Journal of Dentistry 52 (2016) 7078

Contents lists available at ScienceDirect

Journal of Dentistry
journal homepage: www.intl.elsevierhealth.com/journals/jden

Mechanical and structural characterization of discontinuous

ber-reinforced dental resin composite
Jasmina Bijelic-Donova, CDT DDSa,* , Sufyan Garoushia , Lippo V.J. Lassilaa ,
Filip Keulemansb , Pekka K. Vallittua,c
a
Department of Biomaterials Science and Turku Clinical Biomaterials Centre-TCBC, Institute of Dentistry, University of Turku, Itinen Pitkkatu 4 B, 20520
Turku, Finland
b
Dental Materials Science, Dental School, Ghent University, De Pintelaan 185/P8, B-9000 Gent, Belgium
c
City of Turku Welfare Division, Oral Health Care, Lemminkisenkatu 2, 20520 Turku, Finland

A R T I C L E I N F O A B S T R A C T

Article history: Objectives: This study evaluated several ber- and matrix related factors and investigated different
Received 14 March 2016 mechanical properties of discontinuous i.e. short ber-reinforced composite (SFRC) (everX Posterior,
Received in revised form 18 July 2016 eXP). These were compared with three conventional composites, microlled G-nial Anterior (GA),
Accepted 20 July 2016
nanolled Supreme XTE (SXTE) and bulk-ll Filtek Bulk-Fill (FBF).
Methods: Fracture toughness (KIC), exural strength (FS), exural modulus (FM), compressive strength
Keywords: (CS), diametral tensile strength (DTS), apparent horizontal shear strength (AHSS) and fracture work (Wf)
Short ber-reinforced composite
were determined for each composite (n = 8) stored dry or in water. SEM analysis of the ber diameter (df)
Critical ber length
Volume fraction
(n = 6) and orientation (n = 6) were performed. The theoretical critical ber length (lfc) and the aspect
Fiber orientation ratio (l/d) of SFRC were calculated and the volume fraction of discontinuous bers (Vf%) and the ber
Toughening mechanism length (lf) of SFRC were evaluated. The results were statistically analyzed with two-way ANOVA
Aspect ratio (a = 0.05).
Results: The mechanical properties of SFRC (eXP) were generally superior (p < 0.05) compared with
conventional composites. GA had the highest FM (p > 0.05), whereas FBF had the highest AHSS (p < 0.05).
The ber related properties Vf%, l/d, lf, lfc and df of eXP were 7.2%, 18112, 0.31.9 mm, 0.851.09 mm and
17 mm respectively. SEM results suggested an explanation to several toughening mechanisms provided
by the discontinuous bers, which were shown to arrest crack propagation and enable a ductile fracture.
Water exposure weakened the mechanical properties regardless of material type. Wf was unaffected by
the water storage.
Conclusion: The properties of this high aspect ratio SFRC were dependent on the ber geometry (length
and orientation) and matrix ductility.
Clinical signicance: The simultaneous actions of the toughening mechanisms provided by the short bers
accounted for the enhanced toughness of this SFRC, which toughness value matched the toughness of
dentin. Hence, it could yield an inherently uniform distribution of stresses to the hard biological tissues.
2016 Elsevier Ltd. All rights reserved.

1. Introduction
Composites reinforced with discontinuous bers are known as
short ber-reinforced composites (SFRC). Earlier attempts to
improve the mechanical properties of composites by incorporating
short bers had limited success. This was due to the difculty of
maintaining their high aspect ratio (l/d), which is important for the
* Corresponding author at: TCBC, Institute of Dentistry, University of Turku,
Turku, Finland. Itinen Pitkkatu 4 B, FI-20520 Turku, Finland.
E-mail addresses: jasmina.bijelic@utu., jasminabijelic@gmail.com
(J. Bijelic-Donova).
high strength performance of the composite [1] and the ber
length above its critical value [2]. Indeed, some SFRCs contain
microber llers (60120 mm) [3,4], which are well below the
critical ber length (0.51.6 mm) [2]. Due to the insufcient ber
length, micrometer bers act as microller particles instead of
ber ller. This gives the SFRC properties comparable with
particulate ller composites (PFCs) [5,6].
Interpretation of the properties of SFRCs is more difcult than is
for unidirectional bers and composites reinforced with them. This
could be due to the fact that unidirectional bers are assumed to be
straight, perfectly circular at the cross-section, to have same mean
diameter [7] and to be closely packed [8]. In contrast, short bers

http://dx.doi.org/10.1016/j.jdent.2016.07.009
0300-5712/ 2016 Elsevier Ltd. All rights reserved.
 J. Bijelic-Donova et al. / Journal of Dentistry 52 (2016) 7078 71

and Vf, equally important is the length of the short bers. Indeed,
Nomenclature
these factors together determine the ber debonding from the
matrix during loading of the SFRC structure. As an example, the
eXP or SFRC Millimeter-scale short ber-reinforced com- debonding for longer aligned bers is greater than for random
posite or high aspect ratio short ber-rein- ones, which undergo only limited debonding. In other words, the
forced composite load bearing capacity of SFRCs is more effective with short random
PFC Particulate ller composite than with longer aligned short bers [15]. Furthermore, favorably
GA or MFR G-nial Anterior, microlled hybrid restorative oriented bers would allow the bridging phenomenon, which is
composite material important ber toughening mechanism. Fiber bridging is most
SXTE or NF Supreme XTE, nanolled restorative composite effective for transversally aligned bers, intermediate for random-
material ly orientated bers and absent for non-reinforced systems. Fiber
FBF Filtek Bulk Fill, owable bulk ll restorative breakage, on the other hand, is important as it introduces new
composite material length in the system, while the Vf remains unchanged [14].
KIC Fracture toughness Apparently, among the factors that determine the mechanical
FS Flexural strength performance of SFRC are the length [2,16], type, arrangement and
FM Flexural modulus distribution of the bers [1]. Other factors, known as intrinsic
CS Compressive strength (ber related) and extrinsic (matrix related) factors, also affect the
DTS Diametral tensile strength SFRCs properties [17]. The ber related factors of SFRCs could be
AHSS Apparent horizontal shear strength considered to be the lf, df, Vf, strength as well as the interfacial
Wf Fracture work ber-matrix bonding, whereas the matrix related factor could be
SEM Scanning electron microscopy considered to be the FOD [17]. In addition, the random distribution
FOD Fiber orientation distribution of short bers could be compromised or intentionally affected
Wt% Weight fraction during the manipulation process. If discontinuous bers are long
Vol% Volume fraction enough, they behave as long bers giving anisotropic mechanical
Vf% Fiber volume fraction properties [10,11,18]. To date, there exist some research on the
df Fiber diameter mechanical properties of recently introduced millimeter-scale
lf Fiber length discontinuous ber-reinforced composite (SFRC) (everX Posterior,
lfc Critical ber length GC Corp., Tokyo, Japan) [1823], but none on its ber related and
lfo Optimal ber length matrix related factors. Based on its ber l/d, this SFRC could be
l/d Aspect ratio classied as high aspect ratio SFRC i.e. millimeter-scale SFRC.
Consequently, the present study was designed to rstly
investigate several mechanical properties of this millimeter-scale
SRFC in comparison with three conventional PFCs (G-nial
are often assumed to be randomly distributed, to have a cylindrical Anterior, GC; Supreme XTE, 3M Espe, MN, USA and Filtek Bulk Fill,
shape and an elliptical cross section [9]. Nevertheless, the short 3M Espe) (Table 1) and secondly to evaluate both ber related and
bers could also be transversally aligned (i.e. in plane). The matrix related factors of the SFRC. The selected properties were
mechanical properties of SFRCs essentially depend on the ber fracture toughness (KIC), exural strength (FS), exural modulus
orientation distribution (FOD), aspect ratio (l/d) and their volume (FM), compressive strength (CS), diametral tensile strength (DTS),
fraction (Vf) [10,11]. There is, however, critical l/d for the apparent horizontal shear strength (AHSS) and fracture work (Wf).
transversally orientated short bers above which further improve- The ber related factors of the millimeter-scale SFRC were
ment of their mechanical properties declines [10]. For engineering analyzed on both a theoretical basis and experimentally. On a
SFRC this is between 6 and 9 [10], whereas the lowest l/d for dental theoretical basis, the volume fraction of the short bers (Vf%) and
SFRC is 5.2 [12]. With increasing l/d they start behaving as long the theoretical critical ber length (lfc) were calculated. Experi-
bers [10,11], which macroscopically are transversally isotropic mentally determined ber related factors were the ber diameter
materials [13], although in an x-y system they show anisotropic (df), the ber length (lf) and the aspect ratio (l/d), whereas the
properties [14]. Similarly, transversally aligned short bers have experimentally determined matrix related factor was the ber
also been described as macroscopically transversally isotropic orientation distribution (FOD). The null hypotheses were that the
materials [11]. By contrast, the properties of randomly orientated mechanical properties of the millimeter-scale SFRC will not differ
discontinuous bers are mostly affected by their Vf [10]. Thus, from the PFCs properties and that short-term water storage would
these are macroscopically isotropic materials, with most effective not affect the mechanical properties of the investigated materials.
mechanical properties at Vf of 10% [11]. Consequently, if the bers
are homogeneously distributed with random orientation, the SFRC 2. Materials and methods
is macroscopically homogenous and isotropic, whereas if the bers
are homogeneously distributed with transverse (in plane) orien- FS, FM, DTS, CS, AHSS and KIC were evaluated. Eight specimens
tation, the SFRC is homogeneous and anisotropic i.e. transversally (n = 8) were prepared per restorative composite material listed in
isotropic. Furthermore, in order to evaluate various physicome- Table 1 for each test type and each storage condition. The storage
chanical properties of the SFRCs on macroscopic level, numerous conditions were dry and in water at 37
C for 7 and 30 days,
theoretical, computational based models have been proposed. respectively. Testing procedures were performed at room temper-
Such models are mechanical-based micromechanical numerical ature (23  1
C). The ber diameter (n = 6) and orientation (n = 6)
model [14], energy-based homogenization model [15] and were determined immediately upon preparation of the specimens.
mesostructured with work-energy approach model [13]. These The ber volume fraction, ber length, critical ber length and
models give an insight on what circumstances initiate fracture in aspect ratio were additionally computed.
the SFRC structure, what factors inuence its progression, what FS and FM were evaluated by the three-point bending test.
causes the ber debonding or breakage, how the FOD affects the Specimens were prepared, polymerized and tested following the
stiffness of the SFRC system and how FOD interacts with the protocol previously described [19].
ber l/d. These models have shown that in addition to the FOD, l/d
72 J. Bijelic-Donova et al. / Journal of Dentistry 52 (2016) 7078

Table 1
Materials used in the study.

Brand Type Lot nos Filler Resin Filler load Wt%

*
everX Posterior Millimeter-scale SFRC Silicon dioxide (max. 5 wt%), Barium glass (max. 70 wt%) E-glass Bis-GMA, Average ller load
(GC Corp., short ber- ber (max. 15 wt%) TEGDMA, 67.7
Tokyo, Japan) reinforced PMMA Average glass ber
composite load 8.6
G-nial Anterior Microlled resin MFR 1305311 Pre-polymerized: silica containing & strontium and lanthanoid UDMA, 7080
(GC Corp., hybrid uoride containing ller 1617 mm, Silica ller >100 nm and dimethacrylate
Tokyo, Japan) Fumed silica ller <100 nm co-monomers
Filtek Nanolled NF N649930 Silica: 520 nm nanoparticle Bis-GMA, Silica nanoparticle 8
Supreme XTE Zirconia/silica: 0.61.4 mm nanocluster UDMA, Bis- Zirconia/silica
(3M Espe, St. EMA, TEGDMA nanocluster 71
Paul, MN, USA)
Filtek Flowable bulk ll FBF N600679 Zirconia/silica: 0.013.5 mm Bis-GMA, 64.5
Bulk Fill Ytterbium triuoride (YbF3): 0.15.0 mm UDMA, Bis-
(3M Espe, St. EMA, Procrylat
Paul, MN, USA)

Bis-GMA: bisphenol-A-diglycidyl dimethacrylate; TEGDMA: triethylene glycol dimethacrylate; PMMA: polymethyl methacrylate; UDMA: urethane dimethacrylate; Bis-EMA:
bisphenol-A-polyethylene-glycol-diether dimethacrylate.
*
Eight (8) batches of everX Posterior were used, 1301232, 1301252, 1306101, 1306262, 1308261, 1309191, 1307082 and 1308292.

Cylindrical specimens, 6 mm in height and 4 mm in diameter,
were used to evaluate DTS and CS. The fabrication and testing
procedures of the specimens are described elsewhere [19]. DTS in
megapascals (MPa) was calculated using Eq. (1) [24]:
DTS= 2 F/p l D
where: DTS is the diametral tensile strength, F is the maximum
applied load in newtons (N); D is the diameter of the specimens in
mm and l is the length of the specimen in mm. CS was calculated in
megapascals (MPa) using Eq. (2) [24]:
CS = 4 F/p D2
where: CS is the compressive strength, F is the maximum applied
load in N and D is the diameter of the specimen in mm.
AHSS was determined by the short-beam method, according to
ASTM D 447596 [25]. A translucent mold with an inner diameter
of 3.6 mm and a length of 14.5 mm (0.2 mm), which is one
diameter greater than the test span was used to fabricate the rod
shaped specimens. The mold was placed on a Mylar-strip-covered
glass slide and the restorative material was packed into the mold.
The top surface was attened with Mylar strip and glass slide and
each specimen was polymerized on both sides for 20 s by ve
overlapping irradiations with a light-curing polymerization unit
(Elipar S10, 3M ESPE, St. Paul, MN, USA) at an intensity of
1600 mW/cm2 and wavelength ranging from 430 to 480 nm. The
irradiance of the light-curing polymerization unit was measured
with dental radiometer (Demetron L.E.D. Radiometer, Kerr,
Middleton, WI, USA). Upon removal from the mold, each
longitudinal side was further polymerized for 20 s in ve
overlapping places. The rod specimen was placed with its long
axis perpendicular and its midpoint centered to the loading nose.
The side supports were adjusted to a span length of 11 mm
(0.2 mm) and specimens were loaded by a universal material
testing machine (Lloyd model LRX, Lloyd Instruments Ltd.,
Fareham, UK) at a crosshead speed of 1.0 mm/min until failure.
The diameter of each specimen was measured prior to testing with
a digital caliper and the apparent horizontal shear strength in MPa
was calculated in accordance with Eq. (3):
AHSS= 0.849 P/d2
where AHSS is the apparent horizontal shear strength; P is the
breaking load in N and d is the specimen diameter in meters (m).
Fracture toughness (KIC) was evaluated using single-edge
notched bend (SENB) specimens (2.5  5  25 mm) according to
the adapted ISO 20795-1 standard method [26], and as previously
described [19,20]. The SENB specimens were used for determining
the total fracture work (Wf) in J/m2, by using Eq. (4):
Wf = U/[2B (H  a)] 1000 (4)
(1)
where Wf is the fracture work, U is the recorded area under the
load-deection curve and represents the energy required to break
R
the whole specimen. U = PdD in newton millimeters (Nmm),
where D is the measured deection for load P in N; B is the
specimen width in mm, H is the specimen height in mm, and a is
the crack length in mm.
(2)
Fiber volume fraction (Vf %) was calculated using Eq. (5) [27]:
Wf =rf
Vf %  100 5
Wf =rf Wm =rm
where W is the weight fraction (wt%), r is the density, f represents
the ber and m the matrix. Wf = 8.6 wt% (Wf of the short E- glass
bers in everX Posterior) and rf = 2.56 g/cm3 [27]. The matrix
comprises resin and particulate llers with r(resin) = 1.23 g/cm3
[28], W(resin) = 23.7 wt%, r(particulate ller) = 2.8 g/cm3 and w(particulate
ller) = 67.7 wt%.
The ber length (lf) was measured using the direct technique, by
separating the bers from the matrix, following the technique
previously described [20]. Eight batches of millimeter-scale SFRC
(everX Posterior) were evaluated. Fibers were deposited onto glass
slide and examined with light microscopy (Leica; Leica Mycro-
systems, Wild, Heerbrugg, Switzerland). The average ber length
was then dened as LN = S Ni Li/S Ni, where LN is the number
average of the ber length, Ni is the number of bers with length Li
[27]. Four thousand bers were measured in total (500/batch).
When the ber diameter (df), the ultimate tensile strength (sfu) of
the glass ber and the interfacial shear strength (ti) of the glass
ber reinforced composite are known, then the theoretical critical
ber length (lfc) can be calculated from lfc = sfudf/2 ti[29]. The
tensile strength of E-glass ber is estimated at 2.0 GPa, whereas the
interfacial shear stress (ti) for a glass ber reinforced composite
varies between 17.6 and 33.8 MPa depending on the ber
conditioning and the testing environment [30], but typically is
20 MPa [27]. The diameter of the individual glass ber in our study
(3)
was 17 mm, whereas the interfacial shear stress evaluated by
macroscopic short-beam test was 15.6 MPa. This experimentally
obtained result was then compared with the typical values
presented in the literature [27].
 J. Bijelic-Donova et al. / Journal of Dentistry 52 (2016) 7078 73

In the present investigation the ber orientation distribution Table 3

Fiber related properties of the millimeter-scale SFRC (everX Posterior) investigated
(FOD) was visually studied by sectioning the SENB specimens (12
in this study.
slices obtained from n = 6) along their long-axis plane with a high
performance precision cut-off saw at 300 rpm (Struers Secotom 50, Vf % df lfc lfc lfo lf l/d
(mm) (mm) (mm) (mm) (mm)
Struers, Ballerup, Denmark) and analyzing their images by
scanning electron microscopy (SEM, JSM 5500, Jeol Ltd., Tokyo, everX Posterior 7.2 17 0.85a 1.09b 1.01.3c 0.31.9 18112
Japan). The specimens were ground wet (Stuers LabPol-21, Stuers Vf%: volume fraction of short bers in SFRC (everX Posterior); df: ber diameter; lfc:
A/S, Copenhagen, Denmark) with silicon carbide papers of critical ber length; lfo: optimal ber length; lf: ber length and l/d (measured as
decreasing abrasiveness (1000-, 1200-, 4000-grit) and gold sputter lf/df): aspect ratio.
a
Interfacial shear stress (ti) for a typical glass ber reinforced polymer composite
coated before the SEM examination. After testing the specimens,
is 20 MPa [27].
their fracture surfaces were examined with SEM. b
Interfacial shear stress (ti) evaluated by the macroscopic short-beam test used
The ber diameter (df) was also determined from SEM images, in the present investigation.
c
which were processed with the Image J program (National lfo is the optimal ber length measured according to lfo  1.2lfc [46].
Institutes of Health, USA). A cross section of a ber bundle was
examined with SEM for this purpose. The ber diameter was
determined as a mean value calculated from one hundred
0.3 mm and 1.9 mm; 71% of the ber lengths varied between
measurements per specimen (n = 6).
0.4 mm and 1.0 mm, 11% were between 1.1 mm and 1.9 mm and 18%
The data were statistically analyzed with SPSS version 23 (SPSS,
were 0.3 mm short E-glass bers. The average length of the short
IBM Corp.). Homogeneity of variance was analyzed by the Levene
bers varied among the batches. The shortest average ber length
test (p = 0.05). Two-way ANOVA (a = 0.05) followed by a Tukey post
was 0.7 mm and the longest average ber length was 0.9 mm. The
hoc test was used to determine if the mechanical properties (FS,
critical ber length was between 0.85 mm and 1.09 mm. The
FM, DTS, CS, AHSS, KIC, Wf) were different for the materials
theoretical value of 0.85 mm was obtained on a theoretical basis by
investigated in this study (SFRC, MFR, NF, FBF) and for the type of
using the value of 20 MPa for the interfacial shear stress reported in
storage condition (dry, water). Pearson correlation was used to
the literature. When the average value derived from our
analyze the correlation among the mechanical properties.
macroscopic AHSS test was used, a critical ber length value
1.09 mm was obtained.
3. Results
The SEM investigation (Figs. 1 and 2) of the ber orientation
distribution showed the orientation paths of the short bers
Table 2 displays the results for the mechanical properties (FM,
around the crack notch, which are also presented schematically in
FS, DTS, CS, AHSS, KIC and Wf) investigated in this study. In general,
Fig. 3. The short bers were aligned a. parallel to the crack notch,
water storage weakened the mechanical properties of all materials.
which is along to the long axis of the crack notch (perpendicular to
This reduction signicantly affected the FS regardless of the
the pressing direction); b. normal to the crack notch; c. transverse
material type (p = 0.001), whereas the water effect on FM, DTS,
(perpendicular) to the long axis of the crack notch; and d. a radial
AHSS and KIC was material dependent. The material type also had a
ber alignment was observed at the crack notch tip. The radial
major effect on the mechanical properties. An exception was the
alignment of the bers was obtained as the paths of transversally
CS, which was not inuenced by the material type (p = 0.947). The
and normally oriented bers cross each other (Fig. 2). The SEM
millimeter-scale SFRC (everX Posterior) had signicantly higher KIC
analysis of the tested SENB specimens (Fig. 4) showed that bers
for both storage conditions (p = 0.001), compared to conventional
bridged the crack. Fiber pullout and breakage were also observed.
PFCs. KIC of the millimeter-scale SFRC insignicantly decreased
The pullout effect occurred only for the shorter segment of the
(p = 0.051) after water storage. The millimeter-scale SFRC had
embedded ber length, which is typical in a well-bonded system.
signicantly higher Wf (p = 0.001), for both storage conditions
The longer bers, on the other hand, remained intact in the matrix.
(p = 0.001), than the conventional PFCs.
The SEM analysis of fractured SENB specimens (Fig. 5) showed
Strong, statistically signicant correlation was observed only
pulled out and protruded ber ends. Protruded bers ends were
between the KIC and the Wf (r2 = 0.816), and moderate between KIC
covered with resin, which also indicates a well-bonded system. In
and FS (r2 = 0.475), KIC and DTS (r2 = 0.440) as well as between FM
addition, imprints of debonded bers were observed in the resin
and AHSS (r2 = 0.568).
matrix.
Table 3 displays the results for the Vf%, df, lf, lfc, lfo (lfo is the
Fig. 6 shows typical load-deection curves for the tested
optimal ber length) and l/d of the short bers within the
materials. For the millimeter-scale SFRC multiple cracking of the
millimeter-scale SFRC. The Vf% of the short bers was 7.2%. The
matrix occurred during the non-elastic range, which is the portion
average ber diameter was 17 mm. The ber length varied between
between the rst cracking and the peak. The brittle matrix of the

Table 2
Mechanical properties of investigated materials.
1
Material Storage FM (GPa) FS (MPa) DTS (MPa) CS (MPa) AHSS (MPa) KIC (MPa m /2) Wf (J/m2)
MFR dry 15.0a#(1.7) 90a*(7) 39a#(4) 233a*(32) 14.1a#(0.8) 0.9a#  0.1 55a*(9)
water 14.4a#(2.0) 78a^(6) 38a#(5) 174a^(39) 13.6a#(0.6) 0.8a#  0.1 164a^(35)
NF dry 3.9b#(1.0) 104b*(11) 37a*(3) 219a#(79) 18.4bc*(5.2) 1.1a#  0.1 53a#(7)
water 3.5b#(0.6) 64b^(8) 29b^(5) 191a#(60) 16.0b^(2.5) 0.9a#  0.1 146a#(18)
FBF dry 5.7b#(1.2) 119c*(13) 43ab#(8) 217a#(25) 19.8c#(2.1) 1.1a#  0.1 88a*(14)
water 4.7b#(0.9) 83a^(7) 39a#(4) 200a#(34) 19.9c#(1.1) 0.9a#  0.1 203a^(9)
SFRC dry 12.6a*(2.8) 119c*(5) 46b*(5) 235a#(33) 15.6ab#(1.3) 2.4b*  0.5 546b#(241)
water 9.8c^(1.0) 85a^(8) 39a^(3) 190a#(31) 14.7ab#(1.5) 1.7b^  0.2 580b#(102)

Same superscript letter and sign above the values indicates groups that were statistically similar (p > 0.05). Letters describe the statistical difference among the materials
across each storage medium, whereas the signs represent the statistical difference between dry and water storage mediums across each material category level.
74 J. Bijelic-Donova et al. / Journal of Dentistry 52 (2016) 7078
Fig. 1. SEM picture of the ber orientation distribution. Black arrows indicate the
ber orientated transverse (perpendicular) to the long axis of the crack notch, white
arrows indicate ber orientation parallel to the crack notch and red arrows indicate
ber orientation in plane normal to the crack notch. (For interpretation of the
references to colour in this gure legend, the reader is referred to the web version of
this article.)
conventional PFCs cracked at the rst peak. This was followed by
composite fracture.
4. Discussion
Investigated properties of the millimeter-scale SFRC differed
from the properties of conventional composites and all properties
were affected by the storage medium, regardless of the material
type. Thus, both null hypotheses were rejected.
Fig. 2. SEM picture of the ber orientation distribution around the crack notch. X-ber orientation enabled by crossing paths of normally orientated (red arrow) and
transversely orientated (black arrow) bers. (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)
The effect of water was deterioration for all mechanical
properties evaluated at one month aging time for all investigated
composite materials, which is in good agreement with the
literature ndings [3133]. Firstly, the hydrolytic degradation in
water is presumably due to the hydrolysis of the silane coupling
agent on the surfaces of both E-glass ber and particulate llers.
Consequently, hydrolytic degradation of the polysiloxane network
between the E-glass and the matrix, as well as between the ller
and matrix occur [34]. Furthermore, water uptake is a slow, time-
dependent process, and thus the aging time of the composite in
water affects the saturation level. The period of one month could
be considered as a short-term aging. In short-term aging time, the
specimens do not sufciently equilibrate with water. Water
absorption reduces the mechanical properties of the composite
material until complete water saturation of the composite material
has been reached, which is approximately 46 weeks [32]. This is
due to the plasticization effect of water [34]. After this, however,
further reduction stops and in the long term, the mechanical
properties remain unchanged [35].
The fracture work was the only property which increased
following water storage irrespectively of the material. Although it
cannot be fully explained, this could in part be due to the water
used as storage medium, which allows plasticization of the
polymer network. Indeed, the presence of small plasticizing
molecules, such as water, may induce some exibility of the
composite matrix and consequently, it may allow pronounced
bending of the water stored specimens during testing. However,
the fracture work depends also on the microstructural ller
parameters (size or shape) [36] that could be affected by the plastic
deformation of the matrix around them.
Likewise, ller related properties such as type, size, shape and
load are important for the overall properties of the composite
materials. Higher ller loads are generally associated with
improved properties. Nonetheless, critical ller level exists above
which any addition of ller could include more defects into the
resin matrix and might have an inverse effect on the mechanical
properties. The critical ller load level has been reported to be
approximately 60 vol% (about 80 wt%) [37]. The optimal SFRC
properties, on the other hand, are reached at ber ller load level of
10 vol% [11]. Filtek Bulk Fill had lowest ller fraction among the
studied materials, whereas the inorganic fraction of the other
restorative materials was at or approached the critical ller load
level (Table 1).
 J. Bijelic-Donova et al. / Journal of Dentistry 52 (2016) 7078
Fig. 3. Schematic drawing of the ber orientation around the crack notch. Red line
shows ber orientation in plane normal to the crack notch and green parallel to the
crack notch. Blue line shows the ber orientated transverse (perpendicular) to the
long axis of the crack notch. Grey lines show the radial orientation of the bers
around the crack tip. (For interpretation of the references to colour in this gure
legend, the reader is referred to the web version of this article.)
Fig. 4. SEM picture of tested SENB specimen. Fiber bridging phenomenon (black
arrow), ber breakage (red arrow) and ber pull out (white arrow) were observed.
(For interpretation of the references to colour in this gure legend, the reader is
referred to the web version of this article.)
Fig. 5. SEM picture of fractured surface of tested SENB specimen. Protruded ber
covered with resin (black arrow), pulled out ber ends (white arrow) and in plane
normal to the crack notch orientated ber (red arrow). (For interpretation of the
references to colour in this gure legend, the reader is referred to the web version of
this article.)
75
The nanolled composite in this study had the lowest FM,
comparable mechanical properties to the other investigated PFCs
and weaker mechanical properties compared to the millimeter-
scale SFRC. The water induced interfacial debonding was shown to
reduce the strength and the fracture resistance of the nanolled
composite investigated here [38,39]. This is consistent with the
ndings of the present study. Deterioration of the mechanical
properties of nanolled (NF) composite (Supreme XTE) is due to
the loss of the structural integrity of the nanocluster, the increased
ller surfaces and the large intersurface area between ller and
organic matrix, which is the weakest link in the composite material
and has high potential to hydrolytic degradation [40,41]. The
properties of this nanolled composite were shown to be related to
the inherent porosity and the inconsistency of the nanoclusters
[40].
Except for the FS, the bulk ll composite evaluated in this
investigation, revealed comparable mechanical properties to the
NF composite, but less optimal mechanical properties compared to
the millimeter-scale SFRC and the microlled hybrid (MFR)
composite. This is in good accordance with other studies
[21,42,43], which showed that this could be due to the different
chemistry of the resin matrix of Filtek Bulk Fill that encompasses
Bis-EMA and UDMA monomer mixture [42] accompanied with the
presence of plastifying monomers [21]. Both constituents contrib-
ute to the reduction of the shrinkage stress [21,42]. However, the
lower ller fraction and the lower FM [43] could explain the
weaker mechanical properties of this bulk-ll composite.
The microlled hybrid composite (MFR) in this study showed
good mechanical properties, which could be related to the
presence of pre-polymerized particle llers, which are polymer-
ized and milled polymers heavily loaded with submicron sized
ller particles. This allows for improved resistance to fracture even
if the ller load of the composite is low, and is due to three effects
known as ller addition, stress relaxation and the branching effect
[44].
Generally, the millimeter-scale SFRC performed better in
comparison to the PFCs. High fracture toughness values were
observed for this short ber composite in this and other studies
[19,20]. The millimeter-scale short bers and the semi-IPN (semi
interpenetrating network) structure are the two main factors that
toughen this particular SFRC. The toughening mechanisms
provided by the millimeter bers are a result of their ability to
stretch, to deect crack propagation, to bridge, resist the opening
and propagation of the crack, consequently inducing a closure
force on the crack [45] as well as to reduce the stress intensity at
the crack tip. The stretching capacity of bers may enable the crack
bridging and the crack blunting mechanisms to occur (Fig. 4).
During the crack-bridging, discontinuous bers would probably
stretch between the edges of the propagating crack decreasing the
notch sensitivity and causing blunting of the initially sharp crack.
This would lower the stress concentration at the crack tip and
possibly slow down or impede the crack propagation. However, as
the millimeter-scale SFRC contains range of short ber lengths, it is
expected for the shorter ber segments to pull out and for the
longer segments to actively carry the bridging. In addition, the
pulled out bers will have various lengths because the short bers
have different embedded lengths, and the bridging will be
sustained only by the longer bridging bers [17]. These longer
bers will be responsible for increased fractured energy [46]. In
addition, the gradual ber length might contribute to the
toughening effect. Indeed, recent study showed that dual ber
length distribution enhanced SFRC properties [47]. The silanization
of the bers with methacrylate-based silane, compatible with both
methacrylates and dimethacrylates, and the presence of PMMA
(polymethyl acrylate) within the semi-IPN structure, which
reduces the stiffness of the cross-linked resin monomer, should
76 J. Bijelic-Donova et al. / Journal of Dentistry 52 (2016) 7078
Fig. 6. Typical load-deection curves of investigated materials. 1: everX Posterior; 2: Supreme XTE; 3: G-nial Anterior and 4:Filtek Bulk Fill.
be also mentioned as factors contributing to the toughening effect.
Hence, it becomes clear that the different toughening mechanisms
operate simultaneously and lead to improvement in the toughness
of this SFRC. In addition, the toughness of the investigated SFRC
matches the toughness values reported for the dentin [48], which
is important, because inherently the distribution of the stresses
between the biological hard tissue and the restorative material
substituting it would be uniform. Moreover, both have common
toughening mechanisms such as crack deection and the crack
bridging phenomena [48]. Structural similarity between the dentin
and the millimeter-scale SFRC is also observed. The collagen bers
provide the toughness of dentin similar to short bers, which
provide the toughness of the millimeter-scale SFRC.
Together, the fracture work and the fracture toughness, depict
the fracture resistance ability of the material. As these properties
were signicantly improved for the investigated SFRC, it can be
anticipated that the endurance of the material in static or fatigue
load could also be enhanced. However, it should be born on mind
that this brous material should clinically be protected with PFC
layer, and the fatigue endurance of the material combination will
signicantly be affected by the thickness of this protective layer
[49].
The reinforcement efciency of SFRCs depend on the ber l/d
and the Vf. If the ber l/d is low, the ber ends create discontinuity
in the resin matrix, act as stress concentration sides and weaken
the composite [12]. Experimental versions of high aspect ratio
SFRCs indicated that bers with lengths from 2 mm to 5 mm
provide almost equal reinforcement capacity, and that optimal
physicomechanical properties were obtained with formulations
between 8.5 and 10 Vf%. Other studies have reported comparable
ndings [23,50]. Furthermore, the lower limit for the ber aspect
ratio in dental SFRC is 5.2 [12]. Consequently, it could be concluded
that the upper limit for the length of the short ber within SFRC
systems could be considered 2 mm, optimal ber quantity could be
10 Vf%, and lower limit for bers aspect ratio 5.2. In the present
study, the gradual discontinuous bers lengths for the everX
Posterior were measured between 0.3 mm and 1.9 mm, which
makes the lowest aspect ratio 18 and greatest 112 (Table 3). The
critical ber length for the everX Posterior was measured to be
between 0.85 and 1.09 mm, which is in the range of the reported
critical ber length [2]. This critical length of the bers is
important, because short bers tend to pullout. For the longer
embedded length, only the short segment will pull out, as seen in
Fig. 4. Pulled out bers indicate that the composite continues to
carry the load, although its strength does not increase. Neverthe-
less, the composites fracture toughness, at this stage, still
continues to increase substantially [1]. If, however, the ber
length is twice the critical length, then the bers will break and the
bridging energy will decline. Thus, an optimum ber length should
be maintained, in order to allow the bridging phenomenon [46].
The optimum ber length (lfo) for composites reinforced with short
bers was determined to be approximately 1.2 times the critical
ber length lfc (lfo  1.2lfc) [46]. The optimal lfo for everX Posterior
would be approximately 1.15 mm (1.01.3 mm). It follows then,
that the short ber lengths within the investigated SFRC should be
within the range between the theoretical ber length (0.85 mm)
and the optimal ber length (1.15 mm). When reporting a value for
short ber length, a range of values rather than a xed value should
be used, because of the variations of the average short ber values,
which seemed to be consignment dependent. This is important,
 J. Bijelic-Donova et al. / Journal of Dentistry 52 (2016) 7078
because these variations could explain discrepancies in the
properties reported for the millimeter-scale SFRC. Earlier, a short
ber length between 1.3 and 2.0 mm was reported for the everX
Posterior [20], whereas recently two different ranges of short ber
length values were reported for the same material, namely 0.3
1.5 mm [47] and 1.02.0 mm [51]. The difference between the
present and those studies [20,47,51] might be due to the processing
operations of the millimeter-scale SFRC regulated by the manu-
facturer. Indeed, this type of SFRC is subjected to the mixing
method of the fabrication process, which comprises blending
reinforcing bers, llers, resin matrix and additives together. This
process generates ber fractures and results in a wider range of
millimeter-scale bers, which likely altered the mechanical
properties in the nal SFRC [19,20,47,51].
The presence of shorter and longer short bers in the
millimeter-scale SFRC system is important also for the FOD, ber
debonding and consequently reinforcement efciency. Indeed, the
longer short bers will tend to align in plane providing anisotropic
reinforcement, and shorter ber will be randomly distributed
offering an isotropic reinforcement. However, the debonding for
random bers is limited, which allows them to enhance the load
bearing capacity of the SFRC better than longer aligned short bers
[15].
Measuring the ber orientation distribution is often difcult. A
convenient method for characterizing the ber orientation
distribution of short ber composite has not yet been developed,
which is why the SEM microstructural examination was conducted
here (Fig. 1). Fiber orientation analysis supports the hypothesis of
the importance of the ber orientation in enhancing the materials
toughness. The crossing paths of transversally and normally
oriented bers enabled radial alignment of the short bers. This
type of orientation of the short bers is known as X-orientation
and is important for arresting the crack propagation at the crack
notch tip (Fig. 2). However, it seems that the ber orientation
distribution is not only dependent on the lf, df and Vf, but also on
the mold size and the placement technique. A pressing technique
would produce a planar ber orientation, implying that the short
bers would adjust against the surface they are pressed to and
adapt individually according to its shape.
Failure types for the millimeter-scale SFRC involved matrix
cracking, ber fracture and ber pull out. They all seemed to be
related to factors such as FOD and lf. Indeed, post-failure
examination of both the fractured and tested single-edge notched
bend specimens surfaces suggests ductile behavior of the bers
with failure occurrence in the matrix. This was evident at the
protruded ber ends covered with resin (Fig. 5) as well as in the
bridging phenomenon (Fig. 4), as the bers bridge the matrix crack.
The nding is supported by the load-deection curve (Fig. 6),
suggesting that this would be possible only if the pullout resistance
of bers at the rst crack is greater than the load at the rst
cracking. It seems that owing to the presence of short bers, SFRC
continues to carry tensile stresses after the rst cracking [1].
Contrary to this, the conventional PFCs fracture immediately at the
rst peak, which indicates matrix cracking. Fiber breakage in this
study was also observed, probably at those bers in which length
surpassed the optimal ber length (>1.15 mm). This nding could
be however advantageous when repairing SFRC restorations. In
addition, the SEM investigation revealed uniform ber distribu-
tion, which is rstly important for avoiding localized domination of
matrix properties (at regions with less bers) and secondly for
avoiding physical contact between bers [27].
Mechanical testing (Table 2), the ber properties characteriza-
tion (Table 3), and the SEM analysis (Figs. 1, 2, 4 and 5) suggest that
1) longer discontinuous bers (30%) would enable bridging; 2)
short discontinuous bers (18%) would behave as particulate llers
instead of ber llers; 3) some of the discontinuous bers in the
77
middle range will provide pullout toughening mechanism (0.4
0.5 mm) and the longer ones (0.6 mm0.7 mm) would actively
carry the loading; and 4) small portion (6%) of the bers exceeding
the optimal ber length could break or then behave as continuous
bers. The bers with varying length distributions affect the FOD
and contribute to the toughening effect.
Clinically, a material that has high probability of stopping an
initiated fracture would preserve the tooth structure, because it
will absorb the dissipated energy released during cracking and
stop the progression of the fracture toward the tooth structure. The
material itself would chip, delaminate or crack, but all these are
favorable fracture types. On contrary, a material with low or no
ability of stopping the progression of the initiated fracture will
fracture. This is also accompanied by fracture of the tooth
structure. These phenomena are more visible for restorations
where most of the tooth structure is already lost and replaced with
a restoration. In such situations, the selection of the material to
replace the missing tooth tissue is of paramount importance and a
material with high fracture toughness should be preferred.
Although the results of this investigation cannot be used to infer
how the ber ller volume or ber ller size affect the fracture
toughness, the present study ascertains that the short E-glass
bers affect signicantly the fracture toughness. The ber
orientation distribution obviously warrants further research and
the ber orientation in various dental cavities prepared on teeth,
approximal box and mesial-occlusal-distal (MOD) cavity could be
studied next.
5. Conclusion
In conclusion, the brous composite comprising millimeter-
scale discontinuous E-glass bers llers with a semi-IPN matrix
showed improved tolerance to crack propagation (fracture
toughness) compared to the tested conventional composites
containing particulate llers. The toughening mechanisms provid-
ed by the short bers were the reduction of the stress intensity at
the crack tip, crack blunting and bridging phenomena. The
properties of this millimeter-scale SFRC are dependent on groups
of factors: 1) Fiber strength and structure related factors: breaking
of the embedded short bers during manufacturing process, and
orientation and length of the short bers; 2) Matrix related factors:
viscosity affecting the short bers orientation and distribution,
and porosity of the matrix such as air bubbles inclusions, pores or
voids and 3) Interfacial factors: silane coupling agent type of the
bers and the particulate llers, and reaction with matrix.
Conicts of interest
The authors report no conicts of interests. Author PK Vallittu
consults Stick Tech Ltd., a member of the GC Group in RD and
training.
Acknowledgements
The authors thank Keme Scientic for the English language
editing of this manuscript. The authors also thank 3M ESPE and
Stick Tech Ltd. (member of GC group), for generously providing
materials used in this investigation. This study is part of the BioCity
Turku Biomaterials Research Program (www.biomaterials.utu.).
References
[1] S.M. Lee, Handbook of Composite, Reinforcements, rst ed., VCH Publishers,
Inc, New York, USA, 1992.
78 J. Bijelic-Donova et al. / Journal of Dentistry 52 (2016) 7078
[2] P. Vallittu, High-aspect ratio llers: ber-reinforced composites and their
anisotropic properties, Dent. Mater. 31 (2015) 17.
[3] J.L. Drummond, L. Lin, K.J. Miescke, Evaluation of fracture toughness of a ber
containing dental composite after exural fatigue, Dent. Mater. 20 (2004) 591
599.
[4] J. Manhart, K.H. Kunzelmann, H.Y. Chen, R. Hickel, Mechanical properties of
new composite restorative materials, J. Biomed. Mater. Res. 53 (2000) 353
361.
[5] J.W. van Dijken, K. Sunnegrdh-Grnberg, Fiber-reinforced packable resin
composites in class II cavities, J. Dent. 34 (2006) 763769.
[6] T.C. Fagundes, T.J. Barata, C.A. Carvalho, E.B. Franco, J.W. van Dijken, M.F.L.
Navarro, Clinical evaluation of two packable posterior composites: a ve-year
follow-up, J. Am. Dent. Assoc. 140 (2009) 447454.
[7] S.W. Yurgartis, Measurement of small angle ber misalignments in continuous
ber composites, Compos. Sci. Technol. 30 (1987) 279293.
[8] A.P. Wilczynski, A basic theory of reinforcement for unidirectional brous
composites, Compos. Sci. Technol. 38 (1990) 327337.
[9] Y.T. Zhu, W.R. Blumenthal, T.C. Lowe, Determination on non-symmetric 3-D
ber-orientation distribution and average ber length in short-ber
composites, J. Compos. Mater. 31 (1997) 12871301.
[10] S. Kari, H. Berger, U. Gabbert, Numerical evaluation of effective material
properties of randomly distributed short cylindrical bre composites, Comput.
Mater. Sci. 39 (2007) 198204.
[11] H. Berger, S. Kari, U. Gabbert, R. Rodriguez-Ramos, J. Bravo-Castillero, R.
Guinovart-Diaz, Evaluation of effective material properties of randomly
distributed short cylindrical ber composites using a numerical
homogenization tenchnique, J. Mech. Mater. Struct. 2 (2007) 15611570.
[12] P. Shouha, M. Swain, A. Ellakwa, The effect of ber aspect ratio and volume
loading on the exural properties of owable dental composite, Dent. Mater.
30 (2014) 12341244.
[13] A.L. Kalamkarov, H.Q. Liu, A new model for the multiphase ber-matrix
composite materials, Compos. Part B Eng. 29B (1998) 643653.
[14] R. Brighenti, D. Scorza, Numerical modelling of the fracture behaviour of brittle
materials reinforced with unidirectional or randomly distributed bres, Mech.
Mater. 52 (2012) 1227.
[15] R. Brighenti, D. Scorza, A micro-mechanical model for statistically
unidirectional and randomly distributed bre-reinforced solids, Math.
Mech. Solids 17 (2012) 876893.
[16] R.C. Petersen, Discontinuous ber-reinforced composites above critical length,
J. Dent. Res. 84 (2005) 365370.
[17] L.K. Jain, R.C. Wetherhold, Effect of ber orientation on the fracture toughness
of brittle matrix composites, Acta Metall. Mater. 40 (1992) 11351143.
[18] S. Garoushi, L.V. Lassila, P.K. Vallittu, The effect of span length of exural
testing on properties of short ber reinforced composite, J. Mater. Sci. Mater.
Med. 23 (2012) 325328.
[19] J. Bijelic-Donova, S. Garoushi, P.K. Vallittu, L.V. Lassila, Mechanical properties
fracture resistance, and fatigue limits of short ber reinforced dental
composite resin, J. Prosthet. Dent. 115 (2016) 95102.
[20] S. Garoushi, E. Silynoja, P.K. Vallittu, L. Lassila, Physical properties and depth
of cure of a new short ber reinforced composite, Dent. Mater. 29 (2013) 835
841.
[21] J.G. Leprince, W.M. Palin, J. Vanacker, J. Sabbagh, J. Devaux, G. Leloup, Physico-
mechanical characteristics of commercially available bulk-ll composites, J.
Dent. 42 (2014) 9931000.
[22] C. Goracci, M. Cadenaro, L. Fontanive, G. Giangrosso, J. Juloski, A. Vichi, M.
Ferrari, Polymerization efciency and exural strength of low-stress
restorative composites, Dent. Mater. 30 (2014) 688694.
[23] K. Al-Ahdal, N. Ilie, N. Silikas, D.C. Watts, Polymerization kinetics and impact of
post polymerization on the degree of conversion of bulk-ll resin-composite at
clinically relevant depth, Dent. Mater. 31 (2015) 12071213.
[24] International Organization for Standardisation, ISO 4104. Dental zinc
Polycarboxylate Cements, ISO, Geneva, 1984.
[25] Americal Society for Testing, Materials, ASTM D4475-96, Standard Test
Method for Apparent Horizontal Shear Strength of Pultruded Reinforced
Plastic Rods by the Short-beam Method, ASTM, Philadelphia, 2008 Available
at: http://www.astm.org/DATABASE.CART/HISTORICAL/D4475-96.htm.
[26] International Organization for Standardisation. ISO 20795-1. DentistryBase
polymersPart 1: Denture base polymers. ISO, Geneva, 2013. Available at:
http://www.iso.org/iso/store.htm.
[27] D. Hull, T.W. Clyne, An Introduction to Composite Materials, second ed.,
Cambridge University Press, Cambridge, 1996.
[28] M. Obukuro, Y. Takahashi, H. Shimizu, Effect of diameter of glass bers on
exural properties of ber-reinforced composites, Dent. Mater. J. 27 (2008)
541548.
[29] D.A. Norman, R.E. Robertson, The effect of ber orientation on the toughening
of short ber-reinforced polymers, J. Appl. Polym. Sci. 90 (2003) 27402751.
[30] W.G. McDonough, J.M. Antonucci, J.P. Dunkers, Interfacial shear strengths of
dental resin-glass bers by the microbond test, Dent. Mater. 17 (2001) 492
498.
[31] J.L. Ferracane, H.X. Berge, J.R. Condon, In vitro aging of dental composites in
watereffect of degree of conversion ller volume, and ller/matrix coupling,
J. Biomed. Mater. Res. 42 (1998) 465472.
[32] J.L. Ferracane, V.A. Marker, Solvent degradation and reduced fracture
toughness in aged composites, J. Dent. Res. 71 (1992) 1319.
[33] I.S. Medeiros, M.N. Gomes, A.D. Loquercio, L.E. Filho, Diametral tensile strength
and Vickers hardness of a composite after storage in different solutions, J. Oral
Sci. 49 (2007) 6166.
[34] L.V. Lassila, T. Nohrstrm, P.K. Vallittu, The inuence of short-term water
storage on the exural properties of unidirectional glass ber-reinforced
composites, Biomaterials 23 (2002) 22212229.
[35] P.K. Vallittu, Effect of 10 years of in vitro aging on the exural properties of
ber-reinforced resin composites, Int. J. Prosthodont. 20 (2007) 4345.
[36] C. Baudin, R. Osorio, M. Toledano, S. de Aza, Work of fracture of a composite
resin: fracture-toughening mechanisms, J. Biomed. Mater. Res. 89A (2009)
751758.
[37] N. Ilie, R. Hickel, Investigations on mechanical behaviour of dental composites,
Clin. Oral Invest. 13 (2009) 427438.
[38] M.B. Shan, J.L. Ferracane, J.J. Kruzic, R-curve behavior and toughening
mechanisms of resin-based dental composites: effects of hydration and
post-cure heat treatment, Dent. Mater. 25 (2009) 760770.
[39] A.R. Curtis, A.C. Shortall, P.M. Marquis, W.M. Palin, Water uptake and strength
characteristics of a nanolled resin-based composite, J. Dent. 36 (2008) 186
193.
[40] A.R. Curtis, W.M. Palin, G.J. Fleming, A.C. Shortall, P.M. Marquis, The
mechanical properties of nanolled resin-based composites: characterizing
discrete ller particles and agglomerates using a micromanipulation
technique, Dent. Mater. 25 (2009) 180187.
[41] N. Ilie, R. Hickel, Macro-, mico- and nano-mechanical investigations on
silorane and methacrylate-based composites, Dent. Mater. 25 (2009) 810819.
[42] R.Z. Alshali, N. Silikas, J.D. Satterthwaite, Degree of conversion of bulk-ll
compared to conventional resin-composites at two time intervals, Dent. Mater.
29 (2013) e213217.
[43] H.M. El-Damanhoury, J. Platt, Polymerization shrinkage stress kinetics and
related properties of bulk-ll resin composites, Oper. Dent. 39 (2014) 374382.
[44] K.H. Kim, Y.B. Kim, O. Okuno, Microfracture mechanisms of composite resins
containing prepolymerized particle llers, Dent. Mater. J. 19 (2000) 2233.
[45] S.H. Kim, D.C. Watts, Effect of glass-ber reinforcement and water storage on
fracture toughness (KIC) of polymer-based provisional crown and FPD
material, Int. J. Prosthodont. 17 (2004) 318322.
[46] R.C. Wetherhold, L.K. Jain, The toughness of brittle matrix composites
reinforced with discontinuous bers, Mater. Sci. Eng. 151A (1992) 169177.
[47] L. Lassila, S. Garoushi, P.K. Vallittu, E. Silynoja, Mechanical properties of ber
reinforced restorative composite with two distinguished ber length
distribution, J. Mech. Behav. Biomed. Mater. 60 (2016) 331338.
[48] R.K. Nalla, J.H. Kinney, R.O. Ritchie, Effect of orientation on the in vitro fracture
toughness of dentin: the role of toughening mechanisms, Biomaterials 24
(2003) 39553968.
[49] G.T. Rocca, C.M. Saratti, M. Cattani-Lorente, A.J. Feilzer, S. Scherrer, I. Krejci, The
effect of a ber reinforced cavity conguration on load bearing capacity and
failure mode of endodontically treated molars restored with CAD/CAM resin
composite overlay restorations, J. Dent. 43 (2015) 11061115.
[50] R.B. Fonseca, A.S. Marques, O. Bernades Kde, H.L. Carlo, L.Z. Naves, Effect of
glass ber incorporation on exural properties of experimental composites,
Biomed. Res. Int. (2014) 2014:542678.
[51] H. Abouelleil, N. Pradelle, C. Villat, N. Attik, P. Colon, B. Grosgogeat, Comparison
of mechanical properties of a new ber reinforced composite and bulk lling
composites, Restor. Dent. Endod. 40 (2015) 262270.