Biomaterials 24 (2003) 2773–2778

Thermal and mechanical characteristics of poly(l-lactic acid)nanocomposite scaffold

Jong Hoon Leea, Tae Gwan Parkb, Ho Sik Parka, Doo Sung Leea, Young Kwan Leec,Sung Chul Yoond, Jae-Do Nama,*

aDepartment of Polymer Science and Engineering, University of Sung Kyun Kwan, Suwon 440-746, South KoreabDepartment of Biological Sciences, Korea Advanced Institute of Science and Technology, Taejon 305-701, South Korea

cDepartment of Chemical Engineering, Division of Applied Chemistry, University of Sung Kyun Kwan, Suwon 440-746, South KoreadDivision of Life Science at the College of Natural Sciences and Division of Applied Life Sciences at the Graduate School,

Gyeongsang National University, Chinju 660-701, South Korea

Received 23 October 2002; accepted 29 January 2003

Abstract

Inorganic nanosized silicate nanoplatelets were incorporated into biodegradable poly(l-lactic acid) (PLLA) for the purpose of

tailoring mechanical stiffness of PLLA porous scaffold systems. Increasing the nucleation density around the foreign body surfaces,

the montmorillonite (MMT) nanoplatelets modified with dimethyl dihydrogenated tallow ammonium cations decreased the glass

transition temperature and the degree of PLLA crystallinity, which seemingly caused the accelerated biodegradation rate of PLLA

nanocomposites due to the enhanced segmental mobility of backbone chains and the expanded amorphous region of PLLA matrix.

The tensile modulus was increased from 121.2MPa of pristine polymer scaffold to 170.1MPa of MMT/PLLA nanocomposite

scaffold (ca. 40% increment) by the addition of small amount of MMT platelets (5.79 vol%) acting as a mechanical reinforcement of

polymer chains in the nanoscale molecular level. Overall, the nanotechnology used in this study may be applied to various scaffold

systems of biodegradable polymers and hard/soft scaffold structures requiring critical control and design characteristics of

mechanical stiffness and biodegradation rate.

r 2003 Elsevier Science Ltd. All rights reserved.

Keywords: Nanocomposite; PLLA; Montmorillonite; Modulus; Biodegradation rate; Scaffold

1. Introduction

Porous biodegradable scaffolds have been studied forthe application as a three-dimensional template forinitial cell attachment and subsequent tissue formationboth in vitro and in vivo. Various tissues such ascartilage, bone, heart, valves, nerves, muscle, bladder,liver, etc. have been produced using the scaffolds. Thescaffolds should be highly porous to allow cells to beseeded and to permit the facile invasion of blood vesselsfor the supply of nutrients to the transplanted cells [1,2].Currently, nonwoven polyglycolic acid (PGA) fibermesh reinforced with poly(l-lactic acid) (PLLA) andpoly(d,l-lactic-co-glycolic acid) (PLGA) are most com-monly used for soft-tissue cell transplantation [3–5].

However, it has been pointed out that these systems arenot suitable for the purpose of hard-tissue regenerationdue to its weak mechanical properties [5–9]. As thepolymer scaffold matrix gradually degrades and simul-taneously the resulting space is filled with the growingtissue, the scaffold structure should provide sufficienttemporary mechanical support to withstand in vitro andin vivo stresses and loading in applications such as boneregeneration [10–12].The mechanical property of the scaffold can be

changed by the structure, porosity, fabrication method,crystallinity, etc. [1]. However, it is more likelyassociated with the intrinsic stiffness of the polymermaterial. Recently, nanocomposite technology hasdeveloped in many areas of polymer application.Adding a small quantity of layered silicate particleswith a high aspect ratio can increase mechanical andphysical properties including higher strength, greater

*Corresponding author. Fax: +82-31-292-870.

E-mail address: jdnam@skku.ac.kr (J.D. Nam).

0142-9612/03/$ - see front matter r 2003 Elsevier Science Ltd. All rights reserved.

doi:10.1016/S0142-9612(03)00080-2

stiffness, higher heat resistance, higher UV resistance,while maintaining transparency and impact property[13,14]. For example, the moduli of polypropylene andnylon 6 nanocompoistes have been reported to increasefrom 1.1 to 2.1GPa (91%) and from 1.98 to 3.12GPa(58%), respectively, by the addition of 2–8wt% oflayered silicate nanoparticles [15]. Similar results havebeen observed in other polymer systems such aspolyurethatne nanocomposite from 3 to 7MPa (133%)and epoxy nanocomposite from 1.1 to 2.3GPa (109%)[15,16]. Recently, it has been reported that the storagemodulus of biodegradable polylactide (PLA)/layeredsilicate nanocomposites is increased from 1.63 to2.32GPa (42%), and the biodegradation rate of thenanocomposite is significantly accelerated in compar-ison with pristine polymer in terms of the weight lossand molecular weight [17].In this study, PLLA nanocomposite scaffold was

prepared by the exfoliation–adsorption method using anappropriate solvent, where PLLA and layered silicateare homogeneously dissolved and dispersed, respectively[13,14,18]. The exfoliated PLLA nanocomposite scaffoldwas successfully prepared, and the thermal and mechan-ical properties of the PLLA nanocomposite wereinvestigated. The extent of crystallinity of PLLAnanocomposite was decreased and the tensile modulusof the nanocomposite scaffold was increased by theincorporation of a small amount of montmorillonite(MMT) nanoplatelets, which may to be used to tailorthe structural stiffness and biodegradation rate to fit therequirement of a broad range of hard and soft scaffoldapplications.

2. Experimental

2.1. Materials

Poly(l-lactic acid) was purchased from Shimadzu Co.The number-average molecular weight (Mn) was218,000, and the ratio of weight-average molecularweight (Mw) to Mn was 1.55. Chloroform (CHCl3),ammonium bicarbonate (NH4HCO3) and sodium chlor-ide (NaCl) were obtained from Sigma Aldrich.The MMT ([MxAl4�xMgx]Si8O20(OH)4, xE0:67)

used in our experiment belongs to a family ofphyllosilicates (2:1) that comprises two tetrahedral silicathin layers with a central octahedral sheet of magnesia.Isomorphic substitution within the layers gives intrinsicsurface negative charge, and thus dimethyl dihydroge-nated tallow ammonium (2M2HT) was used as a cationintercalent (95meq of 2M2HT/100 g of MMT: 20Atprovided by Southern Clay Products), where tallow ispredominantly octadecyl chains with small amounts oflower homologues (B65% if C18, B30% of C16 andB5% of C14). The layer thickness is around 1 nm and

the lateral dimensions vary from several nanometers tomicrons.

2.2. Scaffold preparation

The scaffolds were prepared by a salt leaching/gasfoaming method according to the previous report [18]. Aviscous polymer solution with a concentration of 0.1 g/ml was prepared by dissolving PLLA polymer inchloroform. NH4HCO3/NaCl salt particles sieved inthe range of 150–300 mm and 2M2HT–MMT clays wereadded to the PLA solution and thoroughly mixed. Theamount of the 2M2HT–MMT clay was 2.14, 3.58, and5.79 vol% to PLLA. The paste mixture of polymer/salts/solvent was then cast into a special device equipped witha glass slide as a sheet mold. The cast sheet (approxi-mately 3mm thick) was obtained after being air-driedunder atmospheric pressure for 2 h. After the sheet wassemi-solidified, a two-step salt leaching was performed.The sheet was first immersed in 90�C hot water(approximately 10min) to leach out the NH4HCO3particles, concomitantly generating gaseous ammoniaand carbon dioxide in the polymer matrix. After no gasbubbles were generated, the sheet was subsequentlyimmersed into another beaker containing 60�C water(approximately 30min) to leach out the remaining NaClparticles, and then freeze dried for 2 days.

2.3. Characterization

The interlayer spacing (d0 0 1) was examined by wide-angle X-ray diffraction (Mac Science, Mac-18xhf). TheCuKa radiation (l ¼ 0:154 nm) and curved graphitecrystal monochromator were used in the measurement.The applied voltage and current of the X-ray tubes were30 kV and 100mA, respectively. The 2y was scannedbetween 1.5� and 20� at 2�/min. The crystalline meltingand recrystallization behavior was measured by adifferential scanning calorimetry (TA InstrumentsDSC 2910) at a heating rate of 10�C/min in nitrogenenvironment. The nanocomposite polymer scaffoldswere cut in the size of 1 cm� 4 cm for tensile tests. Bothends of tensile specimens were epoxy molded andclipped by a special gripper. The tensile strength andmodulus of the scaffolds were measured by a tensilemachine (Toyoseiki, Stro-Graph V1-C) with a load cellof 10N, and the cross-head speed was 5mm/min atroom temperature. Nine specimens were tested for eachnanocomposite system and the average was taken asmodulus values after eliminating the maximum andminimum values. The molded scaffold surface andfractured cross sections were observed by scanningelectron microscopy (SEM, Hitachi S-2140). The por-osity of scaffolds was estimated by using the rule ofmixture converting the mass to volume of compositecomponents by means of densities of MMT and PLLA,

J.H. Lee et al. / Biomaterials 24 (2003) 2773–27782774

1.77 and 1.25, respectively. And the average pore sizeand its wall thickness were estimated from the SEMmicrograph images.

3. Results and discussion

Fig. 1 compares the XRD patterns of PLLA,2M2HT/MMT, and PLLA/2M2HT/MMT nanocom-posites. 2M2HT/MMT clay shows interlayer spacing at2"e ¼ 4:56� (d0 0 1 ¼ 1:94 nm), but no diffraction peak isvisible in the XRD diffractograms of PLLA and PLLA/2M2HT/MMT nanocomposite either because of a muchtoo large spacing between the layers (i.e. exceeding 8 nmin the ordered exfoliated structure) or because thenanocomposite has no ordered layer structure. In anycases, disappearance of the d0 0 1 interlayer spacing inthe XRD diffractogram may be regarded as a formationof homogeneously dispersed nanocomposites, whichoften give unusual improvement in material propertiesdue to an extremely large surface area of nanosizedplatelets.The DSC thermograms of PLLA nanocomposites are

shown in Fig. 2 and the characteristic quantities aresummarized in Table 1. The glass transition temperatureof pristine PLLA is 63.4�C, but it decreases with theaddition of the MMT clay. For example, the 5.79 vol%of MMT/PLLA nanocomposite system exhibits 57.6�Cof glass transition temperature. The glass transitiontemperature depends primarily on chain flexibility,molecular weight, branching/crosslinking, intermolecu-lar attraction and steric effects, etc. [19,20]. In ournanocomposite system, the intermolecular attractions ofPLLA segments seem to be interrupted by the chargedMMT layers, and subsequently the PLLA backbonechains additionally gain the segmental mobility. It canalso be addressed that the MMT nanoparticles may

provide steric factors that seemingly increase the chainflexibility of PLLA backbones. Conclusively, the MMTnanoclay imposes more flexibility and mobility to thePLLA backbones resulting in the decreased glasstransition temperatures. It should be mentioned thatthe biodegradation rates of the engineered scaffoldscould be affected by the chain stiffness because themobility of reactive ions and molecules in the hydrolysisdegradation reactions could be strongly affected by thechain stiffness and intermolecular attraction of degrad-ing PLLA molecules.Comparing the heat of crystallization, heat of melting,

and recrystallization temperature (Tc) of quenchedPLLA and its nanocomposite systems, they decreaseby the addition of MMT clay. The nanosized layeredMMT plates provide large surface area due to theirsmall size and thus it is reasonable to consider that theclay particles could act as effective nucleating sites ofPLLA crystallization. The increased nucleating sites arelikely to facilitate the PLLA crystallization process inthe nanocomposite systems, and thus Tc is decreased bythe addition of nanoclays. The foreign bodies involvedin crystallizable polymers often favor heterogeneousnucleation exhibiting transcrystalline zone along theforeign surface with a sufficiently high crystalline density[20–28]. Although the transcrystallization is not alwaysobserved around the foreign-body surface, the nuclea-tion density is generally increased due to the foreign-body surface. The number of nucleation sites oftendetermines the morphology of growing crystallites

Fig. 1. Wide-angle XRD comparing MMT clay, PLLA, and PLLA/

MMT nanocomposite.

Fig. 2. DSC thermograms of quenched specimens of PLLA/MMT

nanocomposite containing different amounts of MMT nanoplatelet

measured at 10�C/min.

J.H. Lee et al. / Biomaterials 24 (2003) 2773–2778 2775

because a large number of nucleation centers would leadto a large number of small crystallites, which subse-quently give a low degree of overall crystallinity [27,28].In Table 1, it can be seen that the heat of crystallizationand heat of melting of the nanocomposite systems arelower than the pristine PLLA polymer seeminglybecause the incorporated nanoparticles increase thenumber of small crystallites and subsequently lowerthe overall crystallinity of PLLA.The hydrolysis biodegradation reaction is affected by

the overall extent of polymer crystallinity. Specifically,the biodegradation takes place in the amorphous regionfirst, and then the reaction zone moves to crystallineregion in most semi-crystalline polyester biodegradablepolymers [29,30]. Therefore, it is reasonable to addressthat the biodegradation rate of our nanocompositesystem may be faster than the pristine PLLA polymerdue to the expanded amorphous region of PLLA. Theprevious research has reported that the biodegradabilityof polylactide/MMT nanocomposites is significantlyenhanced in terms of weight loss and molecular weightespecially in the later stage of biodegradation [17], whichhas not been related to the crystallinity or chain stiffnessof PLLA matrix. The effect of MMT nanoparticles andPLLA crystallinity on the accelerated biodegradationrates and hydrolysis reaction mechanisms is underinvestigation. Further study should also be performedto identify the biocompatibility and biodegradability ofMMT composite systems, which are inert by nature andmay be controlled to be desirably small enough to beabsorbed into the circulatory systems.Fig. 3 shows the pore structure of the prepared

nanocomposite scaffold observed by SEM. The porositywas 91–92% and the pores seem to be well-connected inthe scale of 100–300 mm of pore size. Comparing themorphology of pristine PLLA and PLLA nanocompo-site scaffolds, there were little difference in light ofporosity configuration and characteristic feature ofscaffold walls and overall morphology. For the purposeof sustaining the structural integrity during cell growth,the stiffness of scaffold should be the most importantmechanical property. As can be seen in Fig. 3, thescaffolds are composed of very thin walls in the range of5–10 mm, which should withstand gravitational andinternal stresses generated by a relatively large amountof growing tissue (ca. 90% of porosity to be filled withgrowing tissue) for the duration of implantation. Thetheoretical values of gravitational force and internalstresses generated by the growing tissue may be small, ifany, not enough to break down the initial thickness ofengineered scaffold walls at once. However, it is ratherreasonable to consider that the scaffold walls woulddeform gradually by the degrading scaffold walls andgrowing cells, and finally become unsustainable tomaintain the overall structure. In this sense, the modulusmay be a key factor in designing the scaffold structureT

able1

SummaryofthermalandmechanicalpropertiesofPLLA/MMTnanocompositesystems

Claycontent

(vol%/wt%

)

Porosity(%

)Glasstransition

temperature(T

g)

(�C)

Recrystallization

temperature(T

c)

(�C)

Crystalline

melting

temperature(T

m)

(�C)

Heatofcrystalline

melting(H

m)

(J/ga)

Heatof

recrystallization(H

c)

(J/ga)

Tensilestrength

(MPa)

Tensilemodulus

(MPa)

0/0

91.0

63.4

110.8

175.6

37.2

29.30

47.9

121.2

2.14/3.0

92.9

59.6

98.9

169.5

34.2

26.61

41.9

133.8

3.58/5.0

91.9

52.5

87.1

163.7

30.9

24.77

37.9

144.3

5.79/8.0

92.5

57.6

89.9

166.3

33.2

25.54

22.7

170.1

aWeightofPLLAexcludingMMT.

J.H. Lee et al. / Biomaterials 24 (2003) 2773–27782776

by tailoring the mechanical properties to fit thestructural requirement of scaffold application.Fig. 4 shows the tensile modulus of PLLA nanocom-

posite scaffold plotted as a function of MMT claycontent. The modulus of pristine PLLA is 121.2MPaand increases up to 170.1MPa for 5.79 vol% of MMTnanocomposite giving ca. 40% increment of modulus.As discussed earlier, the crystallinity and the glasstransition temperature of PLLA nanocomposites arelower than pristine PLLA, but the modulus is signifi-cantly increased by the incorporation of a small amountof nanosized MMT clay. We believe that the nanoclaysact as a mechanical reinforcement of polymer chains inthe nanoscale molecular level and thus give highermodulus values. In a specific application for bone tissueengineering, the tensile modulus of bone is in the rangeof several hundred MPa [31]. In our nanocompositescaffold system having about 92% of porosity, themodulus seems to be within the range of intact-bonestiffness and furthermore it may be adjusted bychanging the pore morphology, the extent of porosity,and intercalation/exfoliation technique in nanoscalemanipulation.

When the scaffold modulus is much higher than theintact bone, the bone stress is not transferred andsubsequently hardly distributed to the scaffold, oftenresulting in a stress shielding (or stress protection)problem [32,33]. In addition, it should be consideredthat the overall scaffold modulus changes as thebiodegradation reactions proceed, and the regeneratedbone shares the imposed stresses with the scaffold.Therefore, the optimal stiffness of bone scaffolds shouldbe determined in the relation between the biodegrada-tion rates and the stiffness changes desirably minimizingthe stress shielding effects. In this sense, the nanocom-posite technology presented in this study exhibited agreat potential that could be applied to different systemsof biodegradable polymers and scaffold structurestailoring the mechanical stiffness and biodegradationrate, which likely results from nanoscale reinforcementof nanoplatelets, chain flexibility of backbone chains,and expanded amorphous morphology of PLLA matrix.

4. Conclusions

Nanosized platelets of MMT clay were incorporatedwith PLLA for the purpose of tailoring the scaffoldmechanical properties and identifying the crystallizationcharacteristics. The exfoliated PLLA nanocompositewas successfully prepared and lowered the glass transi-tion temperature and degree of crystallinity. Acting as amechanical reinforcement, the prepared PLLA nano-composite scaffolds gave improved tensile modulusvalues up to 40% by increasing the stiffness of scaffoldwalls. The nanocomposite technology presented in thispaper could affect biodegradation rate due to the gainedchain flexibility and decreased crystallinity of PLLAnanocomposite systems.

Acknowledgements

This work was supported by a grant from the KoreaResearch Foundation (KRF-2001-005-E00006)

Fig. 3. SEM micrograph PLLA/MMT nanocomposite scaffold containing 3.58 vol% of MMT: (a) � 100 and (b) � 200.

Fig. 4. Tensile modulus of PLLA/MMT nanocomposite as a function

of MMT content.

J.H. Lee et al. / Biomaterials 24 (2003) 2773–2778 2777

References

[1] Nam YS, Park TG. J Biomed Mater Res 1999;47(1):7–17.[2] Hutmacher DW. J Biomater Sci Polym Ed 2001;12(1):107–24.

[3] Mikos AG, Bao Y, Cima LG, Ingber DE, Vacanti JP, Langer R.

Biomater 1993;14:323–30.

[4] Cima LG, Vacanti JP, Vacanti C, Ingber DE, Mooney DJ,

Langer R. J Biomech Eng 1991;113:143–51.

[5] Freed L, Marquis JC, Nohria A, Emmanual J, Mikos AG.

J Biomed Mater Res 1993;27:11–23.

[6] Mikos AG, Bao Y, Cima LG, Ingber DE, Vacanti JP, Langer R.

J Biomed Mater Res 1993;27:183–9.

[7] Nam YS, Park TG. Biomater 1999;20:1783–90.

[8] Freyman TM, Yannas IV, Gibson LJ. Prog Mater Sci

2001;46:273–82.

[9] Nam YS, Yoon JJ, Park TG. J Biomed Mater Res 2000;53:1–7.

[10] Hua FJ, Nam JD, Lee DS. Macromol Rapid Commun

2001;22:1053–7.

[11] Kikuchi M, Itoh S, Ichinose S, Shinomiya K, Tanaka J.

Biomaterials 2001;22:1705–11.

[12] Kikuchi M, Tanaka J, J Ceram Soc Jpn 2000;108:643–45

[13] Alexandre M, Dubois P. Mater Sci Eng 2000;28:1–63.

[14] Kato M, Usuki A. Poymer–clay nanocomposites. In: Pinnavaia

TJ, Beall GW, editors. Polymer–clay nanocomposites. New York:

Wiley, 2000.

[15] Wang X, Pinnavaia TJ. Chem Mater 1998;10:1820–6.

[16] Lan T, Kaviratna PD, Pinnavaia TJ. ChemMater 1995;7:2144–50.

[17] Ray SH, Yamada K, Okamoto M, Ueda K. Polylactide-layered

silicate nanocomposite: a novel biodegradable material. Nano

Lett 2002;2(10):1093–6.

[18] Yoon JJ, Park TG. Degradation behaviors of biodegradable

macroporous scaffolds prepared by gas foaming of effervescent

salts. J Biomed Mater Res 2001;55:401–8.

[19] Cowie JMG. Polymers: chemistry & physics of modern materials.

New York: Chapman & Hall, 1991.

[20] Burton RH, Folkes MJ. In: Clegg DW, Collyer AA, editors.

Mechanical properties of reinforced thermoplastics. London:

Elsevier, 1986.

[21] Campell D, Qayyum MM. J Polym Sci: Polym Phys Ed

1984;18:83.

[22] Ishida H, Bussi P. Macromolecules 1991;24:3569.

[23] Lustiger A. Polym Compos 1992;13:408.

[24] Thomason JL, Van Rooyen AA. J Mater Sci 1992;27:889.

[25] Varga J, Karger-Kocsis J. Polym Bull 1993;30:105–18.

[26] Wu CM, Chen M, Karger-Kocsis J. Polym Bull 1998;41:

239–45.

[27] Park CS, Lee KJ, Kim SW, Lee YK, Nam JD. J Appl Polym Sci

2002;86(2):478–88.

[28] Park CS, Lee KJ, Nam JD, Kim SW. J Appl Polym Sci

2000;78(3):576–85.

[29] Scott G, Gilead D. Degradable polymers. London: Chapman &

Hall, 1995.

[30] Doi Y, Kumagai Y, Tanahashi N, Nukui K. Biodegradable

polymers and plastics. London: Royal Society of Chemistry,

1992.

[31] Lim JR, Kim SH, Kim YH. Polym Sci Technol 2002;13(1):

15–22.

[32] Moyen BJL, Lahey PL, Weinberg EH, Harris WH. J Bone Jt Surg

1978;60A:940–7.

[33] Uhthoff HK, Finnegan M. J Bone Jt Surg 1983;65B:66.

J.H. Lee et al. / Biomaterials 24 (2003) 2773–27782778