DENTAL-2837; No.

of Pages 8
ARTICLE IN PRESS
d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e8

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.intl.elsevierhealth.com/journals/dema

Bioactive composites containing

TEGDMA-functionalized calcium phosphate
particles: Degree of conversion, fracture strength
and ion release evaluation

Yvette Alania a , Marina D.S. Chiari a , Marcela C. Rodrigues a ,

Victor E. Arana-Chavez a , Ana Helena A. Bressiani b , Flavio M. Vichi c ,
Roberto R. Braga a,∗
a Department of Biomaterials and Oral Biology, University of São Paulo, School of Dentistry, Av. Prof. Lineu Prestes
2227, São Paulo, SP 05508-000, Brazil
b Materials Science and Technology Center, Energy and Nuclear Research Institute, Av. Prof. Lineu Prestes 2242, São

Paulo, SP 05508-000, Brazil

c Department of Fundamental Chemistry, Institute of Chemistry, University of São Paulo, Av. Prof. Lineu Prestes 748,

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 strength and ion release of experimental composites containing
Received 26 October 2015 TEGDMA-functionalized calcium phosphate particles.
Received in revised form Methods. Seven composites containing equal parts (in mols) of BisGMA and TEGDMA and
22 June 2016 60 vol% of fillers were manipulated. Filler phase was constituted by silanized barium glass
Accepted 3 September 2016 and 0% (control), 10% or 20% (volume) of dicalcium phosphate dihydrate (DPCD) particles,
Available online xxx either non-functionalized or functionalized with two different TEDGMA contents. DCPD
particles were synthesized and characterized by X-ray diffraction (XRD), elemental analysis,
Keywords: surface area and dynamic light scattering. Composites were tested for degree of conversion
Calcium phosphate (DC) by near-FTIR. Biaxial flexural strength (BFS) was determined after 24 h and 28 days in
Mechanical properties water. Calcium and phosphate release after 7 days was assessed using inductively coupled
Ion release plasma optical emission spectrometry (ICP-OES). Data were analyzed by ANOVA/Tukey test
Resin composite (alpha:5%).
Results. XRD confirmed the crystalline structure corresponding to DCPD. Elemental analysis
revealed particles with zero, 14% or 22% TEGDMA, with similar D50 (around 19 ␮m) and
surface areas from 3.5 to 11.4 m2 /g. The presence of DCPD did not reduce DC. After 24 h,
functionalization (both 14% and 22% TEGDMA) improved composite strength in comparison
to non-functionalized DCPD, both at 10% and 20% levels. After 28 days, BFS of materials
containing 10% functionalized DCPD were statistically similar to the control containing only
barium glass. Among composites containing 10% DCPD, particle functionalization with 14%
TEGDMA did not jeopardize ion release.

∗
Corresponding author at: Depto. de Biomateriais e Biologia Oral da FOUSP, Av. Prof. Lineu Prestes, 2227, São Paulo, SP 05508-000, Brazil.
E-mail address: [email protected] (R.R. Braga).
http://dx.doi.org/10.1016/j.dental.2016.09.021
0109-5641/© 2016 The Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Alania Y, et al. Bioactive composites containing TEGDMA-functionalized calcium phosphate particles: Degree
of conversion, fracture strength and ion release evaluation. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.021
DENTAL-2837; No. of Pages 8
ARTICLE IN PRESS
xxx.e2 d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e8

Significance. At 10 vol%, the use of TEGDMA-functionalized CaP particles improved composite

strength in relation to non-functionalized particles, while maintaining similar ion release
levels.
© 2016 The Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

1. Introduction

Over the years, the incorporation of calcium orthophosphates

[1–4] and bioactive glass particles [5,6] in resin-based materials
has been studied as a possible path to promote tooth rem-
ineralization and reduce the risk of secondary caries at the Fig. 1 – Structural formula of triethylene glycol
tooth/restoration interface [1,7–10]. These particles behave as dimethacrylate (TEGDMA) used as functionalizing agent.
ion sources, releasing calcium and phosphate to the surround-
ing medium and creating a supersaturated environment that
favors mineral deposition on the tooth.
Unfortunately, the addition of CaP particles to unfilled tion of TEGDMA-functionalized DCPD in a BisGMA:TEGDMA
resins significantly lowers their strength [11,12]. This impaired resin matrix increased biaxial flexural strength by 32% in
mechanical behavior is ascribed to the lack of a strong chem- comparison with a material containing non-functionalized
ical bond between the bioactive particles and the resin phase, particles.
as well as the low cohesive strength of the CaP agglomerate This investigation is a follow-up from a previous study
itself [11–13]. Several approaches have been tested to increase where the substitution of silanized barium glass with pro-
the mechanical properties of bioactive composites. For exam- prietary, non-functionalized DCPD nano-structured agglom-
ple, an experimental composite containing hydroxyapatite erates in experimental composites resulted in up to 35%
particles functionalized with acrylic acid showed a signifi- reduction in fracture strength [13]. In the present investi-
cantly higher flexural strength in comparison with a control gation, TEGDMA-functionalized DCPD particles were tested
material containing unmodified CaP particles [14]. Silaniza- partially replacing the reinforcing fillers. The ultimate goal
tion of CaP particles was also tested and, though composite was to determine if the use of functionalized DCPD parti-
strength increased in comparison to the use of non-silanized cles would improve composite strength (in comparison to
particles, ion release was negatively affected [15,16]. the material containing non-functionalized particles) with-
The association between reinforcing fillers and bioactive out jeopardizing ion release. Additionally, the effect of the
particles seems to offset some of the negative effect of the lat- amount of TEGDMA in the functionalizing layer was also veri-
ter on composite mechanical properties. For example, when fied. The null hypotheses were: (1) replacing 10 vol% or 20 vol%
40 wt% of tetracalcium phosphate (TTCP) was added to an of silanized glass fillers by DCPD particles does not affect com-
unfilled resin, 20 wt% of silanated glass had to be added in posite degree of conversion and strength (prior and after aging
order to recover the flexural strength of the unfilled resin [12]. in water for 28 days), regardless of the DCPD content or the
In another study, the addition of 10 wt% reinforcing fillers to presence of a TEGDMA functionalizing layer, (2) ion release is
an experimental material containing 40 wt% amorphous cal- not affected by DCPD content or by functionalization of the
cium phosphate (ACP) also contributed to improve composite DCPD particles.
strength, with no adverse effect on ion release [17].
Recently, DCPD nanoparticles functionalized with triethy-
2. Methods
lene glycol dimethacrylate (TEGDMA) were synthesized [18].
Among the different calcium orthophosphate phases, dical-
2.1. Synthesis and characterization of dicalcium
cium phosphate dihydrate (DCPD, CaHPO4 .2H2 O) presents a
phosphate dihydrate (DCPD) nanoparticles
relatively high solubility [19], which makes it more suitable as
ion source than less soluble phases, such as hydroxyapatite
DCPD (CaHPO4 .2H2 O) nanoparticles were synthesized through
(HA). Also, its refractive index (1.54–1.55) [20] is similar to bar-
the reaction between ammonium phosphate (NH4 )2 HPO4,
ium glass (1.53) [21]. HA, for example, has a refractive index of
and calcium nitrate Ca(NO3 )2 .4H2 O (both from Sigma–Aldrich
1.63–1.67, which compromises light transmission during pho-
Co., St. Louis, MO, USA) by a sol–gel process [18]. Equimo-
toactivation, as well as esthetics [14]. By adding the monomer
lar solutions (0.078 mol/L) were prepared with deionized
to one of the reacting solutions (i.e., prior to DCPD crystalliza-
water at room temperature and mixed with a magnetic
tion), it would attach to the Ca2+ at the nanoparticle surface
stirrer. The ammonium phosphate solution received 6.86 g
at early stages of its growth, when reactivity is at its maxi-
(0.024 mol) of triethylene glycol dimethacrylate (Fig. 1,
mum. Also, the TEGDMA in the functionalizing layer would
TEGDMA, Mw = 286 g/mol, Sigma–Aldrich). 400 mL of the cal-
co-polymerize with monomers of the resin matrix, increasing
cium nitrate solution was added to the same volume of the
the compatibility between the nanostructured agglomerates
ammonium phosphate/TEGDMA solution using a peristaltic
and the resin. Preliminary studies showed that the incorpora-
pump (9 mL/min). The resulting mixture was kept under stir-

Please cite this article in press as: Alania Y, et al. Bioactive composites containing TEGDMA-functionalized calcium phosphate particles: Degree
of conversion, fracture strength and ion release evaluation. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.021
DENTAL-2837; No. of Pages 8
ARTICLE IN PRESS
d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e8 xxx.e3

ring for 30 min, followed by 30 min of decantation. Then, the co-addition of 32 scans with 4 cm−1 resolution (Vertex 70;
supernatant was removed, and the gel was washed by adding Bruker Optik GmbH, Ettlingen, Germany). The composite was
deionized water and centrifuging it (1085 × g, Sorvall RC6 Plus, then photoactivated through the glass slide (Radii Cal, SDI,
Thermo Fisher Scientific, Asheville, NC, USA). After centrifu- Bayswater, Australia), receiving 24 J/cm2 (1200 mW/cm2 for
gation, the TEGDMA-rich supernatant was removed and a 20 s) and the whole set was dry-stored at 37 ◦ C for 24 h. Then, a
new washing cycle initiated. TEGDMA content on the func- new spectrum was obtained and the degree of conversion was
tionalizing layer was modulated by varying the number of calculated as the ratio between the areas of the absorption
washing cycles. The first two batches were washed two and band located at 6165 cm−1 (corresponding to the absorption of
seven times, respectively, while in the last synthesis the gel the C H2 group) of the polymerized and the unpolymerized
was washed four times using water plus one cycle using a material.
TEGDMA/water mixture. The pastes were freeze-dried and a
white powder was obtained. 2.4. Biaxial flexural test
Nanoparticles were characterized according to different
methods. In order to identify the calcium orthophosphate Disk-shaped specimens (12 mm × 1 mm, n = 20) were built
phase, X-ray diffraction (XRD) measurements were taken using a stainless steel split mold. Half of the specimens were
using nickel CuK␣ radiation at 40 kV and 30 mA (MultiFlex, stored in water at 37 ◦ C for 24 h and the other half for 28 days.
Rigaku Corp., Tokyo, Japan). The equipment geometry was Photoactivation was performed by irradiating each quadrant
/2. Readings were continuous from 10◦ to 60◦ at 0.05◦ inter- for 10 s. After storage, the specimens were placed on a “pis-
vals, 10 s per interval. Elemental analysis (2400 CHN Elemental ton on three balls” device positioned on a universal testing
Analyzer, PerkinElmer, Waltham, MA, USA) was used to deter- machine (Kratos KE series, Cotia, Sao Paulo, Brazil) at a cross-
mine TEGDMA content, based on the percentage of carbon speed of 0.5 mm/min. The discs were positioned with the
detected in the sample. Approximately 1 mg of each powder irradiated surface facing the flat piston and loaded until frac-
was heated to 925 ◦ C under pure oxygen. Resultant gases (CO2 , ture. Biaxial flexural strength (BFS), in MPa, was calculated
H2 O and N2 ) were homogenized and quantified with a resolu- according to the following equations:
tion of 0.3%.
−0.2387P(X − Y)
Particle surface area was determined after nitrogen adsorp- BFS =
b2
tion isotherms were obtained using a volumetric adsorption
analyzer (Quantachrome Instruments, Boynton Beach, FL,  r 2  1 −    r 2
2 2
USA). In order to maintain the organic coating around the X = (1 − )ln +
r3 2 r3
particles, samples were purged at 50 ◦ C for 14 h. Surface area
was determined according to the Brunauer, Emmet and Teller
  r 2   r 2
1 1
(BET) method (NOVAWin, Quantachrome Instruments, Boyn- Y = (1 − ) 1 − ln + (1 − )
r2 r3
ton Beach, FL, USA). Size distribution of the nanoparticle
agglomerates was estimated using a laser diffraction particle where P is the fracture load (in Newtons); b is the specimen
analyzer (Mastersizer 2000 Ver. 5.54, Malvern Instruments Ltd., thickness (in mm); ␯ is the Poisson’s ratio; r1 is the radius
Worcestershire, UK). of the supporting spheres circumference (5 mm); r2 is the
radius of the loading piston (0.6 mm): r3 is the specimen radius
2.2. Composite formulation (in mm). The Poisson’s ratio was set for 0.30 for all composites.
Fractured surfaces were gold-sputtered and observed under
Seven composites were prepared, with the organic phase scanning electron microscopy (LEO 430, Electron Microscopy
containing 1:1 in mols of bisphenol-A glycidyl dimethacry- Ltd., Cambridge, UK).
late (BisGMA) and TEGDMA, plus 0.5 wt% of camphorquinone
and ethyl-4-dimethylamino benzoate (EDMAB, all compo- 2.5. Ion release
nents from Sigma–Aldrich) as photoinitiators. All composites
had a total filler of 60 vol%. Filler phase was constituted by 40, Disk-shaped specimens (5 mm × 1 mm, n = 3) were dry-stored
50 or 60 vol% of silanated barium glass and 20, 10 or 0 vol% for 24 h at 37 ◦ C and then immersed individually for 7 days
of DCPD particles. Three different DPCD particles with dif- in 5 mL of NaCl solution (133 mmol/L) buffered to pH = 7
ferent TEGDMA contents (obtained as previously described) with 50 mmol/L HEPES. The concentrations of calcium and
were used. Composites were mechanically mixed under vac- phosphate ions released in the solution were determined
uum (Speedmixer DAC 150.1 FVZ-K, FlackTek Inc., Landrum, by inductively coupled plasma-optical emission spectrome-
SC, USA) and kept under refrigeration until 2 h before use. try (ICP-OES, 700, Agilent Technologies, Santa Clara, CA, USA).
Prior to analysis, the specimen was removed from the vial and
2.3. Degree of conversion the solution was filtered (pore size: 3 ␮m) before being acidi-
fied with 5 mL of 10% HNO3 (pH = 0.2). ICP-OES uses photon
Degree of conversion (n = 3) was determined by Fourier- detection (emitted by the elements of interest when decaying
transformed near-infrared spectroscopy (near-FTIR) [22]. from an excited state reached after collision with ions from
Unpolymerized composite was placed in a silicon mold an argon plasma) to quantify element concentration in the
(7 mm diameter and 1 mm thickness) pressed between two solution. Calcium and phosphorous readings were obtained
microscopy glass slides. The spectrum of the unpolymerized simultaneously, based on the emission wavelengths of 184 nm
material was obtained between 4000 and 10,000 cm−1 by the and 177 nm, respectively.

Please cite this article in press as: Alania Y, et al. Bioactive composites containing TEGDMA-functionalized calcium phosphate particles: Degree
of conversion, fracture strength and ion release evaluation. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.021
DENTAL-2837; No. of Pages 8
ARTICLE IN PRESS
xxx.e4 d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e8

Fig. 2 – Difractograms of the calcium phosphate Fig. 3 – Filler size distributions for barium glass and
nanoparticles (Top: DCPD functionalized with 22% TEGDMA, functionalized DCPD particles.
middle: 14% TEGDMA, bottom: non-functionalized DCPD).

erates were quite similar among the different powders

(non-functionalized: 17 ␮m, DCPD 14: 20 ␮m and DCPD 22:
2.6. Statistical analysis
19 ␮m). Barium glass filler showed a D50 of 0.5 ␮m.

Data were subjected to either one-way ANOVA (degree of con-

3.2. Degree of conversion
version) or two-way ANOVA (biaxial flexural strength and ion
release). For the mechanical testing, the main factors were
Degree of conversion (DC) results are shown in Table 1.
DCPD content (seven levels, i.e., six combinations between
The replacement of barium glass by non-functionalized or
two DCPD levels and three TEGDMA contents, plus the con-
DCPD 22 particles did not significantly affect DC, regard-
trol material without DCPD) and storage time (24 h or 28 days),
less of the DCPD content. However, the use of DCPD 14
while for ion release the main factors were DCPD content (10%
resulted in higher DC compared to the composites contain-
or 20%) and particle TEGDMA content. In all cases, Tukey test
ing only silanized glass or non-functionalized DCPD particles
was used for pair-wise comparisons. The global significance
(p < 0.001).
level (˛) was set at 0.05.

3.3. Biaxial flexural test

3. Results
BFS results are shown in Table 1. After 24 h storage, the
3.1. Particles characterization replacement of barium glass by calcium phosphate parti-
cles resulted in significant reductions in fracture strength
Difractograms (Fig. 2) identified the crystalline structure of (p < 0.05). More severe reductions were observed for compos-
DCPD for the three synthesized calcium phosphate powders ites containing non-functionalized nanoparticles (10% DCPD:
(International Center for Diffraction Data/Joint Committee 41%, 20% DCPD: 71%) than those containing DCPD 14 (10%
on Powder Diffraction Standards, ICDD/JCPDS, PDF 09-0077). DCPD: 28%, 20% DCPD: 36%) and DCPD 22 (10% DCPD: 17%,
Additionally, the powder obtained in the synthesis that used 20% DCPD: 37%). Among composites containing 10% DCPD,
two washing cycles also revealed the presence of DCPA (dical- strength increased with the particle TEGDMA content. For
cium phosphate anhydrous, ICDD/JCPDS PDF 090080). materials containing 20% DCPD, particle functionalization
Elemental analysis revealed only traces of carbon in the increased strength; however, no difference was observed
powder obtained after seven washing cycles and 8% of carbon between DCPD 14 and DCPD 22. Strength values of compos-
in the powder subjected to two washing cycles, which corre- ites containing 10% and 20% DCPD 14 were not statistically
sponds to a TEGDMA content of 14%. The powder resulting different.
from the synthesis where the last of four washing cycles used After 28 days in water, all composites showed reductions in
a TEGDMA/water mixture revealed a carbon content of 13%, strength ranging between 21% and 39%. Among composites
corresponding to 22% of TEGDMA. Therefore, DCPD particles containing 10% DCPD, no statistically significant difference
with 0% TEGDMA will be referred to as “non-functionalized was found between DCPD 14 and DCPD 22. Noteworthy, their
particles”, while those with 8% and 13% carbon will be referred strength values were statistically similar to that of the control
to as “DCPD 14” and “DCPD 22”, respectively. According to the containing only barium glass. For the composites contain-
BET method, the surface areas of the particles were 11 m2 /g ing 20% DCPD, particle functionalization did not lead to
(non-functionalized), 5 m2 /g (DCPD 14), and 4 m2 /g (DCPD 22). improved strength after 28 days. Finally, no difference was
Particle size distributions are shown in Fig. 3. The observed between composites containing 10% or 20% of non-
median particle sizes (D50 ) of the nanoparticles agglom- functionalized particles.

Please cite this article in press as: Alania Y, et al. Bioactive composites containing TEGDMA-functionalized calcium phosphate particles: Degree
of conversion, fracture strength and ion release evaluation. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.021
DENTAL-2837; No. of Pages 8
ARTICLE IN PRESS
d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e8 xxx.e5

Table 1 – Degree of conversion and biaxial flexural strength results (average and standard deviations) of experimental
composites as a function of DCPD content and amount of TEGDMA in the functionalizing layer, and storage period
(strength only). Similar letters in a given test indicate lack of statistically significant differences (ANOVA/Tukey test,
p > 0.05).
DCPD content (vol%) TEGDMA in the Degree of conversion (%) Biaxial flexural strength (MPa)
particle (wt%)

24 h 28 days
C A
0 (control) 78.1 (0.1) 161.4 (12.7) 97.9 (10.5) DE

10 0 77.7 (0.6) C 95.4 (14.4) DE 65.0 (6.1) GH

14 83.2 (1.3) B 116.8 (10.3) C 86.9 (11.5) EF
22 79.6 (0.5) BC 133.6 (15.0) B 96.7 (9.3) DE

20 0 78.3 (0.7) BC 76.2 (9.9) FG 60.5 (4.6) H

14 88.7 (4.6) A 103.5 ± 9.5CD 65.8 ± 13.4GH
22 81.7 (0.2) BC 101.2 ± 11.8DE 69.2 ± 8.1GH

Fig. 4 – Backscattered SEM images of composite fractured surfaces. White arrows indicate the DCPD agglomerates, asterisk
indicate empty spaces where the agglomerates were displaced upon fracture and the black arrow shows the imprint left by
an agglomerate on the resin matrix. A: Control, 28 days; B: 20% non-functionalized DCPD, 28 days; C: 20% DCPD 14, 28 days;
D: 20% DCPD 22, 28 days.

SEM images of fractured surfaces after 28 days storage in 3.4. Calcium and phosphate release
water are shown in Fig. 4. The control composite (Fig. 4A)
revealed exposed particles suggesting that fractured occurred Calcium release values are presented in Fig. 5A. The inter-
at the matrix/particle interface. The composite containing action between DCPD percentage and TEGDMA content was
non-functionalized DCPD (Fig. 4B) showed several plate-like not statistically significant (p = 0.275), though both factors
DCPD agglomerates protruding from the matrix, as well as slit- influenced calcium release (p < 0.001). Composites contain-
shaped spaces suggesting that agglomerates detached from ing 20% DCPD released 23% more Ca2+ then those with 10%
the matrix upon fracture. Composites containing functional- DCPD (20%: 0.242 ± 0.051 mmol/L; 10%: 0.186 ± 0.031 mmol/L).
ized DCPD (Figs. 4C and D) presented a similar aspect, except Overall, regardless of the DCPD content, particle functional-
for the smoother agglomerate edges, in comparison to the ization reduced Ca2+ release between 22% and 32%, while
uneven, rugged edges displayed by the non-functionalized TEGDMA content in the particle did not significantly affect
DCPD. No differences were observed between specimens frac- Ca2+ release (0% TEGDMA: 0.261 ± 0.046 mml/L, DCPD 14:
tured after 24 h and 28 days. 0.203 ± 0.033 mmol/L, DCPD 22: 0.178 ± 0.030 mmol/L). How-

Please cite this article in press as: Alania Y, et al. Bioactive composites containing TEGDMA-functionalized calcium phosphate particles: Degree
of conversion, fracture strength and ion release evaluation. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.021
DENTAL-2837; No. of Pages 8
ARTICLE IN PRESS
xxx.e6 d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e8

Fig. 5 – Ion release (in mmol/L) as a function of DCPD percentage and TEGDMA content (A: calcium, B: hydrogen phosphate).
Same letters indicate lack of statistically significant differences (p > 0.05). Please, notice the difference in scale of y-axes
between both graphics.

ever, at 10% DCPD, the difference between the material increase in polymerization reaction termination through seg-
containing non-functionalized particles and the composite mental diffusion, also associated with high TEGDMA fractions
with DCPD 14 was not statistically significant. [29].
Phosphate release after 7 days is shown in Fig. 5B. The inter- Composites containing DCPD presented significantly lower
action between bioactive filler level and TEGDMA content was flexural strength after 24 h than the control. In agreement
statistically significant (p < 0.001). Only for composites with with previous investigations [12,13], increasing DCPD content
10% DCPD, there was no difference in HPO4 2− release between from 10% to 20% also reduced strength, except for DCPD 14
the materials containing non-functionalized and DCPD 14 where the reduction was only numerical. The lower strength
particles. Regardless of the DCPD level, TEGDMA content in observed with the addition of calcium orthophosphates par-
the particle did not significantly affect HPO4 2− release. ticles occurs due to different reasons. First, while silanized
glass particles increase composite fracture strength [30–32],
calcium orthophosphate particles are nano-particles agglom-
erates with much lower cohesive strength than glass and, as
4. Discussion
such, they are much less effective as toughening agents. Sec-
ond, due to the lack of a coupling interphase between DCPD
Materials containing CaP particles have demonstrated poten-
and the organic matrix, these particles behave as large inclu-
tial for remineralization of enamel and dentin in vitro
sions, increasing the risk of crack initiation at low stress levels
[1,5,23,24]. However, since CaP particles are not cohesively
[11,12,33–35].
strong nor have a chemical interaction with the resin matrix,
Notwithstanding, functionalization of DCPD particles sig-
they show a deleterious effect on the composite strength [13].
nificantly reduced the detrimental effect resulting from the
Functionalization did not contribute to reduce nanoparticles
replacement the silanized glass by non-functionalized parti-
agglomeration. Therefore, in practical terms, they behave as
cles. Compared to the control (without DCPD), the difference
“micro” particles or “nano-particles agglomerates”, with plate-
in strength observed with a 10% replacement decreased from
like morphology. Surface area values are low because the
41% (non-functionalized) to 17% (with DCPD 22). For a 20%
functionalizing layer was not removed prior to analysis, and
replacement, the difference in strength with the addition of
TEGDMA fills the pores and surface irregularities of the CaP
DCPD was reduced from 71% to 36% (with DCPD 14). The supe-
core. Non-functionalized and both functionalized DCPD par-
rior performance of composites containing functionalized
ticles presented agglomerates with similar median (D50 ) sizes
DCPD can be explained by the incorporation of the TEGDMA
and particle size distribution. Since mechanical properties are
monomers bonded to the DCPD core into the polymer network.
influenced by the size of the agglomerates [25,26], this find-
Only for composites containing DCPD 14 (both at 10% and 20%
ing, though unforeseen, is useful to highlight the influence of
levels), the gain in strength could also be attributed to their
particle functionalization on the strength results.
higher DC; however, the fact that composites with DCPD 22 did
Composites containing DCPD 14 presented statistically
not present higher DC than those with non-functionalized par-
higher DC compared to those containing non-functionalized
ticles highlights the importance of the functionalizing layer. It
particles. Increasing TEGDMA content in resin mixtures was
is interesting to point out that the amount of TEGDMA around
shown to increase DC due to its small size and flexible
the particle was not relevant for the strength of composites
structure [27,28]. Therefore, it could be hypothesized that
containing 20% DCPD. The higher DC reached by the composite
TEGDMA-rich domains around functionalized DCPD particles
with DCPD 14 associated with the low reinforcing filler con-
were responsible for the increase in DC. Interestingly, this
tent could explain the similarity in strength with the DCPD 22
increase was slightly lower in composites containing DCPD 22.
composite.
It can only be speculated that the effect of a higher mobility
of the reactive species in TEGDMA-rich areas was offset by an

Please cite this article in press as: Alania Y, et al. Bioactive composites containing TEGDMA-functionalized calcium phosphate particles: Degree
of conversion, fracture strength and ion release evaluation. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.021
DENTAL-2837; No. of Pages 8
ARTICLE IN PRESS
d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e8 xxx.e7

After 28 days of storage in water, all materials showed phosphate release values are in agreement with previous
statistically significant reductions in strength. Composites studies [12,39], which can be attributed to the phosphate
containing 10% DCPD presented similar strength reductions, tetrahedral structure, which is not easily removed from the
between 26% and 32%. The lower strengths verified after crystalline lattice [13].
prolonged water immersion can be attributed to hydrolytic In conclusion, the use of TEGDMA-functionalized DCPD
degradation of the matrix, as well as partial dissolution of particles significantly improved the strength of ion-releasing
the CaP particles [36]. For composites with 20% DCPD, how- composites, in relation to the use of non-functionalized
ever, a more severe strength reduction after water immersion DCPD. Among the tested materials, the composite containing
was observed with the use of functionalized DCPD (between 10% DCPD 14 showed the best compromise between fracture
32% and 36%), in comparison to the material containing non- strength and ion release, including strength statistically sim-
functionalized DCPD (21% reduction). A possible explanation ilar to the control composite after 28 days in water. Further
for this finding involves the higher TEGDMA content in com- research is needed to evaluate the remineralization poten-
posites with a higher fraction of functionalized nanoparticles tial of composites containing TEGDMA-functionalized DCPD
leading to the formation of a more heterogeneous network, particles, as well as to optimize parameters such as filler size
which facilitates water sorption [37]. That, associated with distribution and TEGDMA level in the functionalizing layer in
TEGDMA relatively high hydrophilicity [38] would lead to order to improve mechanical and biological properties.
increased composite degradation. Also noteworthy was the
fact that composites containing 10% of functionalized DCPD
presented strength values statistically similar to the control
Acknowledgements
material after 28-day storage. Albeit encouraging, this finding
must be taken carefully, as the control composite presented
This study was supported by FAPESP (The State of São Paulo
an unexpectedly high reduction in strength, when compared
Research Foundation), grant 2012/25253-6, CAPES (Coordina-
with a previous study testing the same formulation (24 h:
tion for the Improvement of Higher Education Personnel). The
157 MPa; 28 days: 127 MPa) [13].
authors would like to thank Douglas Nesadal de Souza and
Clinically, a sustained ion release is desirable for a long-
Dr. Alyne Simões (Department of Biomaterials and Oral Biol-
term caries inhibition effect. Therefore, it becomes important
ogy, University of São Paulo) for the technical support with
to know the kinetics of ion release over time. The avail-
the ICP-OES equipment and FGM Produtos Odontológicos for
able literature shows different release profiles, depending on
donating the barium glass particles.
the particle characteristics and content, pH of the immer-
sion medium and hydrophilicity of the resin matrix [17,39].
In the present study, however, measurements were restricted
references
to seven days, as the main focus was to verify if function-
alization could impair ion release. In fact, ion release was
influenced by DCPD level and functionalization. The higher
[1] Wang C, Li SJ, Wu ZQ, Xu JJ, Chen HY, Xia XH. Study on the
ion release verified with composites containing a higher DCPD
kinetics of homogeneous enzyme reactions in a
fraction agrees with previous studies [40,41]. Overall, the
micro/nanofluidics device. Lab Chip 2010;10(5):639–46.
use of functionalized DCPD particles reduced ion release, [2] Weir MD, Chow LC, Xu HH. Remineralization of
suggesting that the interaction by chemisorption between demineralized enamel via calcium phosphate
TEGDMA and calcium in the DCPD structure [18] represents nanocomposite. J Dent Res 2012;91(10):979–84.
an extra hindrance for ions to be released from the DCPD crys- [3] Besinis A, van Noort R, Martin N. Remineralization potential
talline structure. Notwithstanding, the composite containing of fully demineralized dentin infiltrated with silica and
hydroxyapatite nanoparticles. Dent Mater 2014;30(3):249–62.
10 vol% of DCPD 14 released calcium and phosphate in lev-
[4] Rodrigues MC, Natale LC, Arana-Chaves VE, Braga RR.
els similar to the composite containing non-functionalized Calcium and phosphate release from resin-based materials
particles and concentrations similar to those from previous containing different calcium orthophosphate nanoparticles.
reports (Ca2+ : 0.18 mmol/L; HPO4 2− : 0.015 mmol/L) [13,42]. Ion J Biomed Mater Res B Appl Biomater 2015;103(8):1670–8.
release levels reported in in vitro remineralization studies [5] Narayana SS, Deepa VK, Ahamed S, Sathish ES, Meyappan R,
(Ca2+ : 0.3–0.5 mmol/L; HPO4 2− : 0.1–0.3 mmol/L) [7,24] are in the Satheesh Kumar KS. Remineralization efficiency of bioactive
glass on artificially induced carious lesion an in-vitro study. J
same range of those obtained with the composites contain-
Indian Soc Pedod Prev Dent 2014;32(1):19–25.
ing 20 vol% of non-functionalized DCPD, but higher than those
[6] Taubock TT, Zehnder M, Schweizer T, Stark WJ, Attin T,
obtained with 10 vol% DCPD. Therefore, it is important to ver- Mohn D. Functionalizing a dentin bonding resin to become
ify in future investigations if the concentrations released from bioactive. Dent Mater 2014;30(8):868–75.
materials containing functionalized DCPD particles are capa- [7] Dickens SH, Flaim GM, Takagi S. Mechanical properties and
ble of promoting mineral recovery in initial caries lesions. biochemical activity of remineralizing resin-based Ca-PO4
Silanization of calcium phosphate particles has demon- cements. Dent Mater 2003;19(6):558–66.
[8] Peters MC, Bresciani E, Barata TJ, Fagundes TC, Navarro RL,
strated to negatively affect ion release because of the
Navarro MF, et al. In vivo dentin remineralization by
hydrophobic nature of silane coating [16,25]. On the
calcium-phosphate cement. J Dent Res 2010;89(3):286–91.
other hand, the similar calcium release among non- [9] Davis HB, Gwinner F, Mitchell JC, Ferracane JL. Ion release
functionalized DCPD and DCPD 14 could be explained by from, and fluoride recharge of a composite with a
TEGDMA hydrophilic character [16,37,38]. HPO4 2− release fol- fluoride-containing bioactive glass. Dent Mater
lowed the same patterns displayed by Ca2+ . Lower hydrogen 2014;30(10):1187–94.

Please cite this article in press as: Alania Y, et al. Bioactive composites containing TEGDMA-functionalized calcium phosphate particles: Degree
of conversion, fracture strength and ion release evaluation. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.021
DENTAL-2837; No. of Pages 8
ARTICLE IN PRESS
xxx.e8 d e n t a l m a t e r i a l s x x x ( 2 0 1 6 ) xxx.e1–xxx.e8

[10] Melo MA, Weir MD, Rodrigues LK, Xu HH. Novel calcium [26] Xu HH, Weir MD, Sun L. Calcium and phosphate ion
phosphate nanocomposite with caries-inhibition in a releasing composite: effect of pH on release and mechanical
human in situ model. Dent Mater 2013;29(2):231–40. properties. Dent Mater 2009;25(4):535–42.
[11] Skrtic D, Antonucci JM, Eanes ED. Improved properties of [27] Lovell LG, Newman SM, Bowman CN. The effects of light
amorphous calcium phosphate fillers in remineralizing intensity, temperature, and comonomer composition on the
resin composites. Dent Mater 1996;12(5):295–301. polymerization behavior of dimethacrylate dental resins. J
[12] Xu HH, Moreau JL. Dental glass-reinforced composite for Dent Res 1999;78(8):1469–76.
caries inhibition: calcium phosphate ion release and [28] Floyd CJ, Dickens SH. Network structure of bis-GMA- and
mechanical properties. J Biomed Mater Res B Appl Biomater UDMA-based resin systems. Dent Mater 2006;22(12):1143–9.
2010;92(2):332–40. [29] Elliott JE, Lovell LG, Bowman CN. Primary cyclization in the
[13] Chiari MD, Rodrigues MC, Xavier TA, de Souza EM, polymerization of bis-GMA and TEGDMA: a modeling
Arana-Chavez VE, Braga RR. Mechanical properties and ion approach to understanding the cure of dental resins. Dent
release from bioactive restorative composites containing Mater 2001;17(3):221–9.
glass fillers and calcium phosphate nano-structured [30] Ferracane JL, Berge HX, Condon JR. In vitro aging of dental
particles. Dent Mater 2015;31(6):726–33. composites in water-effect of degree of conversion, filler
[14] Arcis RW, Lopez-Macipe A, Toledano M, Osorio E, volume, and filler/matrix coupling. J Biomed Mater Res
Rodriguez-Clemente R, Murtra J, et al. Mechanical properties 1998;42(3):465–72.
of visible light-cured resins reinforced with hydroxyapatite [31] Drummond JL. Degradation, fatigue, and failure of resin
for dental restoration. Dent Mater 2002;18(1):49–57. dental composite materials. J Dent Res 2008;87(8):710–9.
[15] Xu HH, Sun L, Weir MD, Takagi S, Chow LC, Hockey B. Effects [32] Marovic D, Tarle Z, Hiller KA, Muller R, Ristic M, Rosentritt M,
of incorporating nanosized calcium phosphate particles on et al. Effect of silanized nanosilica addition on
properties of whisker-reinforced dental composites. J remineralizing and mechanical properties of experimental
Biomed Mater Res B Appl Biomater 2007;81(1):116–25. composite materials with amorphous calcium phosphate.
[16] O’Donnell JN, Schumacher GE, Antonucci JM, Skrtic D. Clin Oral Invest 2014;18(3):783–92.
Structure-composition-property relationships in polymeric [33] Antonucci JM, Fowler BO, Venz S. Filler systems based on
amorphous calcium phosphate-based dental composites. calcium metaphosphates. Dent Mater 1991;7(2):124–9.
Materials (Basel) 2009;2(4):1929–59. [34] O’Donnell JN, Langhorst SE, Fow MD, Antonucci JM, Skrtic D.
[17] Marovic D, Tarle Z, Hiller KA, Muller R, Rosentritt M, Skrtic D, Light-cured dimethacrylate-based resins and their
et al. Reinforcement of experimental composite materials composites: comparative study of mechanical strength,
based on amorphous calcium phosphate with inert fillers. water sorption and ion release. J Bioact Compat Polym
Dent Mater 2014;30(9):1052–60. 2008;23(3):207–26.
[18] Rodrigues MC, Hewer TL, Brito GE, Arana-Chavez VE, Braga [35] Zhao J, Liu Y, Sun WB, Zhang H. Amorphous calcium
RR. Calcium phosphate nanoparticles functionalized with a phosphate and its application in dentistry. Chem Cent J
dimethacrylate monomer. Mater Sci Eng C Mater Biol Appl 2011;5:40.
2014;45:122–6. [36] Skrtic D, Antonucci JM. Effect of bifunctional comonomers
[19] Dorozhkin SV. Calcium orthophosphates in dentistry. J on mechanical strength and water sorption of amorphous
Mater Sci Mater Med 2013;24(6):1335–63. calcium phosphate- and silanized glass-filled
[20] Anthony JW, Bideaux RA, Bladh KW, Nichols MC. Handbook bis-GMA-based composites. Biomaterials 2003;24(17):2881–8.
of mineralogy. Mineralogical Society of America; 2000. [37] Sideridou I, Tserki V, Papanastasiou G. Study of water
[21] Shortall AC, Palin WM, Burtscher P. Refractive index sorption, solubility and modulus of elasticity of light-cured
mismatch and monomer reactivity influence composite dimethacrylate-based dental resins. Biomaterials
curing depth. J Dent Res 2008;87(1):84–8. 2003;24(4):655–65.
[22] Stansbury JW, Dickens SH. Determination of double bond [38] Kalachandra S, Kusy RP. Comparison of water sorption by
conversion in dental resins by near infrared spectroscopy. methacrylate and dimethacrylate monomers and their
Dent Mater 2001;17(1):71–9. corresponding polymers. Polymer 1991;32(13):2428–34.
[23] Skrtic D, Hailer AW, Takagi S, Antonucci JM, Eanes ED. [39] Xu HH, Moreau JL, Sun L, Chow LC. Nanocomposite
Quantitative assessment of the efficacy of amorphous containing amorphous calcium phosphate nanoparticles for
calcium phosphate/methacrylate composites in caries inhibition. Dent Mater 2011;27(8):762–9.
remineralizing caries-like lesions artificially produced in [40] Cheng L, Weir MD, Zhang K, Xu SM, Chen Q, Zhou X, et al.
bovine enamel. J Dent Res 1996;75(9):1679–86. Antibacterial nanocomposite with calcium phosphate and
[24] Langhorst SE, O’Donnell JN, Skrtic D. In vitro quaternary ammonium. J Dent Res 2012;91(5):460–6.
remineralization of enamel by polymeric amorphous [41] Chen C, Weir MD, Cheng L, Lin NJ, Lin-Gibson S, Chow LC,
calcium phosphate composite: quantitative et al. Antibacterial activity and ion release of bonding agent
microradiographic study. Dent Mater 2009;25(7):884–91. containing amorphous calcium phosphate nanoparticles.
[25] Xu HH, Weir MD, Sun L. Nanocomposites with Ca and PO4 Dent Mater 2014;30(8):891–901.
release: effects of reinforcement, dicalcium phosphate [42] Xu HH, Weir MD, Sun L, Takagi S, Chow LC. Effects of
particle size and silanization. Dent Mater calcium phosphate nanoparticles on Ca-PO4 composite. J
2007;23(12):1482–91. Dent Res 2007;86(4):378–83.

Please cite this article in press as: Alania Y, et al. Bioactive composites containing TEGDMA-functionalized calcium phosphate particles: Degree
of conversion, fracture strength and ion release evaluation. Dent Mater (2016), http://dx.doi.org/10.1016/j.dental.2016.09.021