Seediscussions,stats,andauthorprofilesforthispublicationat:http://www.researchgate.net/publication/259298514

Fabricationofnanofibrousmacroporousscaffoldsofpoly(lacticacid)incorporatingbioactiveglassnanoparticlesbycamphene-assistedphaseseparation

ARTICLEinMATERIALSCHEMISTRYANDPHYSICS·NOVEMBER2013

ImpactFactor:2.13·DOI:10.1016/j.matchemphys.2013.11.009

CITATION

1

DOWNLOADS

10

VIEWS

33

4AUTHORS,INCLUDING:

JungjuKim

DankookUniversity

16PUBLICATIONS64CITATIONS

SEEPROFILE

AhmedEl-Fiqi

DankookUniversity

20PUBLICATIONS82CITATIONS

SEEPROFILE

Hae-WonKim

DankookUniversity

283PUBLICATIONS5,171CITATIONS

SEEPROFILE

Availablefrom:AhmedEl-Fiqi

Retrievedon:24June2015

lable at ScienceDirect

Materials Chemistry and Physics 143 (2014) 1092e1101

Contents lists avai

Materials Chemistry and Physics

journal homepage: www.elsevier .com/locate/matchemphys

Fabrication of nanofibrous macroporous scaffolds of poly(lactic acid)incorporating bioactive glass nanoparticles by camphene-assistedphase separation

Jung-Ju Kim a,b,1, So-Hee Bang a,b,1, Ahmed El-Fiqi a,b, Hae-Won Kim a,b,c,*

a Institute of Tissue Regeneration Engineering (ITREN), Dankook University, South KoreabDepartment of Nanobiomedical Science & BK21 PLUS NBM Global Research Center for Regenerative Medicine, Dankook University, South KoreacDepartment of Biomaterials Science, College of Dentistry, Dankook University, South Korea

h i g h l i g h t s

� Exploring macroporous scaffolds with nanofibrous structuring.� Incorporating bioactive glass nanoparticles to produce nanocomposite scaffolds.� Showing excellent in vitro bioactivity useful for bone tissue engineering.

a r t i c l e i n f o

Article history:Received 4 May 2013Received in revised form30 August 2013Accepted 3 November 2013

Keywords:Composite materialsBiomaterialsSolidificationGlassesMicroporous materials

* Corresponding author. Institute of Tissue RegenDankook University, Shinbu, Republic of Korea. Tel.: þfax: þ82 41 550 3085.

E-mail address: kimhw@dku.edu (H.-W. Kim).1 Both authors contributed equally to this work.

0254-0584/$ e see front matter � 2013 Elsevier B.V.http://dx.doi.org/10.1016/j.matchemphys.2013.11.009

a b s t r a c t

Here we produced macroporous and nanofibrous scaffolds with bioactive nanocomposite composition,poly(lactic acid) (PLA) incorporating bioactive glass nanoparticles (BGnp) up to 30 wt%, targeting boneregeneration. In particular, the nanofibrous structure in the scaffolds was generated by using a bicyclicmonoterpene, camphene (C10H16), through a phase-separation process with PLA-BGnp phase in chlo-roform/1,4-dioxane co-solvent. Furthermore, macropores were produced by the impregnation of saltparticles and their subsequent leaching out, followed by freezing and lyophilization processes. Theproduced PLA-BGnp scaffolds presented highly porous and nanofibrous structure with porosities of 90e95% and pore sizes of over hundreds of micrometers. BGnp with sizes of w90 nm were also evenlyimpregnated within the PLA matrix, featuring a nanocomposite structure. The nanofibrous scaffoldsexhibited enhanced hydrophilicity and more rapid hydrolytic degradation as the incorporated BGnpcontent increased. The bone-bioactivity of the scaffolds was substantially improved with the incorpo-ration of BGnp, exhibiting rapid formation of apatite throughout the scaffolds in a simulated body fluid.The developed macroporous and nanofibrous scaffolds with PLA-BGnp bioactive composition areconsidered as a novel 3D matrix potentially useful for bone tissue engineering.

� 2013 Elsevier B.V. All rights reserved.

1. Introduction

Nanocomposite three-dimensional (3D) scaffolds made ofdegradable polymers with the incorporation of bioactive inorganicmaterials have shown great potential for application to the field ofhard tissue regeneration [1e3]. Scaffolds have facilitated functionalrecovery of damaged and diseased tissues by recruiting a popula-tion of surrounding stem or progenitor cells and guiding their

eration Engineering (ITREN),82 41 550 3081;

All rights reserved.

regenerative processes [3e6]. Furthermore, ex vivo culture of suchcandidate cells to engineer tissue-mimicking structures also re-quires the use of scaffolds with appropriate mechanical andphysico-chemical properties [7].

The native bone extracellular matrix (ECM) is a sort ofnanocomposite comprising of a collagen fibrous network withembedded hydroxyapatite inorganic nanocrystals. This nano-composite structure of ECM provides proper mechanical func-tionality to bone, as well as high strength and toughness, andconsequent resistance to catastrophic failure. From the viewpointof biomaterials, the use of bioactive inorganics significantly im-proves cell responses required for bone regeneration, such asosteoblast maturation and bone ECM syntheses [8e11]. The useof biopolymer is also greatly helpful for the processing of

J.-J. Kim et al. / Materials Chemistry and Physics 143 (2014) 1092e1101 1093

inorganic materials, making it possible to apply polymer tech-nology involving low-temperatures and molding and shaping,which is a particular merit of the porous scaffold processing[12,13].

Along with the compositional aspect, the surface physicalproperties also dictate the biological reactions, affecting proteinadsorption, adhesion and growth of cells, and their tissue-specificdifferentiation. As the surface is the first point of contact for cells,the surface micro/nanostructure is critical for regulating cellularprocesses and subsequent tissue formation [14e17]. Among thephysical parameters, morphological traits, mainly those relating tonanoscale morphology, have recently been demonstrated todominate many cellular responses. One of the most attractivemorphological traits is ‘nanofibrous’, and the nanofibrous struc-ture has mainly been implicated in native tissue ECM proteinssuch as collagen and elastin, and thus, has been considered as anECM-mimicking nanostructure [17e19]. This is why cells recognizewell the nanofibrous structured surface, even better than the flatsurface, and behave more profoundly and specifically to secretECM molecules [16e19]. In general, the nanofibrous-tailoredsubstrates possess large surface areas that allow the adsorptionof biological proteins, such as cell adhesive proteins, improving theinitial cellular recognition of the surface. Moreover, the nano-fibrous surfaces have demonstrated better cell proliferation andosteogenic differentiation with respect to flat substrates [16,17].Therefore, the nanofibrous materials are considered as a promisingplatform for the repair and regenerative matrices of tissues,including bone.

One interesting and potential area of utilizing this nanofibrousstructured platform is in electrospinning. In other words, throughelectrospinning processes, nanofibrous matrices can easily beimplemented [18,19]. A range of biomaterials, including degradablepolymers, have been electrospun into a nanofibrous form, findingextensive utility for the repair of damaged tissues, such as skin,blood vessel, cartilage, nerve and bone [18e20]. While the elec-trospunmaterials provide excellent culture substrate conditions formany different sorts of cells, one major drawback is the difficulty inproviding 3D macropores within the nanofibrous structure, andshaping it into complex forms [19]. Although some efforts havebeen exerted by modifying the collecting part of the electro-spinning apparatus, or incorporating porogen during electro-spinning process, these have limitations in terms of processibilityand tunability over shape and porosity [18e20]. This issue con-fronted in electrospinning is particularly in great need when wecompare the process with other 3D scaffolding techniques, such asgas foaming, template replication, and rapid prototyping. There-fore, it is highly recommended to develop a methodology to createnanofibrous and macroporous scaffolds by utilizing possible alter-native methods.

Here we focus on the phase-separation method, wherein anappropriate use of solvents and freezing conditions allows for thegeneration of a nanofibrous network of biopolymers with a porousstructure. Previously, Ma et al. has been working on methodologyto produce nanofibrous polymer scaffolds by means of a phase-separation method [17,21,22]. The method is a solvent-basedprocess requiring a freezing step to separate the two phases ofmaterial and solvent, which is ascribed to the immiscibility be-tween the phases, and a subsequent removal-off the crystallizedsolvent phase. The resultant material part exerts various nano/microstructures, like nanofibrous morphology. The nanofibrousscaffolds of poly(lactic acid) (PLA) or PLA-gelatin using organicsolvents have been successfully produced, and have shown torecruit several cell types and to stimulate differentiation intoosteoblastic lineage [17,21,22], suggesting that the nanofibroussurface is particularly interesting as a scaffold for bone

regeneration. Recently, we have also investigated the generation ofa nanofibrous biopolymer, poly(hydroxyl butyrate-co-valerate)(PHBV), by the phase-separation method [16]. In particular, a novelsolvent, camphene, was used to create nanofibrous structuredPHBV. Because of the high freezing-point and ease-of-sublimation,the camphene phase interconnecting the material part could beliberated easily under relatively mild conditions, suggesting it tobe a promising candidate source for creating nanofibrous structureof biopolymers [16].

We aimed to utilize the camphene-assisted method in the cre-ation of novel nanofibrous 3D scaffolds made of PLA and bioactiveglass nanoparticles (BGnp) for bone regeneration. The BGnp wereprepared by a solegel method with sizes of less than a hundrednanometers. Thus, the prepared BGnp are considered to provideexcellent bioactivity for bone cell functions and bone formation,compared to the hydrophobic and less bioactive PLA polymericscaffold. Macropore generation was exploited by inducing freezingtemperatures as high as possible with proper choice of solvents andcamphene, and additionally through salt impregnation e leachingout method. Here we report first on the preparation of the mac-roporous nanofibrous PLA scaffolds with the incorporation ofdifferent concentrations of BGnp, and thenwe briefly examined theproperties, particularly those of in vitro bone-bioactivity (apatiteforming ability) in a simulated body fluid.

2. Experimental procedures

2.1. Bioactive glass nanoparticles (BGnp)

The BGnp were prepared by a solegel method, as describedelsewhere [23]. In brief, a binary composition 85SiO2/15CaO (mol%) of BGnp was synthesized by a ultra-sound assisted base-catalyzed solegel method using poly(ethylene glycol), (PEG;(C2H4)nH2O, Mn: 10,000) as a template. PEG of 5 g was completelydissolved in methanol at pH adjusted to 12.5 by adding NH4OH.0.179 g of Ca (NO3)2$4H2O of was then dissolved in the solution,and 0.895 g tetraethyl orthosilicate (TEOS) was then diluted withmethanol and added to the solution, with the application of ahigh-power ultra-sound using a Sonoreactor, LH700S ultra-sonicgenerator (Ulsso Hitech) operating at 20 kHz and 700 W. Theoutput power was 220 W in a 10-s on/10-s off cycle for 20 min.After 24 h, the white precipitate was separated by centrifugationat 5000 rpm using a Mega17 R centrifuge (Hanil Science), and wasthen washed and re-dispersed with water and ethanol. The pre-cipitate was dried at 70 �C and was then calcined at 700 �C underair for 5 h.

2.2. Preparation of nanofibrous PLA-BGnp macroporous scaffolds

For the preparation of the nanofibrous structured scaffolds,the camphene-assisted phase-separation method was intro-duced. First, pure PLA composition was used to prepare a nano-fibrous structure by varying the concentration of camphene(C10H16, SigmaeAldrich). For this, PLA was dissolved at 3 wt% in aco-solvent of chloroform/1,4-dioxane of 1/4, within whichdifferent concentrations of camphene (camphene/PLA ratio of 0,2, and 4 by weight) were dissolved. The mixture solution waspoured into a plastic mould, and was then allowed to freezeat �20 �C overnight, during which the camphene-PLA solutionwas frozen solid. The frozen samples were freeze-dried for 3 daysto sublimate the solution and camphene, and to create theporous-structured scaffold. In order to provide macroporeswithin the scaffold, salt-particles were added to the PLA-camphene solution. NaCl particles of 200e500 mm diameterwere first added to a plastic mould, followed by the addition of

Table 1Summary of relative amounts of the materials (PLA and BGnp), camphene, andsolvents (chloroform and 1,4-dioxane) used in this study.

Materialscomposition(BGnp w.r.t PLA)

Camphene w.r.tmaterials (camphene/material ratio)

Materialsw.r.tsolvent

Solvent composition(Chloroform w.r.t. 1,4-dioxane)

10, 20, 30% 0, 200, 400%(0,2,4)

3% 25%

J.-J. Kim et al. / Materials Chemistry and Physics 143 (2014) 1092e11011094

PLA-camphene solution, which was then frozen at �20 �C over-night. After freeze-drying, the samples were thoroughly washedwith DW to leach out the embedded NaCl particles, and werethen dried to prepare macroporous scaffolds.

Based on experiments of pure PLA scaffolds, BGnp added PLAnanocomposite scaffolds were prepared. The BGnp were dispersedin a co-solvent of chloroform/1,4-dioxane (1/4 by wt via ultra-sonicagitation. After confirming a homogeneous and stable emulsionstatus, the BGnp solution was mixed with the PLA-camphene so-lution. The camphene-to-materials ratio was fixed at 4, and theBGnp amount was varied at 10, 20 and 30% by wt relative to PLA.The BGnp-PLA solutions were also processed into macroporousscaffolds using NaCl particles in the same manner as PLA macro-porous scaffolds. Table 1 summarizes the amount of PLA, BGnp,camphene, and co-solvent used in this study.

A schematic illustration of the preparation of PLA or PLA-BGnp nanocomposite scaffolds with the addition of camphenein order to get nanofibrous structured 3D scaffolds is shown inFig. 1. The use of a co-solvent aids freezing at around �20 �Cwhile preserving sufficient solubility of the PLA (a). The mixturewas shaped in a mold which contained NaCl particles withselected sizes of 200e500 mm and was then compacted. Freezingat �20 �C allows for the solidification of the solvents andcamphene, during which time phase separation occurred amongPLA, camphene and solvent (b). Subsequent freeze-drying allowsfor the sublimation of solvents and camphene which create alevel of porosity (c). In particular, the structure of the scaffoldbecame nanofibrous with the assistance of camphene sublima-tion. Salt-leaching in distilled water generated macropores in thescaffolds (d). Finally, the PLA or PLA-BGnp nanocomposite scaf-folds with macroporous morphology and nanofibrous stemstructure were produced.

Fig. 1. Schematic illustration of the preparation of PLA-BGnp nanocomposite nanofibrouschloroform/1,4-dioxane. (b) Mixture is shaped in a mold which contained NaCl particles of 2below �20 �C allows for the solidification of the solvents and camphene, during which timdrying enables the sublimation of solvents and camphene, creating a level of porosity. Incamphene sublimation. Salt-leaching in distilled water generates macropores in the scaffolstructure are produced.

2.3. Characterizations

The morphology of the scaffolds was observed by field-emission scanning electron microscopy (SEM; TESCAN, MIRA IILMH). BGnp were examined by high-resolution transmissionelectron microscopy (TEM; JEM-3010, JEOL). The phase of thesamples was analyzed with X-ray diffraction (XRD; Ultima IV,Rigaku) using CuKa radiation (l ¼ 1.5418 �A). Infra-red spectra ofthe samples were obtained with a resolution of 4 cm�1 on a Varian640-IR Fourier transform infra-red (FT-IR) spectrometer (Varian,Australia) in the range of 2000e400 cm�1 [24]. Porosity of thescaffolds was measured by the mercury intrusion method(Micromeritics).

2.4. In vitro degradation and apatite forming ability tests

Hydrolytic degradation of the scaffolds was investigated byimmersion in phosphate buffered saline (PBS). Each scaffoldsample was soaked in PBS at 37 �C for different periods, up to 28days. At each selected time point, samples were gently washedwith DW and ethanol, and were then dried. The weight of eachsample was measured before and after the PBS-immersion test.Degradation of each sample was recorded as follows: Weight loss% ¼ [(Wi � Wp)/Wi] � 100, where Wi ¼ initial weight, andWp ¼ weight after PBS test.

The in vitro hydroxyapatite forming ability of the preparedscaffolds was tested in a simulated body fluid (SBF) [25]. SBF hasan ionic composition and concentration similar to those of hu-man body plasma. Briefly, SBF was prepared by dissolving NaCl,NaHCO3, KCl, K2HPO4$3H2O, MgCl2$6H2O, CaCl2, and Na2SO4 indeionized water and buffering at pH 7.4 with tris(hydrox-ymethyl) aminomethane (HOCH2)CNH2 and HCl. Each scaffoldsample was immersed in SBF and was then placed in an incu-bator at 37 �C for different periods, up to 28 days. At eachselected time point, samples were gently washed with DW andethanol, and were then dried.

Theweight gain of each samplewas recorded as follows:Weightgain % ¼ [(Ws � Wi)/Wi] � 100, where Wi ¼ initial weight, andWs ¼ weight after SBS test. The morphology of samples wasexamined by scanning electron microscopy (SEM) to detect thecrystallite formation. The calcium and phosphate atoms weredetected by energy dispersive spectroscopy (EDS) attached to the

scaffolds: (a) Camphene is mixed with the PLA-BGnp solution in the co-solvent of00e500 mm, to provide macrochannels in the scaffolds after NaCl dissolution. Freezinge phase separation occurs between materials and camphene. (c) Subsequent freeze-particular, the structure of the scaffold becomes nanofibrous with the assistance ofds. (d) Finally, PLA-BGnp nanocomposite scaffolds with macroporous and nanofibrous

J.-J. Kim et al. / Materials Chemistry and Physics 143 (2014) 1092e1101 1095

SEM. The change in chemical structure of the SBF-treated sampleswas examined by FT-IR.

3. Results and discussion

3.1. Camphene effects on nanofibrous structuring of PLA scaffolds

First, we observed that the addition of chloroform up to 25%(relative to 1,4-dioxane) was effective for freezing of the PLA-camphene mixture solutions at �20 �C, which was considered tobe highly useful for applying the process at relatively mild tem-perature conditions. Next, we observed the effects of camphene onthe generation of a nanofibrous structure of PLA via the phase-separation process, using different PLA-camphene solutions in theco-solvent. The ratio of camphene-to-PLA was varied at 0, 2 and 4.The PLA-camphene solutions in the co-solvent were frozenat �20 �C and were then lyophilized.

Fig. 2. SEM morphologies of the PLA scaffolds with varying additions of camphene: the ramagnification. The solvent used to dissolve PLA and camphene was a mixture of 20% chlorothen freeze-dried. Pores (wtens of micrometers in diameter) possibly interconnected wereseparated solvents during freezing and subsequent sublimation during freeze-drying. NoticeaPLA (ratio ¼ 4).

Fig. 2 presents the SEM morphologies of the PLA produced afterthe freeze-drying process, while varying the camphene/PLA ratios.In pure PLA (at a ratio of 0), large interconnected pores (tens ofmicrometers) were generated throughout the scaffolds (Fig. 2(a)),and the framework surface was highly dense and smooth(Fig. 2(b)). The macropores should be formed by the sublimation ofthe solvent iced-crystals during the lyophilization process, whichwere phase-separated from PLA under freezing at �20 �C. Thismacroporous structure was similarly noticed in other ratios(Fig. 2(c,e)). When the camphene concentration was low (at a ratioof 2), a nanofibrous morphology started to be observed (Fig. 2(d)).At a higher camphene concentration (at a ratio of 4), the scaffoldsurface became completely nanofibrous, while maintaining themacroporous structure (Fig. 2(f)). It is, thus, clear that thecamphene addition significantly assisted the generation of ananofibrous structure in the porous PLA scaffolds, and particularly,at a camphene/PLA ratio of 4, completely nanofibrous PLA scaffoldscould be generated.

tio of camphene-to-PLA was 0 (a,b), 2 (c,d), and 4 (e,f) at low (a,c,e) and high (b,d,f)form and 80% 1,4-dioxane. PLA-camphene solutions were frozen at �20 �C, and weregenerated throughout the scaffolds for all compositions, due to the crystals of phase-ble was the highly nanofibrous structure created on the scaffolds of camphene-assisted

J.-J. Kim et al. / Materials Chemistry and Physics 143 (2014) 1092e11011096

Moreover, a macroporous structure could be implemented bythe process. From a mercury intrusion method, we observed thatthe PLA scaffolds had porosities of 87e94%, and the pore sizeswere approximated 20e50 mm. Moreover, the pores were totallyinterconnected as they were formed by the phase-separationof the bi-continuous phases (PLA and camphene) and the sub-sequent complete sublimation and the resultant remaining ofPLA.

When we performed the freezing process at temperaturesbelow �20 �C (e.g., �60 �C in a deep freezer), the macropore sizesbecame significantly decreased (far less than a few tens of mi-crometers). Lowering the freezing-point below the melting tem-perature (which is considered as the ‘degree of freezing’)pronounces nucleation of solvent iced-crystals with smaller buthigher numbered nuclei, which further hampers ice crystalgrowth, and thus, decreases the crystal size [26]. The sublimationof the small-sized iced-crystals, thus, results in small-sized porestructures. Therefore, in order to increase the pore size thefreezing step must be performed at a temperature as high aspossible. Due to the use of 1,4-dioxane solvent, the freezing

Fig. 3. SEM morphologies of the PLA scaffolds obtained by salt impregnation-leaching methdissolved in a co-solvent of chloroform/1,4-dioxane was poured into a mold containing NaClby leaching of NaCl particles to produce macropores. On the surface of the camphene-assis

temperature could be set as high as �20 �C, where we were able toproduce PLA scaffolds with macropores as large as 20e50 mm.Increasing the macropore size is particularly important for theproduction of tissue engineering scaffolds, where macroporesallow for the migration of cells and the ingrowth of blood vesselsand tissues [27]. Although we did not test the biological perfor-mance of the PLA nanofibrous macroporous scaffolds produced inthis study, it is thought that an increase in macropore size over100 mm might be helpful for applying the scaffolds effectively fortissue engineering purposes [28e30].

3.2. Application of salt impregnation-leaching out method

We therefore utilized the salt impregnation-leaching outmethod to increase themacropore size. NaCl particles selectedwithsizes of 200e500 mm were used to mold the scaffold, and afterfreezing and lyophilization, the NaCl particles were leached out indistilled water to produce macropores. Fig. 3 shows the SEM mor-phologies of the PLA scaffolds produced with (a ratio of 4) orwithout camphene, bymeans of the salt impregnation-leaching out

od: the camphene-to-PLA ratio was (a,b) 0 and (cef) 4. PLA or PLA-camphene solutionparticles of 200e500 mm, which was then frozen at �20 �C and freeze-dried, followedted PLA scaffold a highly nanofibrous structure was also created.

Fig. 4. Bioactive glass nanoparticles (BGnp) used for the production of PLA-BGnp nanocomposite: (a) High-resolution SEM and (b) TEM morphology.

J.-J. Kim et al. / Materials Chemistry and Physics 143 (2014) 1092e1101 1097

method. In pure PLA, the scaffold showed many large-sized mac-ropores replicating the NaCl particles (Fig. 3(a)). The scaffoldframework appeared to be somewhat macro-fibrous but the sur-face of the framework was almost dense (Fig. 3(b)). When

Fig. 5. SEM morphologies of the PLA-BGnp nanocomposite scaffolds with the assistance of ca salt impregnation-leaching method. The amount of BGnp added to PLA was varied: (a,d,composite scaffolds were obtained (aec). Closer examination of the surfaces revealed a nanoscaffolds with 10 and 20%BGnp addition, some agglomerated forms of nanoparticles with size30%BGnp addition.

camphene was added (a ratio of 4), the NaCl-leached macroporestructure was also shown to be well preserved (Fig. 3(c)). Moreover,the framework surface became highly porous with well developednanofibrous morphology (Fig. 3(def)).

amphene (camphene-to-material ratio was 4). Macroporous scaffolds were obtained byg) 10%, (b,e,h) 20% and (c,f,i) 30% by weight. For all cases, highly macroporous nano-fibrous structuring (gei). While the nanoparticles were evenly distributed in the case ofs of hundreds of nanometers (arrowed and enlarged in (i)) were observed in the case of

Fig. 6. Characteristics of the BGnp-PLA nanocomposite scaffolds with nanofibrousstructure; (a) FT-IR spectrum of the scaffolds, showing the development of chemicalbands related to each composition. (b) Porosity of the scaffolds, as measured by amercury intrusion method. The nanofibrous scaffolds using camphene showed highporosities of over 91w95%, which is significantly higher than those obtained in theconventional manner, without using camphene (87e89%). The porosity was slightlydecreased with increasing BGnp content. (c) Hydrolytic degradation of the nanofibrousscaffolds in PBS, during immersion at 37 �C for periods of up to 4 weeks. Sample wasimmersed in PBS and the weight loss was measured. Data are mean � std of threedifferent samples. Degradation occurred almost linearly with time, and the rate washigher as the BGnp content increased. After 28 days, the degraded weight was w40e50% of initial.

J.-J. Kim et al. / Materials Chemistry and Physics 143 (2014) 1092e11011098

When compared to the case of PLA scaffolds obtained withoutNaCl particles (as shown in Fig. 2(f)), the scaffolds obtained withNaCl particles showed some disintegration of the nanofibrousstructure. This was due, in part, to the several additional processingstages, involving NaCl leaching out in water and lyophilization, asthe water contact over hours may affect the degradation/disinte-gration of the nanofibrous structure. Moreover, the differentinterfacial properties provided by the NaCl particles should alterthe phase-separation phenomenon of the mixture solution duringthe freezing stage, leading to a different distribution of each phaseat the framework surface (interface between NaCl and solution)and the consequent solvent-sublimed nanostructures. Althoughthe degree of nanofibrous structuring decreased slightly, the use ofNaCl particles was shown to be effective in creating macroporesover hundreds of micrometers, facilitating the usefulness of thescaffolds in tissue engineering purposes. Therefore, we furtherapplied this method to the production of PLA-BGnp nanocompositescaffolds.

3.3. PLA-BGnp nanofibrous macroporous scaffolds

As the bioactive component, we incorporated BGnp within thePLA scaffolds. The BGnps were produced by a solegel method withsizes of less than a hundred of nanometers. PEG was used as thetemplate for nanoparticle formation, and a basic catalysis conditionwas used. Fig. 4 shows the high-resolution SEM (Fig. 4(a)) and TEM(Fig. 4(b)) images of the BGnps used for the production of PLA-BGnpnanocomposite scaffolds. Discrete spherical particles, with sizes ofw90e100 nm, were well developed. The average size of BGnp wasmeasured to be 89 nm (�15 nm), and the BET analysis showed thatthe BGnp were highly mesoporous, with a specific surface area of890m2 g�1, a pore volume of 0.48 cm3 g�1, and an average pore sizeof 3.8 nm [23], which is typical of the silica-based nanoparticles bya surfactant-mediated solegel approach. Here we added the BGnpwithin the PLA/camphene solution at different contents (up to30 wt% with respect to PLA). Macroporous scaffolds of PLA-BGnpwere produced in a similar manner to the pure PLA scaffolds byusing the NaCl impregnation-leaching out method.

The SEM morphologies of the PLA-BGnp nanocomposite scaf-folds (10, 20 and 30%BGnp) with the assistance of camphene(camphene/material ratio was 4) are presented in Fig. 5. For allcompositions, highly macroporous scaffolds were obtained(Fig. 5(aec)). A close examination of the scaffolds revealed a porousand nanofibrous structuring of the framework surface (Fig. 5(dei)).In 10BGnp scaffold, the ultrafine BGnp were evenly distributed inthe nanofibrous network of PLA matrix, well preserving thenanofibrous morphology similar to that of the pure PLA scaffold(Fig. 5(g)). In the 20BGnp scaffold, the nanofibrous morphology wasalso well developed (Fig. 5(h)). However, in the 30BGnp scaffoldsome parts revealed large-sized particulates (larger than the indi-vidual BGnp), possibly due to the agglomeration of BGnp (arrowedand enlarged in Fig. 5(i)). It is thus considered that the addition ofBGnp within PLA up to 20wt% appeared to result in a gooddispersion of nanoparticles within the PLA matrix, which washelpful for preserving the nanofibrous structure of the scaffolds.

This novel way of producing the nanofibrous structure usingcamphene-assisted phase-separation is believed to have significantmerit over conventional methods, such as electrospinning tech-niques. Electrospinning has been regarded as the most powerfulmethod to produce nanofiber of biopolymer materials, includingPLA and its composites [31,32]; however, 3D structuring andproviding large macropores or formulating different shapes hasbeen a significant challenge, which is primarily due to the line-of-sight process character [6]. Compared to this, the camphene-assisted phase separation is solvent-based, and thus, can easily be

Fig. 7. Apatite forming ability of the PLA and PLA-BGnp nanocomposite scaffolds, as compared by the SEM morphological changes of the surface during incubation for 1, 3 and 7days in SBF at 37 �C, taken at different magnification; nanocomposite of (a) 10BGnp, (b) 20BGnp and (c) 30BGnp, versus (d) pure PLA. A rapid induction of apatite nanocrystalliteshappened in the BGnp added nanocomposite scaffolds, and was more pronounced as the BGnp content increased.

J.-J. Kim et al. / Materials Chemistry and Physics 143 (2014) 1092e1101 1099

Fig. 8. Characterization of the scaffolds after the apatite forming ability test in SBF; (a)weight gain of the scaffolds due to apatite formation, as measured up to 4 weeks.While the BGnp-PLA nanocomposite scaffolds showed substantial weight gain as earlyas 3 days, the PLA showed a rapid increase at w28 days. (b) EDS analysis of Ca and Ppeaks (10BGn case). Both Ca and P atomic percentage increased with time.

J.-J. Kim et al. / Materials Chemistry and Physics 143 (2014) 1092e11011100

molded to produce complex-shaped 3D structures with variousdimensions. Moreover, by introducing porogen materials like saltwithin the solution, tunable-sized macropores, which are effec-tively useful for tissue engineering purposes, can be easily imple-mented. Particularly, here we produced macroporous nanofibrousscaffolds with a novel bioactive composition of PLA-BGnp targetingbone tissue. As deduced from the previous reports on BG and itsnanoparticulate form [33e36], the effective physico-chemical andbiological roles of the additives (BGnp) are considered to be huge.Among the possible effects and roles, here we briefly examined theacellular properties of the scaffolds, including hydrolytic degrada-tion and apatite forming ability in vitro.

3.4. Properties of the PLA-BGnp scaffolds

First, the chemical bonds of the PLA-BGnp scaffolds werecharacterized with FT-IR (Fig. 6(a)). Bands related with PLA (1130e1210 cm�1 CeO, 1380e1460 cm�1 CeOeH, 1470 cm�1 CH2, and1760 cm�1 C]O) and BGnp (470 cm�1 SieOeSi, 1100 SieOeSi, and1635 HeOH) were representatively developed in the composites.The porosity of the PLA-BGnp nanofibrous scaffolds was measuredby a mercury intrusion method (Fig. 6(b)). For all the camphene-assisted nanocomposite scaffolds, the porosities were over 90%(ranging 91w95%), and the increase in BGnp addition appeared toslightly decrease the porosity. The conventional scaffolds, pro-duced without the use of camphene (thus not nanofibrous struc-tured), showed porosities slightly lower (87e89%), demonstratingthat the nanofibrous structure affected somewhat the porositylevel. Rather than the porosity, the significant change in surfacetopography, and thus, the increase in surface area (whichproviding important binding sites for biological molecules) shouldbe envisaged. Further study thus remains as to this effect onbiological properties, including protein adsorption and cell adhe-sion to the nanofibrous scaffolds.

We next investigated the degradation behavior of the nano-fibrous scaffolds with different compositions during immersion inPBS for up to 4 weeks (Fig. 6(c)). Scaffolds presented a gradualdecrease in weight with time. As the BGnp content increased, thedegradation rate increased, suggesting the role of BGnp in accel-erating the degradation of the scaffolds. First, it is considered thatthe higher hydrophilicity of the scaffolds with BGnp additionwouldincrease the water diffusion and hydrolytic degradation of the PLApolymer chains. Furthermore, the ionic dissolution of BGnp shouldalso increase the total weight loss of the scaffolds.

We further sought to find the apatite forming ability of thenanocomposite scaffolds in SBF. As the BGnp have been observed todegrade, the ions released from the nanoparticles should affect theprecipitation of calcium phosphate crystals. Fig. 7 presents the SEMmorphologies of the scaffolds after SBF-immersion for differentperiods (1, 3 and 7 days). At day 1, some tiny crystallites appeared toform only in the PLA-30%BGnp scaffold. At day 3, there was sig-nificant induction of nanocrystallites on the surface of scaffolds forall the nanocomposite compositions, which, however, was notreadily recognized in the pure PLA scaffold. At day 7, the nano-crystallites covered the surface of scaffolds almost completely,forming thick crystal layers, and the PLA scaffold had just started toshow crystal formation.

The apatite crystal formation was accompanied by a weightincrease of the scaffolds (Fig. 8(a)). High levels of weight gain werenoticed as early as 3 days for all the nanocomposite scaffolds, whichshowed continuous increases up to 28 days. More weight gain wasrecorded in the scaffolds containing higher BGnp contents. In thepure PLA scaffold, weight gain was pronounced only at 28 days.Therefore, apatite induction was more significant in the PLA-BGnpnanocomposite scaffolds, and the degree was more substantial as

the BGnp content increased. EDS analysis of the apatite crystalformed on the scaffolds (10BGnp used) also showed a continualincrease in Ca and P atomic percentage with SBF-immersion time,due to the time-dependent increase in the apatite formation on thescaffolds. Furthermore, the Ca/P ratio changed greatly during theSBF-immersion (2.22 at day 1, 2.03 at day 3, and 1.66 at day 7), in amanner of matching well with the stoichiometry of HA with im-mersion time, which suggests the development of CaP crystals intoan apatite phase with increasing immersion time, particularly after7 days. The SBF-immersion test clearly demonstrated the effectiveroles of BGnp incorporation within the PLA scaffold in improvingthe apatite forming ability in vitro, and thus, possible bone-bioactivity. Although this is not directly correlated with thecellular mineralization ability of the scaffolds, it allows one tosurmise the potential of the scaffolds in bone repair and regener-ation applications. Further studies remain as to the cell responses tothe developed bioactive nanocomposite scaffolds, including cellproliferation and osteogenic differentiation processes. It is hopedthat the simultaneous effects of the nanofibrous structure, as wellas the bioactive composition, are both envisaged on the BGnp-PLAnanofibrousmatrices, which are particularly effective for utilizationas bone tissue engineering scaffolds.

J.-J. Kim et al. / Materials Chemistry and Physics 143 (2014) 1092e1101 1101

4. Conclusions

Nanocomposite scaffolds of PLA-BGnp were produced to have ananofibrous structure through camphene-assisted phase-separa-tion method. Freezing of the PLA-BGnp/camphene mixture inchloroform/1,4-dioxane at �20 �C followed by a lyophilizationgenerated highly interconnected pores and nanofibrous surfacemorphology. Production of macropores over hundreds of micro-meters was enabled by combining the process with saltimpregnation-leaching out method. The incorporation of BGnp upto 30wt% accelerated the degradation of scaffolds and significantlyenhanced the bone-bioactivity, i.e., apatite forming ability in SBF.While further studies remain as to the cell and tissue responses ofthe scaffolds, the macroporous and nanofibrous nanocompositescaffolds are considered as a novel 3D matrix useful for bone tissueengineering.

Acknowledgments

This work was supported by the research fund of DankookUniversity in 2012. Authors thank K. Dashnyam for her kind assis-tance in characterization of the materials.

References

[1] K.Rezwan,Q.Z.Chen, J.J. Blaker,A.R.Boccaccini,Biomaterials27 (2006)3413e3431.[2] B. Gupta, N. Revagade, J. Hilborn, Prog. Polym. Sci. (2007) 455e482.[3] B. Susmita, R. Mangal, B. Amit, Trends Biotech. 30 (2012) 546e554.[4] J.H. Kim, T.H. Kim, G.Z. Jin, J.H. Park, Y.R. Yun, J.H. Jang, H.W. Kim, J. Biomed.

Mater. Res. A (2013) 1447e1455.[5] S.A. Oh, J.E. Won, H.W. Kim, J. Biomat. Appl. (2012) 413e422.[6] J.J. Kim, G.Z. Jin, H.S. Yu, S.J. Choi, H.W. Kim, I.B. Wall, Mater. Sci. Eng. C 32

(2012) 2545e2551.[7] M.M. Stevens, Mater. Today 11 (2008) 18e25.[8] H.W. Kim, H.E. Kim, V. Salih, Biomaterials 26 (2005) 5221.[9] R. Khanna-Jain, B. Mannerstrom, A. Vuorinen, G.K.B. Sandor, R. Suuronen,

S. Miettinen, J. Tissue Eng. 3 (2012), http://dx.doi.org/10.1177/2041731412467998.

[10] H.W. Kim, J.C. Knowles, H.E. Kim, J. Biomed. Mater. Res. B Appl. Biomater. 70(2004) 240.

[11] H.H. Lee, U.S. Shin, J.H. Lee, H.W. Kim, J. Biomed. Mater. Res. Part B (2011)246e254.

[12] H.Y. Lee, G.Z. Jin, U.S. Shin, J.H. Kim, H.W. Kim, J. Mater. Sci. Mater. Med.(2012) 1271.

[13] K. Fahsai, M.E. Robertson, N. Gadegaard, R.O.C. Oreffo, R.J. Burchmore,M.J. Dalby, J. Tissue Eng. 2 (2011), http://dx.doi.org/10.4061/2011/534603.

[14] R.A. Perez, J.E. Won, J.C. Knowles, H.W. Kim, Adv. Drug Deliv. Rev. 65 (2013)471e496.

[15] M.B. Rahmany, M. Van Dyke, Acta Biomater. 9 (2013) 5431e5437.[16] S.H. Bang, T.H. Kim, H.Y. Lee, U.S. Shin, H.W. Kim, J. Mater. Chem. (2011)

4523e4530.[17] J. Hu, X. Liu, P.X. Ma, Biomaterials 29 (2008) 3815e3821.[18] S.H. Shin, P. Odnoo, O. Castino, J.A. Planell, H.W. Kim, J. Tissue Eng. 3 (2012),

http://dx.doi.org/10.1177/2041731412443530.[19] J.H. Jang, O. Castano, H.W. Kim, Adv. Drug Deliv. Rev. 6 (2009) 1065e1083.[20] Z.M. Huang, Y.Z. Zhang, M. Kotaki, Compos. Sci. Technol. 63 (2003) 2223e

2253.[21] X. Liu, P.X. Ma, Biomaterials 30 (2009) 4094e4103.[22] L.A. Smith, X. Liu, P.X. Ma, Soft Mat. 4 (2008) 2144e2149.[23] A. El-Fiqi, T.H. Kim, M. Kim, M. Eltohamy, J.E. Won, H.W. Kim, Nanoscale 4 (23)

(2012) 7475e7488.[24] A.A. Ahmed, A. A Ali, A.M. El-Fiqi, Solid State Sci. 13 (2011) 981e992.[25] T. Kokubo, H. Takadama, Biomaterials 27 (2006) 2907.[26] M.G. Haugh, C.M. Murphy, F.J. O’Brien, Tissue Eng. Part C (2010) 1887e1894.[27] C. Marlene de Barros, P.M. Marivalda, J. Biomed. Mater. Res. B 75 (2005)

451e456.[28] Y.Z. Cai, G.R. Zhang, L.L. Wang, Y.Z. Jiang, H.W. Ouyang, X.H. Zou, J. Biomed.

Mater. Res. Part A 100A (2012) 1187e1194.[29] K. Vassilis, K. David, Biomaterials 26 (2005) 5474e5491.[30] W. Guobao, X. Ma. Peter, Biomaterials 25 (2004) 4749e4757.[31] X. Jia, Z. Jinhui, G. Weiquan, L. Hongwei, W. Hongyan, L. Junfeng, Mater. Lett.

63 (2009) 658e660.[32] K.T. Noh, S.H. Jegal, H.H. Lee, U.S. Shin, H.W. Kim, Mater. Lett. 64 (2010)

802e805.[33] Z. Kai, W. Yunbing, A.H. Marc, F.F. Lorraine, Biomaterials 25 (2004) 2489e

2500.[34] W.J. Bae, K.S. Min, J.J. Kim, H.W. Kim, E.C. Kim, Dental Mater. 28 (2012) 1271e

1279.[35] B. Dorj, J.H. Park, H.W. Kim, Mater. Lett. 73 (2012) 119e122.[36] K.D. Patel, A. El-Fiqi, H.Y. Lee, R.K. Singh, D.A. Kim, H.H. Lee, H.W. Kim, J. Mater.

Chem. 22 (47) (2012) 24945e24956.