Acta Biomater. doi: 10.1016/ j.actbio.2015.12.037 In vitro degradation of biodegradable polylactic acid/magnesium composites: relevance of Mg particle shape
S.C. Cifuentes1, R. Gavilán1, M. Lieblich1,*, R. Benavente2, J.L. González-Carrasco1,3 1
Centro Nacional de Investigaciones Metalúrgicas, CENIM-CSIC, Avda. Gregorio del Amo
8, 28040 Madrid, Spain 2
Instituto de Ciencia y Tecnología de Polímeros ICTP-CSIC, Juan de la Cierva 3, 28006
Madrid, Spain 3
Centro de Investigación Biomédica en Red de Bioingeniería, Biomateriales y Nanomedicina
CIBER-BBN, Spain
[email protected],
[email protected],
[email protected] *
[email protected], tel. +34915538900 Abstract Absorbable medical devices must be developed in order to have an appropriate degradation rate in agreement with the healing rate of bone in the implantation site. In this work, biodegradable composites formed by a polylactic acid matrix reinforced with 10% wt. magnesium microparticles were processed and their in vitro degradation investigated during 28 days. A joint analysis of the amount of H2 released, the changes in pH in buffered (PBS) and non-buffered media (distilled water), the variations in mass, microstructure and the mechanical performance of the specimens was developed. The main aim was to elucidate the relevance of Mg particles shape on tailoring the degradation kinetics of these novel composites. The results show that the shape of the Mg reinforcing particles plays a crucial role in the degradation rate of PLA/Mg composites, with spherical particles promoting a
1
lower degradation rate than irregular particles. This fact is only partially due to the smaller surface area to volume ratio of the spherical particles. Irregular particles promote a faster formation of cracks and, therefore, an increasingly faster degradation of the polymeric matrix. In every case, the amount of H2 released by the composites was well below that released by monolithic Mg. The pH of PBS during degradation remained always within 7.2 and 7.4. PLA/Mg reinforced with spherical particles retains more than 90% of its mechanical properties after 7 days of immersion and more than 60% after 28 days. Keywords: polylactic acid, Mg, hydrogen release, mechanical properties, in vitro biodegradation
1. INTRODUCTION The increasing demand for orthopaedic implants is the driving force to seek new strategies to decrease costs and simultaneously improve patients comfort as well as simplify surgical procedures. This is especially critical for temporary implants, where current metallic (stainless steel, titanium and its alloys) prostheses must be removed by a second operation procedure, being this a clear inconvenience, apart from risky and costly. Bioabsorbable and biocompatible materials suitable to be replaced by mature bone without transient loss of mechanical support are the logical alternative to avoid a second operation. Nowadays most of the available bioabsorbable osteosynthesis materials are polymeric implants based on poly (α-hydroxy acids) - polylactic acid, polyglicolic acid and their copolymers [1]. The main disadvantages of this type of materials are related to their lack of bioactivity, the foreign-body reactions and their relatively low mechanical properties. The latter problem forces to design thick implants, which need more time to biodegrade and
2
increase the risk of appearance of adverse tissue reactions [2]. Moreover, inflammatory response could also be accompanied with a significant decrease in the pH, which would accelerate the initial degradation rate of the implant. Several studies adding ceramic reinforcements or carbon nanotubes into biodegradable matrices try to overcome the main drawbacks of polymeric implants [3]. However, as these types of reinforcement are biostable or dissolve very slowly under physiological conditions, some concerns appear regarding their long-term effects in the living tissues. Therefore, the development of materials where both phases are completely degradable becomes important [4]. In opposition to polymers, Mg alloys have emerged as promising biodegradable materials [5] due to their higher mechanical strength, stiffness [6, 7], their biocompatibility and their beneficial role enhancing osteoblastic response [8-10]. However, their use for implants has been limited by their fast degradation rate in the physiological environment [11]. Mg degrades forming hydroxide and releasing hydrogen. Gaseous hydrogen, and the increase of pH associated with the corrosion process, are a source of irritation for the injured tissue, and can strongly deteriorate the biocompatibility and stability of the implant. Therefore, the driven force to develop biodegradable and biocompatible Mg alloys is to slow down the in vivo corrosion rate of Mg. Current trends have focused on the use of alloying elements [7, 12], grain size refinement [13], surface treatments [14, 15] and coatings [16, 17]. Very recently, a novel approach has emerged to face the drawbacks of both materials: poly (αhydroxy acids) and Mg alloys. The new strategy consists on the development of polymer/Mg composites as new bioabsorbable biomaterials for osteosynthesis and other temporary medical applications [18-21]. It has been demonstrated that in this type of materials the polymer benefits from the reinforcing effect of the stiffer and stronger Mg particles, enhancing its mechanical properties [19, 20] and also Mg improves the cell viability [20, 21]. It is also
3
expected that the Mg particles will benefit from the protective surroundings given by the polymer, so that the implants will exhibit tailored degradation rates. Degradation kinetics of absorbable medical devices must be controlled in order to meet the healing rate of bone in the implantation site. Kinetics can be tailored if the chemical reactions that are involved in the degradation of the composite are understood. Hydrolysis of the polymeric matrix and corrosion of Mg are the two main reactions that lead to the breaking down of the polymer/Mg composites. In order to control in vitro degradation kinetics of these composites, it is necessary to control either the hydrolysis of the matrix and/or the corrosion of Mg. Poly lactic acid belongs to the family of aliphatic polyesters, which means that its ester groups are susceptible to be hydrolytically degraded in the physiological environment according to reaction 1. The hydrolysis of ester bonds leads to cleavage of polymer chains, subsequent decrease in molecular weight, and final diffusion of degradation products into the surroundings. PLA degrades to lactic acid. This acid can be metabolized by the tricarboxylic acid cycle and excreted in the lungs as carbon dioxide and water or in the urine [22-24].
− COO − + H 2 O → −COOH + HO −
Reaction 1
Mg degradation rate depends on its purity, alloying elements and grain size [25, 26], it corrodes in aqueous environments forming hydroxides and releasing hydrogen according to reaction 2 [25]: Mg + 2 H 2O → Mg 2+ + 2(OH ) + H 2 −
Reaction 2
The existence of chloride (Cl-), phosphates (PO43-) and calcium (Ca2+) ions in body fluids, complicates the corrosion process. Chloride ions transform Mg hydroxide into soluble MgCl2;
4
resulting in excess OH- ions that eventually raise pH. Mg2+ ions dissolve into the solution, react with PO43- and Ca2+ and form phosphates that precipitate on the surface [25, 27, 28]. Understanding the in vitro degradation of a composite is a complex task since the characteristics of the matrix can modify the degradation behaviour of the reinforcement and vice versa. Specifically, in the case of the filler, incorporation of Mg particles in a poly(αhydroxy acid) matrix may alter the hydrolytic degradation of the polymer. Moreover, it is highly likely that the dispersion of the reinforcing particles, their volume fraction and morphology will play a role on the degradation rate of the composite as a whole. Therefore, it is foreseen that the degradation behaviour of polymer/Mg composites can be tailored by controlling the characteristics of the filler. Currently, the studies that explain the in vitro degradation of polymer composites reinforced with particles of metallic Mg are scarce [21] and do not consider hydrogen release studies. The main concerns in the applications of Mg-containing materials in the biomedical field are the generation of hydrogen and the local alkalinization. Fast Mg corrosion forms hydrogen bubbles that are accumulated at the implant surroundings and may cause clinical problems [29]. In addition, hydroxides that are produced by Mg corrosion increase the local in vivo pH. If the pH exceeds 7.8, the balance of physiological reactions that depend on pH can be affected [29, 30]. The aim of this work is to study the in vitro degradation of polymer/Mg composites by means of a joint analysis of the amount of H2 released, the changes in pH in buffered and nonbuffered media, variations in mass and morphology, and mechanical performance. The study leads to elucidate the relevance of Mg particles shape on tailoring the degradation kinetics of these novel composites, as different particle shapes may modify the rate of chemical reactions occurring between the implant and the surroundings. This work helps to further understand
5
the in vitro degradation of polymer/Mg composites and could lead to the improvement of bioabsorbable materials design. 2. EXPERIMENTAL PROCEDURE 2.1 Materials For this study two composites have been prepared using for the matrix a poly-(L,D-lactic) acid (PLA2002D) from Natureworks with a D- isomer content of 4.25%, molecular weight of 105 kDa and melt flow index (210ºC/2.16 kg) of 35.8 g / 10 min, and as reinforcement, spherical and irregular flake-like Mg particles of less than 50 µm in size. The amount of reinforcing particles was 10% in mass in both cases, or equivalently, 7.17% in volume (PLDA density:1.21, Mg density: 1.74 g cm-3). The composite reinforced by irregular flake-like particles will be named hereafter PLDA/Mg-IRR, and the composite reinforced by spherical particles will be named hereafter PLDA/Mg-SPH. The irregular particles were delivered by Goodfellow, with a purity of 99.9% and a length to width ratio of 1.6:1, with a median particle size of 24.6 µm. The spherical particles were obtained by centrifugal atomization (TLS Technik, Germany) of ingots of pure Mg (99.9%Mg; 0.05% Cu, Fe, Si and Al; 0.001% Ni). The atomized powder was sieved to retain particles of less than 50 µm, resulting a median particle diameter of 31.4 µm. Particle size and aspect ratio were measured by a Malvern 2000 laser-scattering particle size analyser. Figure 1 presents images of both powders and their size distribution. For comparison purposes, unreinforced PLDA and cast pure Mg cylinders were also processed and submitted to the same testing routine as the composites. 2.2 Composites fabrication
6
The PLDA pellets and Mg particles were mixed at ambient temperature, dried, and then melt blended at 180ºC in a Rondol co-rotating twin-screw continuous extruder (screw diameter 10 mm, length/diameter ratio 20). Temperatures of the barrel from the nozzle to the feed zone were: 180 ºC / 180 ºC / 160 ºC / 118 ºC. Screw speed was set at 40 rpm. The estimated residence time was 3 minutes. The resulting filaments, of about 2 mm in diameter, showed a homogeneous aspect. Afterwards, filaments were cut into pellets by using a grinder and then introduced into an OPAL 460 automatic hot mounting press containing a mould with multiple holes of 9 mm high and 6 mm diameter. Temperature of the press was increased until 190ºC and kept during 15 min to allow homogenization of temperature. After compression at 13 MPa, the mould was cooled down to ambient temperature at a fast rate and then specimens demoulded. 2.3 Hydrogen release and pH evolution The in vitro degradation kinetics was assessed by immersion of the specimens in Phosphate Buffered Saline solution (PBS). Although PBS does not accurately simulate body conditions, it is suitable for comparative studies on the hydrolytic degradation of PLA-based materials [24]. Amount of H2 released and pH of the medium were measured as a function of immersion time. In addition, distilled water was also used to study the nature of degradation products by measuring pH changes in a non buffered media. For the measurement of the H2 release (in the PBS solution), a ratio of volume of media (ml) to specimen surface (cm2) of 20:1 was used. Each specimen was placed under an inverted glass funnel with a burette over the top of each funnel to capture the hydrogen. The solution level in each burette was measured twice a day and related to the sample area. Three specimens for each condition were used. Data are presented as mean value ± standard deviation.
7
For the study of pH evolution, specimens were individually introduced into a test tube containing the media in the same ratio of 20 ml/cm2, and then closed to avoid evaporation. The media were fully renovated every seven days. Experiments were run at a constant temperature of 37 ± 1ºC, by introducing the test tubes in a thermostatic bath. The results correspond to average values of three specimens. pH was recorded with a Lazar equipment. The electrode was calibrated every day. Measurements were performed after each hour for the first 10 h and then once a day until completion of the experiment. Experiments were performed twice: during 7 days and during 28 days. Data are presented as mean value ± standard deviation. 2.4 Mass variation and water intake The change in mass was measured for the samples immersed in PBS. Wet samples were weighted immediately after removing from the solution and drying their surface with a paper towel; dry samples were measured after keeping them 2 weeks in a desiccator under vacuum. Two types of measurements were carried out: mass variation and water intake. In the mass variation type, the percentage of mass gain or loss, i.e. mass difference between the dried specimen at time t, Md(t), and the initial specimen, Md(t0), with respect to the initial weight, Equation 1, has been obtained. ((Md(t) - Md(t0))/ Md(t0))*100
Eq. 1
The water intake measurement was obtained through the difference between the mass of the wet specimen Mw(t) and the mass of the same specimen after letting it dry Md(t), related to the dried specimen mass, Equation 2: ((Mw(t) - Md(t))/ Md(t))*100
Eq. 2
A precision balance was employed to weight all samples within an error of 0.0001g.
8
Experiments were conducted in triplicate. Data are presented as mean value ± standard deviation. 2.5 Microstructure The cylinders surfaces and longitudinal sections were characterised by scanning electron microscopy (SEM) equipped with energy dispersive X-ray spectroscopy (EDS). Longitudinal diametric sections were cut and mechanically polished with 9 µm diamond paste; a thin graphite layer was deposited on the samples for SEM examination. 2.6 Mechanical performance The effect of in vitro degradation on the mechanical properties of composites was assessed by compression tests. The compressive mechanical behaviour was studied in a universal machine EM2/100/FR-10kN Micro Tests at ambient conditions. Cylindrical samples tested for pH measurements in PBS were used for compression tests after 7 and 28 days of immersion. The planar surfaces of these specimens were slightly ground to assure the necessary condition for compression tests of parallel plane bases. True dimensions of immersed specimens, once ground, were measured before each compression test. Experiments were conducted in triplicate using a strain rate of 5x10-3 s-1 (crosshead speed 2.6±0.07 mm/min, sample height 8.7±0.22 mm). Data are presented as mean value ± standard deviation.
3. RESULTS 3.1 Macroscopic observation Figure 2 shows the cylinders of the PLDA/Mg-IRR and PLDA/Mg-SPH composites asprocessed and after 7 and 28 days of immersion in PBS. It is remarkable the different aspect of the two degraded Mg reinforced composites. The irregular shaped particles lead to the 9
formation of deep and long cracks and the enlargement of the specimens, whereas the composites with spherical particles do not present cracks and only a slight modification of the cylinder shape. 3.2 Hydrogen Release Figure 3 shows the accumulative hydrogen release of both composites in PBS as a function of immersion time. During the first 5-6 days, the hydrogen released is only slightly higher for the composite reinforced with irregular particles (day 6: 1.60 ml H2/cm2 for IRR versus 1.09 ml H2/cm2 for SPH). From then on, PLDA/Mg-IRR releases increasingly higher amount of H2 than PLDA/Mg-SPH. For both composites, the H2 release starts to stabilize after about day 20. At that moment, the H2 released by PLDA/Mg-IRR (7.19 ml H2/cm2) is 1.5 times the amount released by PLDA/Mg-SPH (4.68 ml H2/cm2) and at the end of the test, i.e. day 28, the total amount of hydrogen released by PLDA/Mg-IRR is 7.49 vs. 4.93 ml H2/cm2 of PLDA/Mg-SPH. The relevance of these data become more evident when they are compared with the total amount of H2 that is expected to be released by the composite samples after the total degradation of all their Mg particles. The volume of hydrogen can be stoichiometrically related with the Mg that reacted. Taking into account the total mass of Mg in the specimen, the cylinder surface and reaction 2, the expected amount of hydrogen released is 16.58 ml H2/cm2. In the case of PLDA/Mg-IRR, at least 50% of all Mg in the IRR particles has been degraded, while the degradation decreases to less than 30% of Mg in the case of the PLDA/Mg-SPH composite. Degradation kinetics of PLDA/Mg composites may be divided into three ranges, as marked by the lines depicted in Figure 3; a first one (I) with a relatively small slope, a second one (II) with a larger slope and a third one (III) with a slope tending to zero. These regions differ from one material to the other in both the limits and the slopes of the lines. In the case of the
10
irregular particles, region I lasts until about day 6 and has a slope that is 1.65 times that of the composite with spherical particles, for which this region lasts until about day 12 of immersion in PBS. Region II, correspondingly, starts much earlier in the PLDA/Mg-IRR composite, with a slope 1.4 times that of the slope of PLDA/Mg-SPH. Finally, region III is similar for both composites, with a very low rate of hydrogen released. The comparison of composites behaviour with that of cylindrical monolithic Mg specimens of the same size indicates that in only five days the pure metallic sample releases as much H2 as the PLDA/Mg-IRR composite after 28 days, which results in a rate of 1.8 ml H2/cm2 per day. At the end of the test, the Mg cylinder released 14.2 H2/cm2 (not shown in the figure), that is twice the volume released by PLDA/Mg-IRR composite and four times that released by the PLDA/Mg-SPH composite. 3.3 pH evolution Variation of pH in the medium containing the composites was monitored in both distilled water and PBS (with a renewal of solution each week). Figure 4 shows the measured pH values. For PBS, the pH value stays stable between 7.2 and 7.4 for both types of composites, with no apparent difference between them. On the contrary, pure Mg cylinders degrade so much and so rapidly that pH increases above 10 in only one day (not shown in the figure), in spite of the fact that PBS is a buffered solution. Immersion of composites in distilled water raises the pH sharply up to about 10 during the first 1.5 day and then diminishes it steadily. This is the expected behaviour considering that Mg degradation would produce hydroxides that raise the pH of the solution whereas PLDA degradation would decrease it as the main degradation product is lactic acid. In our case, with pure Mg, pH of distilled water rises above 11, whereas for unreinforced PLDA cylinders pH
11
varies between 6.3 and 5.0 (not shown in the figure). Worthwhile to note is that pH for PLDA/Mg-IRR remains always slightly lower than that for PLDA/Mg-SPH. This difference increases with time, so that after 28 days of immersion, the water of the composite reinforced with irregular particles has a pH of 7.4 versus 8.9 of the water of the composite reinforced with spherical particles. 3.4 Variation of mass and water intake The weights of the cylinders were measured before beginning the experiments and after 7 and 28 days in PBS. Two types of measurements were carried out: mass variation and water intake. The results are given in % of mass with respect to the reference sample and are depicted in Figure 5. As it can be seen in Figure 5a, after 7 days there is an increase in mass percentage (mass gain) for both composites, more evident for PLDA/Mg-IRR. This gain indicates that most reaction products have remained in the samples. On the contrary, after 28 days both composites lose mass, being PLDA/Mg-IRR the one that loses the highest amount (about 12% versus 3%). This confirms that the composite reinforced with the irregular particles degrades much faster than the one reinforced with the spherical particles. On the other hand, mass loss of pure PLDA cylinders resulted negligible (0.26% after 28 days). Figure 5b shows the water intake results for both composites after 7 and 28 days of immersion in PBS. The most remarkable outcome extracted from this figure is that PLDA/Mg-IRR retains more than twice the liquid retained by PLDA/Mg-SPH after 7 days (5.2 vs. 2.1%), which increases to more than three times after 28 days (23.3 vs. 6.9%), which reflects its much higher degradation and degradation rate. PLDA retains the same amount of liquid after 7 and 28 days and it does not exceed the 1%.
12
3.5 Microstructure Cross sections of tested cylinders were observed by scanning electron microscopy to analyze the evolution of magnesium particles and the integrity of the cylinders. Figure 6 presents backscattered images of both composites as-processed and after 7 days of immersion in PBS solution. Micrographs in Figure 6a and d are taken at lower magnification than the others in order to show the distribution of Mg particles within the matrix. At the centre of the cylinders, Figures 6b and e, only Mg particles (bright contrast) and the PLDA continuous matrix (dark contrast) are seen, whereas at the border, Figures 6c and f, where the solution is in direct contact with the specimens from the beginning of the experiment, reaction phases are detected (gray contrast) in the area close to the border, some of them are arrowed in the figure. There are not very significant differences between both composites, with the exception of a somewhat larger amount of reaction phase in PLDA/Mg-IRR than in PLDA/Mg-SPH. With regards to the chemical composition, EDS taken on particles at the centre of the cylinders only reveal the presence of Mg, Figure 7a, whereas particles close to the border seem to contain more oxygen, Figure 7b. On the other hand, the reaction phase shows a much higher amount of O, due to the formation of Mg(OH)2, and some amounts of P and Cl, that are components of the PBS, Figure 7c. After 28 days of immersion in PBS, some particles at the centre of the specimens appear partially transformed, as can be seen in Figures 8a and c (some of them arrowed). This partial transformation may be more abundant in the case of irregular particles. Close to the border, there is a clear difference between composites. In PLDA/Mg-IRR, almost every particle appears fully transformed, with the reaction phase broken down into small chips reflecting the brittle nature of Mg(OH)2, Figure 8b. In the case of PLDA/Mg-SPH, particles with a Mg core, still untransformed, are more frequently seen, as shown in the example of Figure 8d.
13
3.6 Mechanical properties Mechanical properties were tested in PLDA/Mg-SPH and in PLDA as a function of degradation time in PBS. In the case of PLDA/Mg-IRR, as shown in Figure 2 these specimens suffered a drastic change on their morphology, which makes impossible the performance of compression tests after 7 and 28 days. Figure 9a shows the stress vs. strain curves of the three materials at time zero and PLDA/Mg-SPH after 7 and 28 days. Results of Young’s modulus and compressive strength at yield as a function of immersion time are shown in Figure 9b and c respectively. At time zero the three materials present similar mechanical properties, Young´s Modulus between 1.7 and 2 GPa and compressive strength at Yield between 105 and 120 MPa. Regarding PLDA/Mg-SPH, it is observed that a week in PBS leads to a reduction of Young´s modulus and compressive strength of only 0.1 GPa and 5 MPa respectively. This implies a loss of 7% in stiffness and 4% in resistance. After near a month, the loss of mechanical properties becomes greater. The modulus and the compressive strength at yield are reduced by 40% compared with the original material. PLDA maintains its mechanical performance during the whole experiment. 4. DISCUSSION The different slopes that describe the degradation behaviour of the PLDA/Mg composites indicate a change in the mechanism that governs hydrogen release. Different slope values can be related with the changing microstructure of the samples with immersion time. Photographs of the cylinders in Figure 2 show the formation of deep cracks in the degraded PLDA/MgIRR composite, already after 7 days of immersion.
14
The presence of these cracks, also shown in Figure 10a (some of them arrowed), helps to understand the faster degradation rate of PLDA/Mg-IRR in comparison with PLDA/Mg-SPH. Cracks act as fast paths for the diffusion of the solution into the material, which would accelerate the degradation rate and would explain the higher slope in region II of Figure 3 of accumulated hydrogen release vs. days of immersion. Cracks in the PLDA/Mg-SPH are not macroscopically evident but can be microscopically observed in the cross sections of samples after 28 days immersed in PBS (Figure 10b). In both materials cracks appear at the polymer/particle interface and then progress. As the degradation reaction evolves the available amount of Mg decreases, and so does the amount of released H2, which explains the lower slope of region III. The fact that pH of distilled water with PLDA/Mg-SPH results higher than that with PLDA/Mg-IRR is striking given its lower rate of degradation, i.e. hydrogen release, which indicates a lower rate of reaction of the spherical Mg particles with water. This result may be understood, however, by considering the deep and long cracks that appear in the PLDA/MgIRR specimens. These cracks expose new polymer surface to the water, which by degrading releases lactic acid that reduces the pH. The pH increment that is expected from the faster reaction of the irregular Mg particles is therefore compensated by the higher amount of lactic acid released by the polymer. A possible concern regarding the application of these composites refers to the acidification of the media once Mg particles have completely reacted, due to the remaining PLDA matrix, which degrades much more slowly. In the authors’ opinion, this would represent a less critical problem compared with the actual polylactic acid-based implants in use, which are made of monolithic polylactic acid. Moreover, according to our studies, this acidification would happen only after more than 28 days.
15
The pH behaviour of the buffered solution containing pure Mg cylinders is consistent with the literature [31-33] and elucidates one important concern regarding the applicability of Mg alloys in the biomedical field, the possible alkalinization of the physiological environment. Measurements in buffered media containing the PLDA/Mg composites indicate that a neutral pH is maintained during the whole test. The polymeric matrix is playing an important role controlling the corrosion of Mg particles and hindering the alkalinization of the media. The water intake measurement gives the amount of solution that is absorbed by the cylinders at time t. The higher the water retained, the higher the degradation of the composite. The amount of water retained by a polymer depends on factors such as its molecular structure, crystallinity degree or the incorporation of additives and fillers. PLDA water retention falls within the range of the equilibrium water content of poly(α –hydroxy acids), which goes from 0.3% to 2.6% depending on the monomer unit [24], and its value do not change with immersion time Figure 5b. Mg particles increase the water diffusion through the polymeric matrix and accelerate its hydrolytic degradation. It has been demonstrated that the incorporation of fillers within PLA alters its hydrolytic degradation [22, 24]. Hydrophilic fillers accelerate the diffusion of water into the material, which leads to increase the degradation rate of the polymer [34, 35]. The increment of water content is also an indicative of pores, cracks and capillars formation [22]. These defects can be produced by the reaction of Mg particles. As water diffuses within the matrix, it reaches the surface of Mg particles leading to the corrosion of the metal. When Mg reacts with the aqueous environment, it releases hydrogen and produces hydroxides that lead to a local alkalinization in the polymer/particle interface. Given that a high concentration of hydroxide ions can accelerate the hydrolytic degradation of poly(α-hydroxy acids) [36-38], the ester bonds of those polymer
16
chains that are subjected to a high pH environment are more vulnerable to be attacked. Therefore, degradation of the polymer at the interface will occur faster. The microstructure of the degraded composites shows that particles that are close to the surface react before those that are at the centre. This indicates that, although degradation in these polymers occurs through bulk erosion [39], the degradation of the composite does not occur simultaneously in the whole volume, but it is controlled by the diffusion rate of water/PBS within the polymeric matrix. Effectively, the solution needs time to diffuse among the polymer chains and reach the Mg particles inside the specimen, with the consequent of higher erosion of Mg particles that are closer to the specimen surface. A question that remains to answer is why there are more cracks in PLDA/Mg-IRR than in PLDA/Mg-SPH. A possible origin of the formation of cracks could be the expansion of the polymer matrix produced by the hydrogen molecules released at the particles surface due to the chemical reaction between Mg and the aqueous solution. Keeping in mind that any irregular particle has a larger surface area than a spherical one of the same volume, together with the fact that the irregular particles employed in this work have a lower median size than the spherical ones (24.6 vs. 31.4 µm), it results that, for the same mass % of particles, the total surface area of the Mg irregular particles is larger than that of the spherical ones. Therefore, there are more places available for the reaction between Mg and PBS in the PLDA/Mg-IRR composite, which would explain the faster hydrogen release, which in turn may be the cause of the formation of more and earlier cracks. Once the cracks are formed, it results increasingly easier for the solution to reach the Mg particles, thus accelerating the rate of degradation. Apart from the effect of hydrogen, another feature that may contribute to the appearance of cracks is the formation of a Mg(OH)2 interphase at the particle/matrix interface. This phase
17
produces an increase of particle volume that has to be accommodated by the matrix. This may cause internal stresses that may help to promote the formation of cracks. Another factor that contributes to PLDA/Mg-IRR faster degradation is that corrosion of Mg particles alters the hydrolysis of the polymer. Given that IRR particles degrade faster than SPH, the alkalinization of the polymer/Mg interface within PLDA/Mg-IRR composite is greater. Therefore, the hydrolytic degradation of PLDA in these composites is enhanced promoting the formation of cracks. Regarding the mechanical performance of PLDA/Mg-SPH, this material meets the compressive strength of cortical bone (95 – 130 MPa) but not its stiffness (Young´s modulus: 3 – 23 GPa). However, the compressive modulus is within the range of cancellous bone (50 – 800 MPa) [40]. Mg content of 10 wt.% does not impair the initial mechanical performance of PLDA but reduces its strength retention. The decrease in compressive resistance with degradation time can be explained by the degradation of the matrix and the particles. In addition, corrosion of Mg particles forms a Mg(OH)2 interphase and increments the hydrolysis at polymer/matrix interface, leading to the deterioration of the union between particles and the polymeric matrix. To our knowledge, this work reports for the first time results regarding the retention of mechanical properties of PLA/Mg composites. PLDA/Mg-SPH shows retention of compressive strength of 96% after 7 days, and 60% after 28 days. This retention of mechanical strength is higher or similar to that reported in the literature regarding bioabsorbable composites based on polylactic acid reinforced with bio-glass particles or fibres for which a mechanical strength loss greater than 50% was reported after only few days [4143].
18
In order to apply this material in the fabrication of implants, research is in progress to develop composites that meet cortical bone stiffness and exhibit higher and longer retention of mechanical properties by using coupling agents between Mg particles and the matrix. Further validation of these composites should also include the evaluation of bending and tensile properties. Future device design should consider the effect of the shape of the reinforcing filler on the degradation rate of the implant as well as on the mechanical properties.
SUMMARY AND CONCLUSIONS The in vitro degradation of biodegradable composites formed by a polylactic acid matrix reinforced with 10% wt. Mg microparticles was investigated in order to determine the relevance of Mg particles shape (flake-like and spherical) on tailoring the degradation kinetics of these composites. It was found that: - The irregular shaped particles lead to the earlier formation of deeper and longer cracks and the swelling of the specimens, in comparison with the spherical ones. - Hydrogen release rate of PLDA/Mg composites can be divided into three ranges; a first one (I) with a relatively small slope, a second one (II) with a larger slope and a third one (III) with a slope tending to cero. This is explained by the formation of cracks that act as fast paths for the diffusion of the solution into the material. - PLDA/Mg composites release hydrogen at a much slower rate than monolithic Mg, and PLDA/Mg-SPH releases hydrogen at a slower rate than PLDA/Mg-IRR. In every case, the amount of H2 released by the composites after 28 days is between a third and a half of that released by monolithic Mg.
19
- The pH value in PBS stays stable between 7.2 and 7.4 for both types of composites. On the contrary, pure Mg cylinders increase pH of the solution above 10 in only one day. In water, pH for PLDA/Mg-IRR remains slightly lower than that for PLDA/Mg-SPH due to the slower degradation rate of the matrix in the PLDA/Mg-SPH composites. - After 28 days of immersion in PBS, PLDA/Mg-IRR loses more weight than PLDA/Mg-SPH (about 12% versus 3%). On the other hand, PLDA/Mg-IRR retains more than twice the liquid of that retained by PLDA/Mg-SPH after 7 days, and more than three times after 28 days. - During degradation, transformation of Mg into Mg(OH)2 occurs. In PLDA/Mg-IRR, almost every particle appears fully transformed after 28 days, whereas in PLDA/Mg-SPH, particles with an untransformed Mg core still remain. - With regard to the mechanical integrity of PLDA/Mg-SPH composite, it is observed that after 7 days in PBS the composite retains the 93% of its stiffness and 96% of its compressive strength, whereas after 28 days, the material retains the 60% of its mechanical strength. The main conclusion that can be withdrawn from the present study is that the shape of the Mg reinforcing particles plays a crucial role in the degradation rate of PLA/Mg composites, being the spherical particles the ones that promote better behaviour than the flake-like particles. The reason for this behaviour is only partially due to the smaller surface area to volume ratio of the spherical particles. The other important reason is that irregular particles promote an accelerated degradation rate by releasing hydrogen at a faster rate, which in turn provokes the earlier formation of cracks and, therefore, an increasingly faster degradation of the polymeric matrix. These findings help to further understand the in vitro degradation of polymer/Mg composites and could lead to the improvement of bioabsorbable materials design.
20
ACKNOWLEDGEMENTS Thanks are due to MAT2012-37736-C01 and MAT2013-47972-C2-1-P (MICCIN) for financial support and to CSIC and European Social Fund for JAE-I3P Grant (S.C. Cifuentes). Special thanks are deserved to Prof. V. Lorenzo (UPM) for his advice and help with the processing. Technical advice of Jesús Chao with the mechanical tests, and technical assistance of Amalia San Román are greatly acknowledged.
DISCLOSURE The authors do not have any financial/personal conflict of interest that could affect their objectivity, or inappropriately influence their actions. REFERENCES [1] Maurus PB, Kaeding CC. Bioabsorbable implant material review. Operative Techniques in Sports Medicine 2004;12:158-60. [2] Böstman OM, Pihlajamäki HK. Adverse tissue reactions to bioabsorbable fixation devices. Clin Orthop Relat Res 2000;371:216-27. [3] Rezwan K, Chen QZ, Blaker JJ, Boccaccini AR. Biodegradable and bioactive porous polymer/inorganic
composite
scaffolds
for
bone
tissue
engineering.
Biomaterials
2006;27:3413-31. [4] Navarro M, Ginebra MP, Planell JA, Barrias CC, Barbosa MA. In vitro degradation behavior of a novel bioresorbable composite material based on PLA and a soluble CaP glass. Acta Biomater 2005;1:411-9. [5] Witte F, Hort N, Vogt C, Cohen S, Kainer KU, Willumeit R, et al. Degradable biomaterials based on magnesium corrosion. Current Opinion in Solid State and Materials Science 2008;12:63-72.
21
[6] Staiger MP, Pietak AM, Huadmai J, Dias G. Magnesium and its alloys as orthopedic biomaterials: A review. Biomaterials 2006;27:1728-34. [7] Müller WD, Nascimento ML, Zeddies M, Córsico M. Magnesium and its Alloys as Degradable Biomaterials. Corrosion Studies Using Potentiodynamic and EIS Electrochemical Techniques. Materials Research 2007;10:5-10. [8] Zreiqat H, Howlett CR, Zannettino A, Evans P, Schulze-Tanzil G, Knabe C, et al. Mechanisms of magnesium-stimulated adhesion of osteoblastic cells to commonly used orthopaedic implants. Journal of Biomedical Materials Research 2002;62:175-84. [9] Revell P, Damien E, Zhang X, Evans P, Howlett C. The effect of magnesium ions on bone bonding to hydroxyapatite coating on titanium alloy implants. Key Enginnering Materials 2004;254 - 256. [10] Janning C, Willbold E, Vogt C, Nellesen J, Meyer-Lindenberg A, Windhagen H, et al. Magnesium hydroxide temporarily enhancing osteoblast activity and decreasing the osteoclast number in peri-implant bone remodelling. Acta Biomater 2010;6:1861-8. [11] Witte F, Kaese V, Haferkamp H, Switzer E, Meyer-Lindenberg A, Wirth CJ, et al. In vivo corrosion of four magnesium alloys and the associated bone response. Biomaterials 2005;26:3557-63. [12] Castellani C, Lindtner RA, Hausbrandt P, Tschegg E, Stanzl-Tschegg SE, Zanoni G, et al. Bone–implant interface strength and osseointegration: Biodegradable magnesium alloy versus standard titanium control. Acta Biomater 2011;7:432-40. [13] Alvarez-Lopez M, Pereda MD, del Valle JA, Fernandez-Lorenzo M, Garcia-Alonso MC, Ruano OA, et al. Corrosion behaviour of AZ31 magnesium alloy with different grain sizes in simulated biological fluids. Acta Biomater 2010;6:1763-71. [14] Carboneras M, García-Alonso MC, Escudero ML. Biodegradation kinetics of modified magnesium-based materials in cell culture medium. Corros Sci 2011;53:1433-9. [15] Carboneras M, Hernández LS, del Valle JA, García-Alonso MC, Escudero ML. Corrosion protection of different environmentally friendly coatings on powder metallurgy magnesium. Journal of Alloys and Compounds 2010;496:442-8.
22
[16] Wong HM, Yeung KWK, Lam KO, Tam V, Chu PK, Luk KDK, et al. A biodegradable polymer-based coating to control the performance of magnesium alloy orthopaedic implants. Biomaterials 2010;31:2084-96. [17] Degner J, Singer F, Cordero L, Boccaccini AR, Virtanen S. Electrochemical investigations of magnesium in DMEM with biodegradable polycaprolactone coating as corrosion barrier. Applied Surface Science 2013;282:264-70. [18] González-Carrasco JL, Multigner M, Lieblich M, Muñoz M, Frutos E, Saldaña L, et al. Polymer and magnesium particle material for biomedical applications WO 2011/161292 A1. In: internacional O, editor.2011. [19] Cifuentes SC, Frutos E, González-Carrasco JL, Muñoz M, Multigner M, Chao J, et al. Novel PLLA/magnesium composite for orthopedic applications: A proof of concept. Materials Letters 2012;74:239-42. [20] Brown A, Zaky S, Ray Jr H, Sfeir C. Porous magnesium/PLGA composite scaffolds for enhanced bone regeneration following tooth extraction. Acta Biomater 2015;11:543-53. [21] Wan P, Yuan C, Tan L, Li Q, Yang K. Fabrication and evaluation of bioresorbable PLLA/magnesium and PLLA/magnesium fluoride hybrid composites for orthopedic implants. Composites Sci Technol 2014;98:36-43. [22] Lehtonen TJ, Tuominen JU, Hiekkanen E. Resorbable composites with bioresorbable glass fibers for load-bearing applications. In vitro degradation and degradation mechanism. Acta Biomater 2013;9:4868-77. [23] Eglin D, Alini M. Degradable polymeric materials for osteosynthesis: tutorial. European Cells & Materials 2008;16:80-91. [24] Auras R, Lim LT, Selke SEM, Tsuji H. Poly-lactic acid. Synthesis, Structures, Properties, and Applications. New Jersey: John Wiley & Sons, Inc.; 2010. [25] Zhang S, Zhang X, Zhao C, Li J, Song Y, Xie C, et al. Research on an Mg–Zn alloy as a degradable biomaterial. Acta Biomater 2010;6:626-40. [26] Mueller W-D, Lucia Nascimento M, Lorenzo de Mele MF. Critical discussion of the results from different corrosion studies of Mg and Mg alloys for biomaterial applications. Acta Biomater 2010;6:1749-55.
23
[27] Gunde P. Ph D Thesis: Biodegradable magnesium alloys for osteosynthesis - Alloy development and surface modifications. Zurich: ETH 2010. [28] Li J, Cao P, Zhang X, Zhang S, He Y. In vitro degradation and cell attachment of a PLGA coated biodegradable Mg6Zn based alloy. Journal of Materials Science 2010;45:6038 45. [29] Song G. Control of biodegradation of biocompatible magnesium alloys. Corros Sci 2007;49:1696 - 701. [30] Song G. Recent Progress in Corrosion and Protection of Magnesium Alloys. Advanced Engineering Materials 2005;7:563-86. [31] Lorenz C, Brunner JG, Kollmannsberger P, Jaafar L, Fabry B, Virtanen S. Effect of surface pre-treatments on biocompatibility of magnesium. Acta Biomater 2009;5:2783-9. [32] Schinhammer M, Hofstetter J, Wegmann C, Moszner F, Löffler JF, Uggowitzer PJ. On the Immersion Testing of Degradable Implant Materials in Simulated Body Fluid: Active pH Regulation Using CO2. Advanced Engineering Materials 2013:1 - 8. [33] Song G, Song S-Z. Corrosion Behaviour of Pure Magnesium in a Simulated Body Fluid. Acta Physico-Chimica Sinica 2006;22:1222 - 6. [34] Meer SATvd, Wijn JRd, Wolke JGC. The influence of basic filler materials on the degradation of amorphous D- and L-lactide copolymer. J Mater Sci Mater Med 1996;7:359 61 [35] Ara M, Watanabe M, Imai Y. Effect of blending calcium compounds on hydrolytic degradation of poly(dl-lactic acid-co-glycolic acid). Biomaterials 2002;23:2479-83. [36] Tsuji H, Ikada Y. Properties and morphology of poly(L-lactide). II. Hydrolysis in alkaline solution. Journal of Polymer Science Part A: Polymer Chemistry 1998;36:59-66. [37] Cam D, Hyon S-h, Ikada Y. Degradation of high molecular weight poly(l-lactide) in alkaline medium. Biomaterials 1995;16:833-43. [38] Yuan X, Mak AFT, Yao K. Surface degradation of poly(l-lactic acid) fibres in a concentrated alkaline solution. Polym Degradation Stab 2003;79:45-52. [39] Burkersroda Fv, Schedl L, Göpferich A. Why degradable polymers undergo surface erosion or bulk erosion. Biomaterials 2002;23:4221-31.
24
[40] Huiskes R, Mow VC. Biomechanics of bone. In: Mow VC, Huiskes R, editors. Basic orthopaedic biomechanics and mechanobiology. Philadelphia: Lippincott Williams & Wilkins; 2005. [41] Vergnol G, Ginsac N, Rivory P, Meille S, Chenal JM, Balvay S, et al. In vitro and in vivo evaluation of a polylactic acid-bioactive glass composite for bone fixation devices. J Biomed Mater Res Part B 2015:000-. [42] Ahmed I, Cronin PS, Abou Neel EA, Parsons AJ, Knowles JC, Rudd CD. Retention of mechanical properties and cytocompatibility of a phosphate-based glass fiber/polylactic acid composite. Journal of Biomedical Materials Research Part B: Applied Biomaterials 2009;89B:18-27. [43] Felfel RM, Ahmed I, Parsons AJ, Walker GS, Rudd CD. In vitro degradation, flexural, compressive and shear properties of fully bioresorbable composite rods. Journal of the Mechanical Behavior of Biomedical Materials 2011;4:1462-72.
25
Figure 1. SEM images (left) and particle size distribution (right) of a) irregular flake like and b) spherical Mg particles
Figure 2. Photographs of the PLDA/Mg-IRR (left) and PLDA/Mg-SPH (right) cylinders asprocessed (original) and after immersion during 7 and 28 days in PBS solution.
26
Figure 3. Accumulated amount of hydrogen released as a function of immersion time in PBS. The straight lines (dashed and continuous) delimit regions of different slopes, which values are indicated by the numbers (in ml/cm2day).
Figure 4. Evolution of pH of the composites in distilled water and PBS solution.
Figure 5. a) Percentage of mass variation of dry specimens after 7 and 28 days of immersion in PBS. b) Percentage of mass gain due to water intake after 7 and 28 days of immersion in PBS.
27
Figure 6. Backscattered images of longitudinal sections of: a) PLDA/Mg-IRR composite asprocessed, b) central area after 7 days of immersion in PBS, and c) border after 7 days of immersion in PBS; d) PLDA/Mg-SPH composite as-processed, e) central area after 7 days of immersion in PBS, and f) border after 7 days of immersion in PBS.
28
Figure 7. Representative EDS spectra of PLDA/Mg-IRR and PLDA/Mg-SPH composites taken on Mg reinforcing particles after immersion in PBS: a) particle at the centre of the cylinder, 7 days, b) particle at the border of the cylinder, 7 days, and c) reaction phase of a particle at the border, 7 days.
Figure 8. Backscattered images of longitudinal sections of the composites after 28 days of immersion in PBS. a) Centre and b) border of PLDA/Mg-IRR, c) centre and d) border of PLDA/Mg-SPH.
29
Figure 9. a) Stress-strain curves of compression tests and b) resulting Young’s modulus, and c) yield strength as a function of immersion time.
30
Figure 10. Backscattered images of longitudinal sections of the composites after 28 days of immersion in PBS. a) PLDA/Mg-IRR, b) PLDA/Mg-SPH.
31
