d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.intl.elsevierhealth.com/journals/dema
Mechanical properties and ion release from bioactive restorative composites containing glass fillers and calcium phosphate nano-structured particles Marina D.S. Chiari, Marcela C. Rodrigues, Tathy A. Xavier, Eugen M.N. de Souza, Victor E. Arana-Chavez, Roberto R. Braga ∗ Department of Biomaterials and Oral Biology, Univ. of São Paulo School of Dentistry, Av. Prof. Lineu Prestes, 2227, São Paulo, SP 05508-000, Brazil
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objective. To evaluate the effect of the replacement of barium glass by dicalcium phosphate
Received 29 September 2014
dihydrate (DCPD) particles on the mechanical properties and degree of conversion (DC) of
Received in revised form
composites. Additionally, calcium and hydrogen phosphate (HPO4 2− ) release were followed
2 December 2014
for 28 days.
Accepted 30 March 2015
Methods. Nine composites containing equal parts (in mols) of BisGMA and TEGDMA and 40, 50 or 60 vol% of total filler were manipulated. Filler phase was constituted by silanated barium glass and 0%, 10% or 20% of DCPD particles. DC was determined by near-FTIR. Biaxial
Keywords:
flexural strength (BFS) and modulus (E) were tested using the “piston on three balls” method,
Calcium phosphate
while fracture toughness (KIc ) used the “single edge notched beam” method. Specimens were
Mechanical properties
tested after 24 h and 28 days in water. Ion release was determined using inductively coupled
Ion release
plasma optical emission spectrometry (ICP-OES). Data were analyzed by ANOVA/Tukey (DC
Resin composite
and ion release) or Kruskal–Wallis/Mann–Whitney (mechanical properties; alpha: 5%). Results. DC was not affected by DCPD. The presence of DCPD reduced BFS for both storage times, while differences in E became evident after 28 days. After 24 h, KIc increased with the addition of DCPD; after 28 days, however, KIc decreased only for DCPD-containing composites. Calcium release was similar for both DCPD contents and remained fairly constant during the 28-day period. Overall, HPO4 2− release was higher at 7 days and did not decrease after 14 days. Significance. The composite with the highest filler level and 10% DCPD represented the best compromise between mechanical properties after aging in water and ion release. © 2015 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
∗ Corresponding author at: Faculdade de Odontologia da USP, Depto. de Biomateriais e Biologia Oral, Av. Prof. Lineu Prestes, 2227, São Paulo, SP 05508-000, Brazil. Tel.: +55 11 3091 7840x224; fax: +55 11 3091 7840x224. E-mail address: [email protected] (R.R. Braga).
http://dx.doi.org/10.1016/j.dental.2015.03.015 0109-5641/© 2015 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
1.
Introduction
The development of permanent restorative materials with remineralizing or antimicrobial activity has been an aspiration in restorative dentistry as a possible path to reduce the risk of secondary caries [1–3], one of the main causes of restoration failure and replacement [4,5]. The first attempts to incorporate calcium orthophosphate particles in dimethacrylate resins as ion source for remineralization date back several decades [6], but only recently orthodontic cements and sealants containing amorphous calcium phosphate (ACP) nanoparticles became available for clinical use [7,8]. Early on, the difficulties of combining high fracture strength with high levels of ion release became evident. Due to the lack of a chemical bond between the calcium phosphate particles and the resin phase, the former behave as large flaws (i.e. several micrometers in width), concentrating stress and acting as crack initiation sites. In fact, when 40% (by weight) of calcium phosphate particles were added to a dimethacrylate resin, its fracture strength was reduced by approximately 50% [9,10]. Interestingly, elastic modulus is not influenced by the bond (or the lack of thereof) between resin and particle phases, as successive additions of tetracalcium phosphate (TTCP) particles to an unfilled resin led to cumulative increases in modulus [10]. Several attempts were made to improve the fracture strength of resin-based materials containing calcium phosphates. In broad terms, the tested approaches can be divided into two groups, namely, modification/functionalization of calcium phosphate particles and association of ion-releasing fillers with a reinforcing phase. The addition of the glass forming compounds tetraethoxysilane (TEOS) or zirconyl chloride in the synthesis of ACP resulted in 25–33% increase in the fracture strength of materials containing 40 wt% of filler, in comparison with those containing unmodified ACP particles [9]. Hydroxyapatite particles functionalized with acrylic acid increased the flexural strength of a BisGMA/TEGDMA-based material by 75% in comparison to the use of non-functionalized particles [11]. Finally, several studies tested the effect of silanization of calcium phosphate particles. Though these studies consistently reported higher fracture strengths for composites containing silanized calcium phosphate particles in comparison to unmodified particles [12–16], some of them also found that silanization hindered ion release [12,14,16]. The association of calcium phosphate particles with a reinforcing phase in resin-based materials involved the use of silanated, silica-fused silicon carbide or silicon nitride whiskers [15–20], or silanated glass particles [10,21–25]. The use of whiskers as reinforcing phase in resin-based materials containing calcium phosphates was very effective from a mechanical standpoint [16,17]. In fact, the flexural strength of composites containing equal mass fractions of calcium phosphate and whiskers matched that of a commercial hybrid composite with similar filler fraction, while their flexural modulus was actually higher [18–20]. However, these materials have the drawback of being opaque and, therefore, had to be formulated as two-paste systems.
727
The weakening effect of calcium phosphate particles on resins can be exemplified by a study reporting that 20–30 wt% of silanated barium glass had to be added to a resin matrix containing 40 wt% TTCP in order to bring its strength to the level of the unfilled resin [10]. Only a few studies tested the effect of replacing glass fillers by calcium phosphate particles. For instance, the replacement of 15% of glass particles by ACP in experimental composites containing 75 wt% of filler led to increased wear [23], and a significant reduction in strength, but had no effect on elastic modulus [21]. Another important issue is the effect of calcium phosphate particles on composite degradation over time. Available information in the literature suggests that prolonged water immersion had similar effects on composites containing only glass particles (75 wt%) or up to 20 wt% of ACP particles [23]. Considering the effectiveness of glass fillers as reinforcing phase for stress-bearing restorative composites, it is important to determine the percentage of glass fillers that could be replaced by calcium phosphate particles without jeopardizing the mechanical properties and, at the same time, ensuring enough ion release for remineralization of caries lesions. Based on the scant amount of information on the effects of replacing silanated glass fillers with bioactive particles, the objective of the present study was to evaluate how degree of conversion, mechanical properties and ion release of experimental composites were affected when dicalcium phosphate dihydrate (DCPD) particles were incorporated in the resin matrix at the expense of reinforcing fillers. Different total filler levels were tested, as well as the effect of prolonged aging in water. The null hypotheses were: (1) the replacement of silanated glass fillers by DCPD does not affect the degree of conversion or the mechanical properties, regardless of the total filler content or DCPD levels considered, (2) the effects of prolonged immersion in water on mechanical properties are not influenced by the total filler content or by the presence of DCPD, and (3) ion release is similar among total filler content and DCPD levels.
2.
Method
2.1.
Composite formulation
Nine composites were prepared, constituted by a 1:1 ratio, in mols, of bisphenol-A glycidyl dimethacrylate (BisGMA) and triethylene glycol dimethacrylate (TEGDMA), 0.5% (by weight) of camphorquinone and ethyl-4-dimethylamino benzoate (EDMAB, all components from Sigma-Aldrich, St. Louis, MO, USA). The experimental design was based on a factorial model with total filler level (40%, 50% or 60% by volume) and DCPD content (0%, 10% or 20% by volume) as main factors. Silanated barium glass (D50 : 0.5 m) and proprietary DCPD particles (D50 : 8.0 m, CaHPO4 ·2H2 O, Labsynth, Diadema, SP, Brazil) were used. Their filler size distributions were determined by laser light scattering (Mastersizer 2000, Malvern Instruments Ltd., Malvern, UK) and are shown in Fig. 1. A transmission electron microscopy image of part of a nanostructured agglomerate is shown in Fig. 2. Particle densities (barium glass: 2.4 g/cm3 ; DCPD: 2.7 g/cm3 ), necessary to calculate the amounts used in the tested formulations, were
728
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
the absorption of the C–H2 group) of the polymerized and the unpolimerized material.
2.3.
Fig. 1 – Filler size distributions for barium glass and DCPD particles.
Biaxial flexural test
Disk-shaped specimens (12 mm × 1.2 mm, n = 20) were built using a stainless steel split mold. Photoactivation was performed by overlapping four 10-s irradiations, one per quadrant. Half of the specimens were stored in water at 37 ◦ C for 24 h, while the other half was stored for 28 days. After storage, the specimens were placed on a “piston on three balls” setup positioned on a universal testing machine (model 5565, Instron Corp., Norwood, MA, USA). In this setup, the specimen was supported by three 2.5-mm diameter steel spheres distributed on a 10-mm diameter circumference 120◦ apart from each other, and the load was applied by a flat piston with 1.2 mm in diameter. The discs were positioned in a way that the irradiated surface would be under compressive stresses and fractured at a cross-head speed of 0.5 mm/min. Deflection at the center of the specimen was monitored by a deflectometer (model W-E401-E, Instron Corp, USA). Biaxial flexural strength (BFS) in MPa, was calculated according to the following equations: BFS =
−0.2387P(X − Y) b2
X = (1 + v) ln
r 2 2
r3
Y = (1 − v) 1 − ln Fig. 2 – Transmission electron microscopy (TEM) image showing part of a DCPD agglomerate. Plate-like nanoparticles are distinguishable at the agglomerate’s fringe (200,000× magnification).
determined in a Helium picnometer (Ultrapyc 1200e, Quantachrome Instruments, Boynton Beach, FL, USA). Composites were mechanically mixed under vacuum (Speedmixer DAC 150.1 FVZ-K, FlackTek Inc., Landrum, SC, USA) and kept under refrigeration until 2 h before use.
2.2.
Degree of conversion
Degree of conversion (n = 3) was determined by Fouriertransformed near-infrared spectroscopy (near-FTIR). The unpolimerized material was placed in a silicone mold with 7 mm in diameter and 1 mm in thickness and both were pressed between two microscopy glass slides. The spectrum of the unpolimerized material was obtained between 4000 and 10,000 cm−1 by the co-addition of 32 scans with 4 cm−1 resolution (Vertex 70; Bruker Optik GmbH, Ettlingen, Germany). Then, the composite was photoactivated through the glass slide (Radii Cal, SDI, Bayswater, Australia), receiving 24 J/cm2 (1200 mW/cm2 for 20 s) and the whole setup was dry-stored at 37 ◦ C. After 24 h, a new spectrum was obtained. Degree of conversion was calculated as the ratio between the areas of the absorption band located at 6165 cm−1 (corresponding to
+
1 − v r 2
r 2 1
r2
2
2 + (1 − v)
r3
r 2 1
r3
where P is the fracture load (in Newtons); b is the specimen thickness (in mm); v Poisson’s ratio; r1 is the radius of the supporting spheres circumference (5.0 mm); r2 is the radius of the loading piston (0.6 mm); r3 is the specimen radius (in mm). A Poisson’s ratio of 0.3 was adopted for all composites [26]. Flexural modulus was calculated according to the following equation:
E=
ˇP˛2 ωh3
× 0.001
where E is the flexural modulus (in GPa), ˇ is a constant related to the deflection at the center of the disk (0.509), P is the load (in Newtons), ˛ is the disk radius (in mm), ω is the deflection corresponding to P and h is the disk thickness (in mm). Surface fractures were gold-sputtered and observed under scanning electron microscopy (LEO 430, LEO Electron Microscopy Ltd., Cambridge, UK).
2.4.
Fracture toughness
Fracture toughness (KIc ) was determined using the “singleedge notched beam” method (SENB). Specimens were built using a stainless steel split mold (25.0 mm in length, 5.0 mm in width and 2.8 mm in thickness, n = 20) with a razor blade inserted to produce a sharp notch at mid-point of specimen’s length. The composite was covered by a mylar strip and photoactivated by three adjacent exposures of 40 s (Radii-Cal, SDI). After being stored in distilled water at 37 ◦ C for 24 h or 28 days,
729
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
specimens were tested in three-point bending (span: 20 mm) in a universal testing machine (model 5565, Instron Corp.) at a cross-head speed of 0.13 mm/min, loading the notched surface in tension. The fractured surfaces were analyzed by a stereomicroscope with a CCD camera (model SZ61, Olympus, Tokyo, Japan) at 90× magnification. The width (w) and thickness (b) of the specimen, and the notch length (a) were measured using the ImageJ software (National Institute of Health, Bethesda, USA). Six notch length measurements (three on each fracture surface) were averaged and only the specimens with a/w ratio between 0.45 and 0.50 were used. KIc , in MPa m0.5 , was calculated using the following equations: KIc =
P×S b × w1.5
×f
a w
where P is the fracture load (in Newtons) and S is the span (in meters). The f(a/w) value was calculated as follows: f
a w
2.5.
Total filler content (vol%)
40 50 60
DCPD content (vol%) 0
10
77.9 (2.6) A 81.0 (3.5) A 78.5 (0.4) A
80.0 (2.7) A 78.0 (3.2) A 77.2 (0.8) A
20 78.3 (2.8) A 79.4 (1.8) A 75.1 (0.8) A
levels and both storage times evaluated. Further reductions were observed between composites containing 10% and 20%, except after 24 h at 60% total filler level (p < 0.001). Prolonged storage significantly reduced BFS for all the tested materials. After 24-h storage, the replacement of reinforcing fillers by
2 1.99 − (a/w) × [1 − (a/w)] × 2.15 − 3.93(a/w) + 2.7(a/w) q 3 =
3/2 w
2 × [1 + 2(a/w)] × 1 − (a/w)
Ion release
Specimens (5 mm in diameter, 1 mm in thickness, n = 3) were built and dry-stored at 37 ◦ C for 24 h. Then, they were individually immersed in 5 mL of sodium chloride 133 mmol/L buffered to pH 7.0 using 50 mmol/L of HEPES solution [19]. The immersion medium was changed weekly and the concentrations of Ca2+ and HPO4 2− were determined, non-cumulatively, after 7, 14, 21 and 28 days using inductively coupled plasma optical emission spectrometry (ICP-OES, Agilent Technologies, Santa Clara, CA, USA). Calibration curves for calcium and phosphorous were prepared using different dilutions of standard solutions (SpecSol Quimlab, Jacareí, SP, Brazil). Previous to the analysis, the samples were acidified using a 10% HNO3 solution (1:1 in volume) and filtered. Calcium and phosphorous readings were obtained simultaneously using the spectral emissions of 184 nm and 177 nm, respectively.
2.6.
Table 1 – Averages and standard deviations of degree of conversion (in %) as a function of total filler and DCPD content (ANOVA/Tukey test, p > 0.05).
DCPD did not reduce E in any of the filler levels (Fig. 4). However, after 28 days, the composites containing 40% filler and 10% or 20% DCPD, as well as the composite with 60% total filler and 20% DCPD presented statistically lower E than the respective controls (p < 0.001). SEM images of the fracture surfaces of DCPD-containing composites show large regularly shaped flaws varying from a few micrometers up to 30 m, corresponding to sites where the DCPD agglomerates detached from the resin (Fig. 5). At higher magnification, it is possible to notice that most of these spaces contain lamellar structures, corresponding to areas of transparticle fracture. No marked differences were observed between specimens fractured at different storage periods. KIc averages and standard deviations are shown in Fig. 6. For the control groups, after 24 h KIc increased significantly only between 40% and 50% filler. Also in the 24-h groups, there was
Statistical analysis
Data were subjected to normality and homoscedasticity tests. Degree of conversion and ion release were subjected to two-way and three-way ANOVA, respectively, and multiple comparisons were performed using Tukey’s test. Due to the lack of homoscedasticity, data from mechanical tests were analyzed using non-parametric tests (Kruskal–Wallis and Mann–Whitney for pair-wise comparisons). In all cases, the global significance level was 5%.
3.
Results
Degree of conversion results are shown in Table 1. No statistically significant differences were observed among experimental groups, with average values ranging from 75% to 81% (p = 0.161). Biaxial flexural strength (BFS) results are shown in Fig. 3. The replacement of 10% barium glass by DCPD resulted in statistically significant reductions in fracture strength in all filler
Fig. 3 – Biaxial flexural strength results (averages and standard deviations) for composites containing different total filler levels and DCPD contents. The same letters indicate lack of statistically significant differences (Kruskal–Wallis/Mann–Whitney tests, p > 0.05).
730
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
Fig. 4 – Flexural modulus results (averages and standard deviations) for composites containing different total filler levels and DCPD contents. The same letters indicate lack of statistically significant differences (Kruskal–Wallis/Mann–Whitney tests, p > 0.05).
an overall trend for KIc to increase with the replacement of glass fillers by DCPD. At 40% filler level, successive increases in KIc were observed with 10% and 20% DCPD replacement, while at 60% filler KIc increased significantly only between 0% and 20% DCPD. For composites with 50% filler, the increase was only numerical. No statistically significant differences were observed among materials with similar DCPD contents. After 28 days, statistically significant reductions in KIc were observed only for DCPD-containing composites. All materials containing DCPD presented similar KIc levels, regardless of filler level or DCPD content. For composites with 40% filler,
Fig. 6 – Fracture toughness results (averages and standard deviations) for composites containing different total filler levels and DCPD contents. The same letters indicate lack of statistically significant differences (Kruskal–Wallis/Mann–Whitney tests, p > 0.05).
there was no difference between the control and the DCPD containing composites after 28 days, while at 50% filler the control showed higher KIc than the 20% DCPD and at 60% filler, the control showed higher KIc than the 10% and 20% DCPD composites. Weekly non-cumulative calcium release data presented a large scattering, with coefficients of variation between 31% and 136%. None of the interactions were statistically significant (p > 0.05). Among the main factors, only immersion period had a significant effect on calcium release (p < 0.01, Table 2), meaning that at a given immersion period all
Fig. 5 – Scanning electron microscopy (SEM) images of fractured surfaces of 60% filler content composites. A, B: control; C, D: 10% DCPD; E, F: 20% DCPD.
731
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
Table 2 – Calcium (Ca2+ ) release (in mmol/L) as a function of immersion time and particle content of experimental composites. Due to the high scattering of the data, Tukey test did not detect any statistically significant differences between groups. Please refer to text for further details on the statistical analysis. Time (days)
7 14 21 28
Glass/DCPD (vol%)
Pooled average
30/10
20/20
40/10
30/20
50/10
0.283 (0.227) 0.042 (0.029) 0.240 (0.245) 0.183 (0.205)
0.280 (0.287) 0.050 (0.019) 0.264 (0.287) 0.156 (0.212)
0.298 (0.309) 0.032 (0.019) 0.155 (0.203) 0.169 (0.209)
0.325 (0.223) 0.050 (0.031) 0.112 (0.040) 0.196 (0.214)
0.182 (0.145) 0.048 (0.021) 0.285 (0.321) 0.171 (0.204)
40/20 0.356 (0.261) 0.059 (0.018) 0.209 (0.138) 0.218 (0.243)
0.287 (0.215) A 0.047 (0.021) B 0.211 (0.200) AB 0.182 (0.182) AB
Table 3 – Hydrogen phosphate (HPO4 2− ) release (in mmol/L) as a function of immersion time and particle content of experimental composites. Similar upper-case letters in the same column and lower-case letter in the same row indicate lack of statistically significant differences (three-way ANOVA/Tukey test, p > 0.05). Time (days)
7 14 21 28
Glass/DCPD (vol%) 30/10
20/20
40/10
30/20
50/10
0.083 (0.034) Ab 0.040 (0.006) Ba 0.035 (0.002) Ba 0.040 (0.003) Ba
0.083 (0.013) Ab 0.046 (0.003) Ba 0.037 (0.005) Ba 0.043 (0.005) Ba
0.065 (0.013) Ab 0.033 (0.002) Ba 0.037 (0.004) ABa 0.039 (0.002) ABa
0.092 (0.013) Ab 0.044 (0.005) Ba 0.036 (0.004) Ba 0.054 (0.011) Ba
0.066 (0.006) Ab 0.041 (0.004) Aa 0.050 (0.006) Aa 0.040 (0.004) Aa
composites released similar calcium concentrations, regardless of total filler content or DCPD level. Calcium release decreased significantly between 7 and 14 days. At 21 and 28 days, concentrations were intermediate and statistically similar to those observed in the first two weeks. Hydrogen phosphate (HPO4 2− ) release presented lower values than Ca2+ . The second-order interaction was statistically significant (p < 0.001). With a few exceptions, HPO4 2− release was statistically higher at 7 days and did not decrease further after 14 days. At 7 days only, the composite with 60% total filler and 20% DCPD released more HPO4 2− than the other composites (Table 3).
4.
Discussion
Several in vitro studies have demonstrated the possibility of remineralizing non-cavitated enamel lesions or demineralized dentin surfaces with the use of resin-based materials containing calcium phosphates [27–30]. More recently, an in situ study showed that the enamel adjacent to a composite containing 40 wt% ACP suffered significantly less mineral loss in the presence of biofilm than the enamel next to a nonbioactive control [31]. In spite of these encouraging results, there are many other aspects to be observed in the development of an ion-releasing restorative material. For example, as calcium and phosphate release decrease substantially after a few weeks [9,16,30], their long-term efficacy remains unclear. Also, it is important to evaluate if materials containing bioactive fillers are more prone to degradation than conventional composites, as a result of calcium phosphate particles erosion caused by their contact with solvents. In the present study, DCPD was used as bioactive filler due to its relatively high solubility, granted by a Ca/P = 1 and by the presence of structural water molecules [32]. The proprietary DCPD particles have surface area similar to other commercial nano-structured calcium phosphates (unpublished data). Another interesting characteristic of DCPD concerning its use as bioactive filler in restorative composites is the fact its
40/20 0.166 (0.011) Aa 0.049 (0.004) Ba 0.045 (0.004) Ba 0.060 (0.004) Ba
refractive index (1.54–1.55) [33] is very close to that of barium glass (1.53) [34]. Hydroxyapatite, for example, has a refractive index of 1.63–1.67, which poses a problem both in terms of esthetics and also for photopolymerization [11]. In the present study, the replacement of barium by DCPD did not affect the degree of conversion. This is an important point, as differences found in ion release and mechanical properties, either after 24 h or 28 days, cannot be ascribed to differences in conversion, but to differences in filler type and content. Due to the weakening effect of calcium phosphate particles on the resin matrix [9,10], it is important to keep its content as low as possible without jeopardizing the materials’s remineralizing potential. The replacement of reinforcing fillers by calcium phosphate particles led to reductions in the 24-h fracture strength for the three filler levels tested. Compared to the controls, larger reductions were observed at 20% DCPD (28–35%), compared to 10% DCPD (17–22%). As evidenced by the SEM images of fractured surfaces, the lack of a strong bond between the DCPD particles and the resin cause them to behave as large flaws, facilitating crack initiation and propagation under relatively low loads. Moreover, there were evidences of transparticle fracture, suggestive of low cohesion strength of the agglomerate. Flexural modulus after 24 h was not affected by the replacement of barium glass by DCPD particles, confirming that the bond between the resin matrix and the filler phase is not determinant for the material’s stiffness [35]. In fact, a previous study observed that successive additions of TTCP (tetracalcium phosphate) to an unfilled resin led to a monotonic increase in modulus and, at the same time, reduced the material’s strength [10]. Interestingly, KIc at 24 h increased when barium glass was replaced by DCPD, most noticeably for the 40% filler level. A possible explanation for this finding is the presence of large DCPD agglomerates that, in spite of the lack of chemical bonding with the resin matrix, would cause the deflection of the crack front initiated from the specimen notch. This change of direction reduces the stress intensity factor (K) at the crack tip, postponing its unstable growth [36].
732
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
Additionally, the evidence of transparticle fracture suggests that the energy needed to break the DCPD agglomerate could also contribute to increase KIc . It is important to point out that the opposite effects of DCPD particles on strength and fracture toughness are not contradictory. Strength is influenced by characteristics of the material’s flaw population (size, geometry and location) in relation to tensile stress distribution upon loading. Fracture toughness, on the other hand, is an intrinsic property defined by the material’s microstructure [36]. Therefore, the presence of large DCPD agglomerates in areas subjected to tensile stresses facilitates crack initiation leading to fracture of the biaxial strength specimen, while for the crack initiated from the notch in the KIc specimen, the agglomerate represents an obstacle to its propagation. If after 24 h the replacement of barium glass by DCPD did not show a negative effect on flexural modulus and KIc , the same is not true the aged specimens. Statistically significant reductions in flexural modulus (between 19% and 24%) were observed at both DCPD levels for 40% filler composites and also at 20% DCPD for the 50% and 60% filler materials. While the KIc of the controls were not affected by prolonged storage, the average reductions in KIc for composites containing 10% and 20% DCPD were 24% and 35%, respectively. Fracture strength also dropped after 28 days. In general, the higher the DCPD content, the higher was the reduction in strength (controls: 13–16%, 10% DCPD: 16–28%, 20% DCPD: 27–33%). These results agree with previous investigations showing that the use of non-coupled calcium phosphate particles instead of silanized fillers increases water uptake [37,38]. Consequently, reductions in mechanical properties could be ascribed both to the hydrolytic degradation of the resin matrix and the partial solubilization of the DCPD particles. Calcium orthophosphates can present different Ca/P ratios depending on the valence of the orthophosphate anion (H2 PO4 − , HPO4 2− or PO4 3− ). As previously mentioned, in general, the lower the Ca/P of the compound, the higher is its solubility [39]. However, there are other aspects that define the particle’s efficacy as ion source. A high surface area, for example, is important to optimize ion release [16]. Actually, it seems more important than the type of calcium phosphate used (unpublished data). Besides the surface area of the agglomerate, ion release from resin-based materials containing calcium phosphates is influenced by percentage of bioactive particles in the material [9], pH of the immersion medium [19], and also to the hydrophilicity of the resin matrix [40]. Previous studies found that ion release was not linearly related to particle content [41]. In fact, the present study showed similar calcium release among composites containing 10% and 20% DCPD, regardless of the total filler level. The average calcium concentration released after 7 days (0.287 mmol/L) was similar to those reported in previous studies using ACP and more hydrophilic matrices [10,21]. The lack of statistically significant differences among materials may be ascribed to the high scattering of the data found in some experimental groups, which, in turn, could be attributed to the inhomogeneous dispersion of the agglomerates in the resin, associated to the fact that calcium ions are more easily released from the structural phosphate tetrahedra. Is it important to point out that, though calcium release was statistically similar among 7 and 28 days, a numerical reduction of 37% was observed. This
trend of reduction in calcium release raises the concern if in the long term calcium release would be enough to grant any protection to the tooth. In agreement with previous studies, the concentration of hydrogen phosphate released was lower than calcium [10,21]. This can be explained by its polyatomic nature, as well as its structural role in the compound. In summary, DC was not affected by the presence of DCPD. After 24 h the replacement of barium glass by DCPD reduced strength for all filler levels, but had no effect on flexural modulus and actually increased KIc . Therefore, the first hypothesis cannot be fully rejected. More importantly, after 28 days storage in water, the DCPD-containing composites presented more severe reductions in properties than the controls. Overall, the higher the DCPD content, the higher was the reduction in mechanical properties. Regardless of the DCPD level and total filler content, similar concentrations of calcium and (with a few exceptions) hydrogen phosphate were released by the composites. Based on the above, the second and the third hypotheses can be rejected. Finally, it is possible to conclude that the best compromise between ion release and overall mechanical properties was represented by the composite containing the highest total filler level and the lowest bioactive filler content.
Acknowledgements The authors would like to thank FAPESP (The State of São Paulo Research Foundation), grants 2012/04532-4 and 2012/25253-6, and CNPq (National Council for Scientific and Technological Development), grant 300548/2012-5.
references
[1] Jandt KD, Sigusch BW. Future perspectives of resin-based dental materials. Dent Mater 2009;25:1001–6. [2] Chen MH. Update on dental nanocomposites. J Dent Res 2010;89:549–60. [3] Ferracane JL. Resin composite–state of the art. Dent Mater 2011;27:29–38. [4] Demarco FF, Correa MB, Cenci MS, Moraes RR, Opdam NJ. Longevity of posterior composite restorations: not only a matter of materials. Dent Mater 2012;28:87–101. [5] Rasines Alcaraz MG, Veitz-Keenan A, Sahrmann P, Schmidlin PR, Davis D, Iheozor-Ejiofor Z. Direct composite resin fillings versus amalgam fillings for permanent or adult posterior teeth. Cochrane Database Syst Rev 2014;3:CD005620. [6] Antonucci JM, Misra DN, Peckoo RJ, The Accelerative. Adhesive bonding capabilities of surface-active accelerators. J Dent Res 1981;60:1332–42. [7] Dunn WJ. Shear bond strength of an amorphous calcium-phosphate-containing orthodontic resin cement. Am J Orthod Dentofacial Orthop 2007;131:243–7. [8] Silva KG, Pedrini D, Delbem AC, Ferreira L, Cannon M. In situ evaluation of the remineralizing capacity of pit and fissure sealants containing amorphous calcium phosphate and/or fluoride. Acta Odontol Scand 2010;68:11–8. [9] Skrtic D, Antonucci JM, Eanes ED. Improved properties of amorphous calcium phosphate fillers in remineralizing resin composites. Dent Mater 1996;12:295–301. [10] Xu HH, Moreau JL. Dental glass-reinforced composite for caries inhibition: calcium phosphate ion release and
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
[11]
[12]
[13] [14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
mechanical properties. J Biomed Mater Res B: Appl Biomater 2010;92:332–40. Arcis RW, Lopez-Macipe A, Toledano M, Osorio E, Rodriguez-Clemente R, Murtra J, et al. Mechanical properties of visible light-cured resins reinforced with hydroxyapatite for dental restoration. Dent Mater 2002;18:49–57. Chen WC, Wu HY, Chen HS. Evaluation of reinforced strength and remineralized potential of resins with nanocrystallites and silica modified filler surfaces. Mater Sci Eng C Mater Biol Appl 2013;33:1143–51. Labella R, Braden M, Deb S. Novel hydroxyapatite-based dental composites. Biomaterials 1994;15:1197–200. O’Donnell JN, Schumacher GE, Antonucci JM, Skrtic D. Structure-composition-property relationships in polymeric amorphous calcium phosphate-based dental composites. Materials (Basel) 2009;2:1929–59. Xu HH, Quinn JB. Whisker-reinforced bioactive composites containing calcium phosphate cement fillers: effects of filler ratio and surface treatments on mechanical properties. J Biomed Mater Res 2001;57:165–74. Xu HH, Weir MD, Sun L. Nanocomposites with Ca and PO4 release: effects of reinforcement, dicalcium phosphate particle size and silanization. Dent Mater 2007;23:1482–91. Xu HH, Sun L, Weir MD, Antonucci JM, Takagi S, Chow LC, et al. Nano DCPA-whisker composites with high strength and Ca and PO(4) release. J Dent Res 2006;85: 722–7. Xu HH, Weir MD, Sun L, Takagi S, Chow LC. Effects of calcium phosphate nanoparticles on Ca-PO4 composite. J Dent Res 2007;86:378–83. Xu HH, Weir MD, Sun L. Calcium and phosphate ion releasing composite: effect of pH on release and mechanical properties. Dent Mater 2009;25:535–42. Xu HH, Weir MD, Sun L, Ngai S, Takagi S, Chow LC. Effect of filler level and particle size on dental caries-inhibiting Ca-PO(4) composite. J Mater Sci Mater Med 2009;20: 1771–9. Xu HH, Moreau JL, Sun L, Chow LC. Nanocomposite containing amorphous calcium phosphate nanoparticles for caries inhibition. Dent Mater 2011;27:762–9. Moreau JL, Sun L, Chow LC, Xu HH. Mechanical and acid neutralizing properties and bacteria inhibition of amorphous calcium phosphate dental nanocomposite. J Biomed Mater Res B: Appl Biomater 2011;98:80–8. Moreau JL, Weir MD, Giuseppetti AA, Chow LC, Antonucci JM, Xu HH. Long-term mechanical durability of dental nanocomposites containing amorphous calcium phosphate nanoparticles. J Biomed Mater Res B: Appl Biomater 2012;100:1264–73. Marovic D, Tarle Z, Hiller KA, Muller R, Ristic M, Rosentritt M, et al. Effect of silanized nanosilica addition on remineralizing and mechanical properties of experimental composite materials with amorphous calcium phosphate. Clin Oral Invest 2014;18:783–92. Marovic D, Tarle Z, Hiller KA, Muller R, Rosentritt M, Skrtic D, et al. Reinforcement of experimental composite materials
[26] [27]
[28]
[29]
[30]
[31]
[32] [33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
733
based on amorphous calcium phosphate with inert fillers. Dent Mater 2014;30:1052–60. Craig RG. Restorative dental materials. 7th ed St. Louis: Mosby; 1985. Langhorst SE, O’Donnell JN, Skrtic D. In vitro remineralization of enamel by polymeric amorphous calcium phosphate composite: quantitative microradiographic study. Dent Mater 2009;25:884–91. Weir MD, Chow LC, Xu HH. Remineralization of demineralized enamel via calcium phosphate nanocomposite. J Dent Res 2012;91:979–84. Skrtic D, Hailer AW, Takagi S, Antonucci JM, Eanes ED. Quantitative assessment of the efficacy of amorphous calcium phosphate/methacrylate composites in remineralizing caries-like lesions artificially produced in bovine enamel. J Dent Res 1996;75:1679–86. Dickens SH, Flaim GM, Takagi S. Mechanical properties and biochemical activity of remineralizing resin-based Ca-PO4 cements. Dent Mater 2003;19:558–66. Melo MA, Weir MD, Rodrigues LK, Xu HH. Novel calcium phosphate nanocomposite with caries-inhibition in a human in situ model. Dent Mater 2013;29:231–40. LeGeros RZ. Calcium phosphates in oral biology and medicine. Basel, Switzerland: Karger; 1991. Anthony JW, Bideaux RA, Bladh KW, Nichols MC, editors. Handbook of Mineralogy. Chantilly, VA, USA: Mineralogical Society of America; 2000. http://www. handbookofmineralogy.org/ Shortall AC, Palin WM, Burtscher P. Refractive index mismatch and monomer reactivity influence composite curing depth. J Dent Res 2008;87:84–8. Beatty MW, Swartz ML, Moore BK, Phillips RW, Roberts TA. Effect of microfiller fraction and silane treatment on resin composite properties. J Biomed Mater Res 1998;40:12–23. Cesar PF, Yoshimura HN, Miranda Jr WG, Miyazaki CL, Muta LM, Rodrigues Filho LE. Relationship between fracture toughness and flexural strength in dental porcelains. J Biomed Mater Res B: Appl Biomater 2006;78:265–73. Skrtic D, Antonucci JM. Effect of bifunctional comonomers on mechanical strength and water sorption of amorphous calcium phosphate- and silanized glass-filled Bis-GMA-based composites. Biomaterials 2003;24:2881–8. Antonucci JM, Skrtic D. Fine-tuning of polymeric resins and their interfaces with amorphous calcium phosphate. A strategy for designing effective remineralizing dental composites. Polymers (Basel) 2010;2:378–92. LeGeros RZ. Calcium phosphates in demineralization/remineralization processes. J Clin Dent 1999;10:65–73. Skrtic D, Antonucci JM. Dental composites based on amorphous calcium phosphate-resin composition/physicochemical properties study. J Biomater Appl 2007;21:375–93. Xu HH, Weir MD, Sun L, Moreau JL, Takagi S, Chow LC, et al. Strong nanocomposites with Ca, PO(4), and F release for caries inhibition. J Dent Res 2010;89:19–28.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.intl.elsevierhealth.com/journals/dema
Mechanical properties and ion release from bioactive restorative composites containing glass fillers and calcium phosphate nano-structured particles Marina D.S. Chiari, Marcela C. Rodrigues, Tathy A. Xavier, Eugen M.N. de Souza, Victor E. Arana-Chavez, Roberto R. Braga ∗ Department of Biomaterials and Oral Biology, Univ. of São Paulo School of Dentistry, Av. Prof. Lineu Prestes, 2227, São Paulo, SP 05508-000, Brazil
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objective. To evaluate the effect of the replacement of barium glass by dicalcium phosphate
Received 29 September 2014
dihydrate (DCPD) particles on the mechanical properties and degree of conversion (DC) of
Received in revised form
composites. Additionally, calcium and hydrogen phosphate (HPO4 2− ) release were followed
2 December 2014
for 28 days.
Accepted 30 March 2015
Methods. Nine composites containing equal parts (in mols) of BisGMA and TEGDMA and 40, 50 or 60 vol% of total filler were manipulated. Filler phase was constituted by silanated barium glass and 0%, 10% or 20% of DCPD particles. DC was determined by near-FTIR. Biaxial
Keywords:
flexural strength (BFS) and modulus (E) were tested using the “piston on three balls” method,
Calcium phosphate
while fracture toughness (KIc ) used the “single edge notched beam” method. Specimens were
Mechanical properties
tested after 24 h and 28 days in water. Ion release was determined using inductively coupled
Ion release
plasma optical emission spectrometry (ICP-OES). Data were analyzed by ANOVA/Tukey (DC
Resin composite
and ion release) or Kruskal–Wallis/Mann–Whitney (mechanical properties; alpha: 5%). Results. DC was not affected by DCPD. The presence of DCPD reduced BFS for both storage times, while differences in E became evident after 28 days. After 24 h, KIc increased with the addition of DCPD; after 28 days, however, KIc decreased only for DCPD-containing composites. Calcium release was similar for both DCPD contents and remained fairly constant during the 28-day period. Overall, HPO4 2− release was higher at 7 days and did not decrease after 14 days. Significance. The composite with the highest filler level and 10% DCPD represented the best compromise between mechanical properties after aging in water and ion release. © 2015 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
∗ Corresponding author at: Faculdade de Odontologia da USP, Depto. de Biomateriais e Biologia Oral, Av. Prof. Lineu Prestes, 2227, São Paulo, SP 05508-000, Brazil. Tel.: +55 11 3091 7840x224; fax: +55 11 3091 7840x224. E-mail address: [email protected] (R.R. Braga).
http://dx.doi.org/10.1016/j.dental.2015.03.015 0109-5641/© 2015 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
1.
Introduction
The development of permanent restorative materials with remineralizing or antimicrobial activity has been an aspiration in restorative dentistry as a possible path to reduce the risk of secondary caries [1–3], one of the main causes of restoration failure and replacement [4,5]. The first attempts to incorporate calcium orthophosphate particles in dimethacrylate resins as ion source for remineralization date back several decades [6], but only recently orthodontic cements and sealants containing amorphous calcium phosphate (ACP) nanoparticles became available for clinical use [7,8]. Early on, the difficulties of combining high fracture strength with high levels of ion release became evident. Due to the lack of a chemical bond between the calcium phosphate particles and the resin phase, the former behave as large flaws (i.e. several micrometers in width), concentrating stress and acting as crack initiation sites. In fact, when 40% (by weight) of calcium phosphate particles were added to a dimethacrylate resin, its fracture strength was reduced by approximately 50% [9,10]. Interestingly, elastic modulus is not influenced by the bond (or the lack of thereof) between resin and particle phases, as successive additions of tetracalcium phosphate (TTCP) particles to an unfilled resin led to cumulative increases in modulus [10]. Several attempts were made to improve the fracture strength of resin-based materials containing calcium phosphates. In broad terms, the tested approaches can be divided into two groups, namely, modification/functionalization of calcium phosphate particles and association of ion-releasing fillers with a reinforcing phase. The addition of the glass forming compounds tetraethoxysilane (TEOS) or zirconyl chloride in the synthesis of ACP resulted in 25–33% increase in the fracture strength of materials containing 40 wt% of filler, in comparison with those containing unmodified ACP particles [9]. Hydroxyapatite particles functionalized with acrylic acid increased the flexural strength of a BisGMA/TEGDMA-based material by 75% in comparison to the use of non-functionalized particles [11]. Finally, several studies tested the effect of silanization of calcium phosphate particles. Though these studies consistently reported higher fracture strengths for composites containing silanized calcium phosphate particles in comparison to unmodified particles [12–16], some of them also found that silanization hindered ion release [12,14,16]. The association of calcium phosphate particles with a reinforcing phase in resin-based materials involved the use of silanated, silica-fused silicon carbide or silicon nitride whiskers [15–20], or silanated glass particles [10,21–25]. The use of whiskers as reinforcing phase in resin-based materials containing calcium phosphates was very effective from a mechanical standpoint [16,17]. In fact, the flexural strength of composites containing equal mass fractions of calcium phosphate and whiskers matched that of a commercial hybrid composite with similar filler fraction, while their flexural modulus was actually higher [18–20]. However, these materials have the drawback of being opaque and, therefore, had to be formulated as two-paste systems.
727
The weakening effect of calcium phosphate particles on resins can be exemplified by a study reporting that 20–30 wt% of silanated barium glass had to be added to a resin matrix containing 40 wt% TTCP in order to bring its strength to the level of the unfilled resin [10]. Only a few studies tested the effect of replacing glass fillers by calcium phosphate particles. For instance, the replacement of 15% of glass particles by ACP in experimental composites containing 75 wt% of filler led to increased wear [23], and a significant reduction in strength, but had no effect on elastic modulus [21]. Another important issue is the effect of calcium phosphate particles on composite degradation over time. Available information in the literature suggests that prolonged water immersion had similar effects on composites containing only glass particles (75 wt%) or up to 20 wt% of ACP particles [23]. Considering the effectiveness of glass fillers as reinforcing phase for stress-bearing restorative composites, it is important to determine the percentage of glass fillers that could be replaced by calcium phosphate particles without jeopardizing the mechanical properties and, at the same time, ensuring enough ion release for remineralization of caries lesions. Based on the scant amount of information on the effects of replacing silanated glass fillers with bioactive particles, the objective of the present study was to evaluate how degree of conversion, mechanical properties and ion release of experimental composites were affected when dicalcium phosphate dihydrate (DCPD) particles were incorporated in the resin matrix at the expense of reinforcing fillers. Different total filler levels were tested, as well as the effect of prolonged aging in water. The null hypotheses were: (1) the replacement of silanated glass fillers by DCPD does not affect the degree of conversion or the mechanical properties, regardless of the total filler content or DCPD levels considered, (2) the effects of prolonged immersion in water on mechanical properties are not influenced by the total filler content or by the presence of DCPD, and (3) ion release is similar among total filler content and DCPD levels.
2.
Method
2.1.
Composite formulation
Nine composites were prepared, constituted by a 1:1 ratio, in mols, of bisphenol-A glycidyl dimethacrylate (BisGMA) and triethylene glycol dimethacrylate (TEGDMA), 0.5% (by weight) of camphorquinone and ethyl-4-dimethylamino benzoate (EDMAB, all components from Sigma-Aldrich, St. Louis, MO, USA). The experimental design was based on a factorial model with total filler level (40%, 50% or 60% by volume) and DCPD content (0%, 10% or 20% by volume) as main factors. Silanated barium glass (D50 : 0.5 m) and proprietary DCPD particles (D50 : 8.0 m, CaHPO4 ·2H2 O, Labsynth, Diadema, SP, Brazil) were used. Their filler size distributions were determined by laser light scattering (Mastersizer 2000, Malvern Instruments Ltd., Malvern, UK) and are shown in Fig. 1. A transmission electron microscopy image of part of a nanostructured agglomerate is shown in Fig. 2. Particle densities (barium glass: 2.4 g/cm3 ; DCPD: 2.7 g/cm3 ), necessary to calculate the amounts used in the tested formulations, were
728
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
the absorption of the C–H2 group) of the polymerized and the unpolimerized material.
2.3.
Fig. 1 – Filler size distributions for barium glass and DCPD particles.
Biaxial flexural test
Disk-shaped specimens (12 mm × 1.2 mm, n = 20) were built using a stainless steel split mold. Photoactivation was performed by overlapping four 10-s irradiations, one per quadrant. Half of the specimens were stored in water at 37 ◦ C for 24 h, while the other half was stored for 28 days. After storage, the specimens were placed on a “piston on three balls” setup positioned on a universal testing machine (model 5565, Instron Corp., Norwood, MA, USA). In this setup, the specimen was supported by three 2.5-mm diameter steel spheres distributed on a 10-mm diameter circumference 120◦ apart from each other, and the load was applied by a flat piston with 1.2 mm in diameter. The discs were positioned in a way that the irradiated surface would be under compressive stresses and fractured at a cross-head speed of 0.5 mm/min. Deflection at the center of the specimen was monitored by a deflectometer (model W-E401-E, Instron Corp, USA). Biaxial flexural strength (BFS) in MPa, was calculated according to the following equations: BFS =
−0.2387P(X − Y) b2
X = (1 + v) ln
r 2 2
r3
Y = (1 − v) 1 − ln Fig. 2 – Transmission electron microscopy (TEM) image showing part of a DCPD agglomerate. Plate-like nanoparticles are distinguishable at the agglomerate’s fringe (200,000× magnification).
determined in a Helium picnometer (Ultrapyc 1200e, Quantachrome Instruments, Boynton Beach, FL, USA). Composites were mechanically mixed under vacuum (Speedmixer DAC 150.1 FVZ-K, FlackTek Inc., Landrum, SC, USA) and kept under refrigeration until 2 h before use.
2.2.
Degree of conversion
Degree of conversion (n = 3) was determined by Fouriertransformed near-infrared spectroscopy (near-FTIR). The unpolimerized material was placed in a silicone mold with 7 mm in diameter and 1 mm in thickness and both were pressed between two microscopy glass slides. The spectrum of the unpolimerized material was obtained between 4000 and 10,000 cm−1 by the co-addition of 32 scans with 4 cm−1 resolution (Vertex 70; Bruker Optik GmbH, Ettlingen, Germany). Then, the composite was photoactivated through the glass slide (Radii Cal, SDI, Bayswater, Australia), receiving 24 J/cm2 (1200 mW/cm2 for 20 s) and the whole setup was dry-stored at 37 ◦ C. After 24 h, a new spectrum was obtained. Degree of conversion was calculated as the ratio between the areas of the absorption band located at 6165 cm−1 (corresponding to
+
1 − v r 2
r 2 1
r2
2
2 + (1 − v)
r3
r 2 1
r3
where P is the fracture load (in Newtons); b is the specimen thickness (in mm); v Poisson’s ratio; r1 is the radius of the supporting spheres circumference (5.0 mm); r2 is the radius of the loading piston (0.6 mm); r3 is the specimen radius (in mm). A Poisson’s ratio of 0.3 was adopted for all composites [26]. Flexural modulus was calculated according to the following equation:
E=
ˇP˛2 ωh3
× 0.001
where E is the flexural modulus (in GPa), ˇ is a constant related to the deflection at the center of the disk (0.509), P is the load (in Newtons), ˛ is the disk radius (in mm), ω is the deflection corresponding to P and h is the disk thickness (in mm). Surface fractures were gold-sputtered and observed under scanning electron microscopy (LEO 430, LEO Electron Microscopy Ltd., Cambridge, UK).
2.4.
Fracture toughness
Fracture toughness (KIc ) was determined using the “singleedge notched beam” method (SENB). Specimens were built using a stainless steel split mold (25.0 mm in length, 5.0 mm in width and 2.8 mm in thickness, n = 20) with a razor blade inserted to produce a sharp notch at mid-point of specimen’s length. The composite was covered by a mylar strip and photoactivated by three adjacent exposures of 40 s (Radii-Cal, SDI). After being stored in distilled water at 37 ◦ C for 24 h or 28 days,
729
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
specimens were tested in three-point bending (span: 20 mm) in a universal testing machine (model 5565, Instron Corp.) at a cross-head speed of 0.13 mm/min, loading the notched surface in tension. The fractured surfaces were analyzed by a stereomicroscope with a CCD camera (model SZ61, Olympus, Tokyo, Japan) at 90× magnification. The width (w) and thickness (b) of the specimen, and the notch length (a) were measured using the ImageJ software (National Institute of Health, Bethesda, USA). Six notch length measurements (three on each fracture surface) were averaged and only the specimens with a/w ratio between 0.45 and 0.50 were used. KIc , in MPa m0.5 , was calculated using the following equations: KIc =
P×S b × w1.5
×f
a w
where P is the fracture load (in Newtons) and S is the span (in meters). The f(a/w) value was calculated as follows: f
a w
2.5.
Total filler content (vol%)
40 50 60
DCPD content (vol%) 0
10
77.9 (2.6) A 81.0 (3.5) A 78.5 (0.4) A
80.0 (2.7) A 78.0 (3.2) A 77.2 (0.8) A
20 78.3 (2.8) A 79.4 (1.8) A 75.1 (0.8) A
levels and both storage times evaluated. Further reductions were observed between composites containing 10% and 20%, except after 24 h at 60% total filler level (p < 0.001). Prolonged storage significantly reduced BFS for all the tested materials. After 24-h storage, the replacement of reinforcing fillers by
2 1.99 − (a/w) × [1 − (a/w)] × 2.15 − 3.93(a/w) + 2.7(a/w) q 3 =
3/2 w
2 × [1 + 2(a/w)] × 1 − (a/w)
Ion release
Specimens (5 mm in diameter, 1 mm in thickness, n = 3) were built and dry-stored at 37 ◦ C for 24 h. Then, they were individually immersed in 5 mL of sodium chloride 133 mmol/L buffered to pH 7.0 using 50 mmol/L of HEPES solution [19]. The immersion medium was changed weekly and the concentrations of Ca2+ and HPO4 2− were determined, non-cumulatively, after 7, 14, 21 and 28 days using inductively coupled plasma optical emission spectrometry (ICP-OES, Agilent Technologies, Santa Clara, CA, USA). Calibration curves for calcium and phosphorous were prepared using different dilutions of standard solutions (SpecSol Quimlab, Jacareí, SP, Brazil). Previous to the analysis, the samples were acidified using a 10% HNO3 solution (1:1 in volume) and filtered. Calcium and phosphorous readings were obtained simultaneously using the spectral emissions of 184 nm and 177 nm, respectively.
2.6.
Table 1 – Averages and standard deviations of degree of conversion (in %) as a function of total filler and DCPD content (ANOVA/Tukey test, p > 0.05).
DCPD did not reduce E in any of the filler levels (Fig. 4). However, after 28 days, the composites containing 40% filler and 10% or 20% DCPD, as well as the composite with 60% total filler and 20% DCPD presented statistically lower E than the respective controls (p < 0.001). SEM images of the fracture surfaces of DCPD-containing composites show large regularly shaped flaws varying from a few micrometers up to 30 m, corresponding to sites where the DCPD agglomerates detached from the resin (Fig. 5). At higher magnification, it is possible to notice that most of these spaces contain lamellar structures, corresponding to areas of transparticle fracture. No marked differences were observed between specimens fractured at different storage periods. KIc averages and standard deviations are shown in Fig. 6. For the control groups, after 24 h KIc increased significantly only between 40% and 50% filler. Also in the 24-h groups, there was
Statistical analysis
Data were subjected to normality and homoscedasticity tests. Degree of conversion and ion release were subjected to two-way and three-way ANOVA, respectively, and multiple comparisons were performed using Tukey’s test. Due to the lack of homoscedasticity, data from mechanical tests were analyzed using non-parametric tests (Kruskal–Wallis and Mann–Whitney for pair-wise comparisons). In all cases, the global significance level was 5%.
3.
Results
Degree of conversion results are shown in Table 1. No statistically significant differences were observed among experimental groups, with average values ranging from 75% to 81% (p = 0.161). Biaxial flexural strength (BFS) results are shown in Fig. 3. The replacement of 10% barium glass by DCPD resulted in statistically significant reductions in fracture strength in all filler
Fig. 3 – Biaxial flexural strength results (averages and standard deviations) for composites containing different total filler levels and DCPD contents. The same letters indicate lack of statistically significant differences (Kruskal–Wallis/Mann–Whitney tests, p > 0.05).
730
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
Fig. 4 – Flexural modulus results (averages and standard deviations) for composites containing different total filler levels and DCPD contents. The same letters indicate lack of statistically significant differences (Kruskal–Wallis/Mann–Whitney tests, p > 0.05).
an overall trend for KIc to increase with the replacement of glass fillers by DCPD. At 40% filler level, successive increases in KIc were observed with 10% and 20% DCPD replacement, while at 60% filler KIc increased significantly only between 0% and 20% DCPD. For composites with 50% filler, the increase was only numerical. No statistically significant differences were observed among materials with similar DCPD contents. After 28 days, statistically significant reductions in KIc were observed only for DCPD-containing composites. All materials containing DCPD presented similar KIc levels, regardless of filler level or DCPD content. For composites with 40% filler,
Fig. 6 – Fracture toughness results (averages and standard deviations) for composites containing different total filler levels and DCPD contents. The same letters indicate lack of statistically significant differences (Kruskal–Wallis/Mann–Whitney tests, p > 0.05).
there was no difference between the control and the DCPD containing composites after 28 days, while at 50% filler the control showed higher KIc than the 20% DCPD and at 60% filler, the control showed higher KIc than the 10% and 20% DCPD composites. Weekly non-cumulative calcium release data presented a large scattering, with coefficients of variation between 31% and 136%. None of the interactions were statistically significant (p > 0.05). Among the main factors, only immersion period had a significant effect on calcium release (p < 0.01, Table 2), meaning that at a given immersion period all
Fig. 5 – Scanning electron microscopy (SEM) images of fractured surfaces of 60% filler content composites. A, B: control; C, D: 10% DCPD; E, F: 20% DCPD.
731
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
Table 2 – Calcium (Ca2+ ) release (in mmol/L) as a function of immersion time and particle content of experimental composites. Due to the high scattering of the data, Tukey test did not detect any statistically significant differences between groups. Please refer to text for further details on the statistical analysis. Time (days)
7 14 21 28
Glass/DCPD (vol%)
Pooled average
30/10
20/20
40/10
30/20
50/10
0.283 (0.227) 0.042 (0.029) 0.240 (0.245) 0.183 (0.205)
0.280 (0.287) 0.050 (0.019) 0.264 (0.287) 0.156 (0.212)
0.298 (0.309) 0.032 (0.019) 0.155 (0.203) 0.169 (0.209)
0.325 (0.223) 0.050 (0.031) 0.112 (0.040) 0.196 (0.214)
0.182 (0.145) 0.048 (0.021) 0.285 (0.321) 0.171 (0.204)
40/20 0.356 (0.261) 0.059 (0.018) 0.209 (0.138) 0.218 (0.243)
0.287 (0.215) A 0.047 (0.021) B 0.211 (0.200) AB 0.182 (0.182) AB
Table 3 – Hydrogen phosphate (HPO4 2− ) release (in mmol/L) as a function of immersion time and particle content of experimental composites. Similar upper-case letters in the same column and lower-case letter in the same row indicate lack of statistically significant differences (three-way ANOVA/Tukey test, p > 0.05). Time (days)
7 14 21 28
Glass/DCPD (vol%) 30/10
20/20
40/10
30/20
50/10
0.083 (0.034) Ab 0.040 (0.006) Ba 0.035 (0.002) Ba 0.040 (0.003) Ba
0.083 (0.013) Ab 0.046 (0.003) Ba 0.037 (0.005) Ba 0.043 (0.005) Ba
0.065 (0.013) Ab 0.033 (0.002) Ba 0.037 (0.004) ABa 0.039 (0.002) ABa
0.092 (0.013) Ab 0.044 (0.005) Ba 0.036 (0.004) Ba 0.054 (0.011) Ba
0.066 (0.006) Ab 0.041 (0.004) Aa 0.050 (0.006) Aa 0.040 (0.004) Aa
composites released similar calcium concentrations, regardless of total filler content or DCPD level. Calcium release decreased significantly between 7 and 14 days. At 21 and 28 days, concentrations were intermediate and statistically similar to those observed in the first two weeks. Hydrogen phosphate (HPO4 2− ) release presented lower values than Ca2+ . The second-order interaction was statistically significant (p < 0.001). With a few exceptions, HPO4 2− release was statistically higher at 7 days and did not decrease further after 14 days. At 7 days only, the composite with 60% total filler and 20% DCPD released more HPO4 2− than the other composites (Table 3).
4.
Discussion
Several in vitro studies have demonstrated the possibility of remineralizing non-cavitated enamel lesions or demineralized dentin surfaces with the use of resin-based materials containing calcium phosphates [27–30]. More recently, an in situ study showed that the enamel adjacent to a composite containing 40 wt% ACP suffered significantly less mineral loss in the presence of biofilm than the enamel next to a nonbioactive control [31]. In spite of these encouraging results, there are many other aspects to be observed in the development of an ion-releasing restorative material. For example, as calcium and phosphate release decrease substantially after a few weeks [9,16,30], their long-term efficacy remains unclear. Also, it is important to evaluate if materials containing bioactive fillers are more prone to degradation than conventional composites, as a result of calcium phosphate particles erosion caused by their contact with solvents. In the present study, DCPD was used as bioactive filler due to its relatively high solubility, granted by a Ca/P = 1 and by the presence of structural water molecules [32]. The proprietary DCPD particles have surface area similar to other commercial nano-structured calcium phosphates (unpublished data). Another interesting characteristic of DCPD concerning its use as bioactive filler in restorative composites is the fact its
40/20 0.166 (0.011) Aa 0.049 (0.004) Ba 0.045 (0.004) Ba 0.060 (0.004) Ba
refractive index (1.54–1.55) [33] is very close to that of barium glass (1.53) [34]. Hydroxyapatite, for example, has a refractive index of 1.63–1.67, which poses a problem both in terms of esthetics and also for photopolymerization [11]. In the present study, the replacement of barium by DCPD did not affect the degree of conversion. This is an important point, as differences found in ion release and mechanical properties, either after 24 h or 28 days, cannot be ascribed to differences in conversion, but to differences in filler type and content. Due to the weakening effect of calcium phosphate particles on the resin matrix [9,10], it is important to keep its content as low as possible without jeopardizing the materials’s remineralizing potential. The replacement of reinforcing fillers by calcium phosphate particles led to reductions in the 24-h fracture strength for the three filler levels tested. Compared to the controls, larger reductions were observed at 20% DCPD (28–35%), compared to 10% DCPD (17–22%). As evidenced by the SEM images of fractured surfaces, the lack of a strong bond between the DCPD particles and the resin cause them to behave as large flaws, facilitating crack initiation and propagation under relatively low loads. Moreover, there were evidences of transparticle fracture, suggestive of low cohesion strength of the agglomerate. Flexural modulus after 24 h was not affected by the replacement of barium glass by DCPD particles, confirming that the bond between the resin matrix and the filler phase is not determinant for the material’s stiffness [35]. In fact, a previous study observed that successive additions of TTCP (tetracalcium phosphate) to an unfilled resin led to a monotonic increase in modulus and, at the same time, reduced the material’s strength [10]. Interestingly, KIc at 24 h increased when barium glass was replaced by DCPD, most noticeably for the 40% filler level. A possible explanation for this finding is the presence of large DCPD agglomerates that, in spite of the lack of chemical bonding with the resin matrix, would cause the deflection of the crack front initiated from the specimen notch. This change of direction reduces the stress intensity factor (K) at the crack tip, postponing its unstable growth [36].
732
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
Additionally, the evidence of transparticle fracture suggests that the energy needed to break the DCPD agglomerate could also contribute to increase KIc . It is important to point out that the opposite effects of DCPD particles on strength and fracture toughness are not contradictory. Strength is influenced by characteristics of the material’s flaw population (size, geometry and location) in relation to tensile stress distribution upon loading. Fracture toughness, on the other hand, is an intrinsic property defined by the material’s microstructure [36]. Therefore, the presence of large DCPD agglomerates in areas subjected to tensile stresses facilitates crack initiation leading to fracture of the biaxial strength specimen, while for the crack initiated from the notch in the KIc specimen, the agglomerate represents an obstacle to its propagation. If after 24 h the replacement of barium glass by DCPD did not show a negative effect on flexural modulus and KIc , the same is not true the aged specimens. Statistically significant reductions in flexural modulus (between 19% and 24%) were observed at both DCPD levels for 40% filler composites and also at 20% DCPD for the 50% and 60% filler materials. While the KIc of the controls were not affected by prolonged storage, the average reductions in KIc for composites containing 10% and 20% DCPD were 24% and 35%, respectively. Fracture strength also dropped after 28 days. In general, the higher the DCPD content, the higher was the reduction in strength (controls: 13–16%, 10% DCPD: 16–28%, 20% DCPD: 27–33%). These results agree with previous investigations showing that the use of non-coupled calcium phosphate particles instead of silanized fillers increases water uptake [37,38]. Consequently, reductions in mechanical properties could be ascribed both to the hydrolytic degradation of the resin matrix and the partial solubilization of the DCPD particles. Calcium orthophosphates can present different Ca/P ratios depending on the valence of the orthophosphate anion (H2 PO4 − , HPO4 2− or PO4 3− ). As previously mentioned, in general, the lower the Ca/P of the compound, the higher is its solubility [39]. However, there are other aspects that define the particle’s efficacy as ion source. A high surface area, for example, is important to optimize ion release [16]. Actually, it seems more important than the type of calcium phosphate used (unpublished data). Besides the surface area of the agglomerate, ion release from resin-based materials containing calcium phosphates is influenced by percentage of bioactive particles in the material [9], pH of the immersion medium [19], and also to the hydrophilicity of the resin matrix [40]. Previous studies found that ion release was not linearly related to particle content [41]. In fact, the present study showed similar calcium release among composites containing 10% and 20% DCPD, regardless of the total filler level. The average calcium concentration released after 7 days (0.287 mmol/L) was similar to those reported in previous studies using ACP and more hydrophilic matrices [10,21]. The lack of statistically significant differences among materials may be ascribed to the high scattering of the data found in some experimental groups, which, in turn, could be attributed to the inhomogeneous dispersion of the agglomerates in the resin, associated to the fact that calcium ions are more easily released from the structural phosphate tetrahedra. Is it important to point out that, though calcium release was statistically similar among 7 and 28 days, a numerical reduction of 37% was observed. This
trend of reduction in calcium release raises the concern if in the long term calcium release would be enough to grant any protection to the tooth. In agreement with previous studies, the concentration of hydrogen phosphate released was lower than calcium [10,21]. This can be explained by its polyatomic nature, as well as its structural role in the compound. In summary, DC was not affected by the presence of DCPD. After 24 h the replacement of barium glass by DCPD reduced strength for all filler levels, but had no effect on flexural modulus and actually increased KIc . Therefore, the first hypothesis cannot be fully rejected. More importantly, after 28 days storage in water, the DCPD-containing composites presented more severe reductions in properties than the controls. Overall, the higher the DCPD content, the higher was the reduction in mechanical properties. Regardless of the DCPD level and total filler content, similar concentrations of calcium and (with a few exceptions) hydrogen phosphate were released by the composites. Based on the above, the second and the third hypotheses can be rejected. Finally, it is possible to conclude that the best compromise between ion release and overall mechanical properties was represented by the composite containing the highest total filler level and the lowest bioactive filler content.
Acknowledgements The authors would like to thank FAPESP (The State of São Paulo Research Foundation), grants 2012/04532-4 and 2012/25253-6, and CNPq (National Council for Scientific and Technological Development), grant 300548/2012-5.
references
[1] Jandt KD, Sigusch BW. Future perspectives of resin-based dental materials. Dent Mater 2009;25:1001–6. [2] Chen MH. Update on dental nanocomposites. J Dent Res 2010;89:549–60. [3] Ferracane JL. Resin composite–state of the art. Dent Mater 2011;27:29–38. [4] Demarco FF, Correa MB, Cenci MS, Moraes RR, Opdam NJ. Longevity of posterior composite restorations: not only a matter of materials. Dent Mater 2012;28:87–101. [5] Rasines Alcaraz MG, Veitz-Keenan A, Sahrmann P, Schmidlin PR, Davis D, Iheozor-Ejiofor Z. Direct composite resin fillings versus amalgam fillings for permanent or adult posterior teeth. Cochrane Database Syst Rev 2014;3:CD005620. [6] Antonucci JM, Misra DN, Peckoo RJ, The Accelerative. Adhesive bonding capabilities of surface-active accelerators. J Dent Res 1981;60:1332–42. [7] Dunn WJ. Shear bond strength of an amorphous calcium-phosphate-containing orthodontic resin cement. Am J Orthod Dentofacial Orthop 2007;131:243–7. [8] Silva KG, Pedrini D, Delbem AC, Ferreira L, Cannon M. In situ evaluation of the remineralizing capacity of pit and fissure sealants containing amorphous calcium phosphate and/or fluoride. Acta Odontol Scand 2010;68:11–8. [9] Skrtic D, Antonucci JM, Eanes ED. Improved properties of amorphous calcium phosphate fillers in remineralizing resin composites. Dent Mater 1996;12:295–301. [10] Xu HH, Moreau JL. Dental glass-reinforced composite for caries inhibition: calcium phosphate ion release and
d e n t a l m a t e r i a l s 3 1 ( 2 0 1 5 ) 726–733
[11]
[12]
[13] [14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
mechanical properties. J Biomed Mater Res B: Appl Biomater 2010;92:332–40. Arcis RW, Lopez-Macipe A, Toledano M, Osorio E, Rodriguez-Clemente R, Murtra J, et al. Mechanical properties of visible light-cured resins reinforced with hydroxyapatite for dental restoration. Dent Mater 2002;18:49–57. Chen WC, Wu HY, Chen HS. Evaluation of reinforced strength and remineralized potential of resins with nanocrystallites and silica modified filler surfaces. Mater Sci Eng C Mater Biol Appl 2013;33:1143–51. Labella R, Braden M, Deb S. Novel hydroxyapatite-based dental composites. Biomaterials 1994;15:1197–200. O’Donnell JN, Schumacher GE, Antonucci JM, Skrtic D. Structure-composition-property relationships in polymeric amorphous calcium phosphate-based dental composites. Materials (Basel) 2009;2:1929–59. Xu HH, Quinn JB. Whisker-reinforced bioactive composites containing calcium phosphate cement fillers: effects of filler ratio and surface treatments on mechanical properties. J Biomed Mater Res 2001;57:165–74. Xu HH, Weir MD, Sun L. Nanocomposites with Ca and PO4 release: effects of reinforcement, dicalcium phosphate particle size and silanization. Dent Mater 2007;23:1482–91. Xu HH, Sun L, Weir MD, Antonucci JM, Takagi S, Chow LC, et al. Nano DCPA-whisker composites with high strength and Ca and PO(4) release. J Dent Res 2006;85: 722–7. Xu HH, Weir MD, Sun L, Takagi S, Chow LC. Effects of calcium phosphate nanoparticles on Ca-PO4 composite. J Dent Res 2007;86:378–83. Xu HH, Weir MD, Sun L. Calcium and phosphate ion releasing composite: effect of pH on release and mechanical properties. Dent Mater 2009;25:535–42. Xu HH, Weir MD, Sun L, Ngai S, Takagi S, Chow LC. Effect of filler level and particle size on dental caries-inhibiting Ca-PO(4) composite. J Mater Sci Mater Med 2009;20: 1771–9. Xu HH, Moreau JL, Sun L, Chow LC. Nanocomposite containing amorphous calcium phosphate nanoparticles for caries inhibition. Dent Mater 2011;27:762–9. Moreau JL, Sun L, Chow LC, Xu HH. Mechanical and acid neutralizing properties and bacteria inhibition of amorphous calcium phosphate dental nanocomposite. J Biomed Mater Res B: Appl Biomater 2011;98:80–8. Moreau JL, Weir MD, Giuseppetti AA, Chow LC, Antonucci JM, Xu HH. Long-term mechanical durability of dental nanocomposites containing amorphous calcium phosphate nanoparticles. J Biomed Mater Res B: Appl Biomater 2012;100:1264–73. Marovic D, Tarle Z, Hiller KA, Muller R, Ristic M, Rosentritt M, et al. Effect of silanized nanosilica addition on remineralizing and mechanical properties of experimental composite materials with amorphous calcium phosphate. Clin Oral Invest 2014;18:783–92. Marovic D, Tarle Z, Hiller KA, Muller R, Rosentritt M, Skrtic D, et al. Reinforcement of experimental composite materials
[26] [27]
[28]
[29]
[30]
[31]
[32] [33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
733
based on amorphous calcium phosphate with inert fillers. Dent Mater 2014;30:1052–60. Craig RG. Restorative dental materials. 7th ed St. Louis: Mosby; 1985. Langhorst SE, O’Donnell JN, Skrtic D. In vitro remineralization of enamel by polymeric amorphous calcium phosphate composite: quantitative microradiographic study. Dent Mater 2009;25:884–91. Weir MD, Chow LC, Xu HH. Remineralization of demineralized enamel via calcium phosphate nanocomposite. J Dent Res 2012;91:979–84. Skrtic D, Hailer AW, Takagi S, Antonucci JM, Eanes ED. Quantitative assessment of the efficacy of amorphous calcium phosphate/methacrylate composites in remineralizing caries-like lesions artificially produced in bovine enamel. J Dent Res 1996;75:1679–86. Dickens SH, Flaim GM, Takagi S. Mechanical properties and biochemical activity of remineralizing resin-based Ca-PO4 cements. Dent Mater 2003;19:558–66. Melo MA, Weir MD, Rodrigues LK, Xu HH. Novel calcium phosphate nanocomposite with caries-inhibition in a human in situ model. Dent Mater 2013;29:231–40. LeGeros RZ. Calcium phosphates in oral biology and medicine. Basel, Switzerland: Karger; 1991. Anthony JW, Bideaux RA, Bladh KW, Nichols MC, editors. Handbook of Mineralogy. Chantilly, VA, USA: Mineralogical Society of America; 2000. http://www. handbookofmineralogy.org/ Shortall AC, Palin WM, Burtscher P. Refractive index mismatch and monomer reactivity influence composite curing depth. J Dent Res 2008;87:84–8. Beatty MW, Swartz ML, Moore BK, Phillips RW, Roberts TA. Effect of microfiller fraction and silane treatment on resin composite properties. J Biomed Mater Res 1998;40:12–23. Cesar PF, Yoshimura HN, Miranda Jr WG, Miyazaki CL, Muta LM, Rodrigues Filho LE. Relationship between fracture toughness and flexural strength in dental porcelains. J Biomed Mater Res B: Appl Biomater 2006;78:265–73. Skrtic D, Antonucci JM. Effect of bifunctional comonomers on mechanical strength and water sorption of amorphous calcium phosphate- and silanized glass-filled Bis-GMA-based composites. Biomaterials 2003;24:2881–8. Antonucci JM, Skrtic D. Fine-tuning of polymeric resins and their interfaces with amorphous calcium phosphate. A strategy for designing effective remineralizing dental composites. Polymers (Basel) 2010;2:378–92. LeGeros RZ. Calcium phosphates in demineralization/remineralization processes. J Clin Dent 1999;10:65–73. Skrtic D, Antonucci JM. Dental composites based on amorphous calcium phosphate-resin composition/physicochemical properties study. J Biomater Appl 2007;21:375–93. Xu HH, Weir MD, Sun L, Moreau JL, Takagi S, Chow LC, et al. Strong nanocomposites with Ca, PO(4), and F release for caries inhibition. J Dent Res 2010;89:19–28.
