Reinforced Degradable Biocomposite by Homogenously Distributed Functionalized Nanodiamond Particles

Reinforced Degradable Biocomposite by

Homogenously Distributed Functionalized

Nanodiamond Particles

Yang Sun Anna Finne-Wistrand Thilo Waag Zhe Xing Mohamed Yassin

Akihito Yamamoto Kamal Mustafa Doris Steinm€uller-Nethl Anke Krueger

Ann-Christine Albertsson*

Severe phase separation was observed in blending nanodiamond particle (n-DP) in poly

(L-lactide-co-e-caprolactone) (poly(LLA-co-CL)) scaffold. In this study we optimized the scaffold

by the addition of 1–50% (w/w) polylactide modified n-DP (n-DP-PLA) or benzoquinone-

modified n-DP (n-DP-BQ). Composed by 10% n-DP-PLA,

composite had 6 times higher E-modulus in tensile test,

whereas the maximum reinforcement can be higher

than 15 times. However, n-DP-BQ composites conserved

the mechanical properties, and thermal properties of

the polymer substrate. The attachment, spreading and

growth of UE7T13 cells on modified n-DP composites

were similar to poly(LLA-co-CL), and independent to

n-DP concentrations. In summary, a proper modified

n-DP is the key to reinforce poly(LLA-co-CL) for tissue

engineering.

1. Introduction

Our previous results have shown that poly(l-lactide-co-e-

caprolactone) (poly(LLA-co-CL)) can be used to engineer

three-dimensional (3D) porous scaffolds that support

proliferation and differentiation of various types of cells

in vitro and in vivo.[1,2] The 75/25 (lactide/e-caprolactone

mole ratio) poly(LLA-co-CL) is a degradable amorphous

copolymer which has great elasticity and relative low

mechanical stiffness when prepared as a porous scaffold.

The low mechanical stiffness is a problem for load bearing

applications. To counter this problem it is possible to prepare

hybrid polymers by melt/solution blending with other

polymers,[3,4] changing the scaffold design[5] or making

composite materials with a mineral component, for instance

b-tricalcium phosphate[6] or hydroxyapatite.[7] Polymer

Prof. A.-C. Albertsson, Dr. Y. Sun, Assoc. Prof. A. Finne-Wistrand

Fibre and Polymer Technology, School of Chemical Science and

Engineering, KTH, Royal Institute of Technology, Teknikringen 56-

58 10044, Stockholm, Sweden

E-mail: [email protected]

Dr. T. Waag, Prof. A. Krueger

Institute for Organic Chemistry, W€urzburg University, Am

Hubland D-97074, W€urzburg, Germany

M. Yassin, Dr. Z. Xing, Prof. K. Mustafa

Department of Clinical Dentistry-Center for Clinical Dental

Research, University of Bergen, N-5009, Bergen, Norway

Dr. A. Yamamoto

Department of Oral and Maxillofacial Surgery, Nagoya University

Graduate School of Medicine, 65 Tsurumai-cho, Showa-ku,

Nagoya 466-8550, Japan

Dr. D. Steinm€uller-Nethl

DiaCoating GmbH, Mitterweg 24, A-6020, Innsbruck, Austria

Full Paper

� 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Macromol. Mater. Eng. 2014, DOI: 10.1002/mame.201400387 1wileyonlinelibrary.com1

Early View Publication; these are NOT the final page numbers, use DOI for citation !! R

nanocomposite materials have great potential for use in

many industrial and biomedical applications. Different

types of nanoparticles have been developed to achieve an

extensive range of chemical, physical or biological proper-

ties for different applications. Among the various organic

and inorganic nanoparticles used in hybrid nanocomposites,

nanodiamond particle (n-DP) is a well-known material with

unique properties, including extraordinary rigidity together

with good biocompatibility, high thermal conductivity and

high chemical stability.[8,9] It has been reported that n-DP

adsorb proteins strongly, including bone morphogenetic

protein-2 (BMP-2) and basic fibroblast growth factor (bFGF),

which can be loaded by physisorption and released due

to pH value change.[10,11] n-DP is also expected to be a good

filler material that improves material properties, especially

mechanical properties, of composite materials. The rein-

forcement effect by n-DP was observed in n-DP/polylactide

composites prepared by extrusion.[12] Similar reinforcement

and improved anti-wear friction–reduction were showed in

the composite of n-DP/poly(methyl methacrylate) which

was synthesized by co-coupling with the monomer[13] and a

hydrophilic composite of n-DP/poly(vinyl alcohol) prepared

by solvent casting.[14]

Significant property improvement usually requires that

the nanoparticles have good dispersibility and strong

interactions with the matrix, in order to avoid agglomer-

ation. Due to their enormous specific surface area,

appropriate surface functional groups are an efficient tool

to manipulate the interactions with the matrix to form

homogeneous nanoparticle/polymer composites. To

achieve high performance polymer nanocomposites it

has therefore been critical to develop an appropriate

surface on the nanoparticles, which usually involves chain

entanglement, anchoring and chemisorption.[15,16] Pristine

n-DP contains many functional groups, promoted by

oxidation and reduction methods.[17] A composite of 10%

octadecylamine modified n-DP in polylactide matrix

resulted in a doubled E-modulus with the maintenance

of fluorescence.[18] References indicate the n-DP or modified

n-DP could work as predominant nanofiller material for not

only the reinforcement of polymer substrate but also to

include other functionalities.

We have previously modified poly(LLA-co-CL) scaffolds

by surface physisorption of n-DP for the purpose of

attracting cell attachment and proliferation. Evaluated in

an in vivo sheep model, this material showed promise for

use in bone regeneration.[19] However, surface physisorp-

tion is not a totally controlled quantitate method, there is a

risk for burst release and severe aggregation. Preparing n-

DP based poly(LLA-co-CL) composites could be a good

alternative and was therefore assessed here. It should be

noted that the production of n-DP-based polymer compo-

sites also faces problems of phase separated composites,

and uncontrolled nanoparticle release. For the direct

blending method, it is hard to achieve a higher n-DP

concentration than 5 wt.-%, unless the polymer matrix is

hydrophilic and water soluble. If a greater concentration is

desired, appropriate modification of the nanoparticle is

required. We have therefore here prepared three types of

modified n-DP and determined the properties of the

composites relative to the particle ratios. In addition to

the oxygen-terminated n-DP which was prepared by

nanomilling, acid-purified detonation of n-DP,[20,21] a

polylactide-modified n-DP (n-DP-PLA)[22] and benzoqui-

none-modified n-DP (n-DP-BQ).[23] n-DP-PLA was produced

by an amphiphilic modification reaction using sodium

hydride-mediated anionic polymerization in anhydrous

tetrahydrofuran (THF).[22] The PLA grafting was expected to

provide effective entanglements between the side chains

and polymer matrix at the particle interface. Activated

benzoquinone (BQ) is a specific agent used for immobiliza-

tion of functional macromolecules[24,25] and might be a

potential carrier for coupling proteins, which in combina-

tion with poly(LLA-co-CL) would be very interesting in

tissue engineering.[26] The aim was to evaluate the

reinforcement effect of three types of n-DP by blending

with poly(LLA-co-CL), and relate this effect to specific

modifications and particle–polymer interactions.

2.. Experimental Section

2.1.. Material Synthesis

Polylactide-functionalize nanodiamond particles (n-DP-PLA) were

prepared using the ‘‘grafting-from’’ approach of anionic polymer-

ization as previously reported.[23] This amphiphilic n-DP was

prepared by functionalizing oxygen-terminated n-DP with poly-

lactide using sodium hydride as the initiator. The n-DP-PLA was

characterized by proton nuclear magnetic resonance spectroscopy

(1H NMR, Bruker AC 400, Bruker, Switzerland), that showed a

molecular weight of about 12 000 g �mol� 1 calculated by end-group

analysis. The surface loading was measured by thermogravimetric

analysis system (TGA, N2 flow) of 75%.

Benzoquinone functionalized nanodiamond particle (n-DP-BQ)

was prepared by adding as follows: 189.0 mg milled detonation

nanodiamond in 7 mL deionized water were mixed with 0.5 mL

half-saturated NaCl solution and centrifuged at 15 000 rpm for

5 min.[21] The supernatant was removed and the precipitate

redispersed in 20 mL phosphate buffered saline (PBS, PAA

Laboratories, UK) buffer (pH 8). After the addition of 150 mg

(1.38 mmol) benzoquinone (Sigma–Aldrich, USA) the suspension

was stirred for 24 h at room temperature. The nanodiamond was

subsequently washed eight times with PBS buffer solution (pH 7.4)

and twice with deionized water. After drying in vacuum, the

powder was redispersed in THF to obtain a reddish brown solution.

FTIR: n¼ 3 349 (s), 2 935 (m), 1 718 (m), 1 650 (s), 1 623 (s), 1 502 (s), 1

452 (s), 1 197 (s), 1 095 (s), 821 (s) cm� 1. Elemental analysis: C 84.40%,

H 1.74%, N 1.98%. Surface loading (TGA, N2 flow): 5.63%. Particle

size: (DMSO, c¼ 0.2 mg �mL� 1): 50% �90 nm, 90% �242 nm; zeta-

potential: –30.8 mV (pH¼6.3).

www.mme-journal.de

Y. Sun et al.

2 Macromol. Mater. Eng. 2014, DOI: 10.1002/mame.201400387

2 � 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim2

www.MaterialsViews.com

REarly View Publication; these are NOT the final page numbers, use DOI for citation !!

The random copolymer poly(l-lactide-co-e-caprolactone) (poly-

(LLA-co-CL)) was synthesized by ring-opening-polymerization. As

previously described, the reaction was catalysed by stannous

octoate with the molar ratio of monomer/catalyst equal to 10

000:1.[2,5] The poly(LLA-co-CL) contained about 25 mol-% e-capro-

lactone, and had an number average molecular weight of about 100

000 Da and a molar-mass dispersity (ÐM) of about 1.3 as measured by

size exclusion chromatography (SEC). The poly(LLA-co-CL) was then

used as the matrix in the composites described below.

2.2. Composites Preparation

All the particles were dispersed into tetrahydrofuran (THF) under

sonication. Four ratios of 1, 5, 10 and 50% w/w of n-DP-PLA and n-

DP-BQ were chosen and blended into poly(LLA-co-CL) solutions to

prepare composite materials. The nanoparticle/polymer film

composites were produced by solvent casting directly in a petri

dish. To prepare the scaffolds, the size-controlled (90–500 mm)

sodium chloride porogen with 0.25 g �mL� 1 composite solution was

first poured into a 15�45 mm2 glass vials (VWR, Sweden). After the

solvent evaporation, samples were removed from the vials, salt-

leached by deionized water, dried and shaped to proper size for

further characterizations.

n-DP physisorbed scaffolds (n-DP-Phys) were prepared by

modifying the 4 mg �mL� 1 n-DP solution for physisorption. An

ultrasonic vacuum drying system were employed to achieve good

particle distribution in the scaffold as described previously.[19] The n-

DP particle size was physically absorbed by burst particle water

suspensions during air pump evacuation. This process was repeated

eight times. After the modification procedure the n-DP modified

scaffolds were rinsed with distilled water and dried in vacuum.

2.3. Material Characterizations

Water contact angle measurements were made on a CAM 200

contact angle meter (KSV Instruments Ltd, Finland) using a drop of

5 mL Milli-Q water. An average was then calculated of five repeat

measurements was then made for each group. The data were

collected and calculated by CAM 200 software version 3.992.

Test of the mechanical property of tensile strength was carried

out on an Instron 5566 instrument (Instron, UK) with the cross-

head speed of 100 mm �min� 1 by a load cell of 50 N. Particle

concentration of 1, 5, 10 and 20% w/w of n-DP-PLA and n-DP-BQ/

poly(LLA-co-CL) were punched into strips 80 mm (length)�5 mm

(width) for the test. The thickness of specimens as measured by a

micrometer (Mitutoyo, Japan) was about 1 mm. The compression

test was then performed using the same machine with a

compressive rate of 10% thickness �min� 1. The size of the scaffolds

for the compression test was about 12 mm (thickness)� 12.5 mm

(diameter). The specimens were first preconditioned under 50%

humidity, 23 8C for 48 h. The stress and strain values at break points,

and the Young’s modulus were recorded for comparison.

Morphology of pristine n-DP and the modified n-DPs was observed

using a JEM-2100F (Jeol, USA) transmission electron microscope (TEM)

at an acceleration voltage of 100 kV. The fracture specimens from

tensile testing were collected and pre-coated with 5 nm thick platinum

layers by an automatic sputter coater (Agar Scientific, Stansted, UK) for

analysis by scanning electron microscopy (SEM) using an S-4300

instrument (Hitachi, Japan) accelerated at 2 kV voltage.

A differential scanning calorimetry instrument (DSC) (Mettler-

Toledo DSC instrument with DSC 820 module, Switzerland), and a

thermogravimetric analysis system (TGA) (Mettler-Toledo, Swit-

zerland) were employed for the thermal analysis. The DSC

measurements were performed under 50 mL �min� 1 nitrogen with

a heating/cooling rate of 10 K �min� 1. The data were recorded from

� 50 to 175 8C for the determination of the glass transition

temperature (Tg), and the degree of crystallinity (Xc). The Xc was

calculated using Equation (1):

Xc ¼DHf

DH�f

� �

� 100 ð1Þ

where the DHf is the heat of fusion of the sample andDH�f is the heat

of fusion for 100% crystalline polymer. For the calculation, the DH�fof poly(l-lactide) is 93.7 J � g� 1.[27] The thermal decomposition

temperature (Td) and the degree of mass loss were determined by

TGA with a heating rate of 10 K �min� 1 under an oxygen flow rate of

50 mL �min� 1.

Micro-computed tomography (mCT) was used to nondestruc-

tively and quantitatively measure the 3D porosity of scaffolds.

Randomly chosen scaffolds were scanned with a SkyScan 1172 mCT

imaging system (Aartselaar, Belgium) at 9 mm resolution using a

voltage of 40 kV, and a current of 250 mA. Image reconstruction and

analysis were conducted using the software package provided by

SkyScan. Recorded images were then reconstructed (NRecon,

version 1.6.9, Skyscan) into 3D images, which were in turn

analyzed using proprietary software (CTAn64, Skyscan) for percent

and total volume porosity as well as number of actual pores.

2.4. Cell Culture

Interactions between various composites with different ratios of n-

DP-PLA/poly(LLA-co-CL) and the stem cell line UE7T-13 (JCRB1154,

Health Science Research Resources Bank, Osaka, Japan) were

studied. The cells were immortalized by retrovirus mediated gene

transfer according to previous protocol,[28,29] and seeded on the top

of the materials of all surfaces and cultured under 37 8C with 5%

CO2. After 1 or 24 h of incubation, the cells were rinsed with PBS.

Cells were fixed in 10% formaldehyde and rinsed three times with

PBS before staining with Rhodamine phalloidin (Cytoskeleton, Inc.

Denver, CO 80223, USA) prepared from a 100 nm working stock in

PBS. The working solution was kept in the dark at room

temperature. Staining was performed by placing a drop of working

solution on a piece of parafilm, and then the membrane with cells

was placed on the drop of working solution. Incubation was done in

the dark for 30 min at room temperature. The membranes were

then washed three times in PBS. Counterstaining DNA for 30 s with

100 nm DAPI (Invitrogen, Carlsbad, CA, USA) in PBS, followed by

washing with PBS. Cell morphology was observed using a Nikon

A1R confocal microscope (Nikon, Japan).

3. Results and Discussion

A composite material receives the positive effects from the

additive itself, or from specific interactions at the interface of

the matrix and the particle surfaces. To obtain a successful

Reinforced Degradable Biocomposite by Homogenously

www.mme-journal.de

Macromol. Mater. Eng. 2014, DOI: 10.1002/mame.201400387 3

� 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 33



composite material requires a homogeneous distribution of

the additive and specific physical or chemical linkages created

in the composite matrix. The challenge here is that pristine n-

DP do not form a stable nanosuspension in solvents such as

chloroform and toluene, which are commonly used for

dissolving polymers.[23,30] We prepared polymer films by

direct blending of n-DP into the poly(LLA-co-CL) matrix. The

dispersed n-DP suspension was gradually added into the

copolymer solutions, and shaken overnight before solvent

casting. The surface morphology of the films composed of

different ratios of n-DP in the poly(LLA-co-CL) matrix are

shown in Figure 1. After solvent evaporation, the cavernous

structure distributed on the surface indicated uneven

composite morphology when the amount of added n-DP

was higher than 5 wt.-% (Figure 1c, d). From the cross-section

image of 5 wt.-% n-DP sample shown in Figure 1e, as well as

the images from the surfaces, it is obvious that the hollow

holes were also formed inside the sample. A possible

explanation of this severe phase segregation phenomenon

is insufficient dispersion of n-DP in the solvent (and hence

agglomerate formation) and poor connectivity between n-DP

and the polymer matrix. Diminished mechanical properties of

the composite may be seen when this occurs. To prevent this,

we designed and prepared two types of chemically modified

n-DPs: (polylactide-modified n-DP) n-DP-PLA and (benzoqui-

none modified n-DP) n-DP-BQ (Figure 2), with the aim to

improve particle dispersibility and cooperative interactions

between the solid particles and flexible polymer chains.

The morphology of pristine and modified particles was

examined by TEM (Figure 2). Benzoquinone and PLA were

covalently bound on the n-DP surfaces (Figure 2a) and it was

obvious that this stabilized the particle to form suspensions

in THF. Representative TEM images (Figure 2b) show n-DP-

PLA was mostly confined within the shadow-shaped cover

of PLA grafting possesses a polymeric shell with a thickness

of about 5–7 nm. The synthesis method of n-DP-BQ yields

very thin layer of BQ (Figure 2c) which is much thinner

compared to n-DP-PLA. Probably there was only one layer of

BQ grafted as no secondary reaction was expected. In

addition, we characterized the particles by TGA for the

purpose of determining particle thermal stability and

grafting properties. In the thermograms (Figure 2d), a mass

loss of about 67% was seen in n-DP-PLA from 307 to 354 8C,

while a mass loss of about 11% from 160 to 386 8C was

observed. This demonstrates surface grafting of both PLA

and BQ respectively and that n-DP-PLA contains a higher

quantity of organic material on the shell compared to n-DP-

BQ. This result corroborated to the finding of different

thickness of contour layers in TEM images (Figure 2b, c).

After surface modification, n-DP-BQ and n-DP-PLA

formed suspensions in THF, and it was possible to fabricate

composite materials by solvent casting method. Images of

the composites are presented in Figure 3a. It is clear that

different sample colours in the image were corresponding

to the particle concentrations and the homogeneous colour

distribution indicates well-dispersed particles in the

horizontal direction. Contact angle measurements showed

slightly increased hydrophilicity of the contact angle

results with the increase of particle concentration of both

n-DP-PLA and n-DP-BQ composites in Figure 3b. This

indicates the existence of small amount of hydrophilic

groups on the particle surfaces. However, the hydrophilicity

of the scaffolds is not significantly changed.

The extraordinary hardness of n-DP is one of the major

reasons to design and use surface modified n-DP in

composite materials. We have now two types of modified

n-DP that form relatively stable suspensions in organic

solvents, resulting in improved particle dispersion during

blending with polymer solutions. However, it remains to be

clarified whether the particles indeed reinforce the compo-

sites. In Figure 4, we compared the tensile properties of the

two composites containing 1, 5, 10 and 50% w/w of particles,

respectively. There is a clear trend towards enhanced E-

modulus with increasing amount of n-DP-PLA. With 50% n-

DP-PLA (Figure 4a) the E-modulus was about 15 times higher

than pristine poly(LLA-co-CL), and for even 1%, the smallest

amount used, the average E-modulus was increased. At the

break point, good elasticity was conserved of around 800% in

n-DP-PLA composite except for the 50%. The value of stress

was also similar between the groups except for a significant

decrease in the 50% n-DP-PLA. In contrast to n-DP-PLA, the n-

DP-BQ had no influence on the E-modulus, there was no

significant changes in the values of stress and strain

between each group measured at the break point. In

summary, n-DP-PLA significantly enhanced stiffness in

the poly(LLA-co-CL) matrix, while the influence of n-DP-BQ

on tensile properties was negligible. This may be due to

different properties of the surface grafting interactions. The

Figure 1. Surface morphology of solvent cast n-DP/poly(LLA-co-

CL) composites with the particle concentration of 0% (a), 1% (b),

5% (c), 10% (d), respectively. A cross-section image of 5% n-DP (e)

shows the porous structure inside the film.

www.mme-journal.de

Y. Sun et al.





PLA grafting has a structure similar to the polymer matrix,

which makes the particle readily miscible, while the PLA

chain from the surface may be entangled with the poly(LLA-

co-CL) chain to establish physical interactions that optimize

mechanical properties, especially stiffness. These properties

are not present in the n-DP-BQ/poly(LLA-co-CL) composites,

and the n-DP-BQ lacks functional groups to form chemical

interactions with the poly(LLA-co-CL) matrix.

Figure 2. Illustration of the modification of n-DP (top). TEM micrographs show particle morphologies of (a) n-DP, (b) n-DP-PLA, and (c) n-DP-

BQ. Thermal gravimetric analysis of the particles processed under oxygen environment is shown in (d).


www.mme-journal.de





To better understand why the E-modulus was enhanced

significantly by n-DP-PLA, the surface and cross-section of

specimens from the tensile test were collected and

evaluated by SEM. We observed that the particles were

separate and evenly distributed, without significant

aggregation (Figure 5c, e, g). The exception was at the

highest particle ratio of 50 wt.-% (Figure 5i), where the

particles formed agglomerates. Images of fracture surfaces

from the tensile test showed significantly different

morphologies at different particle concentrations. The

fracture surface became rougher with increased particles

concentrations (Figure 5b, d, f, h, j). Meanwhile, the particle

agglomeration was observed only with 50% n-DP-PLA (right

edge of Figure 5j) suggesting the most effective particle

concentration would be <50%. This agglomeration could

explain the decreased strain of 50% n-DP-PLA in the tensile

test. Furthermore, from the cross-section images, an

increase in fibrous polymer structures were observed with

increased addition of n-DP-PLA (Figure 5d, f, h, j) after the

tensile tests. Examination of the fibrous structures in 10% n-

DP-PLA cross fracture images by SEM shows the typical

structure of several filaments (Figure 5 k–m). Further

enlargement shows that a filament initiates from an

embedded particle (Figure 5 l), and another image clearly

shows how the n-DP-PLA clusters are linked by filaments

(Figure 5 m). In a specimen of pristine poly(LLA-co-CL)

(Figure 6a) no fibrous structure was observed, and the same

Figure 3. Optical images of the composites (a) showed on the top

of corresponding data points of the contact angle (b).

Figure 4. Summary of the values of E-modulus, stress and strain (break point) from tensile testing (a,b) and compression test (c,d) of the

composites with different particle ratios.

www.mme-journal.de

Y. Sun et al.





was true for the n-DP-BQ based composite fracture surfaces

(Figure 6b–d). Strong agglomeration can be seen in a high

magnification image of the 50 wt.-% n-DP-BQ composite

(Figure 6f). By comparing the fracture surfaces in n-DP-BQ

and n-DP-PLA composites, it appear that n-DP-PLA are

involved in the deform of the whole polymer matrix under

tension, which is responsible for the enhanced stiffness.

This reinforcement is proportional to the amount of n-DP-

PLA up to 50 wt.-%, where the particles start to aggregate.

Besides the benefits of increased resistance to tension

provided by the hard-core particles, the elastic property of

the material was preserved by the amorphous polymer

matrix.

In addition to producing homogeneous film samples, the

ability to manufacture porous scaffolds is essential for

tissue engineering applications. The amorphous copolymer

poly(LLA-co-CL) has been carefully analyzed as a material

for porous scaffolds, and different sizes and shapes have

been produced. n-DP has also been physisorbed in the

porous scaffolds, the combination was successfully used as

in vivo implants in a sheep model.[19] Usually the

introduction of nanoparticles by physisorption or directly

blending brings the potential risk of nanoparticle burst

release into the surrounding medium. For this reason,

modification of n-DP is important as it may help to increase

the miscibility and stability to control the release. The

prepared porous scaffolds are shown in Figure 7c. Like

the film samples, the homogeneous colour comes from

the particles integrated into the scaffolds. For comparison,

we also prepared n-DP-Phys scaffolds by salt-leaching

Figure 5. SEM images of 0–50% n-DP-PLA/poly(LLA-co-CL) surface (a,c,e,g,i) and cross fracture (b,d,f,h,j). The cross fracture images of

reinforced 10% n-DP (k–m) were presented under different magnifications.


www.mme-journal.de





method. The compression tests of the scaffolds showed a

slightly increased E-modulus for the n-DP-PLA composite

compared to pure poly(LLA-co-CL). The E-modulus increased

as the particle concentration increased. For n-DP-BQ

composites, the E-modulus was not consistently related

to the particle concentration. When the scaffolds were

compressed to 60% of their original thickness (Figure 7b),

the stress values of n-DP-PLA scaffolds were similar within

each ratio. However, the n-DP-BQ scaffolds showed a

slightly decreased stress with increasing amounts of n-DP-

BQ due to weak particle–matrix interactions.

In porous scaffolds, the E-modulus may change with the

degree of porosity. We therefore examined the scaffold

porosity by m-CT. All the samples had relatively high

porosity and low closed porosity. Compared with the other

samples in this study, the n-DP-Phys scaffolds had lower

porosity and a higher ratio of closed pores (Figure 7c). This

may result from the physisorption method loading large

amounts of n-DP into the pores. Loaded n-DP might occupy

the effective pore volume and decrease porosity, as well as

blocking and sealing some of the micro-size pores. As with

tensile properties, compressive properties are affected by

the degree of particle dispersion and particle–matrix

interactions. In n-DP-BQ/poly(LLA-co-CL) scaffolds, the

concentration-dependent decrease of stress is primarily

due to the missing interaction between the n-DP-BQ and the

poly(LLA-co-CL) matrix. This weakening effect has also

described when epoxy matrix composites was filled with

Figure 6. SEM images of 0–50 wt.-% n-DP-BQ/poly(LLA-co-CL) cross fracture (a–d) with higher magnification images of the 50% composite (e,f).

www.mme-journal.de

Y. Sun et al.





diamond particles.[31] While the mechanical properties of

aminated n-DP/epoxy composite increased or decreased as

a result of the degree of curing.[32] For our n-DP-PLA

composite scaffold, PLA grafting provided better interac-

tions with the matrix which in turn improved the E-

modulus. However, when the particle agglomerates were

relatively large (like in the 50% composites), this enhance-

ment was limited. Further, the scaffold composites with

insufficient molecular interaction (n-DP-BQ composites)

did not show this enhancement. The lower porosity

explains why a high E-modulus was seen for n-DP-Phys

scaffolds.

When introducing nanoparticles into the polymer

matrix, there is a possibility of polymer crystallinity

change, which may cause a change in mechanical proper-

ties. For example, previous results indicated that n-DP could

increase the rate of crystallization and crystallinity of

ethylene-1-octene copolymer.[33] In addition, octadecyl-

amine-modified n-DP increased the crystallinity of PLA and

thereby improve the mechanical properties.[18] In this

Figure 7. Summary of compressive E-modulus (a), and compressive stress at 60% strain (b) for particle ratios from 0 to 50%. Porosity of

scaffolds and corresponding images are shown in (c).


www.mme-journal.de





study, poly(LLA-co-CL) is an amorphous random copolymer

with a Tg of about 7 8C. However, n-DP-PLA showed similar

thermal characteristics as PLA, with a glass transition

temperature (Tg) at 47 8C, a crystallinity of about 10% and

melting temperature of 143 8C measured by DSC (Figure 8a).

With the addition of n-DP-PLA, Tg gradually increased to

around 18 8C. For 50 wt.-% n-DP-PLA, a melting peak of

about 4% crystallinity was seen. This melting peak is most

likely due to establishment of agglomeration as shown in

Figure 5j.

This result emphasizes the previous conclusion that n-

DP-PLA can be generously mixed into poly(LLA-co-CL)

matrix without causing self-aggregation, up to a certain

concentration, and effectively contribute to the improve-

ment of mechanical properties. In contrast, the n-DP-BQ

composites had a constant Tg around 6–7 8C. The intro-

duction of n-DP-BQ into poly(LLA-co-CL) did not neither

influence the polymer thermal property or mechanical

properties. This ‘‘inert’’ property of n-DP-BQ makes it

interesting for delivery applications; the weak interactions

may facilitate release of the particles from the matrix. For

example, the benzoquinone group could be functionalized

with specific domains to promote cell or protein binding. A

stable release of functionalized n-DP-BQ from the compo-

site material could be a potential candidate as a macro-

molecule carrier platform. Poly(LLA-co-CL), PLA and n-DP

have shown good biocompatibility in previous stud-

ies.[8,34,35] Composites of these materials may therefore

be expected to also show biocompatibility. We tested cell

behaviour on n-DP-PLA/poly(LLA-co-CL) composites of

Figure 8. DSC curves of 0–50 wt.-% n-DP-PLA/poly(LLA-co-CL) (a) and n-DP-PLA-BQ (b) composites.

www.mme-journal.de

Y. Sun et al.





different particle ratios. Cells on top of the different

materials showed similar behaviour both after 1 h

(Figure 9A–D) and 24 h (Figure 9E–H). Cells of the UE7T13

cell line were able to attach to the material and started to

spread in round shape after 1 h, with normal spindle-

shaped cells observed in all groups after 24 h. The concern of

the potential toxicity, especially for the 50% composite, is

not observed which also indicates good biocompatibility of

the n-DP-PLA composite materials.

4.. Conclusion

A reinforced composite with homogenously distributed

functionalized n-DPs was achieved by optimizing the

aggregations and interactions between poly(l-lactide-co-

e-caprolactone) (poly(LLA-co-CL)) matrix and varying ratios

of oxygen-terminated nanodiamond (n-DP), polylactide-

modified nanodiamond (n-DP-PLA) or benzoquinone-modi-

fied nanodiamond (n-DP-BQ). A phase separated composite

was observed in pristine n-DP/poly(LLA-co-CL) composite

while aggregation was avoided and a better dispersion in

the polymer matrix was achieved using the modified

nanodiamond particles n-DP-PLA. The n-DP-BQ did not

change the original mechanical and thermal properties of

the polymer matrix. Whereas, the n-DP-PLA had strong

interactions with the matrix and greatly reinforced the

mechanical properties of the composite, increasing E-

modulus by about six times for 10 wt.-%. Biocompatibility

for all composites was demonstrated using the UE7T13 cell

line and even the 50 wt.-% n-DP-PLA composite showed

comparable biocompatibility with poly(LLA-co-CL)), which

support the potential for use of these composites in bone

tissue engineering.

Acknowledgements: Dr. Michele Fox (University of Arkansas For

Medical Sciences) is gratefully acknowledged for the language

correction and helpful discussions. This research is supported by

the Deutsche Forschungsgemeinschaft (KR3316/1-2) and the

European Commission (FP7 Health project Vascubone, contract

number 242175).

Received: November 5, 2014; Published online:DOI: 10.1002/

mame.201400387

Keywords: composites; reinforcement; grafting from; nanodia-

mond; polylactide

[1] P. Plikk, S. Malberg, A.-C. Albertsson, Biomacromolecules 2009,

10, 1259.

[2] S. Danmark, A. Finne-Wistrand, M. Wendel, K. Arvidson, A.-C.

Albertsson, K. Mustafa, J. Bioact. Compat. Polym. 2010, 25,

207.

[3] B. Gupta, A. R. Ray, J. Appl. Polym. Sci. 2012, 123, -1944.

[4] N. S. Choi, C. H. Kim, K. Y. Cho, J. K. Park, J. Appl. Polym. Sci.

2002, 86, 1892.

[5] Y. Sun, A. Finne-Wistrand, A. C. Albertsson, Z. Xing, K.

Mustafa, W. J. Hendrikson, D. W. Grijpma, L. Moroni, J.

Biomed. Mater. Res. Part A 2012, 100, 2739.

[6] F. J. Mart�ınez-V�azquez, F. H. Perera, P. Miranda, A. Pajares, F.

Guiberteau, Acta Biomater. 2010, 6, 4361.

Figure 9. Cell attachment, spreading, distribution and morphology on pristine poly(LLA-co-CL) (A,E), 1% (B,F), 10% (C,G), 50% (D,H) n-DP-PLA/

poly(LLA-co-CL) composites were shown after 1 h (A–D) and 24 h (E–H).


www.mme-journal.de





[7] G. Wei, P. X. Ma, Biomaterials 2004, 25, 4749.

[8] V. N. Mochalin, O. Shenderova, D. Ho, Y. Gogotsi, Nat.

Nanotechnol. 2011, 7, 11.

[9] A. Krueger, J. Mater. Chem. 2011, 21, 12571.

[10] L. Moore, M. Gatica, H. Kim, E. Osawa, D. Ho, J. Dent. Res. 2013,

92, 976.

[11] D. Steinm€uller-Nethl, F. R. Kloss, M. Najam-Ul-Haq, M. Rainer,

K. Larsson, C. Linsmeier, G. K€ohler, C. Fehrer, G. Lepperdinger,

X. Liu, Biomaterials 2006, 27, 4547.

[12] Y.-Q. Zhao, K.-T. Lau, J.- Kim, C.-L. Xu, D.-D. Zhao, H.-L. Li,

Compos. Part B: Eng. 2010, 41, 646.

[13] Y. L. Hsin, H.-Y. Chu, Y.-R. Jeng, Y.-H. Huang, M. H. Wang, C. K.

Chang, J. Mater. Chem. 2011, 21, 13213.

[14] S. Morimune, M. Kotera, T. Nishino, K. Goto, K. Hata,

Macromolecules 2011, 44, 4415.

[15] D. Ciprari, K. Jacob, R. Tannenbaum, Macromolecules 2006, 39,

6565.

[16] A. C. Balazs, T. Emrick, T. P. Russell, Science 2006, 314, 1107.

[17] A. Krueger, D. Lang, Adv. Funct. Mater. 2012, 22, 890.

[18] Q. Zhang, V. N. Mochalin, I. Neitzel, I. Y. Knoke,J. Han,C. A. Klug, J.

G. Zhou, P. I. Lelkes, Y. Gogotsi, Biomaterials 2011, 32, 87.

[19] Z. Xing, T. O. Pedersen, X. Wu, Y. Xue, Y. Sun, A. Finne-

Wistrand, F. R. Kloss, T. Waag, A. Krueger, D. Steinm€uller-

Nethl, Tissue Eng. Part A 2013, 19, 1783.

[20] A. Kr€uger, F. Kataoka, M. Ozawa, T. Fujino, Y. Suzuki, A.

Aleksenskii, A. Y. Vul, E. sawa, Carbon 2005, 43, 1722.

[21] M. Ozawa, M. Inaguma, M. Takahashi, F. Kataoka, A. Krueger,

E. sawa, Adv. Mater. 2007, 19, 1201.

[22] Y. Sun, P. Ols�en, T. Waag, A. Krueger, D. Steinm€uller-Nethl, A.-

C. Albertsson, A. Finne-Wistrand, Part. Part. Syst. Char. 2014,

DOI: 10.1002/ppsc.201400098

[23] S. Suliman, Z. Xing, X. Wu, Y. Xue, T. O. Pedersen, Y. Sun, A. P.

D�skeland, J. Nickel, T. Waag, H. Lygre, A. Finne-Wistrand, D.

Steinm€uller-Nethl, A. Krueger, K. Mustafa, J. Control. Rel.

2015, 197, 148.

[24] J. Brandt, L.-O. Andersson, J. Porath, Biochim. Biophys. Acta

(BBA)-Protein Struc. 1975, 386, 196.

[25] M.-A. Mateescu, E. Agostinelli, G. Weltrowska, M. Weltrowski,

B. Mondovi, Biol. Met. 1990, 3, 98.

[26] K. Purtov, A. Petunin, A. Burov, A. Puzyr, V. Bondar, Nanoscale

Res. Lett. 2010, 5, 631.

[27] E. W. Fischer, H. Sterzel, G. Wegner, Kolloid-Zeitschrift und

Zeitschrift f€ur Polymere 1973, 251, 980.

[28] H. Ishimine, N. Yamakawa, M. Sasao, M. Tadokoro, D. Kami,

S. Komazaki, M. Tokuhara, H. Takada, Y. Ito, S. Kuno, K.

Yoshimura, A. Umezawa, H. Ohgushi, M. Asashima, A.

Kurisaki, Biochem. Biophys. Res. Commun. 2013, 438, 753.

[29] T. Mori, T. Kiyono, H. Imabayashi, Y. Takeda, K. Tsuchiya, S.

Miyoshi, H. Makino, K. Matsumoto, H. Saito, S. Ogawa, Mol.

Cell. Biol. 2005, 25, 5183.

[30] A. Krueger, T. Boedeker, Diam. Relat. Mater. 2008, 17, 1367.

[31] J. d’Almeida, S. Monteiro, G. Menezes, R. Rodriguez, J. Reinf.

Plast. Compos. 2007, 26, 321.

[32] I. Neitzel, V. Mochalin, J. Niu, J. Cuadra, A. Kontsos, G. Palmese,

Y. Gogotsi, Polymer 2012, 53, 5965.

[33] A. Korobko, N. Bessonova, S. Krasheninnikov, E. Konyukhova,

S. Drozd, S. Chvalun, Diam. Relat. Mater. 2007, 16, 2141.

[34] T. O. Pedersen, A. L. Blois, Y. Xue, Z. Xing, Y. Sun, A. Finne-

Wistrand, J. B. Lorens, I. Fristad, K. N. Leknes, K. Mustafa, Stem

Cell Res. Ther. 2014, 5, 23.

[35] Y. Sun, Z. Xing, Y. Xue, K. Mustafa, A. Finne-Wistrand, A.-C.

Albertsson, Biomacromolecules 2014, 15, 1259.

www.mme-journal.de

Y. Sun et al.





Reinforced Degradable Biocomposite by Homogenously Distributed Functionalized Nanodiamond Particles

Documents