H.W. Tong and M.Wang*
Department of Mechanical Engineering, Faculty of Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong
Received date: November 11, 2010; Accepted date: December 02, 2010
Visit for more related articles at Bioceramics Development and Applications
Due to osteoconductivity of hydroxyapatite (HA), HA/polymer composite scaffolds have been investigated for bone tissue engineering by various groups. In scaffold fabrication, electrospinning of ultrafine fibrous scaffolds incorporated with substantial amounts of evenly distributed HA nanoparticles is still a challenge. The problem of nanoparticle agglomeration not only reduces the electrospinning efficiency but also undermines the homogeneity of the distribution of the HA nanoparticles along the fibers. In this study, an approach using ultrasonification was developed so as to electrospin fibrous nanocomposite scaffolds having the carbonated HA (CHA) nanosphere content up to about 15 wt% with minimal agglomeration of CHA nanospheres. SEM and EDX results showed an even distribution of CHA nanospheres while FTIR results further confirmed the presence of CHA within the electrospun fibers. The nanocomposite fibrous scaffolds fabricated through this route could be used in further investigations for bone tissue engineering applications.
carbonated hydroxyapatite (CHA); poly(hydroxybutyrate-co-hydroxyvalerate) (PHBV); electrospinning; ultrasonification; nanocomposite; scaffold; tissue engineering
Due to osteoconductivity of bioactive ceramics such as hydroxyapatite (HA), electrospinning of composite fibers with incorporated HA nanoparticles has been investigated for bone tissue engineering [1,4]. However, the agglomeration of the nanoparticles in the polymer solution is a common problem during the electrospinning process. To date, it is still a challenge to incorporate substantial amounts of HA particles into electrospun fibers because fine (micro- or nano-size) HA particles tend to agglomerate together . Jose et al. tried to electrospin poly(lacticco- glycolic acid) (PLGA) fibers with different amounts of HA nanoparticles (1 to 20wt%) and observed severe agglomeration of HA nanoparticles . The problem of HA particle agglomeration not only reduces the electrospinning efficiency but also undermines the homogenous distribution of HA nanoparticles along the fibers. Some researchers tried to mix surfactants with the polymer solution before electrospinning in order to prevent the agglomeration of HA particles . But the use of additives could cause biocompatibility problems due to possible cytotoxicity of the additives. Therefore, it is ideal that the use of additives is kept to minimum or totally avoided when constructing tissue engineering scaffolds. In the present study, techniques to uniformly disperse substantial amounts of carbonated HA (CHA) nanospheres in electrospun poly(hydroxybutyrateco- hydroxyvalerate) (PHBV) fibers without the use of additives were investigated. The CHA nanospheres were incorporated into electrospun fibers through three different routes: (I) traditional mixing, (II) ultrasonification before electrospinning, and (III) ultrasonification during electrospinning. It was believed that the ultrasonificating power could break down nanosphere agglomerates, which in turn could improve the fiber formation process and allow the formation of nanocomposite fibers with a homogenous distribution of the CHA nanospheres in fibers.
PHBV with 5mol% of hydroxyvalerate (Sigma-Aldrich, USA) was dissolved in chloroform (analytical grade) at a polymer concentration of 15% w/v. The CHA nanospheres used for nanocomposite fibers were synthesized inhouse through a nanoemulsion process . PHBV fibers without CHA were electrospun using the facility described previously . CHA/PHBV nanocomposite fibers with CHA content of 5, 10 and 15 wt% were fabricated by electrospinning via three routes (Figure 1). In Route I, the CHA nanospheres were mixed with the polymer solution by a magnetic stirrer and the mixture was then subjected to the electrospinning process. This was a two-step process for producing composite fibers (Figure 1(a)). In Route II, the mixture consisting of the CHA nanospheres and polymer solution was sonicated by ultrasound energy for 30min before electrospinning. This was also a two-step process (Figure 1(b)). In Route III, ultrasound energy was applied directly to the mixture by an ultrasonic cell disruptor during the electrospinning process. This was considered to be a one-step process (Figure 1(c)). The fiber formation efficiencies were compared among these approaches and the CHA/PHBV fibers were analyzed using SEM, EDX, TGA and FTIR.
Figure 2 shows the CHA/PHBV nanocomposite fibers electrospun via Route I. The distribution of CHA nanospheres was not homogenous and agglomerated CHA clusters were frequently observed. The needle was easily blocked during electrospinning when the CHA content was above 10 wt%, resulting in insufficient fiber formation. Route II allowed the formation of nanocomposite fibers with CHA content up to 15 wt%. Figure 3 shows the EDX mapping of CHA (red spots in figures) for CHA/PHBV fibers fabricated via Route II. Compared with the Route I, the CHA nanospheres distribution was relatively homogenous. CHA was found in limited regions only when its content was 5 wt% while some agglomeration problems occurred when its content became 15 wt%. Route III caused rapid evaporation of the solvent and hence solidification of the polymer solution during electrospinning, which ended the electrospinning process within a very short period of time.
In Route I, CHA nanospheres could easily agglomerate during the traditional electrospinning process (Figure 1(a)), resulting in the uneven distribution of the CHA nanospheres within the electrospun fibers. The distribution of red spots in Figure 3(b) indicated that the CHA nanospheres were distributed around the whole fibrous scaffold rather than agglomerated into limited area. In Route II (Figure 1(b)), ultrasonic power was applied to the CHA/PHBV solution during solution preparation so that the CHA nanospheres were dispersed sufficiently in the polymer solution. The solution containing dispersed CHA nanospheres was immediately used for electrospinning. The CHA particles were dispersed reasonably well in resultant composite fibers. Even though CHA nanospheres were dispersed effectively during solution preparation, agglomeration could still occur during the electrospinning process no matter how quickly the electrospinning process took place after solution preparation. To eliminate the time gap between these two processes, Route III (Figure 1(c)) was used. But the rapid evaporation of solvent in Route III totally undermined electrospinning of nanocomposite fibers. It appeared that Route II is the only promising approach for fabricating usable nanocomposite fibrous scaffolds among these three routes.
TGA was conducted for composite fibers. The final weight percentage in a TGA curve was the actual amount of CHA incorporated in the fibrous scaffolds. It was found that the actual CHA content was generally less than the corresponding nominal value, suggesting the loss of CHA nanospheres during the fiber formation process. The deviation for 15 wt% CHA composite fibers was higher than the deviations for either 5 or 10wt% CHA fibers, implying that it was difficult to encapsulate CHA nanospheres in fibers at high CHA content.
FTIR spectra of a PHBV fibrous scaffold and a CHA/PHBV nanocomposite fibrous scaffold are shown in Figure 4. The two spectra exhibited several common peaks due to the matrix polymer of PHBV. These absorption peaks were the asymmetric stretching vibration of C-CH2 at 2918 cm−1, the symmetric stretching vibration of C-CH2 at 2849 cm−1, the carbonyl peak at 1651 cm−1, the methyl peak at 1379 cm−1, and the stretching vibration of C-O at 1219 cm−1. Apart from these common peaks, additional peaks were found in the spectrum for CHA/PHBV. This spectrum also exhibited peaks for the OH− group at 3289 cm−1, theCO32− group at 1717 cm−1 and 1455 cm−1, and the PO43− group at 1054 cm−1. The peaks for the OH− group and the PO43− group confirmed the existence of CHA in the nanocomposite scaffold while the peak for the CO32− group confirmed that the CHA nanospheres were carbonated.
CHA nanospheres may be incorporated into electrospun PHBV fibers through three routes: traditional mixing of CHA nanospheres with the PHBV solution for electrospinning solution preparation (Route I), application of ultrasonic power to break down CHA nanosphere agglomerates during solution preparation (Route II), and application of ultrasonic power to solution during electrospinning (Route III). Route II appeared to be the best route for fabricating usable CHA/PHBV nanocomposite fibrous scaffolds. Through Route II, fibrous CHA/PHBV nanocomposite scaffolds with a homogenous distribution of CHA nanospheres could be electrospun.
This work was supported by The University of Hong Kong through the Nano-biotechnology Strategic Research Theme and by the Research Grants Council of Hong Kong through a GRF grant (HKU 7182/05E). H. W. Tong thanks The University of Hong Kong for providing him with a research studentship.