Development of calcium phosphate/ethylene glycol dimethacrylate

particles for dental applications

Livia C. Natale,1 Marcela C. Rodrigues,2 Yvette Alania,1 Marina D. S. Chiari,1 Handially S. Vilela,1
Douglas N. Vieira,1 Victor Arana-Chavez,1 Marcia M. Meier,3 Flavio M. Vichi,4 Roberto R. Braga 1

1
Department of Biomaterials and Oral Biology, University of Sa ~o Paulo, Sa~o Paulo, Sa~o Paulo, Brazil
2
~ o Paulo, Sa
Cruzeiro do Sul University, Sa ~ o Paulo, Brazil
3
Santa Catarina State University, Department of Chemistry, Floriano  polis, Santa Catarina, Brazil
4
Department of Fundamental Chemistry, University of Sa ~o Paulo, Sa
~o Paulo, Sa~o Paulo, Brazil

Received 4 January 2018; revised 2 April 2018; accepted 22 April 2018

Published online 00 Month 2018 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.34164

Abstract: This study describes the synthesis of dicalcium phos-
phate dihydrate (DCPD) particles in the presence of different
ethylene glycol dimethacrylates (EGDMA, ethylene glycol/EG
units: 1, 2, 3 or 4) at two monomer-to-ammonium phosphate
molar ratios (1:1 and 2:1), as a strategy to develop CaP-
monomer particles with improved interaction with resin matri-
ces. Particles displaying high surface areas and organic con-
tents were added to a photocurable BisGMA-TEGDMA resin
and the resulting materials were tested for degree of conversion
(DC), biaxial flexural strength (BFS), flexural modulus, and ion
release. Data were subjected to one-way ANOVA or Kruskal-
in BFS was obtained with the use of particles with the highest
functionalization level compared to non-functionalized particles
(p < 0.001). The use of functionalized DCPD reduced flexural
modulus in comparison to non-functionalized particles
(p < 0.001). Calcium release was similar among materials and
remained constant during the experiment, while phosphate
release was higher at 7 days in comparison to the remaining
weeks (p < 0.001). In conclusion, diethylene glycol dimethacry-
late resulted in the highest functionalization levels and the high-
est BFS among DCPD-containing materials. Ion release was not
affected by functionalization. VC 2018 Wiley Periodicals, Inc. J Biomed

Wallis/Dunn test (alpha: 0.05). Functionalization with EGDMA Mater Res Part B: Appl Biomater 00B: 000–000, 2018.
derivatives was dependent upon the length of the spacer group
and monomer concentration in the synthesis. No differences in Key Words: calcium phosphate, nanoparticles, dental materi-
DC were observed among materials (p > 0.05). A 39% increase als, ethylene dimethacrylate, bisphenol a-glycidyl methacrylate

How to cite this article: Natale LC, Rodrigues MC, Alania Y, Chiari MDS, Vilela HS, Vieira DN, Arana-Chavez V, Meier MM,
Vichi FM, Braga RR 2018. Development of calcium phosphate/ethylene glycol dimethacrylate particles for dental applications.
J Biomed Mater Res Part B 2018:00B:000–000.

INTRODUCTION drawback, calcium phosphate particles can be modified to

The efficacy of composites containing calcium orthophos-
phates in promoting mineral recovery of initial enamel car-
ies lesions has been demonstrated in vitro.1,2 Also, a
composite material containing amorphous calcium phos-
phate was able to reduce mineral loss of the surrounding
enamel in the presence of cariogenic biofilm in situ.3
Besides creating a supersaturated environment that could
favor remineralization, CaP-containing composites increase
the formation of Ca-F reservoirs in the biofilm, an important
mechanism to reduce demineralization.4,5
Unfortunately, the incorporation of calcium phosphate
particles to resin composites results in significant reduc-
tions in their fracture strength and fracture toughness, as
the lack of an effective interaction between the particles
and the resin phase facilitates crack propagation through
the matrix-particle boundaries.6,7 In order to minimize this
improve micromechanical interlocking with the resin phase8
and/or promote some chemical interaction with monomers
in the matrix, for example, via functionalization with carbox-
ylic acids9 or silanization.10
Recently, the synthesis of dicalcium phosphate dihydrate
particles (DCPD, CaHPO4.2H2O) functionalized with triethy-
lene glycol dimethacrylate (TEGDMA) was described.11
DCPD presents a refractive index similar to barium
glass,12,13 which is advantageous in terms of photoactivation
and overall aesthetics of the restorative material. Also, its
higher solubility in relation to other calcium orthophosphate
phases14 may allow for sufficient ion release at relatively
low volume fractions. During the synthesis, TEGDMA may
bind to Ca21 ions via ion-dipole bonds with oxygen atoms
of the ethylene glycol groups (–O–CH2–CH2–).15 When
TEGDMA-functionalized DCPD particles were added to

Correspondence to: R. R. Braga; e-mail: rrbraga@usp.br

Contract grant sponsor: FAPESP (The State of Sa ~o Paulo Research Foundation); contract grant number: 2015/15019-4; CAPES (Coordenaça
~o de
Aperfeiçoamento de Pessoal de Nivel Superior); and CNPq (Conselho Nacional de Pesquisa)

C 2018 WILEY PERIODICALS, INC.

V 1

FIGURE 1. Structural formula of ethylene glycol dimethacrylate, with
“n” representing the number of ethylene glycol units in the molecule.
n 5 1, ethylene glycol dimethacrylate (EGDMA, 198.22 g/mol); n 5 2,
diethylene glycol dimethacrylate (DEGDMA, 242.27 g/mol); n 5 3, tri-
ethylene glycol dimethacrylate (TEGDMA, 286.32 g/mol) and n 5 4, tet-
raethylene glycol dimethacrylate (TETDMA, 330.37 g/mol).
either unfilled or glass-filled bisphenol A glycidyl dimetha-
crylate (BisGMA)/TEGDMA resins, 30–40% increases in
fracture strength were observed in relation to materials
containing non-functionalized DCPD.16,17 Also, because
TEGDMA-rich network domains are relatively hydro-
philic,18,19 functionalization did not hinder water access to
CaP, keeping ion release at levels similar to those observed
with non-functionalized DCPD.17 Finally, it is licit to assume
that at least a fraction of the particles are actually nanopar-
ticles agglomerates interspaced by the polymer, and these
agglomerates could behave as low-elastic modulus fillers
and contribute for reducing polymerization stress develop-
ment in resin-based materials.20,21
The hydrophilicity of ethylene glycol dimethacrylates
(EGDMA) increases with the number of ethylene glycol (EG)
units.19 Also, the availability of oxygen atoms is affected by
chain spacer length, not only due to the actual number of
atoms in the molecule, but by the fact that longer molecules
can assume more complex tridimensional configurations.
Therefore, the use of different EGDMA monomers may affect
the final monomer content of the particles, as well as ion
release from resin-based materials.
The present study investigated the effect of different
EGDMA derivatives (with EG units between one and four)
as functionalizing agents for DCPD particles. Particles were
synthesized, characterized and added to a dimethacrylate-
based resin matrix. The resulting materials were tested for
mechanical properties, degree of conversion and ion release.
The overall null hypothesis was that DCPD functionalization
did not affect the tested variables.
MATERIALS AND METHODS
Particle synthesis and characterization
DCPD particles were synthesized by a co-precipitation
method using calcium nitrate [Ca(NO3)24H2O] and ammo-
nium dihydrogen phosphate [(NH4)H2PO3] solutions
(0.2 molL21) as precursors (all chemicals from Sigma–
Aldrich, St. Louis, MO). For the syntheses of functionalized
particles, EGDMA derivatives (Figure 1) were added to the
ammonium dihydrogen phosphate solution at molar ratios
of 1:1 or 2:1. The calcium nitrate solution was added drop-
wise to the ammonium dihydrogen phosphate/monomer
solution and kept under stirring for 24 h. An additional syn-
thesis was conducted without the addition of any monomer
to the precursor (non-functionalized control, NF). Reaction
2 NATALE ET AL.
byproducts and excess monomer were removed by filtration
under vacuum. The resulting gel was freeze-dried until a
white powder was obtained.
The calcium orthophosphate phase present in the par-
ticles was determined by X-ray diffraction (XRD). Diffracto-
grams were obtained using nickel-filtered CuKa radiation at
40 kV and 30 mA (XRD-7000 Maxima, Shimadzu, Kyoto,
Japan). The equipment geometry was u/2u, and data were
collected continuously from 108 to 608 at 28/min. Monomer
mass fraction in the particles was calculated based on their
carbon content determined by elemental analysis (model
240 CHN Elemental Analyzer, Perkin-Elmer, Waltham, MA).
Particles were observed under transmission electron
microscopy (80 kV, TEM, JEOL, model 1010, Tokyo, Japan)
and their long axis (n 5 100) was measured using ImageJ
software (National Institute of Health, Bethesda, MD).
Surface area was calculated based on nitrogen adsorp-
tion isotherms using the BET (Brunauer, Emmet and Teller)
method (Nova 1100e, Quantachrome Instruments, Boynton
Beach, FL). In order to determine only the surface area of
the calcium orthophosphate core, particles were sonicated
in acetone to remove their organic constituents. Particle size
distribution was determined using laser light diffraction
(Mastersizer 2000, Malvern Instruments Ltd., Malvern, UK).
Particle micromorphology was observed under scanning
electron microscopy (Quanta FEG600, Eindhoven, The Neth-
erlands) and their true density was determined in a helium
pycnometer (Ultrapyc 1200e, Quantachrome Instruments,
Boynton Beach, FL).
Resin-based materials containing DCPD particles
Particles combining high monomer levels and high surface
area were selected for the second part of the study, along
with the non-functionalized DCPD. A resin mixture contain-
ing equimolar amounts of BisGMA (2,2-bis[4–(2-hydroxy-3-
methacryloxypropoxy)phenyl]-propane) and TEGDMA, and
0.5% (by weight) of both camphorquinone and ethyl-4-
dimethylamino benzoate as photoinitiators was prepared.
Particles were added to this resin at 30 vol % using a
mechanical mixer, under vacuum (Speedmixer DAC 150.1
FVZ-K, Flak Tek Inc., Landrum, SC).
Degree of conversion
Degree of conversion (DC) was determined using mid-
infrared spectroscopy (Vertex 70, Bruker Optics, Germany)
with an attenuated total reflectance accessory (ATR, MIRacle
Pike Technologies, Madison, WI). The non-polymerized
material was inserted in a silicone mold (5 mm in diameter
by 1 mm in thickness) placed on top of the diamond crystal
and its spectrum was obtained by the co-addition of 32
scans, under 4 cm21 resolution. After photoactivation (Blue-
phase, Ivoclar-Vivadent, Schaan, Liechtenstein, 24 J/cm2)
and dry-storage for 24 h at 378C, a new spectrum was
obtained in a similar manner, assuring an intimate contact
between the composite and the ATR crystal by means of the
accessory pressure point. DC was calculated according to
the ratio of the absorption bands corresponding to carbon
DEVELOPMENT OF CAP/ETHYLENE GLYCOL DIMETHACRYLATE PARTICLES
 ORIGINAL RESEARCH REPORT

TABLE I. Characterization of the Particles Synthesized in the Study

Equivalent
spherical
Full width at Nanoparticle diameter (D50,
Functionalizing Monomer: half maximum long axis Organic content Density Surface in mm, D10–D90
agent (NH4)H2PO4 (FWHM)a (in nm)b (in wt%) (g/cm3) area (m2/g) in parentheses)

Non-functionalized N/A 0.150 142 (61) N/A 2.4 28 25 (13–50)

EGDMA 1:1 0.188 111 (69) 0.6 2.5 25 26 (14–51)
2:1 0.254 86 (36) 1.3 2.4 43 25 (12–56)
DEGDMA 1:1 0.181 83 (40) 34.4 1.7 40 12 (6–23)
2:1 0.217 111 (46) 30.3 1.8 66 20 (11–47)
TEGDMA 1:1 0.257 70 (40) 14.1 2.2 16 19 (10–37)
2:1 0.188 136 (60) 27.2 1.9 55 20 (11–52)
TETDMA 1:1 0.179 42 (18) 0.8 2.4 10 17 (9–30)
2:1 0.200 91 (36) 12.9 2.2 63 16 (8–27)

N/A, not applicable.

a
FWHM represents the width of the highest intensity peak of the diffractogram (at 10.58) measured at the y-axis value corresponding to half of
its maximum amplitude. Higher values correspond to smaller crystallite sizes.
b
Obtained from TEM images; average and standard deviation in parentheses (n 5 100).

double bonds located at 1,638 cm21 (aliphatic) and Flexural modulus (E, in GPa) was calculated according
1,610 cm21 (aromatic) according to the formula (n 5 3): to the following equation:
 
polymerized bPa2
DC5 12 3100 E5
non2polymerized xh3

where E is the flexural modulus (in GPa), b is a constant

related to the deflection at the center of the disk (0.509), P
Mechanical properties is the load (in Newtons), a is the disk radius (in mm), x is
Disc-shaped specimens (12 3 1 mm, n 5 10) were made the deflection corresponding to P and h is the disk thickness
using a stainless steel split mold and photoactivating each (in mm).
quadrant for 20 s (24 J/cm2 per quadrant, Bluephase, Ivo-
clar Vivadent). As an additional control, a set of unfilled Ion release
resin specimens was prepared. After 24-h storage in dis- Disc-shaped specimens (5 3 1 mm, n 5 5) were prepared
tilled water at 378C, specimens were fractured on a “piston using a silicone mold. A single 20-s irradiation was used to
on three spheres” testing jig positioned under the crosshead photoactivate the material (24 J/cm2, Bluephase, Ivoclar-
of a universal testing machine (Instron model 5565, Instron Vivadent). After 24-h storage at 378C under 100% relative
Corp, Canton MA). The center of the specimen was loaded humidity, specimens were individually immersed in 5 mL of
at a crosshead speed of 0.5 mm/min and specimen deflec- NaCl solution (133 mmol/L) buffered to pH 7 using HEPES
tion was monitored by a contact transducer (model W- solution (50 mmol/L). The immersion medium was totally
E401-E, Instron). Biaxial flexural strength (BFS, in MPa) was replaced by fresh solution weekly, for four weeks. Calcium
calculated using the equations below: and phosphorous concentrations in the collected solutions
were estimated by colorimetric methods (DU 800 Spectro-
20:2387PðX2Y Þ photometer Beckman Coulter, Brea, CA).
BFS5
b2
Statistical analysis
 2   2 Degree of conversion data was analyzed using one-way
r2 12m r2 ANOVA. Mechanical properties and ion release data were
X5ð12mÞln 1
r3 2 r3 analyzed Kruskal–Wallis test, with multiple comparisons
performed using Dunn test. A significance level of 0.05 was
"  2 #  2 adopted in all cases.
r1 r1
Y 5ð12mÞ 12ln 1ð12mÞ
r2 r3 RESULTS
Particles characterization
where P is the failure load (in Newtons); b is the specimen Characteristics of the synthesized particles are shown in
thickness (in mm); is the specimen Poisson’s ratio (0.3); r1 Table I. Diffractograms (Figure 2) revealed that all particles
is the radius of the circle where the spheres were posi- presented the peaks characteristic of DCPD (International Cen-
tioned (5.0 mm); r2 is the loading piston radius (0.6 mm); ter for Diffraction Data/Joint Committee on Powder Diffrac-
r3 is the specimen radius (in mm). tion, PDF #09–0077). Functionalized particles showed peaks

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | MONTH 2018 VOL 00B, ISSUE 00 3
FIGURE 2. Diffractograms of the calcium phosphate particles obtained with the addition of different ethylene glycol dimethacrylate monomers
to the ammonium phosphate solution. Left: 1: 1 monomer-to-ammonium phosphate molar ratio, right: 2: 1 molar ratio.

with higher full width at half maximum (FWHM) in relation to
NF, an indicative of smaller crystallite size. At the nanoscale,
agglomerates of plate-like nanoparticles were observed, with
long axis ranging from 42 to 142 nm (Figure 3). At the micro-
scale, particles and nanoparticle agglomerates with median
sizes ranging from 12 to 26 mm were observed (Figure 4).
The highest functionalization level was achieved with
the addition of DEGDMA at 1:1 ratio. EGDMA (both 1:1 and
2:1 ratios) and TETDMA at 1:1 showed very low retention
rates compared to the other conditions tested. For
2  EG  4, the organic content on the particles was
inversely related to the length of the spacer group. The use
of a higher monomer ratio resulted in increased surface
area and, except for DEGDMA, lower density.
DCPD-containing resin materials
DC was not affected by the presence of DCPD, with values
ranging from 49 to 56% (p > 0.05, Table II). Flexural

FIGURE 3. Examples of transmission electron microscopy (TEM) images showing plate-like nanoparticles agglomerates obtained with different
syntheses conditions. (A) DEGDMA 1:1, (B) TETDMA 1:1, (C) TEGDMA 2:1, (D) non-functionalized.

4 NATALE ET AL. DEVELOPMENT OF CAP/ETHYLENE GLYCOL DIMETHACRYLATE PARTICLES

 ORIGINAL RESEARCH REPORT

FIGURE 4. Examples of scanning electron microscopy (SEM) images showing particles obtained with different syntheses conditions. (A)
DEGDMA 1:1, (B) TEGDMA 1:1, (C) TETDMA 1:1, (D) non-functionalized.

properties of the resulting resin-based materials are DISCUSSION

shown in Figure 5. Among the DCPD-containing materials,
DEGDMA 2:1 showed the highest average fracture
strength, 39% higher than the material with NF
(p < 0.001). Conversely, the use of functionalized DCPD
resulted in 19–26% lower flexural modulus than those
obtained with NF (p < 0.001).
Calcium release was not affected by particle functionali-
zation and remained relatively steady during the 28-day
period (p > 0.05, Table III). Weekly values ranged from 7 to
18 ppm. Phosphate was released in lower concentrations
than calcium and similar patterns were observed among
materials, with higher concentrations being released at 7
days (average: 4.2 ppm) in comparison to the remaining
weeks (averages between 0.8 and 1.1 ppm). Similar concen-
trations were observed among materials at 7 and 28 days,
while a few differences were observed at 14 and 21 days
(p < 0.001).
Particle characterization
The use of different EG monomers affected the particle syn-
thesis outcomes. Though all particles presented their typical
large tabular habit,11 the addition of EG monomers resulted
in smaller crystallites compared to the non-functionalized
control due to the reduction in surface tension of the reac-
tion medium, which facilitates crystal nucleation.22 The role
of EG monomers as co-solvents during DCPD precipitation
seems independent of their retention on the particle surface
as, for instance, EGDMA reduced crystallite size but did not
bind to Ca21. Also, 25–54% smaller crystallites were
observed for EG  2 added at 1:1 ratio, probably because at
2:1 ratio the increase in the viscosity of the reaction
medium decreases ionic diffusivity, partially offsetting the
effect of the reduced surface tension.
The low monomer retention in particles synthesized in
the presence of EGDMA suggests that the oxygen atoms

TABLE II. Averages and SD for degree of conversion (in %). No statistically significant differences were detected among mate-
rials (one-way ANOVA, p > 0.05)

Non-functionalized DEGDMA 1:1 DEGDMA 2:1 TEGDMA 2:1 TETDMA 2:1

55.9 (6.1) 54.8 (1.7) 51.4 (5.3) 51.4 (2.8) 48.7 (3.7)

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | MONTH 2018 VOL 00B, ISSUE 00 5
FIGURE 5. Averages and standard deviations for biaxial flexural strength (left) and flexural modulus (right) of resin-based materials containing
30 vol % of DCPD particles functionalized with different ethylene glycol derivatives. The unfilled resin and a material containing non-
functionalized particles were used as controls. Values accompanied by the same letter are not statistically different (Kruskal–Wallis/Dunn tests,
p > 0.05).

adjacent to the bulky methacrylate groups are not suffi-
ciently exposed to bind to calcium ions. On the other hand,
TEGDMA and TETDMA (particularly the later) showed sig-
nificant retention only when added at 2:1 ratio possibly
because their complex tridimensional structures reduce the
likelihood of calcium-oxygen interaction. On the other hand,
the high retention rates presented by DEGDMA (30–34 wt
%) even at 1:1 ratio can be ascribed to its intermediate
chain spacer length and the presence of a central oxygen
atom.
Functionalization increased particle surface area only for
those synthesized at 2:1 monomer-to-ammonium dihydro-
gen phosphate ratio. Possibly, the high monomer content
partially inhibited particle growth and/or increased the
space among them, preventing their direct contact and plate
stacking. In fact, density values and TEM images suggest
that the nanoparticles agglomerates of functionalized DCPD
are less packed compared to NF agglomerates (Figure 6).
The low surface areas of TEGDMA- and TETDMA-
functionalized particles synthesized at 1:1 ratio may be the
result of incomplete monomer removal after sonication in
acetone. Though removal was verified using infra-red spec-
trometry, this technique may not be sensitive enough to
detect very small amounts of residual organic material. Not-
withstanding, the relationship between nanoparticle size
and surface area cannot be totally disregarded, as the syn-
thesis using 2:1 EGDMA resulted in nanoparticles 39%
smaller and with 54% higher surface area compared to NF.
In general, particle size distribution showed values simi-
lar to those reported in previous studies.16,17 Lower
medians (between 20 and 50%) was observed for particles
functionalized with monomers with EG  2. The only excep-
tion were the particles synthesized with TETDMA (1:1),
which showed 32% smaller particles than non-
functionalized DCPD even though its organic content was
only 0.8 wt %. At this point, it is important to highlight the
fact that syntheses did not produce only nanoparticles, and
particle size distribution describes both nanoparticles
agglomerates and micron-sized particles. Dimethacrylates
are apolar molecules and, therefore, they do not contribute

TABLE III. Averages and SD for Ca21 and HPO22 4 concentrations (in ppm) released by resin-based materials containing 30 vol
% of DCPD as a function of functionalization and immersion period. Ratios following monomer names refer to the “monomer:
ammonium phosphate” ratio used in the synthesis. No statistically significant differences were detected among groups for
calcium (Ca21) release (p > 0.05). For hydrogen phosphate (HPO22
4 ), similar lower-case letters indicate absence of statistically
significant differences within the same column, and upper-case letters, within the same row (Kruskal–Wallis/Dunn’s test,
p < 0.05)

Immersion (days)
7 14 21 28
21
Ca Non-functionalized 8.5 (1.1) 10.0 (4.7) 11.2 (2.9) 14.3 (3.7)
DEGDMA 1:1 10.9 (3.8) 12.9 (5.8) 17.5 (2.2) 11.4 (2.1)
DEGDMA 2:1 11.2 (2.1) 12.4 (5.1) 10.8 (3.6) 11.3 (2.3)
TEGDMA 2:1 9.7 (2.7) 9.9 (2.9) 12.5 (4.1) 13.7 (3.0)
TETDMA 2:1 11.4 (4.0) 7.0 (1.8) 8.2 (2.7) 9.1 (4.0)
HPO22
4 Non-functionalized 3.9 (1.1) Aa 1.0 (0.2) Bab 1.0 (0.2) Bab 0.9 (0.2) Ba
DEGDMA 1:1 5.0 (1.3) Aa 1.3 (0.4) ABa 1.3 (0.3) ABa 0.9 (0.3) Ba
DEGDMA 2:1 5.2 (0.5) Aa 1.2 (0.3) Ba 0.9 (0.2) Bab 0.8 (0.1) Ba
TEGDMA 2:1 5.3 (1.5) Aa 1.6 (0.4) ABa 1.6 (0.3) ABa 1.0 (0.3) Ba
TETDMA 2:1 1.7 (0.2) Aa 0.6 (0.1) Bb 0.6 (0.1) Bb 0.5 (0.1) Ba

6 NATALE ET AL. DEVELOPMENT OF CAP/ETHYLENE GLYCOL DIMETHACRYLATE PARTICLES


FIGURE 6. Schematic representations of a non-functionalized DCPD agglomerate and a DEGDMA-functionalized DCPD agglomerate. The lower
density and higher surface area of the nanoparticles suggest a less packed structure for the latter.
to reduce agglomeration or prevent particle growth, as
organic species containing ionizable groups such as carbox-
ylic acids, for instance.23
Resin-based materials containing DCPD particles
The relatively low DC values recorded were due to the tech-
nique used to obtain the spectra. Because the IR beam
interacts with the sample at the interface between the ATR
crystal and the composite, the resulting spectrum is slightly
different than those obtained in transmission methods, for
instance. Nevertheless, DC values remain valid for compari-
son purposes.24 Resin materials containing DCPD showed
similar DC to the unfilled resin. Therefore, differences in
mechanical properties and ion release among materials can
be ascribed solely to the characteristics of the particles
combined with the resin. Calcium phosphates are not rein-
forcing fillers and the decrease in strength verified when
DCPD particles were added to the BisGMA-TEGDMA resin
was expected.6 The DCPD-filled materials tested in the pre-
sent study were not modelled after heavily filled, reinforced
restorative composites. Rather, they were prepared as a
vehicle to highlight the possible gain in mechanical proper-
ties in comparison to the use of non-functionalized DCPD
particles. In fact, the use of DEGDMA-functionalized par-
ticles (2:1) led to a 39% increase in strength compared to
the material with NF. The increase in strength is the result
of the co-polymerization of monomers in the particle with
those from the resin matrix. Besides the possibility of ion-
dipole interaction between calcium ions and the oxygen in
the ether groups (which are considered stronger than
dipole-dipole forces, e.g., hydrogen bonding), functionaliza-
tion also reduces the interfacial tension between DCPD and
the dimethacrylates in the matrix, improving the wetting of
the particle surface and, therefore, promoting a better
mechanical interlocking between both phases.
Though a high monomer content seemed to be the pri-
mary factor for increasing fracture strength, it is important
to bring attention to the fact that the higher the organic
content of the particle/agglomerate, the lower is its actual
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | MONTH 2018 VOL 00B, ISSUE 00
ORIGINAL RESEARCH REPORT
DCPD content. In other words, particle volume (Vp) can be
expressed as follows:
Vp 5VDCPD 1 Vmonomer 5ðm=qÞDCPD 1ðm=qÞmonomer
where m is the mass fraction and q is the density (in g/
cm3). Based on monomer mass fractions and density val-
ues, it is possible to calculate the effective DCPD content
in the particles. Utilizing the density values of 1.082
(DEGDMA), 1.092 (TEGDMA) and 1.082 (TETDMA, all val-
ues obtained from Sigma-Aldrich) and solving the equation
for a particle mass of 1 g, the effective DCPD contents by
volume in the particles are 46% (DEGDMA 1:1), 51%
(DEGDMA 2:1), 55% (TEGDMA 2:1) and 75% (TETDMA
2:1). Thus, it would be licit to question if differences in
fracture strength are reflecting the differences in effective
DCPD content. However, SEM images show that particles
micromorphology is similar, regardless of their monomer
content. Also, it was verified in previous studies that speci-
men fracture surfaces display similar features, regardless
of particle functionalization.16 Therefore, one must con-
sider that though their effective DCPD content may differ,
they must be considered (and they behave as) individual
particles.
Notwithstanding, flexural modulus results suggest that
functionalization does impart different mechanical behaviors
to DCPD particles and agglomerates. In agreement with pre-
vious studies, the addition of calcium phosphate particles to
the unfilled resin led to significant increases in modulus, as
this property relies on the physical impingement of particles
on one another, regardless of their chemical interaction with
the resin matrix.6,9,17,25 However, materials containing func-
tionalized DCPD particles presented 19–26% lower flexural
modulus in comparison to the material with NF, suggesting
that highly functionalized DCPD behave as low-modulus fill-
ers. Considering that elastic modulus is one of the determi-
nants of composite polymerization stress,26 the reduction in
flexural modulus imparted by these particles may bring the
additional benefit of reducing stress magnitude in restor-
ative composites.
7
 The fact that calcium release did not decay over a 28-
day period was also verified in previous investigations.7,17,27
In fact, cumulative calcium release in a BisGMA-TEGDMA
composite containing 20 wt % of amorphous calcium phos-
phate reached a plateau after 35 days.28 Though ion release
does not vary linearly with calcium phosphate content,29 a
higher release could be expected from the material with 30
vol % of NF DCPD in comparison to that containing
DEGDMA 1:1, where the effective DCPD content is 14 vol %.
However, that was not the case. It is possible that the effect
of a lower DCPD content was offset by the higher surface
area of the particles and agglomerates.10 Phosphate release,
on the other hand, showed much higher values at 7 days in
comparison to the subsequent weeks. Also, released concen-
trations were much lower than for calcium. This seems to
be a characteristic of crystalline calcium phosphates due to
the structural role of the phosphate tetrahedra. Also, in cal-
cium phosphate hydrated forms oxygen atoms establish
hydrogen bonds with the structural water molecules, which
further hinders their release.30 It is important to mention
that experimental composites containing 20 vol % of
TEGDMA-functionalized DCPD (monomer fraction in the
agglomerate: 32 wt %) were able to promote 23% of min-
eral recovery of enamel caries-like lesions subjected to a
pH-cycling regime in vitro (unpublished observations).
In conclusion, among the synthesized DCPD particles,
those functionalized with DEGDMA presented the most favor-
able characteristics in terms of monomer content and particle
size distribution. Besides resulting in increased strength com-
pared to NF particles, functionalization did not impair ion
release. Also, the reduction in flexural modulus may be advan-
tageous as a strategy to develop low-contraction stress mate-
rials and deserves further investigation.
CONFLICTS OF INTEREST
The authors declare no conflicts of interest regarding
authorship and/or information presented in this article.
REFERENCES
1. Langhorst SE, O’Donnell JNR, Skrtic D. In vitro remineralization of
enamel by polymeric amorphous calcium phosphate composite:
Quantitative microradiographic study. Dent Mater 2009;25:884–891.
2. Weir MD, Chow LC, Xu HH. Remineralization of demineralized
enamel via calcium phosphate nanocomposite. J Dent Res 2012;
91:979–984.
3. Melo MA, Weir MD, Rodrigues LK, Xu HH. Novel calcium phos-
phate nanocomposite with caries-inhibition in a human in situ
model. Dent Mater 2013;29:231–240.
4. Vogel GL, Schumacher GE, Chow LC, Takagi S, Carey CM. Ca pre-
rinse greatly increases plaque and plaque fluid F. J Dent Res
2008;87:466–469.
5. Souza JG, Tenuta LM, Del Bel Cury AA, Nobrega DF, Budin RR,
de Queiroz MX, Vogel GL, Cury JA. Calcium prerinse before fluo-
ride rinse reduces enamel demineralization: An in situ caries
study. Caries Res 2016;50:372–377.
6. Xu HH, Moreau JL. Dental glass-reinforced composite for caries
inhibition: calcium phosphate ion release and mechanical proper-
ties. J Biomed Mater Res B Appl Biomater 2010;92:332–340.
7. Chiari MD, Rodrigues MC, Xavier TA, de Souza EM, Arana-Chavez
VE, Braga RR. Mechanical properties and ion release from bioac-
tive restorative composites containing glass fillers and calcium
phosphate nano-structured particles. Dent Mater 2015;31:726–733.
8 NATALE ET AL.
8. Skrtic D, Antonucci JM, Eanes ED. Improved properties of amor-
phous calcium phosphate fillers in remineralizing resin compo-
sites. Dent Mater 1996;12:295–301.
9. Arcis RW, Lopez-Macipe A, Toledano M, Osorio E, Rodriguez-
Clemente R, Murtra J, Fanovich MA, Pascual CD. Mechanical
properties of visible light-cured resins reinforced with hydroxyap-
atite for dental restoration. Dent Mater 2002;18:49–57.
10. Xu HH, Weir MD, Sun L. Nanocomposites with Ca and PO4
release: effects of reinforcement, dicalcium phosphate particle
size and silanization. Dent Mater 2007;23:1482–1491.
11. Rodrigues MC, Hewer TL, Brito GE, Arana-Chavez VE, Braga RR.
Calcium phosphate nanoparticles functionalized with a dimethacry-
late monomer. Mater Sci Eng C Mater Biol Appl 2014;45:122–126.
12. Anthony J, W., Bideaux RA, Bladh KW, Nichols MC. Mineralogical
Society of America In: Chantilly V, editor. Handbook of Mineral-
ogy. Chantilly, VA: Mineralogical Society of America; 2000.
13. Shortall AC, Palin WM, Burtscher P. Refractive index mismatch
and monomer reactivity influence composite curing depth. J Dent
Res 2008;87:84–88.
14. Dorozhkin SV. Calcium orthophosphates in dentistry. J Mater Sci
Mater Med 2013;24:1335–63.
15. Yoshida Z, Kihara S. Polyoxyethylene ether surfactants on ion
transfer across aqueous organic solution interfaces studied by
polarography with the electrolyte dropping electrode.
J Electroanal Chem 1987;227:171–181.
16. Alania Y, Chiari MD, Rodrigues MC, Arana-Chavez VE, Bressiani
AH, Vichi FM, Braga RR. Bioactive composites containing
TEGDMA-functionalized calcium phosphate particles: Degree of
conversion, fracture strength and ion release evaluation. Dent
Mater 2016;32:e374–e381.
17. Rodrigues MC, Xavier TA, Arana-Chavez VE, Braga RR. Polymer-
based material containing calcium phosphate particles functional-
ized with a dimethacrylate monomer for use in restorative den-
tistry. J Biomater Appl 2016;31:871–877.
18. Sideridou I, Tserki V, Papanastasiou G. Study of water sorption,
solubility and modulus of elasticity of light-cured dimethacrylate-
based dental resins. Biomaterials 2003;24:655–665.
19. Kalachandra S, Kusy RP. Comparison of water sorption by meth-
acrylate and dimethacrylate monomers and their corresponding
polymers Polymer 1991;32:7.
20. Condon JR, Ferracane JL. Assessing the effect of composite formu-
lation on polymerization stress. J Am Dent Assoc 2000;131:497–503.
21. Goncalves F, Pfeifer CS, Ferracane JL, Braga RR. Contraction
stress determinants in dimethacrylate composites. J Dent Res
2008;87:367–371.
22. Mehta SK, Chaudhary S, Bhasin KK. Spectral characterization and col-
loidal properties of 1-hexadecylpyridinium chloride in aqueous
binary mixtures of different glycols. J Colloid Interface Sci 2009;333:9.
23. Domingo C, Arcis RW, Lopez-Macipe A, Osorio R, Rodriguez-
Clemente R, Murtra J, Fanovich MA, Toledano M. Dental composites
reinforced with hydroxyapatite: mechanical behavior and absorp-
tion/elution characteristics. J Biomed Mater Res 2001;56:297–305.
24. Rueggeberg FA, Caughman WF, Chan DC. Novel approach to mea-
sure composite conversion kinetics during exposure with stepped
or continuous light-curing. J Esthet Dent 1999;11:197–205.
25. Braga RR, Yamamoto T, Tyler K, Boaro LC, Ferracane JL, Swain
MV. A comparative study between crack analysis and a mechani-
cal test for assessing the polymerization stress of restorative com-
posites. Dent Mater 2012;28:632–641.
26. Ferracane JL, Hilton TJ. Polymerization stress–is it clinically
meaningful? Dent Mater 2016;32:1–10.
27. Xu HH, Moreau JL, Sun L, Chow LC. Nanocomposite containing
amorphous calcium phosphate nanoparticles for caries inhibition.
Dent Mater 2011;27:762–769.
28. Zhang L, Weir MD, Chow LC, Antonucci JM, Chen J, Xu HH.
Novel rechargeable calcium phosphate dental nanocomposite.
Dent Mater 2016;32:285–293.
29. Xu HH, Weir MD, Sun L, Takagi S, Chow LC. Effects of calcium
phosphate nanoparticles on Ca-PO4 composite. J Dent Res 2007;
86:378–383.
30. Arsic J, Kaminski D, Poodt P, Vlieg E. Liquid ordering at the
Brushite-{010}-water interface. Phys Rev B 2004;69.
DEVELOPMENT OF CAP/ETHYLENE GLYCOL DIMETHACRYLATE PARTICLES