Adhesion of Resin Materials to S2-glass

Unidirectional and E-glass Multidirectional

Fiber Reinforced Composites:
Effect of Polymerization Sequence Protocols
Petr Polaceka / Vladimir Pavelkab / Mutlu Özcanc

Purpose: To evaluate the effect of different polymerization sequences employed during application of bis-GMA-
based particulate filler composites (PFC) or a flowable resin (FR) on fiber-reinforced composite (FRC).
Materials and Methods: Unidirectional, pre-impregnated S2-glass fibers (Dentapreg) and multidirectional pre-
impregnated E-glass fibers (Dentapreg) (length: 40 mm; thickness: 0.5 mm) were obtained (N = 144, n = 12 per
group) and embedded in translucent silicone material with the adhesion surface exposed. The resulting specimens
were randomly divided into 12 groups for the following application sequences: a) FRC+PFC (photopolymerized
in one step), b) FRC+FR (photopolymerized in one step), c) FRC+PFC (photopolymerized individually), d) FRC+FR
(photopolymerized individually), e) FRC (photopolymerized)+intermediate adhesive resin and PFC (photopolymer-
ized in one step), f) FRC (photopolymerized)+intermediate adhesive resin and FR (photopolymerized in one step).
The sequences of unidirectional (groups a to f) were repeated for multidirectional (groups g to l) FRCs. PFCs were
debonded from the FRC surfaces using the shear bond test in a universal testing machine (1 mm/min). On ad-
ditional specimens from each FRC type, thermogravimetric analysis (TGA) was performed to characterize the fiber
weight content (Wf) (N = 6, n = 3 per group). After debonding, all specimens were analyzed using SEM to categor-
ize the failure modes. The data were statistically analyzed using 3-way ANOVA and Tukey’s tests (_ = 0.05).
Results: Significant effects of the FRC type (S2 or E-glass) (p < 0.01), resin type (PFC or FR) (p < 0.01) and
polymerization protocol (p < 0.05) were observed on the bond strength (MPa). Interaction terms were also sig-
nificant (p < 0.05). The multidirectional FRC groups (g to l) showed significantly lower bond strengths (2.3 ± 0.2
to 7.3 ± 0.3 MPa) than did the unidirectional FRC groups (a to f) (10.7 ± 0.6 to 24.4 ± 0.8 MPa). Among the
unidirectional FRC groups, the highest values were obtained with protocol f (24.4 ± 0.8), followed by protocol
e (18.6 ± 0.4). PFC adhesion to unidirectional FRC was lower when FRC and PFC were polymerized in one step
(protocol a: 11.3 ± 0.5) than individual polymerization (protocol c: 14.1 ± 0.5). The opposite situation was
true for FR (protocol b: 17.5 ± 0.4; protocol d: 10.7 ± 0.6). Groups a to f presented exclusively mixed failures
(a combination of partial cohesive failure in the PFC and adhesive failure between the FRC and PFC) and groups
g to l demonstrated only adhesive failure (intact FRC no cohesive failure of PFC). TGA revealed 55 ± 3 wt% fiber
content for multidirectional and 60 ± 3 wt% for unidirectional FRCs tested.
Conclusion: Multidirectional pre-impregnated E-glass fibers cannot be recommended in combination with the PFC
and FR materials tested in this study. Application of an intermediate adhesive resin layer increases the adhesion
of both PFC and FR to unidirectional FRC. FRC and FR can be polymerized in one step, but FRC and PFC combina-
tions should be polymerized individually.
Keywords: bond strength, dental materials, intermediate adhesive resin, fiber-reinforced-composites, flowable
resin.

J Adhes Dent 2013; 15: 507–510. Submitted for publication: 24.03.13; accepted for publication: 11.06.13

a Research Fellow, Faculty of Chemistry, Brno University of Technology, Czech
Republic. Performed the experiments, discussed results and commented on
manuscript at all stages.
b
Research Fellow, Faculty of Chemistry, Brno University of Technology, Czech
Republic. Analyzed the data, discussed results and commented on manu-
script at all stages.
c Professor, University of Zurich, Dental Materials Unit, Center for Dental and
Oral Medicine, Clinic for Fixed and Removable Prosthodontics and Dental
Materials Science, Zurich, Switzerland. Designed study, wrote manuscript,
discussed results and commented on manuscript at all stages.
Correspondence: Prof. Dr. med. dent. Mutlu Özcan, Dental Materials Unit,
Center for Dental and Oral Medicine, Clinic for Fixed and Removable Prostho-
dontics and Dental Materials Science, University of Zurich, Plattenstrasse 11,
CH-8032 Zurich, Switzerland. Tel: +41-44-63 45600, Fax: +41-44-63 44305.
e-mail: mutluozcan@hotmail.com

Vol 15, No 6, 2013 507

Polacek et al
Fig 1 Schematic
0.5
drawing of the PFC/
FRC FR (indicated in
blue) and the dimen-
sions of the speci-
mens (3 mm thick
1.3 × 3 mm wide × 4 mm
40
long if applied on
PFC 4 or 5 unidirectional FRC,
and 3 mm thick
× 4 mm wide × 5 mm
long if applied on
3
3 or 4 multidirectional FRC
surfaces).
I nlay-retained fixed dental prostheses (FDPs) made of
fiber-reinforced composites (FRC) are cost effective alter-
natives to their metal-ceramic or all-ceramic counterparts.
Unidirectional or multidirectional FRC materials are recom-
mended depending on the type of loading exerted on the
FDP.1 Their fabrication is typically based on application
procedures of resin-based materials such as flowable resin
(FR) or particulate filler composites (PFC) with and without
intermediate adhesive resin. Polymerization of each of
these resin materials individually would increase the time
required for fabrication while also decreasing the ability of
the polymerized layer to react with the subsequent resin
layer. On the other hand, photopolymerization of all resin
materials in one step during application may decrease
the fabrication time and increase the interfacial adhesion
between each layer.2,7 When building up resin-based ma-
terials, the surface of successive layers is exposed to air.
The presence of oxygen throughout photopolymerization
results in the formation of a resin-rich inhibited layer at
the uppermost surface of the resin. For radical chain po-
lymerizations, oxygen is a powerful inhibitor reacting with
radicals to form unreactive peroxide radicals that retard or
even terminate polymerization by reacting with themselves
or other propagating radicals to form inactive products re-
sulting in a poorly polymerized resin-rich surface layer.7,10
In the incremental build-up technique, new resin can bond
covalently to the already polymerized resin, either through
free radical polymerization between the old and new mono-
mer of the resin, or through a combination of free radical
polymerization and interdiffusion of the monomers of new
resin with the substrate resin. Covalent bonding, based on
free radicals, takes place when unreacted carbon-carbon
double bonds open to form carbon-carbon single bonds.
Bonding associated with interdiffusion occurs if the polym-
erized substrate is partly made of linear polymer and cross-
linked polymer and if the monomers of the new polymer
can dissolve the substrate.4,5 The proper polymerization of
successive resin layers, be they FR or PFC on the FRC, is
crucial for the success of an FRC FDP.
The survival rate of indirect FRC FDPs ranges between
72% and 86% from 2 to 5 years with failures due to
debonding or a combination of chipping and fracture,10
indicating some further potential improvement in their
survival rate. Since the manufacturers of the available
FRC materials advise different polymerization techniques
and protocols during build up of such restorations, some
508
confusion is present regarding clinical procedures. The
objectives of this study were a) to evaluate the effect
of different polymerization sequences employed during
the application of bis-GMA-based PFC or FR on the FRC,
b) to analyze the failure types after debonding, and c) to
determine the fiber weight content in FRC. The following
hypotheses were tested: a) Photopolymerization of FRC
and PFC in one step would result in better adhesion than
individual polymerization of these materials; b) applica-
tion of an intermediate adhesive resin on the FRC would
deliver better adhesion of the PFC and FR on the FRC
compared to direct adhesion of PFC and FR on the FRC.
MATERIALS AND METHODS
Unidirectional, pre-impregnated S2-glass fibers (ADM,
Dentapreg; Brno, Czech Republic) and multidirec-
tional pre-impregnated E-glass fibers (ADM, Dentapreg)
(length: 40 mm; thickness: 0.5 mm) were obtained
(N = 144, n = 12 per group), embedded in translucent
silicone material with the adhesion surface exposed, and
photopolymerized for 5 min (Targis Power, Ivoclar Vivadent;
Schaan, Liechtenstein). The PFC and FR were applied in
a silicone mold of 3 mm thickness × 3 mm width × 4 mm
length if applied on unidirectional FRC, and 3 mm thick-
ness × 4 mm width × 5 mm length if applied on multidi-
rectional FRC, and then photopolymerized. The resulting
specimens were randomly divided into 12 groups for the
following application and polymerization sequences: a)
FRC+PFC (Boston C&B, Arkona; Lublin, Poland), photopo-
lymerized in one step for 10 min; b) FRC+FR (Gradia Direct
Flow, GC; Leuven, Belgium), photopolymerized in one step
for 10 min; c) FRC+PFC, photopolymerized individually
each for 5 min; d) FRC+FR, photopolymerized individually
each for 5 min; e) FRC (photopolymerized) + intermedi-
ate adhesive resin (Adper Single Bond Plus, 3M ESPE;
Seefeld, Germany) and PFC, photopolymerized in one step
for 10 min; f) FRC photopolymerized for 5 min + intermedi-
ate adhesive resin and FR photopolymerized in one step
for 5 min. Photopolymerization was achieved using a la-
boratory polymerization device (Targis Power; wavelength:
400 to 580 nm; power ouput: 1000 mW/cm2).
The sequences were performed for both unidirectional
(groups a to f) and multidirectional FRCs (groups g to l).
The specimen dimensions are presented in Fig 1. PFCs
were debonded from the FR and FRC surface using the
shear bond test in a universal testing machine (1 mm/
min) (Zwick Z010; Ulm, Germany). The bond strength oA
(MPa) was calculated from the following equation:
oA F (1)
SA • SB
where: F = loading force (N)
SA = height of PFC/FR block (mm)
SB = width of PFC/FR block (mm)
After debonding, substrate (FRC) and adherend (PFC/
FR) of all specimens were analyzed using a scanning
The Journal of Adhesive Dentistry
 Polacek et al

Table 1 Bond strength data (MPa) of the studied

groups and the failure types

FRC Type Group Mean bond Failure type

strength (MPa)
Unidirectional a 11.3 ± 0.5B Mixed
b 17.5 ± 0.4D Mixed
c 14.1 ± 0.5C Mixed
a
d 10.7 ± 0.6A Mixed
e 18.6 ± 0.4E Mixed
f 24.4 ± 0.8F Mixed
Multidirectional g 6.0 ± 0.2J Adhesive between
FRC and PFC
h 2.3 ± 0.2G Adhesive between
FRC and FR
i 4.6 ± 0.3I Adhesive between
FRC and PFC
j 3.4 ± 0.4H Adhesive between b
FRC and FR
k 7.3 ± 0.3K Adhesive between
FRC and PFC
l 5.4 ± 0.2J Adhesive between
FRC and FR
*The same superscript letters indicate no significant differences (Tukey’s
test, _ = 0.05)

electron microscope (SEM) (Philips 30; Brno, Czech Re-

public) to categorize the failure modes.
On additional specimens from each FRC type, thermo-
gravimetric analysis (TGA) (Perkin Elmer TGA 6, Perkin
Elmer; Waltham, MA, USA) was performed to determine
the fiber weight content (Wf) in FRC (N = 6, n = 3 per group)
under a nitrogen atmosphere. FRC specimens of 10 to
15 mg were heated to 50°C and equilibrated for 1 min.
Then, the FRC specimens were heated from 50°C up to
550°C at a rate of 25°C/min with a holding time at 550°C
for 10 min. The bond strength data were statistically ana-
lyzed using 3-way ANOVA and Tukey’s tests (_= 0.05).
RESULTS
Significant effects of the FRC type (S2 or E-glass)
(p < 0.01), resin type (PFC or FR) (p < 0.01), and poly-
merization protocol (p < 0.05) were observed on the
bond strength data (MPa). Interaction terms were also
significant (p < 0.05).
Multidirectional FRC groups presented significantly lower
bond strengths in all conditions (2.3 ± 0.2 to 7.3 ± 0.3)
compared to those of unidirectional FRC groups (10.7 ± 0.6
to 24.4 ± 0.8) (Table 1). In combination with the unidirec-
tional FRC, the highest values were obtained with protocol
f (24.4 ± 0.8), followed by protocol e (18.6 ± 0.4). PFC
adhesion to unidirectional FRC was lower when FRC and
PFC were polymerized in one step (protocol a: 11.3 ± 0.5)
compared to when individually polymerized (protocol c:
14.1 ± 0.5). The situation was the opposite for FR (protocol
b: 17.5 ± 0.4; protocol d: 10.7 ± 0.6).
c
Fig 2 Representative SEM images of the unidirectional S2-
glass FRC (50X) and PFC or FR (150X) surfaces after debonding.
a) Mixed type of failure where FRC remained intact with some
cohesive failure in the PFC; b) mixed type of failure in the FRC
with some remnants of FRC on the FR surface accompanied by
some cohesive failures of the FR; c) adhesive failure of between
the FRC and PFC. Note the intact surface of the FRC and no co-
hesive failure in the PFC.
Groups a to f presented exclusively mixed failures
(a combination of partial cohesive failure in the PFC and
adhesive failure between the FRC and PFC) and groups
g to l demonstrated only adhesive failure (intact FRC no
cohesive failure of PFC) (Table 1, Figs 2a to 2c). TGA
revealed 55 ± 3 wt% fiber content for the multidirectional
and 60 ± 3 wt% for unidirectional FRCs tested (Fig 3).
DISCUSSION
The hypothesis that the photopolymerization of FRC and
PFC in one step would result in higher bond strength than
individual polymerization of these materials was accepted
only partially, because the PFC material (groups a and c)
exhibited the opposite trend. Since the application of an
intermediate adhesive resin on the FRC increased the ad-
hesion of the PFC and FR on the FRC compared to direct
adhesion of PFC and FR on the FRC, the second hypoth-
esis was accepted. It must be noted that in this study,

Vol 15, No 6, 2013 509

Polacek et al
Thermogravimetric Analysis
120
100
Weight (%)
80
41.06%
(4.317 mg)
60
40
0 100 200 300 400
Temperature (°C)
Fig 3 Characteristic thermogravimetric analysis of the uni-
directional S2-glass FRCs studied, indicating the loss of the
resin during heating and the amount of fiber left in weight%.
rectangular PFC bars, not cylinders, were bonded to FRC
in order to eliminate the possible stress concentration at
the top (tangent) of the cylinders.6
Interestingly, FR and PFC exhibited 2 to 5 times higher
adhesive strength to unidirectional than to multidirec-
tional FRC. In principle, in terms of chemistry, similar
adhesion could have been expected from PFC and FR
(both bis-GMA based) to both FRC types. The low bond
strength data obtained in the multidirectional FRC group
could be attributed to the difference in the orientation of
the fibers against the exerted loading and possible inferior
wettability compared to the unidirectional FRC. Although
FR has a lower modulus of elasticity than PFC, higher bond
strengths were obtained when the FRs were polymerized in
one step with the FRC both with and without intermediate
adhesive resin, where the latter resulted in even higher
mean bond strength. This could be a consequence of the
lower viscosity of FR vs PFC. Less viscous materials pen-
etrate better into the micropores of the FRC surface than
do more viscous ones.10 In a previous study, load direc-
tion was found to affect the shear bond strength of PFC to
unidirectional FRC preimpregnated with linear polymethyl
methacrylate and dimethacrylate monomer: the highest
results (46.6 MPa) were obtained when shear force was
applied perpendicular to the bonding surface compared to
transverse loading (27.6 MPa). However, in that study FRC
surfaces were ground flat with 1200-grit silicone carbide
that might have increased the surface roughness. Thus,
the results were higher than those obtained in this study
(11.3 to 24.4 MPa).3 In another similar study, the adhe-
sion of low-viscosity diacrylate veneering composite resin
and restorative composite resin to unidirectional FRC with
multiphase polymer matrix was tested.5 The use of an
intermediate adhesive resin containing bis-GMA and TEG-
DMA increased the adhesion of PFC (15 MPa) compared
to the control group, where no adhesive resin was used
(1 MPa).5 Similarly, application of an intermediate adhe-
sive resin in this study significantly increased the bond
strength to unidirectional FRC (groups e and f) compared
to that of other groups (a to d). A similar trend was also
observed when intermediate adhesive resin was used on
510
the multidirectional FRC (groups k,l) with the exception of
0.3 group g. Although oxygen inhibition is defined as a reduc-
tion in the completeness of a free radical polymerization
through reaction with atmospheric oxygen, its effect to
Derivation Weight (%/°C)
0,2 bond strength is not clear.8 Not only the presence or thick-
ness of the oxygen inhibited layer, but also the interplay
between the micromechanical interlocking, wettability of
0.1
the intermediate adhesive resin, FR and PFC, as well as
the elastic modulus of these materials need to be consid-
ered when advising buildup protocols for FRC FDPs.
0.0
The fiber weight fraction results for S2-glass FRCs studied
using TGA (55 wt% to 60 wt%) were in good agreement with
–0.1 the information provided by the manufacturer of the FRC ma-
500 600 terials (50 wt% to 60 wt%) in their material safety data sheet.
The bond strength results and failure types may change
with other material combinations and the results must be
verified for direct FRC FDP applications.
CONCLUSIONS
1. FRC and FR can be polymerized in one step, but FRC and
PFC combinations should be polymerized individually.
2. Application of an intermediate adhesive resin layer
increases the adhesion of both PFC and FR to unidi-
rectional FRC.
3. Multidirectional pre-impregnated E-glass fibers cannot
be recommended in combination with the PFC and FR
materials tested in this study.
REFERENCES
1. Dyer SR, Lassila LV, Jokinen M, Vallittu PK. Effect of fiber position and
orientation on fracture load of fiber-reinforced composite. Dent Mater
2004;20:947-955.
2. Gauthier MA, Stangel I, Ellis TH, Zhu XX. Oxygen inhibition in dental res-
ins. J Dent Res 2005;84:725-729.
3. Lassila LV, Tezvergil A, Dyer SR, Vallittu PK. The bond strength of par-
ticulate-filler composite to differently oriented fiber-reinforced composite
substrate. J Prosthodont 2007;16:10-17.
4. Lastumäki TM, Kallio TT, Vallittu PK. The bond strength of light-curing
composite resin to finally polymerized and aged glass fiber-reinforced
composite substrate. Biomaterials 2002;23:4533-4539.
5. Lastumäki TM, Lassila LV, Vallittu PK. The semi-interpenetrating poly-
mer network matrix of fiber-reinforced composite and its effect on the
surface adhesive properties. J Mater Sci Mater Med 2003;14:803-809.
6. Phrukkanon S, Burrow MF, Tyas MJ. Dent Mater 1998;14:212-221.
7. Shawkat ES, Shortall AC, Addison O, Palin WM. Oxygen inhibition and
incremental layer bond strengths of resin composites. Dent Mater
2009;25:1338-1346.
8. Vallittu PK. Oxygen inhibition of autopolymerization of polymethyl meth-
acrylate-glass fibre composite. J Mater Sci Mater Med 1997;8:489-492.
9. van Heumen CC, Kreulen CM, Creugers NH. Clinical studies of fiber-
reinforced resin-bonded fixed partial dentures: a systematic review. Eur
J Oral Sci 2009;117:1-6.
10. Vankerckhoven H, Lambrechts P, van Beylen M, Davidson CL, Vanherle G.
Unreacted methacrylate groups on the surfaces of composite resins.
J Dent Res 1982;61:791-795.
Clinical relevance: During application procedures on the
fiber-reinforced composite materials after photopolymer-
ization, the use of an intermediate adhesive resin that
can be photopolymerized in one step with the particu-
late filler composite or flowable resin is compulsory.
The Journal of Adhesive Dentistry