Reinforcement of calcium phosphate cement with multi-walled carbon nanotubes and bovine serum albumin for injectable bone substitute applications

J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S

4 (2011) 331–339

available at www.sciencedirect.com

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

Research paper

Reinforcement of calcium phosphate cement with multi-walled carbon nanotubes and bovine serum albumin for injectable bone substitute applications Kean-Khoon Chew a , Kah-Ling Low a , Sharif Hussein Sharif Zein a,∗ , David S. McPhail b , Lutz-Christian Gerhardt b,1 , Judith A. Roether b,2 , Aldo R. Boccaccini b,c a School of Chemical Engineering, Engineering Campus, Universiti Sains Malaysia, Seri Ampangan 14300 Nibong Tebal, Seberang Perai

Selatan, Pulau Pinang, Malaysia b Department of Materials, Imperial College London, Prince Consort Road, London SW7 2AZ, UK c Institute of Biomaterials, Department of Materials Science and Engineering, University of Erlangen-Nuremberg, 91058 Erlangen, Germany

A R T I C L E

I N F O

A B S T R A C T

Article history:

This paper presents the development of novel alternative injectable calcium phosphate

Received 11 August 2010

cement (CPC) composites for orthopaedic applications. The new CPC composites comprise

Received in revised form

β-tri-calcium phosphate (β-TCP) and di-calcium phosphate anhydrous (DCPA) mixed

28 October 2010

with bovine serum albumin (BSA) and incorporated with multi-walled carbon nanotubes

Accepted 31 October 2010

(MWCNTs) or functionalized MWCNTs (MWCNTs–OH and MWCNTs–COOH). Scanning

Published online 12 November 2010

electron microscopy (SEM), compressive strength tests, injectability tests, Fourier transform infrared spectroscopy and X-ray diffraction were used to evaluate the properties of the

Keywords:

final products. Compressive strength tests and SEM observations demonstrated particularly

Calcium phosphate cement (CPC)

that the concomitant admixture of BSA and MWCNT improved the mechanical properties,

Carbon nanotubes (CNTs)

resulting in stronger CPC composites. The presence of MWCNTs and BSA influenced the

Bovine serum albumin (BSA)

morphology of the hydroxyapatite (HA) crystals in the CPC matrix. BSA was found to act

Compressive strength

as a promoter of HA growth when bounded to the surface of CPC grains. MWCNT–OH-

Injectability

containing composites exhibited the highest compressive strengths (16.3 MPa), being in the range of values for trabecular bone (2–12 MPa). c 2010 Elsevier Ltd. All rights reserved. ⃝

1.

Introduction

shown to have important clinical potential for stabilizing such disorders (Mermelstein et al., 1998). Bio-resorbable

The improved life expectancy in developed countries has led to a significant rise in the number of musculoskeletal disorders, such as osteoporosis and osteoarthritis (Bohner, 2000). Minimally invasive surgical techniques have been

calcium phosphate cements (CPCs) represent an interesting alternative to traditional bone graft materials. Moreover, CPC is a highly desirable material for orthopaedic applications due to its mouldability, in situ self-hardening ability, excellent

∗ Corresponding author. Tel.: +60 4 599 6442; fax: +60 4 594 1013. E-mail address: [email protected] (S.H. Sharif Zein). 1 Present address: Technische Universiteit Eindhoven, Biomedical Engineering, PO Box 513, 5600 MB Eindhoven, The Netherlands. 2 Present address: Department of Materials Science and Engineering, University of Erlangen-Nuremberg, 91058 Erlangen, Germany. c 2010 Elsevier Ltd. All rights reserved. 1751-6161/$ - see front matter ⃝ doi:10.1016/j.jmbbm.2010.10.013

332

J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S

osteoconductivity, adjustable resorbability rate and bone replacement capability (Costantino et al., 1992; Friedman et al., 1998; Wang et al., 2007b; Chow, 2009). Chow (2009) has reviewed the current knowledge on CPCs in terms of their setting chemistry, mechanical properties and in vivo tissue response, and has attributed a key role to the ‘injectability’ capability of the formulations for substantially improved next-generation calcium phosphate cements. The state of knowledge and recent developments in carbon nanotube composites and coatings for hard and soft tissue engineering applications has been highlighted by Boccaccini and Gerhardt (2010). The development of self-setting CPCs has extended the application of calcium phosphates to injectable bone substitutes that can be shaped and moulded to fit irregular defects, and exhibit osteo-integrative properties comparable to or better than those of bulk calcium phosphates (Brown and Chow, 1985). A number of CPC formulations are currently available commercially. However, due to their limited compressive strength, CPCs are restricted primarily to non-stress-bearing applications. These include maxillofacial surgery, or the repair of cranial defects and dental fillings (Friedman et al., 1998; Schmitz et al., 1999; Bohner, 2000). Several strategies are being investigated to develop stronger CPC materials; of those, the development of CPC-based composites represents one particular and attractive approach (Fernandez et al., 1998; Watson et al., 1999). In this context, a variety of reinforcing elements ranging from particulate bioceramic inclusions to polymer fibres (Dos Santos et al., 2000) and carbon nanotubes (CNTs) (Wang et al., 2007a) have been considered. The modified interfacial bonding between bio-mineralized (i.e. pre-treated in simulated body fluid (SBF)) CNTs and CPCs accounted for the significantly improved mechanical properties in CPC/CNT composites (Wang et al., 2007a). The use of polyamide fibres also allowed a moderate increase in the compression strength in the CPC composites, where the observed reinforcement mechanism was related to the joining of the fibre to the matrix, and the appearance of cracks in a radial direction to the insertion cavity of the fibres (Dos Santos et al., 2000). Moreover, increasing mechanical properties with decreasing liquid-to-powder (L/P) ratio is normally associated with a lower overall porosity of CPCs (Fernandez et al., 1998; Watson et al., 1999). Since the detailed investigations on carbon nanotubes (CNTs) published by Iijima (1991), an increasing interest in the applications of CNTs has focused on their use as a reinforcement in different matrix materials because of their excellent mechanical properties (Treacy et al., 1996; Wong et al., 1997; Gojny et al., 2003). CNTs are also being investigated for biomedical applications (e.g. neural implants and tissue scaffolds) that utilize their high tensile strength, electrical conductivity and chemical stability (Mattson et al., 2000; Correa-Duarte et al., 2004; Gheith et al., 2005; Boccaccini and Gerhardt, 2010). One particular application of interest is the use of composite materials containing CNTs for bone tissue engineering, leading to novel matrices which can sustain bone cell growth and new bone tissue formation (MacDonald et al., 2005; Marrs et al., 2006; Meng et al., 2006; Shi et al., 2006; Leonor et al., 2009). CNTs have also been

4 (2011) 331–339

suggested as a reinforcement for inorganic bioactive coatings, e.g. hydroxyapatite (Casagrande et al., 2008; Kaya et al., 2008), or as a coating on three-dimensional (3D) scaffolds (Boccaccini et al., 2007). Another strategy to prepare improved calcium phosphate ceramics is their incorporation with proteins, such as bovine serum albumin (BSA) (Theodore, 1996; Zieba et al., 1996; Combes et al., 1999; Burke et al., 2000; Leonor et al., 2009). For example, the addition of low concentrations of BSA has been shown to enhance calcium phosphate crystal growth (being favourable for bone tissue mineralization), whereas higher concentrations inhibited calcium phosphate crystallization (Combes et al., 1999; Leonor et al., 2009). In the present paper, we report on novel CPC composites incorporated with three different types of multi-walled CNTs (MWCNTs) and BSA. The specific aims were to investigate the influence/role of (a) filler content (MWCNT weight percentage), (b) surface functionalization of MWCNT, and (c) BSA admixture on HA formation and the resultant mechanical properties of CPC matrices. The new composites are intended for applications as injectable bone substitutes.

2.

Materials and methods

2.1. Composition of the cement and preparation of materials Pristine MWCNTs, hydroxylated MWCNTs (MWCNTs–OH) and carboxylated MWCNTs (MWCNTs–COOH) with a diameter of 30–50 nm and length of ≈30 µm were provided by the Chinese Academy of Science. The purification processes were carried out according to the experimental protocol presented in our previous work (Hong et al., 2006; Hassan et al., 2007). β-tri-calcium phosphate, β-Ca3 (PO4 )2 , (β-TCP) and dicalcium phosphate anhydrous, CaHPO4 , (DCPA) were supplied by Sigma-Aldrich. The mean diameters of the β-TCP and DCPA particles were measured using a CILAS 1180 laser particle size analyzer and found to be 17.30 µm and 11.50 µm, respectively. Equimolar fractions of β-TCP and DCPA were mixed with deionized water, 0.25–1.0 wt% of MWCNTs and 15 wt% of BSA (supplied by Fluka) to produce the different CPC/MWCNTs/BSA composites to be investigated. The final solution volume was determined by the amount required to produce a workable paste, i.e., a viscous cement with an L/P ratio of 0.27 ml/g (see Section 3.6) was prepared. The paste was blended using a mechanical overhead stirrer at 30–50 rotations per minute until a homogeneous paste was obtained (approximately 1 h) and then firmly packed by manual spatulation into a cylindrical stainless steel mould with a diameter of 25 mm. The packed stainless steel mould was wrapped with a water-soaked wipe to prevent the sample from drying out and was then stored in a Gyro-Rocker Incubator (Model: S170) at 37 ◦ C and 97% humidity for 24 h. All experiments were carried out under controlled conditions at temperatures of 24–26 ◦ C and at a relative humidity of 50–60%.

2.2.

Mechanical testing

The compressive strength of the cylindrical specimens (nominal dimensions: Ø 25 mm, height: 10 mm) was

J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S

4 (2011) 331–339

333

25

determined using an Instron 3367 universal testing machine operating with a crosshead speed of 0.1 mm/min. Before mechanical testing, the sample ends were filed flush to ensure that the test specimens had plane-parallel surfaces.

Structural characterization

2.3.1.

Scanning electron microscopy (SEM)

In order to investigate the microstructure of the as-prepared hardened cements both on the surface and on the inner part of the composite, SEM examinations were performed using a 5 kV accelerating voltage in a Leo Supra 35VP-2458 microscope. All the samples were inspected at various magnifications directly in the pellet form, except for the pure CPC specimen, which was not strong enough to be directly inspected in its pellet form. Thus, this sample was tested in its crushed form, which was spread evenly on top of a doublesided carbon tape attached to an aluminium sample stub.

2.3.2.

10

0 CPC/0.25wt% MWCNTs/BSA

CPC/0.50wt% MWCNTs/BSA

CPC/0.75wt% MWCNTs/BSA

CPC/1.00wt% MWCNTs/BSA

Fig. 1 – The comparison of the compressive strength value of CPC/MWCNTs/BSA composites containing different percentages by weight of pristine MWCNTs. All composites have a BSA content of 15 wt%. Data are presented as mean ± 1 standard deviation (n = 2).

the injectability, which was determined by considering the percentage mass of the CPC paste extruded from the syringe divided by the original mass of the paste inside the syringe (Eq. (1)) (Burguera et al., 2008): Injectability =

Mass of extruded CPC Original mass of CPC inside the syringe × 100[%].

(1)

X-ray diffraction (XRD) analysis

XRD was used to determine the crystalline structure of the cement. The analysis was performed on a Siemens D5000 diffractometer using a diffraction angle 2θ in the range 25–70 degrees at a sweep rate of 0.04◦ /s. The qualitative analysis of different characteristic patterns of the materials investigated was achieved by comparing the peaks of the XRD spectrum with the standard diffraction patterns of specific compounds based on the International Centre for Diffraction Data (ICDD).

2.4.

15

5

Fourier transform infrared (FTIR) spectroscopy

A Fourier transform infrared (FTIR) spectrometer (PerkinElmer FTIR 2000) was employed to characterize the presence of specific surface functional groups in the composites. The FTIR spectra were recorded in the wave number interval 400–4000 cm−1 using the transmission mode. Since possible incomplete mixing of the starting materials could have led to inhomogeneous HA formation and in order to increase the measurement accuracy/sensitivity, ten different locations of the CPC sample were scanned and averaged for further analysis. The resolution of the spectrometer was 4 cm−1 . Before analysis, calibration of the spectrometer was performed by using polystyrene as a control sample. Then, the test sample was mixed with potassium bromide using a weight ratio of approximately 1:10. The mixture was ground to a fine, homogeneous powder, which was then poured into a mould. The powder was densified and compacted using a hydraulic press applying a pressure of ≈600 MPa to form thin pellets (thickness ≈100 µm). The thin and transparent pellet was then placed in the sample holder for analysis.

2.3.3.

Compressive Strength (MPa)

20

2.3.

Injectability tests

Injectability was qualitatively assessed and evaluated by extruding the paste through a disposable syringe. A 10 ml syringe with a diameter of 16 mm and a needle with an inner diameter of 2 mm was filled with CPC paste, which was then extruded from the syringe manually within a few seconds at relatively constant speed. The injectability test was carried out in two parts. The objective of the first part was to examine the L/P ratio required to produce a workable and injectable CPC paste, whilst the second part investigated

3.

Results and discussion

3.1.

Fabrication of CPCs

CPC, the biomaterial for bone repair considered in this study, was created by mixing β-TCP and DCPA with deionized water. After mixing, CPC becomes a paste-like material, which can suitably be shaped and injected according to the contours of a bone defect or to fill miniscule pores and cracks. The material is capable of self-setting at 37 ◦ C (normal body temperature, setting time ≈30 min) and it forms calciumdeficient hydroxyapatite (CDHA) as an end product.

3.2.

Compressive strength

In order to identify the minimum amount of pristine MWCNTs required for the substantial improvement of the mechanical properties of CPC/MWCNTs/BSA composite cement, we investigated their compressive strength as a

334

J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S

4 (2011) 331–339

25

Compressive Strength (MPa)

20

15

10

5

Nil*

0 CPC

CPC/MWCNTs

CPC/BSA CPC/MWCNTs/ CPC/MWCNTs- CPC/MWCNTsBSA OH/BSA COOH/BSA

Fig. 2 – The comparison of the compressive strength values of the materials investigated. The composites have a MWCNT loading of 0.5 wt% and a BSA content of 15 wt%. Data are presented as means ± 1 standard deviation (n = 2). *Note that the compressive strength of the CPC/BSA composite could not be measured because the composite was too weak to form the required shape for compressive test purposes.

function of filler content prior to the reinforcement effects of MWCNT surface functionalization. Fig. 1 shows the results of compressive strength tests on CPC/BSA composites incorporated with 0.25–1 wt% un-functionalized MWCNTs. Compared to the neat CPC (see Fig. 2), the addition of 0.25, 0.50, 0.75, and 1 wt% MWCNTs considerably improved the mechanical properties of the CPC/BSA matrix, resulting in compressive strength values ranging from 6.3 to 12.5 MPa (Fig. 1). For a similar CPC matrix and CNT loading, Wang et al. (2007a) found higher values for the CPC matrix (25 MPa) and for 0.5 wt%-containing bio-mineralized CNT composites (55 MPa). These results can be explained by the pre-treatment of the CNTs with simulated body fluid, as well as by different compressive strength test parameters (0.1 versus 0.5 mm/min), nanotube dimensions (20–30 nm versus 60–100 nm), and composite fabrication methods (manual spatulation versus hydraulic compaction with 700 kPa). In our study, CPC/BSA composites containing 0.5 wt% MWCNTs were further studied in terms of CNT surface functionalization and their effect on the mechanical properties. The results of the compressive strength tests for the neat CPC, CPC/MWCNTs, CPC/BSA, CPC/MWCNTs/BSA, CPC/MWCNTs–OH/BSA and CPC/MWCNTs–COOH/BSA composites are shown in Fig. 2. The addition of MWCNTs–OH gave the highest value of compressive strength (16 ± 4 MPa) in the composite, followed by un-functionalized MWCNTs (12 ± 3MPa) and MWCNTs–COOH (9 ± 4 MPa), as further discussed below. For human bone, compressive strengths in the range 2–12 MPa (trabecular bone) and 100–230 MPa (cortical bone) have been reported (Kokubo et al., 2003). The present results indicate that a further improvement (including a detailed investigation of the effect of MWCNT size and concentration) is necessary to increase the mechanical performance

Fig. 3 – Typical SEM images of (a) CPC/MWCNTs–OH/BSA composite, (b) CPC/MWCNTs–COOH/BSA composite (white arrows are considered to be hydroxyapatite crystals), which was confirmed by FTIR and XRD (Sections 3.3 and 3.4).

towards the level of cortical bone, required to fill bone defects at high load-bearing anatomical sites. This study showed that the concomitant admixture of BSA and MWCNTs to the pure CPC considerably increased the compressive strength of the CPC. In particular, the addition of BSA has contributed to a further increase (by a factor of more than 10) of the compressive strength of CPC/MWCNTs composites. This further improvement in the mechanical properties can be plausibly explained by considering that appropriate amounts of BSA are capable of promoting CPC crystal growth (see Figs. 3 and 4) (Boccaccini et al., 2007). In this research, physical blending was applied as a first approach to create a homogeneous dispersion of MWCNTs in the cement matrix, ensuring uniform properties throughout the CPC composite. However, this aim poses certain challenges due to the high aspect ratio (length to diameter) of MWCNTs, making them hard to mix. In particular, nonfunctionalized MWCNTs are known to have a tendency to agglomerate and form bundles (Cho et al., 2009). In addition, MWCNTs are insoluble in water and organic solvents. Nevertheless, hydroxyl group (–OH) functionalization has been confirmed to enable MWCNTs to disperse more easily in water and organic solvents, and at the same time the surface functionalization improves the interfacial bonding with

J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S

a

b

c

Fig. 4 – Typical SEM images of (a) CPC, (b) CPC/MWCNTs/ BSA composite, (c) CPC/MWCNTs–COOH/BSA composite (solid-line circle: longer needle-like HA crystals exhibiting so-called cauliflower morphology, dashed-line circle: shorter plate-like HA crystals) and (d) CPC/MWCNTs–OH/ BSA composite showing the compact microstructure of fine HA crystals. the matrix (White et al., 2007; Kaya et al., 2008). This effect had probably led to the highest compressive strength in the MWCNTs–OH composites investigated here.

335

In addition, since the MWCNTs are chemically inert and exhibit a highly hydrophobic surface nature, the un-functionalized MWCNTs are expected to have low chemical wettability for their dispersion in CPC matrices (Wang et al., 2007a). It is possible that lack of bonding at the interface and different HA crystal morphologies (as discussed in Section 3.3) have probably resulted in lower compressive strength values for CPC/MWCNTs/BSA and CPC/MWCNTs–COOH/BSA cements in comparison to hydroxylated CPC/MWCNTs–OH/BSA composites. The reasons for this phenomenon require more detailed investigation on the micromechanical mechanisms involved as well as larger sample sizes for reliable statistical analyses. From the point of view of the composite theory, it is well known that the transfer of stress and stiffness of the filler to the matrix depends on the quality of the interfacial bonding between the two phases (MWCNT filler and CPC matrix in the present case) (White et al., 2007), which in our study is influenced by the wettability and interfacial area between MWCNTs and CPC. Thus, strong MWCNT/CPC interfacial bonding is an essential factor and a significant condition for improving the mechanical properties of CPCs enabling load transfer across the MWCNT/CPC interface (Zhao and Gao, 2004; Wang et al., 2007a). However, it should be pointed out that, for cement composites, stronger interfaces can also lead to increased brittleness (White et al., 2007). In this context, relatively weak interfaces might be advantageous in composites with brittle matrices, as toughening mechanisms can be activated, thereby increasing the fracture toughness (Matthews and Rawlings, 2006).

3.3.

d

4 (2011) 331–339

Scanning electron microscopy characterization

Different morphologies were observed in the HA crystal structures of CPC, CPC/MWCNTs/BSA, CPC/MWCNTs–OH/BSA, and CPC/MWCNTs–COOH/BSA (see Figs. 3 and 4). The SEM image of CPC/MWCNTs–OH/BSA composite presented in Fig. 3(a) shows that HA crystals grew on MWCNTs–OH, whereas only a few HA crystals grew on CPC/MWCNTs–COOH/BSA composites, as illustrated by arrows in Fig. 3(b). Similar morphologies of HA crystals of Fig. 3(a–b) and Fig. 4(a–d) were obtained by Ratier et al. (2004), Carey et al. (2005) and Xu et al. (2006, 2008). However, it is still unknown and unclear how the microstructure of cement interfaces and the HA morphology affect the mechanical properties of CPC-based composites. Despite the different HA morphologies (Figs. 3 and 4), relatively similar compressive strength values were found for all the CPC/MWCNTs/BSA composites tested. This observation indicates that the addition of BSA (Fig. 2) caused the improved mechanical strength of the composites and masked the enhancement of the composites normally associated with surface functionalization of CNTs. This finding also suggests that, in CPC composites, BSA addition has a much higher enhancement effect than CNT surface functionalization. In order to exploit the unique mechanical properties of CNTs in composites, the CNTs are usually functionalized to improve and enhance their reactivity, solubility, wettability and interfacial bonding (Niyogi et al., 2002; White et al., 2007; Kaya et al., 2008; Cho et al., 2009). Interestingly, in this study the addition of carboxylated carbon nanotubes (MWCNTs–COOH)

336

J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S

was not particularly effective at improving the compressive strength of CPC/BSA composites. Zhao et al. (2005) showed that –COOH groups present on the surface of SWCNTs were relatively inefficient at nucleating and growing HA, probably due to the impaired capability of attracting both Ca2+ and PO3− ions for nucleating and growing HA crystals. This 4 effect can explain the slightly lower compressive strength for COOH-terminated MWCNTs. On the other hand, the effective attraction of both Ca2+ and PO3− ions by specific 4 functional groups on MWCNTs is expected to enhance mechanical properties (compressive strength), as observed for hydroxylated MWCNTs. The above explanation and in particular the HA crystal morphology (Figs. 3 and 4) of CPC/MWCNTs/BSA composites can be used to plausibly interpret/explain the differences in compressive strength of cements containing hydroxylated and carboxylated MWCNTs, as further discussed below. From Fig. 4(a), one can observe that the HA crystals grown in CPC were shorter, wider, flatter and appear to be more plate-like than needle shaped and less entangled. It is possible that shorter, flatter plate-like HA crystals reduce the compressive strength, as observed for the pure CPC compared to the different CPC composites. For example, the CPC/MWCNTs/BSA composite showed clusters of HA crystals, orientated in the same direction (Fig. 4(b)). The HA crystals formed were thinner and longer than in CPC and existed in a medium size needle-like form. Longer crystals with higher aspect ratios would increase the mechanical properties (Matthews and Rawlings, 2006). However, the interfacial bonding between un-functionalized MWCNTs and the CPC matrix might be lower, as compared to composites with MWCNTs–OH, as previously discussed in relation to the compressive strength results (Section 3.2 and Fig. 2). Fig. 4(c) shows the microstructure of the CPC/MWCNTs–COOH/BSA composite composed of different structures of HA crystals. There were some regions with longer needle-like HA crystals, illustrated by the solid line circle, as well as some regions with shorter plate-like HA crystals demonstrated by the dashed-line circle in Fig. 4(c). These different crystal formations give rise to an inhomogeneous morphology, possibly inducing porosity, which can explain the lower compressive strength of the carboxylated MWCNTs/BSA calcium phosphate composites. On the other hand, the SEM image in Fig. 4(d) shows that in the CPC/MWCNTs–OH/BSA composite, the HA crystals are homogeneously distributed in all directions. The scale of the microstructure is finer than in the previous case, forming a composite of wellinterconnected HA crystals with a high surface/contact area, and the HA crystals appear to be finer and well packed. We hypothesize that this particular microstructure has led to an increased compressive strength of these composites (Fig. 2) and might be also favourable/beneficial to withstand shear stresses. As already mentioned, BSA has the capability of enhancing calcium phosphate crystal growth (Combes et al., 1999). At low concentrations (<10 g/l), BSA has been hypothesized to stabilize nuclei and promote growth of octa-calcium phosphate crystals, while at higher concentrations crystal growth seems to be impeded by high BSA coverage. Although the net charge on BSA at neutral pH is −17 mV, the protein

4 (2011) 331–339

contains both positively and negatively charged residues (Theodore, 1996). The arrangement of these charges on the protein, as well as the complementarities between the charged groups on the protein and the growing apatite surfaces, may influence the crystal growth behaviour and also lead to more cohesive cements for higher BSA contents (Zieba et al., 1996; Burke et al., 2000).

3.4.

Fourier transform infrared analysis

Fig. 5(a–c) shows the FTIR spectra of the CPC/MWCNTs/BSA, CPC/MWCNTs–OH/BSA and CPC/MWCNTs–COOH/BSA composites, respectively. The spectra show absorption bands at 3297–3302 cm−1 which correspond to the strong characteristic peak of the stretching mode of the hydroxyl group (–OH) (Mahabole et al., 2005; Janusz et al., 2008). The characteristic bending mode of intercalated H2 O can be observed at 1655–1656 cm−1 (Mahabole et al., 2005). A very strong, broad phosphate band derived from the P–O asymmetric stretching mode (ν3 ) of the PO3− group was identified in the region 4 943–1128 cm−1 , indicating a deviation of phosphate ions from their ideal tetrahedral structure (Mahabole et al., 2005; Janusz et al., 2008). The absorption bands appearing at about 400–600 cm−1 can be attributed to the triple (ν4 )-degenerated 409, 417, 548, 551, 554, 586, 587 and 604 cm−1 and double (ν2 )-degenerated fundamental bending modes of the PO3− functional group 4 (Mahabole et al., 2005; Janusz et al., 2008). The bands observed at 1543 cm−1 (ν3 mode), 1547 cm−1 (ν3 mode), and 943 cm−1 (ν2 mode) were assigned to the CO2− group (Komath et al., 3 2000). Furthermore, the characteristic bands at 943 cm−1 indicate the presence of HPO2− in the crystal lattice (Tsuchiya 4 et al., 2006). As a result, all the bands discussed above and also their positions in the FTIR spectra confirm the formation of apatite in composites fabricated with both un-functionalized MWCNTs and functionalized MWCNTs. The FTIR results further confirm the observations made on SEM images discussed above. By comparing the results of the different FTIR spectra, it was found that the absorption bands of CPC/MWCNTs/BSA (Fig. 5(a)) and bands of CPC/MWCNTs–OH/BSA (Fig. 5(b)) give rise to sharper peaks at wave numbers of 550, 587, 1065, 1128 and 1655 cm−1 , compared to CPC/MWCNTs–COOH/BSA (Fig. 5(c)). For the CPC/MWCNTs/BSA and CPC/MWCNTs–OH/BSA samples, an extra band was detected in the region of 780 cm−1 , which was not observed in the CPC/MWCNTs –COOH/BSA composite. The origin of this extra band is unclear, but it might arise from the different extents of apatite formation on CPC/MWCNTs/BSA and CPC/MWCNTs–OH/BSA composites compared to CPC/MWCNTs–COOH/BSA composites.

3.5.

X-ray diffraction analysis

The XRD patterns of the CPC/MWCNTs/BSA, CPC/MWCNTs –OH/BSA and CPC/MWCNTs–COOH/BSA composites are shown in Fig. 6(a–c). Diffraction peaks corresponding to HA crystalline phase were detected at 2θ angles of 26◦ , 29◦ , 32◦ , 40◦ and 53◦ . It is therefore evident that it is possible

J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S

337

4 (2011) 331–339

v3PO43v3PO43-

-OH v3CO32-

-OH

v2CO32-

v4PO43-

HPO42-

(a)

(b)

%T

(c)

4000.0

3600

3200

2800

2400

2000

1800

1600

1400

1200

1000

800

600

400.0

cm-1

Fig. 5 – An FTIR spectrum of the investigated calcium phosphate-based ceramic cements (a) CPC/MWCNTs/BSA, (b) CPC/MWCNTs–OH/BSA and (c) CPC/MWCNTs–COOH/BSA (see the discussion in Section 3.4).

to obtain self-setting injectable HA by mixing β-TCP and DCPA with deionized water. The sharp and narrow diffraction peaks observed in the regions of relevance to HA suggest that the HA formed is crystalline, which can be correlated with the crystal morphology observed by SEM (Ratier et al., 2004; Carey et al., 2005; Xu et al., 2006, 2008) (Fig. 4(a–d)). However, from Fig. 6, the XRD analysis also revealed two extra phases of the starting materials corresponding to βTCP and DCPA, indicating that the reaction to form HA is not complete, which might be associated with factors such as the liquid to powder ratio, the hydration time and the hydration environment. As a whole, the XRD, SEM and FTIR results showed that the investigated CPC composites developed a crystalline HA phase, which is in its chemical and crystallographic composition similar to the mineral phase of bone (Suchanek and Yoshimura, 1998).

3.6.

Injectability test

The injectability test was performed only with the CPC/ MWCNTs–OH/BSA composite, because this composite showed the most promising result in terms of compressive strength. The desired physical condition of workable CPC/MWCNTs –OH/BSA composite paste was found at an L/P ratio of 0.27 ml/g, resulting in an injectability of 97%, i.e., 97% of the CPC paste could be extruded. It is important to note here that the maximum percentage of cement paste extruded can never

Fig. 6 – X-ray diffraction patterns of the investigated samples: (a) CPC/MWCNTs/BSA, (b) CPC/MWCNTs–OH/BSA and (c) CPC/MWCNTs–COOH/BSA. The composites have an MWCNT loading of 0.5 wt% and a BSA content of 15 wt%.

338

J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S

achieve 100%, due to small amounts of residual cement paste inside the syringe. It is clear that the injectability of cement pastes can be influenced by varying the L/P ratio. The injectability of cement pastes with an L/P ratio <0.25 ml/g was not tested because the specimen was not workable (too viscous). The injectability of cement pastes with an L/P ratio >0.28 ml/g was not tested because the resulting cement paste was too liquid. For example, Bohner and Baroud (2005) suggested that a well-injectable cement paste should have the capacity to stay homogeneous during injection, independently of the injection force. They suggested that this approach can be achieved by increasing the cement’s L/P ratio. As a result, the ability of cement paste to harden in an aqueous condition will be reduced because the viscosity of the cement paste is reduced at the same time. This reduced stability will cause a total degradation of the cement paste. Summarizing, an L/P ratio of CPC/MWCNTs–OH/BSA composite paste of 0.27 ml/g yielded mechanically strong and injectable CPCs with an injectability of 97%. This material is thus suitable for bone repair applications as an injectable bone substitute.

4.

Conclusion

The present work demonstrated the possibility of developing high compressive strength CPCs by reinforcement with MWCNTs and BSA for use as injectable bone substitute. Drawing on the results from the compressive strength tests, the CPC/MWCNTs–OH/BSA composite exhibited substantially improved compressive strength (≈16 MPa) compared to pure cement (≈1 MPa). Of all MWCNTs studied, functionalized MWCNTs–OH were found to be the most effective to increase the compressive strength of CPC. It was suggested that hydroxyl functional groups on the surface of MWCNTs improved the reactivity and wettability of MWCNTs leading to strong interfacial bonding. In addition, the effective attraction of both Ca2+ and PO3− by the functional groups 4 of MWCNTs–OH is expected to promote the nucleation and growth of HA crystals. The XRD, SEM and FTIR analyses confirmed the formation of crystalline HA during the synthesis of CPC. SEM observations demonstrated that the addition of MWCNTs modifies the morphology of HA crystallites. The HA crystals in CPC/MWCNTs–OH/BSA composites were fine, homogeneously grown and distributed in all directions, forming a well-packed composite microstructure, which resulted in the highest compressive strength. A simple test of injectability on the CPC/MWCNTs–OH/BSA composite demonstrated a high percentage of extrusion (97%), being thus easily workable and clinically applicable as an in situ hardening cement. The promising results presented here in terms of compression strength enhancement must be confirmed with larger sample sizes and a comprehensive study of the influence of CNT surface modification and CPC pre-treatment (e.g. upon soaking in SBF) on the mechanical properties (fracture toughness, elastic modulus, shear strength and elongation at break) to confirm the suitability of the composites for the intended application as

4 (2011) 331–339

bone substitutes. Due to concerns about the biocompatibility and toxicity of CNTs for their use as biomaterials (Boccaccini and Gerhardt, 2010), further research should also focus on in vitro cell biological investigations and on the in vivo performance of the novel CPC containing MWCNTs and BSA developed here. Future experimental work should also include investigations on the optimization of blending properties/characteristics and on investigating the setting time of the present CPC.

Acknowledgements The financial support provided by the British Council through the UK’s Prime Minister’s Initiative for International Education (PMI2) Connect scheme is gratefully acknowledged. Kean Khoon Chew and Kah Ling Low also acknowledge a USM Fellowship for the support for their studies.

REFERENCES

Boccaccini, A.R., Chicatun, F., Cho, J., Bretcanu, O., Roether, J.A., Novak, S., Chen, Q.Z., 2007. Carbon nanotube coatings on bioglass-based tissue engineering scaffolds. Adv. Funct. Mater. 17 (15), 2815–2822. Boccaccini, A.R., Gerhardt, L.C., 2010. Carbon nanotube composite scaffolds and coatings for tissue engineering applications. Key Eng. Mater.: Adv. Bioceram. Med. Appl. 441, 31–52. Bohner, M., 2000. Calcium orthophosphates in medicine: from ceramics to calcium phosphate cements. Injury 31, 37–47. Bohner, M., Baroud, G., 2005. Injectability of calcium phosphate pastes. Biomaterials 26 (13), 1553–1563. Brown, W.E., Chow, L.C., 1985. Dental restorative cement pastes. American Dental Association Health Foundation. Washington, DC, USA. US patent no: 4518430. Burguera, E.F., Xu, H.H.K., Sun, L.M., 2008. Injectable calcium phosphate cement: effects of powder-to-liquid ratio and needle size. J. Biomed. Mater. Res. Part B: Appl. Biomater. 84B (2), 493–502. Burke, E.M., Guo, Y., Colon, L., Rahima, M., Veis, A., Nancollas, G.H., 2000. Influence of polyaspartic acid and phosphophoryn on octacalcium phosphate growth kinetics. Colloid. Surf.: Biointerfaces 17 (1), 49–57. Carey, L.E., Xu, H.H., Simon Jr., C.G., Takagi, S., Chow, L.C., 2005. Premixed rapid-setting calcium phosphate composites for bone repair. Biomaterials 26 (24), 5002–5014. Casagrande, T., Lawson, G., Li, H., Wei, J., Adronov, A., Zhitomirsky, I., 2008. Electrodeposition of composite materials containing functionalized carbon nanotubes. Mater. Chem. Phys. 111 (1), 42–49. Cho, J., Boccaccini, A.R., Shaffer, M.S.P., 2009. Ceramic matrix composites containing carbon nanotubes. J. Mater. Sci. 44 (8), 1934–1951. Chow, L.C., 2009. Next generation calcium phosphate-based biomaterials. Dent. Mater. J. 28 (1), 1–10. Combes, C., Rey, C., Freche, M., 1999. In vitro crystallization of octacalcium phosphate on type I collagen: influence of serum albumin. J. Mater. Sci. Mater. Med. 10 (3), 153–160. Correa-Duarte, M.A., Wagner, N., Rojas-Chapana, J., Morsczeck, C., Thie, M., Giersig, M., 2004. Fabrication and biocompatibility of carbon nanotube-based 3D networks as scaffolds for cell seeding and growth. Nano Lett. 4 (11), 2233–2236.

J O U R N A L O F T H E M E C H A N I C A L B E H AV I O R O F B I O M E D I C A L M AT E R I A L S

Costantino, P.D., Friedman, C.D., Jones, K., Chow, L.C., Sisson Sr., 1992. Experimental hydroxyapatite cement cranioplasty. Plast. Reconstr. Surgr. 90 (2), 174–185. Dos Santos, L.A., De Oliveira, L.C., Rigo, E.C.D., Carrodeguas, R.G., Boschi, A.O., De Arruda, A.C.F., 2000. Fiber reinforced calcium phosphate cement. Artif. Organs 24 (3), 212–216. Fernandez, E., Gil, F.J., Best, S.M., Ginebra, M.P., Driessens, F.C., Planell, J.A., 1998. Improvement of the mechanical properties of new calcium phosphate bone cements in the CaHPO4 -alpha-Ca3 (PO4 )2 system: compressive strength and microstructural development. J. Biomed. Mater. Res. Part A. 41 (4), 560–567. Friedman, C.D., Costantino, P.D., Takagi, S., Chow, L.C., 1998. BoneSource (TM) hydroxyapatite cement: a novel biomaterial for craniofacial skeletal tissue engineering and reconstruction. J. Biomed. Mater. Res. Part B: Appl. Biomater. 43 (4), 428–432. Gheith, M.K., Sinani, V.A., Wicksted, J.P., Matts, R.L., Kotov, N.A., 2005. Single-walled carbon nanotube polyelectrolyte multilayers and freestanding films as a biocompatible platform for neuroprosthetic implants. Adv. Mater. 17 (22), 2663–2670. Gojny, F.H., Nastalczyk, J., Roslaniec, Z., Schulte, K., 2003. Surface modified multi-walled carbon nanotubes in CNT/epoxycomposites. Chem. Phys. Lett. 370 (5–6), 820–824. Hassan, N.H.A., Mohamed, A.R., Zein, S.H.S., 2007. Study of hydrogen storage by carbonaceous material at room temperature. Diamond Relat. Mater. 16 (8), 1517–1523. Hong, K.B., Ismail, A.A.B., Mahayuddin, M.E.M., Mohamed, A.R., Zein, S.H.S., 2006. Production of high purity multi-walled carbon nanotubes from catalytic decomposition of methane. J. Nat. Gas. Chem. 15 (4), 266–270. Iijima, S., 1991. Helical microtubules of graphitic carbon. Nature 354 (6348), 56–58. Janusz, W., Skwarek, E., Pasieczna-Patkowska, S., Slosarczyk, A., Paszkiewicz, Z., Rapacz-Kmita, A., 2008. A study of surface properties of calcium phosphate by means of photoacoustic spectroscopy (FT-IR/PAS), potentiometric titration and electrophoretic measurements. Eur. Phys. J. — Special Topics 154, 329–333. Kaya, C., Singh, I., Boccaccini, A.R., 2008. Multi-walled carbon nanotube-reinforced hydroxyapatite layers on Ti6 Al4 V medical implants by electrophoretic deposition (EPD). Adv. Eng. Mater. 10 (1–2), 131–138. Kokubo, T., Kim, H.M., Kawashita, M., 2003. Novel bioactive materials with different mechanical properties. Biomaterials 24 (13), 2161–2175. Komath, M., Varma, H.K., Sivakumar, R., 2000. On the development of an apatitic calcium phosphate bone cement. Bull. Mater. Sci. 23 (2), 135–140. Leonor, I.B., Alves, C.M., Azevedo, H.S., Reis, R.L., 2009. Effects of protein incorporation on calcium phosphate coating. Mater. Sci. Eng. C—Biomimetic Supramol. Syst. 29 (3), 913–918. MacDonald, R.A., Laurenzi, B.F., Viswanathan, G., Ajayan, P.M., Stegemann, J.P., 2005. Collagen–carbon nanotube composite materials as scaffolds in tissue engineering. J. Biomed. Mater. Res. Part A 74A (3), 489–496. Mahabole, M.P., Aiyer, R.C., Ramakrishna, C.V., Sreedhar, B., Khairnar, R.S., 2005. Synthesis, characterization and gas sensing property of hydroxyapatite ceramic. Bull. Mater. Sci. 28 (6), 535–545. Marrs, B., Andrews, R., Rantell, T., Pienkowski, D., 2006. Augmentation of acrylic bone cement with multiwall carbon nanotubes. J. Biomed. Mater. Res. Part A 77 (2), 269–276. Matthews, F.L., Rawlings, R.D., 2006. Fracture mechanics and toughening mechanisms. In: Matthews, F.L., Rawlings, R.D. (Eds.), Composite Materials: Engineering and Science. Woodhead Publishing Limited, CRC Press, Cambridge, England, pp. 326–362.

4 (2011) 331–339

339

Mattson, M.P., Haddon, R.C., Rao, A.M., 2000. Molecular functionalization of carbon nanotubes and use as substrates for neuronal growth. J. Mol. Neurosci. 14 (3), 175–182. Meng, J., Song, L., Meng, J., Kong, H., Zhu, G., Wang, C., Xu, L., Xie, S., Xu, H., 2006. Using single-walled carbon nanotubes nonwoven films as scaffolds to enhance long-term cell proliferation in vitro. J. Biomed. Mater. Res. Part A 79A (2), 298–306. Mermelstein, L.E., McLain, R.F., Yerby, S.A., 1998. Reinforcement of thoracolumbar burst fractures with calcium phosphate cement — a biomechanical study. Spine 23 (6), 664–670. Niyogi, S., Hamon, M.A., Hu, H., Zhao, B., Bhowmik, P., Sen, R., Itkis, M.E., Haddon, R.C., 2002. Chemistry of single-walled carbon nanotubes. Acc. Chem. Res. 35 (12), 1105–1113. Ratier, A., Freche, M., Lacout, J.L., Rodriguez, F., 2004. Behaviour of an injectable calcium phosphate cement with added tetracycline. Int. J. Pharm. 274 (1–2), 261–268. Schmitz, J.P., Hollinger, J.O., Milam, S.B., 1999. Reconstruction of bone using calcium phosphate bone cements: a critical review. J. Oral. Maxillofac. Surg. 57 (9), 1122–1126. Shi, X.F., Hudson, J.L., Spicer, P.S., Tour, J.M., Krishnamoorti, R.M., Mikos, A.G., 2006. Injectable nanocomposites of single-walled carbon nanotubes and biodegradable polymers for bone tissue engineering. Biomacromolecules 7 (7), 2237–2242. Suchanek, W., Yoshimura, M., 1998. Processing and properties of hydroxyapatite-based biomaterials for use as hard tissue replacement implants. J. Mater. Res. 13 (1), 94–117. Theodore Jr., Peters, 1996. All About Albumin: Biochemistry, Genetics, and Medical Applications. Academic Press, CA, San Diego. Treacy, M.M.J., Ebbesen, T.W., Gibson, J.M., 1996. Exceptionally high Young’s modulus observed for individual carbon nanotubes. Nature 381 (6584), 678–680. Tsuchiya, H., Macak, J.M., Muller, L., Kunze, J., Muller, F., Greil, P., Virtanen, S., Schmuki, P., 2006. Hydroxyapatite growth on anodic TiO2 nanotubes. J. Biomed. Mater. Res. Part A. 77A (3), 534–541. Wang, X.P., Ye, J.D., Wang, Y.J., Chen, L., 2007a. Reinforcement of calcium phosphate cement by bio-mineralized carbon nanotube. J. Am. Ceram. Soc. 90 (3), 962–964. Wang, X.P., Ye, J.D., Wang, Y.J., Wu, X.P., Bo, B., 2007b. Control of crystallinity of hydrated products in a calcium phosphate bone cement. J. Biomed. Mater. Res. Part A 81 (4), 781–790. Watson, K.E., Tenhuisen, K.S., Brown, P.W., 1999. The formation of hydroxyapatite–calcium polyacrylate composites. J. Mater. Sci. Mater. Med. 10 (4), 205–213. White, A.A., Best, S.M., Kinloch, I.A., 2007. Hydroxyapatite–carbon nanotube composites for biomedical applications: a review. Int. J. Appl. Ceram. Tech. 4 (1), 1–13. Wong, E.W., Sheehan, P.E., Lieber, C.M., 1997. Nanobeam mechanics: elasticity, strength, and toughness of nanorods and nanotubes. Science 277 (5334), 1971–1975. Xu, H.H.K., Weir, M.D., Burguera, E.F., Fraser, A.M., 2006. Injectable and macroporous calcium phosphate cement scaffold. Biomaterials 27 (24), 4279–4287. Xu, H.H.K., Weir, M.D., Simon, C.G., 2008. Injectable and strong nano-apatite scaffolds for cell/growth factor delivery and bone regeneration. Dent. Mater. 24 (9), 1212–1222. Zhao, L.P., Gao, L., 2004. Novel in situ synthesis of MWNTs–hydroxyapatite composites. Carbon 42 (2), 423–426. Zhao, B., Hu, H., Mandal, S.K., Haddon, R.C., 2005. A bone mimic based on the self-assembly of hydroxyapatite on chemically functionalized single-walled carbon nanotubes. Chem. Mater. 17 (12), 3235–3241. Zieba, A., Sethurman, G., Perez, F., Nancollas, G.H., Cameron, D., 1996. Influence of organic phosphonates on hydroxyapatite crystal growth kinetics. Langmuir 12 (11), 2853–2858.