irradiation (Bottoms et al., 1966; Fujimori, 2010), plasma treatment (Colombo et al., 2015; Yao et al., 2010), and dehydrothermal treatment (Weadock et al., 1995), and chemical cross-linking by some cross-linking agent (Zhang et al., 2006). The chemical methods take place in aqueous phases, which usually damage severely the fibrous morphology of electrospun scaffolds made of water-soluble natural polymers (Ko et al., 2010; Ratanavaraporn et al., 2010). Furthermore, the physical cross-linking approach may cause uneven cross-linking of interior and exterior of electrospun nanofibers since cross-linkers may not permeate throughout the fibers uniformly (Lin and Tsai, 2013). Therefore, it is desirable to eliminate post-electrospinning treatment and try to cross-link nanofibers in a simple one-step process instead of the previously reported two-step method (Tang et al., 2010). In situ cross-linking is a one-step production method which can lower manufacturing costs and production times. A great deal of studies on in situ cross-linking natural polymers have been reported in the past (Kishan et al., 2015b; Koski et al., 2004; Lin and Tsai, 2013; Stone et al., 2013; Tang et al., 2010), but there have been no researches about in situ cross-linking pullulan nanofibers.

In the present study, pullulan nanofibers with long-term stabilization were fabricated by using glutaraldehyde (GA) as the cross-linking agent and sulfuric acid (H₂SO₄) as the catalyzer via in situ cross-linking electrospinning. The effects of GA content on the morphology, chemical structure, thermal property, water absorption, and water solubility of in situ cross-linking pullulan nanofibers were investigated and presented. More significantly, considering the special adsorption properties and electrochemical properties of pullulan, the cross-linked pullulan nanofibers could be used as promising materials for a variety of potential applications in adsorption and separation field.

Experimental

Materials

Pullulan (Mw = 200,000) was obtained from Tianjin Peiyang Biotrans Biotech Co., Ltd, China. GA (50% aq, analytical grade) and concentrated H₂SO₄ (98% aq) were received from Tianjin Kermel Co., Ltd, China. All of the materials were used as received without further purification.

Fabrication of pullulan nanofibers

Pullulan solution at a concentration of 22 wt.% was prepared by dissolving pullulan powder in deionized water with an appropriate amount of concentrated H₂SO₄ (concentration of 1.0 wt.%), stirring at room temperature until they were homogeneous. The pullulan solution and GA were combined in appropriate proportions at room temperature and stirred for 10 min. The adding volume of GA was selected to be 20, 60, 120, 200, 300, and 400 µL and the corresponding concentrations were 0.12, 0.37, 0.74, 1.24, 1.85, and 2.47 wt.%, respectively. The polymer solution was in a horizontally placed graduated pipette with a metal needle and then attached to a micro-syringe pump, as shown in Figure 1. The electrospinning process was carried out at applied voltage of 25 kV, tip to collector distance of 15 cm and spinning rate of 0.4 mL h⁻¹, respectively. As a control, we also prepared uncross-linked pullulan nanofiber membranes by the same spinning process parameters. The electrospinning temperature was 25±5°C and the humidity was 30±10%. The nanofiber membranes were collected on the surface of a grounded aluminum foil.

Figure 1.

Schematic diagram of electrospinning setup.

Characterization

The viscosity of the solution was measured by digital viscometer (NDJ-8S, Shanghai Fangrui Instrument Co., Ltd, China). The morphology of the nanofibers was observed by field emission scanning electron microscopy (FE-SEM) (S-4800, Hitachi Ltd, Japan). The structural information of nanofiber was characterized by X-ray photoelectron spectroscopy (XPS) (ThermoFisher K-alpha, England) and Fourier transform infrared spectroscopy (FT-IR) (TENSOR37, BRUKER, Germany). Thermal properties of nanofibers were examined with differential scanning calorimetry (DSC) (200 F3, NETZSCH, Germany).

To examine the water absorption and water solubility of nanofibers, small sections of dried electrospun nanofiber membranes were weighed (W _O ) and then immersed in deionized water for 24 h. Surface water on membranes was absorbed by a filter paper. Wet membranes were weighed (W _W ), freeze-dried and then weighed again (W _D ). The percentages of water absorption and weight loss of membranes were calculated using the following equations (Lin and Tsai, 2013)

Water absorption rate (%) = (\frac{W_{W} - W_{D}}{W_{D}}) \times 100

(1)

Weight loss rate (%) = (\frac{W_{O} - W_{D}}{W_{O}}) \times 100

(2)

Results and discussion

Effect of GA content on the morphology of nanofibers

Figure 2 shows the FE-SEM images of in situ cross-linking pullulan nanofiber membranes with different contents of GA placed in a room with 50% humidity for 6 hours. When the content of GA was 20 µL (Figure 2(a)), the nanofibers showed large area of adhesion which was caused by the moisture absorption of nanofibers in moist air, and the nanofiber membrane could dissolve in water easily due to the low degree of cross-linking. Further increasing the content of GA (Figure 2(b) to (e)), the adhesion morphologies disappeared and the nanofibers exhibited uniform morphology. And more important, the stability of fibers in the water gradually increased. When the GA content was 400 µL (Figure 2(f)), the spinnability of the solution deteriorated and the nanofibers showed flattened morphology. Thus, in situ cross-linking caused significant changes in fiber morphology. The morphology of nanofibers transformed from regular round structure to flat ribbon structure which was caused by the increased effective molecular entanglement and molecular weight due to the presence of GA. According to previous studies, when the molecular weight was high, relatively wet fibers were flattened on impact when deposited due to the reduction of solvent evaporation and the increase of viscosity (Koski et al., 2004). To examine this issue, time-dependent rheological behavior of pullulan/GA (200 µL) solution was recorded and the results are shown in Figure 3. The increased viscosity and increased effective molecular weights with time indicated that the intermolecular cross-linking occurred between pullulan and GA prior to electrospinning. Therefore, according to the FE-SEM images and the spinning process, the better addition range of GA was 60–300 µL.

Figure 2.

FE-SEM images of pullulan in situ cross-linking nanofiber membranes with different volumes of GA: (a) 20 µL, (b) 60 µL, (c) 120 µL, (d) 200 µL, (e) 300 µL, and (f) 400 µL. Spinning parameter: applied voltage of 25 kV, tip to collector distance of 15 cm, extrusion rate of 0.4 mL h⁻¹.

Figure 3.

Time-dependent rheological behavior of pullulan/GA (200 µL) solution.

Effect of GA content on the water absorption and water solubility of nanofibers

The optical images and FE-SEM images of the pure pullulan nanofiber membrane (PUL-NM) and cross-linked pullulan/GA nanofiber membrane (PUL/GA-NM) (200 µL) are shown in Figure 4. The pure PUL-NM was readily dissolved in deionized water (Figure 4(a) to (c)), whereas the cross-linked PUL/GA-NM was not dissolved in deionized water (Figure 4(d)). Although swelling occurred for the cross-linked PUL/GA-NM after immersion in deionized water for 24 hours, the morphology of nanofiber remained intact (Figure 4(e)).

Figure 4.

Optical images of PUL/GA-NMs (200 µL) (right) compared to PUL-NMs (left) immersed in deionized water for: (a) 0 second, (b) 1 second, (c) 5 second, (d) 24 hours; and (e): FE-SEM images of PUL/GA-NMs (200 µL) immersed in deionized water for 24 hours.

We know that the swelling behavior is affected by its degree of cross-linking. To study the effect of degree of cross-linking on swelling behaviors, cross-linked PUL-NMs with different GA contents were prepared to examine their water absorption. Considering the water solubility and morphology of pullulan nanofibers, only four GA contents (60, 120, 200, and 300 µL) were studied. Figure 5(a) shows the water absorption rate of pullulan/GA cross-linking nanofiber membranes with different contents of GA. It was clear that with the increase of the GA content, the water absorption rate of PUL-NM increased firstly, and then decreased. Adding 200 µL of GA, the membranes exhibited the largest water absorption rate of 650%. When the content of GA was less than 120 µL, the cross-linking degree was low and a lot of unreacted hydroxyl groups existed, causing dissolving immediately after immersing in deionized water and the water absorption rate was small. Further increasing the content of GA to 300 µL, pullulan nanofibers formed excessive cross-linking and the free volume in the membrane reduced, the containable water also reduced, and at the same time, the intermolecular forces increased, causing the decrease of water absorption rate. Figure 5(b) shows the weight loss rate of cross-linked PUL/GA-NMs. When the content of GA was less, the cross-linking degree was low and a part of unreacted hydroxyl groups existed, causing the weight loss of the membranes. With the increasing of GA content, the weight loss rate of membranes decreased obviously due to the increased degree of cross-linking. When the content of cross-linking was 200 µL, the nanofiber membranes exhibited the minimum weight loss rate of 4.1%. Based on the obtained results, the PUL/GA solution (200 µL) exhibited better spinnability and the PUL/GA-NMs (200 µL) showed the optimal morphology and the lowest water solubility.

Figure 5.

Water absorption rate (a) and weight loss rate and (b) of PUL/GA-NMs with different contents of GA.

FT-IR analysis

The reaction mechanism of pullulan cross-linked by GA is clearly shown in Figure 6. It indicated that the carbonyl group of GA could react with the hydroxyl of pullulan under acid condition. Figure 7 shows the FT-IR spectroscopy of PUL-NMs and PUL/GA-NMs. Strong absorption in 850 cm⁻¹ is characteristic of the α-glucopiranosid units. Absorption in 755 cm⁻¹ indicates the presence of α-(1,4) glucosidic bonds, and spectra in 932 cm⁻¹ proves the presence of α-(1,6) glucosidic bonds. Bands at 2850–3000 cm⁻¹ are due to stretching vibrations of CH and CH₂ groups and bands attribute to CH/CH₂ deformation vibrations are present at 1300–1500 cm⁻¹ range. For PUL/GA-NM, new bands at 1729 cm⁻¹ appeared which was assigned to the carbonyl group arising from the unreacted GA. In addition, the band at about 1207 cm⁻¹ enhanced after cross-linking which was due to the reaction between the carbonyl group of GA and hydroxyl of pullulan. Also very intensive, broad hydroxyl band occurs at 3000–3600 cm⁻¹ (Karim et al., 2009). It was clear that with the addition of GA, some absorption peaks of pullulan became lower in intensity which was because that the hydroxyl groups of pullulan reacted with GA.

Figure 6.

Reaction mechanism of pullulan cross-linked by GA.

Figure 7.

FT-IR spectroscopy for: (a) PUL-NM and (b) PUL/GA-NM.

XPS analysis

The XPS wide scan spectra of PUL-NM and PUL/GA-NM (200 µL) are shown in Figure 8. Compared with PUL-NM, the content of carbon increased from 53.88% to 58.02%, while oxygen content deceased from 46.12% to 41.98%. The reason was attributed to the in situ cross-linking reaction between the aldehyde groups of GA and the hydroxyl groups of pullulan. In order to further illustrate the cross-linking reaction, XPS-peak-differentiation-imitating analysis was applied to study the variation of the chemical shifts of C1s. Figure 8(b) and (c) presents the fitted XPS spectra of the C1s region. Four components at 286.25, 287.07, 285.06, and 286.89 eV in the C1s region were observed for PUL-NM and PUL/GA-NM, which was corresponded to C–C, C–OH, C–H and C–O, respectively (Liu et al., 2014). Their corresponding proportions are listed in Table 1. After cross-linking using GA, the most obvious changes were that the C–OH content decreased while the C–O content increased, indicating part of C–OH transformed into C–O.

Figure 8.

XPS spectra (a) and C1s spectra of PUL-NM (b) and PUL/GA-NM (c).

Table 1.

Relative content of carbon element at different chemical states.

	C–C (%)	C–OH (%)	C–H (%)	C–O (%)
PUL-NM	23.5	26.2	20.6	29.7
PUL/GA-NM	27.1	16.6	20.9	35.4

PUL-NM: pullulan nanofiber membrane; PUL/GA-NM: pullulan/GA nanofiber membrane.

DSC analysis

he DSC analysis curves of PUL-NM and PUL/GA-NM are shown in Figure 9. A clear glass transition step appeared at 95.8°C for PUL-NM (Figure 9(a)). The enthalpy value of PUL/GA-NM decreased obviously, which was caused by the decrease of active ability of cross-linking pullulan molecules. At the same time, a new melting peak appeared at 170°C for PUL/GA-NM, this was because the pullulan macromolecular reacted with the GA to form a stable cross-linking structure and then formed a relatively complete crystal structure after electrospinning. Therefore, the PUL/GA-NMs exhibited more stable thermal properties.

Figure 9.

DSC spectroscopy for: (a) PUL-NM and (b) PUL/GA-NMs (200 µL).

Conclusion

We have incorporated a chemical cross-linker and catalyst to pullulan solutions to perform the in situ cross-linking of nanofibers via one-step electrospinning. The cross-linking reaction causes significant changes in solution rheology and fiber morphology. As the GA content increases, molecular entanglement increases as fibers can transfer from bonded fibers to uniform fibers to flat fibers. In situ cross-linking yields insoluble nanofibers whose weight loss rate can be decreased with the increase of GA content. Based on the obtained results, the PUL/GA-NMs (200 µL) showed the optimal morphology and the lowest water solubility. Therefore, the cross-linked pullulan nanofibers showed good application future in adsorption, separation, food packaging, biomedical and tissue engineering, etc.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to thank China Postdoctoral Science Foundation Grant (2018M630276), the National Natural Science Foundation of China (51673148 and 51573133) and the Science and Technology Plans of Tianjin (No.17PTSYJC00040 and 18PTSYJC00180).

References

Bonino

et al . (2011) Electrospinning alginate-based nanofibers: From blends to crosslinked low molecular weight alginate-only systems. Carbohydrate Polymers 85(1): 111–119.

Bottoms

Cater

Shuster

(1966) Effect of ultraviolet irradiation on skin collagen. Nature 211(5044): 97.

Colombo

et al . (2015) Crosslinking of water-soluble pullulan nanofibrous mats through atmospheric plasma treatment. In: IEEE international conference on plasma sciences May 24-28.

Dang

Leong

(2006) Natural polymers for gene delivery and tissue engineering. Advanced Drug Delivery Reviews 58(4): 487–499.

Enomoto-Rogers

et al . (2015) Synthesis and characterization of pullulan alkyl esters. European Polymer Journal 66: 470–477.

Fujimori

(2010) Ultraviolet light‐induced change in collagen macromolecules. Biopolymers 3(2): 115–119.

Karim

et al . (2009) Preparation and characterization of electrospun pullulan/montmorillonite nanofiber mats in aqueous solution. Carbohydrate Polymers 78(2): 336–342.

Kishan

et al . (2015a) In situ crosslinking of electrospun gelatin for improved fiber morphology retention and tunable degradation. Journal of Materials Chemistry B 3(40): 7930–7938.

Kishan

et al . (2015b) In situ crosslinking of electrospun gelatin for improved fiber morphology retention and tunable degradation. Journal of Materials Chemistry B 3(41): 8212–8212.

10.

et al . (2010) Characterization of cross-linked gelatin nanofibers through electrospinning. Macromolecular Research 18(2): 137–143.

11.

Koski

Yim

Shivkumar

(2004) Effect of molecular weight on fibrous PVA produced by electrospinning. Materials Letters 58(3–4): 493–497.

12.

Kuo

Hung

Hsu

(2014) The effect of elastic biodegradable polyurethane electrospun nanofibers on the differentiation of mesenchymal stem cells. Colloids and Surfaces B: Biointerfaces 122: 414–422.

13.

Leathers

(2003) Biotechnological production and applications of pullulan. Applied Microbiology and Biotechnology 62(5–6): 468–473.

14.

Lee

et al . (2011) Designed hybrid scaffolds consisting of polycaprolactone microstrands and electrospun collagen-nanofibers for bone tissue regeneration. Journal of Biomedical Materials Research Part B: Applied Biomaterials 97B(2): 263–270.

15.

Lin

et al . (2016) Fabrication of multi-biofunctional gelatin-based electrospun fibrous scaffolds for enhancement of osteogenesis of mesenchymal stem cells. Colloids and Surfaces B: Biointerfaces 138: 26–31.

16.

Lin

Tsai

(2013) In situ UV-crosslinking gelatin electrospun fibers for tissue engineering applications. Biofabrication 5(3).

17.

Liu

et al . (2014) Solution blowing of chitosan/PVA hydrogel nanofiber mats. Carbohydrate Polymers 101: 1116–1121.

18.

Ratanavaraporn

et al . (2010) Influences of physical and chemical crosslinking techniques on electrospun type A and B gelatin fiber mats. International Journal of Biological Macromolecules 47(4): 431–438.

19.

Schiffman

Schauer

(2007) One-step electrospinning of cross-linked chitosan fibers. Biomacromolecules 8(9): 2665–2667.

20.

Stone

et al . (2013) In situ citric acid crosslinking of alginate/polyvinyl alcohol electrospun nanofibers. Materials Letters 112: 32–35.

21.

Taepaiboon

Rungsardthong

Supaphol

(2007) Effect of cross-linking on properties and release characteristics of sodium salicylate-loaded electrospun poly(vinyl alcohol) fibre mats. Nanotechnology 18(17): 175102–175112.

22.

Tang

et al . (2010) In situ cross-linking of electrospun poly(vinyl alcohol) nanofibers. Macromolecules 43(2): 630–637.

23.

Weadock

et al . (1995) Physical crosslinking of collagen fibers: Comparison of ultraviolet irradiation and dehydrothermal treatment. Journal of Biomedical Materials Research 29(11): 1373.

24.

Xie

et al . (2015) Electrospun nanofibrous adsorbents for uranium extraction from seawater. Journal of Materials Chemistry A 3(6): 2552–2558.

25.

Yao

Liu

Zhu

(2010) The influence of the surface cross-linking of the high-density polyethylene induced by plasma treatment on the durability of treatment efficiency. Journal of Applied Polymer Science 48(1): 57–65.

26.

Zhang

et al . (2006) Crosslinking of the electrospun gelatin nanofibers. Polymer 47(8): 2911–2917.