The

prim

mor

elem

Ammar A. Musta

aAssistant Professor, Dental MateriaKuantan Campus.bAssociate Professor, Dental MateriacDoctor, Clinical Sciences DepartmedProfessor, Medical Devices TechnolTeknologi, Malaysia.

Mustafa et al

influence of experimental silane

ers on dentin bond strength and

phology: A laboratory and finite

ent analysis study

fa, MSc, PhD,a Jukka P. Matinlinna, MSc, PhD,b

Syafiqah Saidin, PhD,c and M. R. Abdul Kadir, MEng, PhDd

International Islamic University Malaysia, Kuantan Campus, Pahang,Malaysia; Faculty of Dentistry, University of Hong Kong, Hong KongSAR, China; Faculty of Biosciences & Medical Engineering, UniversitiTeknologi, Malaysia, Johor Bahru, Malaysia

Statement of problem. The inconsistency of dentin bonding affects retention and microleakage.

Purpose. The purpose of this laboratory and finite element analysis study was to investigate the effects on the formation of ahybrid layer of an experimental silane coupling agent containing primer solutions composed of different percentages ofhydroxyethyl methacrylate.

Material and methods. A total of 125 sound human premolars were restored in vitro. Simple class I cavities were formed oneach tooth, followed by the application of different compositions of experimental silane primers (0%, 5%, 25%, and 50% ofhydroxyethyl methacrylate), bonding agents, and dental composite resins. Bond strength tests and scanning electron mi-croscopy analyses were performed. The laboratory experimental results were validated with finite element analysis to deter-mine the pattern of stress distribution. Simulations were conducted by placing the restorative composite resin in a premolartooth by imitating simple class I cavities. The laboratory and finite element analysis data were significantly different from eachother, as determined by 1-way ANOVA. A post hoc analysis was conducted on the bond strength data to further clarify theeffects of silane primers.

Results. The strongest bond of hybrid layer (16.96 MPa) was found in the primer with 25% hydroxyethyl methacrylate,suggesting a barely visible hybrid layer barrier. The control specimens without the application of the primer and the primerspecimens with no hydroxyethyl methacrylate exhibited the lowest strength values (8.30 MPa and 11.78 MPa) with inter-mittent and low visibility of the hybrid layer. These results were supported by finite element analysis that suggested an evenlydistributed stress on the model with 25% hydroxyethyl methacrylate.

Conclusions. Different compositions of experimental silane primers affected the formation of the hybrid layer and its resultingbond strength. (J Prosthet Dent 2014;-:---)

Clinical Implications

The use of a novel formula of silane primers may enhance the bondstrength between the tooth substrate and the composite resinrestoration.

The bond between composite resinand dentin is essential to the biome-chanical and biologic properties of a

ls Science

ls Sciencent, Facultyogy Group

bonded restoration. Important factorsinfluencing a bonded system are themechanical interlock between both

Department, Faculty of Dentistry, Internationa

Department, Faculty of Dentistry, University oof Biosciences & Medical Engineering, Univers(MEDITEG), Faculty of Biosciences & Medica

bodies and the infiltration of resin intothe demineralized intertubular dentin,both widely understood to be the

l Islamic University Malaysia,

f Hong Kong.iti Teknologi, Malaysia.l Engineering, Universiti

2 Volume - Issue -

underlying factors of successful bondstrength.1,2 Bond strength is achievedthrough the formation of tags from thecomposite resin into the dentinal tu-bules.3,4 The cavities between the 2bodies should be completely sealed toprevent microleakage that might lead torecurrent caries.5 The bonding of com-posite resin to dentin is more chal-lenging than bonding to enamel.6 Abond to dentin is formed when thecollagen fibril network in dentin isdiffused into the adhesive layer (AL) ofthe composite resin to form a hybridlayer (HL).7 Different bonding patternshave been observed on the dentin sub-strate8 because of the tubular structurethat contains dentinal fluid transuda-tion with the presence of a smear layer.7

In adhesive dentistry, the HL plays acrucial role as a barrier against thedemineralization process caused by cari-ogenic agents.9 This barrier layer shouldbe evenly distributed with a certainthickness to achieve the ultimate benefitof perfect marginal seal and stabilityunder continuous occlusal loading.10

Imparting a high degree of acid resis-tance is necessary.7 An investigation ofthe morphology and thickness of the HLis significant in determining the stabilityand strength of the bonding layerbetween the dentin and AL.11-13 Severalapplications, such as self-etching bon-ding agents, acid etching, primer appli-cation, and Er:YAG laser, have beenintroduced to control the formation andmorphology of the HL.14-16 Variations inprotocols and conditions such as in thematerial composition, pH, concentrationof primers, wettability, volume of water,and viscosity of the composite resin andprimer might affect the thickness of theHL.17 The thickness of the HL is reportedto range from less than 1 mm for all-in-1systems to 5 mm for conventionalsystems.18

Dentin pretreatment with primershas been found to increase the bondstrength of bonding agents.19,20 Theoptimal primers are those with wetta-bility opposite that of the bondingagents, because the combination of ahydrophilic primer solution and a hy-drophobic bonding agent is preferred.18

The Journal of Prosthetic Dentis

The introduction of trialkoxysilanes asreactive monomers in primer solutionshas been reported to be effective,21

particularly on silica or glass,22

because their reactivity with calciumminerals is moderate.23 Silane couplingagents can promote resin substrateadhesion on dental implant surfacessuch as silica-coated titanium24 andsilica-coated zirconia.25 Because of thecomplexity of dentin bonding, resinmust be allowed to penetrate to avoidmicroleakage and restore durability toprevent secondary caries.

The purpose of this laboratory studywas to investigate how different com-positions of experimental silane-basedprimers in an ethanol solution contain-ing 3-isocyanatopropyltrimethoxysilane(ICMS),26 1,2-bis(triethoxysilyl)ethane(BTSE),27 and different percentages ofhydrophilic hydroxyethyl methacrylate(HEMA) might act as novel experimentaldentin bonding agents. Bond strengthtests were performed to evaluate themechanical properties of the bondinglayer, followed by validation with a finiteelement analysis (FEA). The morphol-ogies and thickness of the HL werevisualized to examine the correlationbetween the different silane primers andthe formation of the HL. Theoretically,the strength properties, morphology,and HL thickness of the bonding layercould be affected by different composi-tions of experimental adhesive silaneprimers. The different material proper-ties of each bonding layer would pro-duce different patterns of stress transfer.The hypothesis of this study was that theuse of HEMA-containing primers withsilane coupling agent monomers wouldincrease the bond strength between thedental composite resin and dentin bysignificantly affecting the formation ofthe HL. The null hypothesis was that nosignificant difference would be found inbond strength and HL formation afteradding experimental silane-basedprimers.

MATERIAL AND METHODS

A total of 125 sound human firstpremolar teeth were selected for the

try

study. The teeth had been extractedfrom individuals aged between 15 and22 years for orthodontic treatment, andthey were used after approval by theinstitutional review board of the Kul-liyyah of Dentistry, International IslamicUniversity Malaysia (IIUM/314/14/1/1-EDW-B-0902-217). Other inclusioncriteria included that the teeth bewithout restoration, without decay, andwithout a history of chemical treatmentsuch as teeth bleaching. The selectedteeth were embedded vertically in anacrylic resin block, with the crownsexposed. Class I cavities (3 mm deep)were prepared with a water-cooleddiamond straight-fissure rotary cuttinginstrument (CFKT00046; AbrasiveTechnology Inc).28

The internal walls were etched with37% phosphoric acid (Gel Etchant 3;SDS Kerr) for 15 seconds, rinsed withwater spray for 10 seconds, and driedwith oil-free air for 10 seconds to achalky white appearance.22,29 The cav-ities were restored with an experimentalprimer applied with a small fine brush(Micro Applicators; Ultradent ProductsInc). Based on a previous study, theprimer solutions were prepared in 95%ethanol30 and 5% water.22 The speci-mens were assigned to 5 groups ac-cording to the composition of theexperimental primers (Table I). Abonding agent (OptiBond Solo Plus;Kerr) was applied on the primer layersand light polymerized (XL-3000; 3MESPE) for 10 seconds at an intensity of550 mW/cm2 at a distance of 2 cm.The specimens were restored withcomposite resin (Point 4; Kerr Corp)with a 2-mm incremental technique.

Figure 1 shows the steps in prepar-ing and sectioning the specimens forthe bond strength test. The teeth weresectioned according to the dumbbellshape shown in Figure 1. The sectioningwas performed carefully with asectioning machine (IsoMet 5000;Buehler), avoiding vibrating move-ments. The specimens with internalbubbles were rejected and replacedwith flawless specimens. The lengths ofthe sectioned restoration sides weremeasured with a digital caliper (Absolute;

Mustafa et al

Table I. Specimen groups according to different percentages of HEMAcomposition

Group nICMS(%vol)

BTSE(%vol)

HEMA(%vol)

Primer with no HEMA 25 1.0 0.5 -

Primer with 5% HEMA 25 1.0 0.5 5.0

Primer with 25%HEMA

25 1.0 0.5 25.0

Primer with 50%HEMA

25 1.0 0.5 50.0

Control 25 Without application of primer solution

ICMS, 3-isocyanatopropyltrimethoxysilane; BTSE, bis-1,2-(triethoxysilyl) ethane; HEMA,hydroxyethyl methacrylate; %vol, percentage by volume.

- 2014 3

Mitutoyo Corp). Because the depth ofthe restoration was fixed at 3.00 mm,the volume was calculated based on thevolume formula for cubic structures.Averages of (3.35 �0.29) mm and(34.02 �1.39) mm3 were recorded forthe length and volume of therestoration.

Twenty sectioned specimens fromeach study group were kept in deionizedwater at 37�C for 24 hours. Thesectioned specimens were then ther-mocycled between 5�C and 55�C for500 cycles. The bond strength tests

Premolartooth

Compositeresin

restoration

Acrylic resinblock

Class I cavities

Bond strengthtest

F

F

1 Multistep preparation pro

Mustafa et al

were performed on each sectionedspecimen with a precision universaltester (AGS-X; Shimadzu). Thesectioned specimens were subjected toa continuous load at a crosshead speedof 1.0 mm/min until the first failureoccurred. The data were recorded andcalculated with Windows-compatibleTrapezium X software (Shimadzu).

The morphologies of the HL werefurther investigated by visualizing thecross section of each specimen underscanning electron microscopy (SEM)(EVO 50; Carl Zeiss AG). Five

Removed fromacrylic resinblock and sectioned

Thermocycling process temperature between 5and 55°C for 500 cycleComposite

resinrestoration

Dentin

cesses of restored specimens for bond stre

specimens from each study group weresectioned longitudinally in a mesiodis-tal direction (IsoMet 1000; Buehler).The specimen surfaces were thensputter-coated with a thin gold film.The SEM images were captured at300�, 700�, and 2000� magnifica-tions under an accelerating voltage of10 kV.

In the FEA simulation analysis, a leftpremolar tooth of the mandible waschosen as a region of interest (ROI); 2-dimensional (2D) computed tomogra-phy (CT) images were stacked togetherto develop a 3-dimensional (3D) modelof the ROI. The hard tissues, whichincluded cancellous bone, corticalbone, enamel, and dentin, weremodeled based on the intensity andvisibility in the 3D models of the ROI.The soft tissues such as the mucosa,periodontal ligament (PDL), and pulpwere modeled based on the informa-tion provided in previous studies,because these soft tissues were notvisible in the CT images.31 The thick-ness of the cortical bone constructedfrom the CT images was between 1.25mm and 3.15 mm, whereas the thick-ness of the mucosa and PDL were

De-ionizedwater at 37°Cfor 24 hours

at°Cs

Adhesive layerComposite resin layer

ngth test.

4 Volume - Issue -

manually modeled based on the workof Müller.31 The heights of enamel anddentin were 8.45 mm and 19.35 mm.

The 3D model of the cubic com-posite resin was designed with 3Dcomputer-aided design software (Sol-idWorks 2009; Dassault Systèmes Sol-idWorks Corp) with a length of 3.35mm and a height of 3.00 mm. Thesevalues were obtained from the spec-imen preparation; the same values mustbe assigned to imitate the laboratoryexperiment condition. The compositeresin system consisted of the HL, theAL, and the composite resin layer (CL).The heights of the CL, AL, and HL wereset at 2.00 mm, 996.15 mm, and 3.85mm based on the experimental datafrom the in vitro testing. Figure 2 showsthe 3D model of the ROI, compositeresin, and restored premolar tooth. Themesh size of the 3D models was definedas 0.003 mm for the thin HL and 0.500mm for the remaining regions. Each 3Dmodel was converted into a solid bodywith the generation of a total of320 026 elements.

For the FEA, the simulation wasdivided into 5 study groups as in thein vitro experiment. All of the 3Dmodels were assumed to be linear,

3D model of theregion of interest

Comresinrest

Premolartooth

Sectionpremotooth

Mucosa

Cortical bone

Cancellous bone

Enamel

Dentin

Per

Pul

Com

Adh

Hyb

2 Three-dimensional model of region orestored premolar tooth.

The Journal of Prosthetic Dentis

isotropic, and homogeneous. The ma-terial properties of the 3D models weredefined from previous studies as shownin Table II. The friction coefficient of0.3 was assigned to each 3D modelwith a glue contact between the HL, AL,composite resin, and adjacent regions.An axial force of 230 N and a bucco-lingual force of 100 N (30 degrees fromthe occlusal plane) were loaded on thecenter region of the restored compositeresin to represent the occlusalforce.32,33 The 3D models were con-strained in the inferior, mesial, anddistal directions. The equivalent vonMises stress (EQVM) was used as theanalysis output of the FEA.

All the data from the bond strengthtest and the FEA were judged to besignificantly different with a 1-wayANOVA technique (a¼.05). In deter-mining the significance level of eachpair of bond strength data, the DunnettT3 multiple comparison tests wereperformed as a post hoc analysis.

RESULTS

The descriptive analyses of the bondstrength data with their mean, SD, andmaximum values are shown in Table III.

posite

oration

edlar

Restored premolartooth

Cross section ofthe restored

region of interest

iodontal ligament (PDL)

p

posite resin layer (CL)

esive layer (AL)

rid layer (HL)

f interest, composite resin, and

try

From the descriptive analyses, thevalues of dFw, F, and P were 95, 11.45,and <.001, respectively. The experi-mental primer with 25% HEMA pro-duced the strongest HL bond, with43.97%, 18.44%, and 18.11% higherbond strengths than the experimentalprimers with no HEMA, 5% HEMA, and50% HEMA, respectively. The restora-tion without the application of a primer(control) produced the weakest bondstrength (8.3 MPa) between the dentinand composite resin material. The nu-merical value of the bond strengthcompared with the control specimenswas increased by 100% with the appli-cation of an experimental primer thatcontained 25% HEMA (16.96 MPa).

The Dunnett T3 analysis data inTable III show a significant differencebetween the control and each group ofexperimental primers, except for theexperimental primer with no HEMA.The analyses found that the experi-mental primer with 5% HEMA was notsignificantly different from that of thegroups with 25% and 50% HEMA. Theexperimental primer with 25% and 50%HEMA exhibited a significant differenceat P¼.046, and the experimental primerwith 25% HEMA showed the bestadhesion performance.

Figure 3 shows the SEM images ofHL formation for the control and theexperimental primers with 0%, 5%, 25%,and 50% HEMA. The specimens with anexperimental primer that containedHEMA produced a visible HL barrierbetween the dentin and the restorativematerial. The HL formation did notdepend on the types of primer solution,because its thickness and morphologycould not be categorized clearly. Theuse of an experimental primer solutionwithout HEMA showed an intermittentand nonhomogeneous HL comparedwith that of the experimental primerswith 5%, 25%, and 50% HEMA. The HLbarrier in the control group was evenmore difficult to recognize.

Figure 4 shows the color contour ofthe EQVM for 5 different groups of CL,AL, HL, and dentin. The color contoursof the CL, AL, and HL were presentedfrom the buccal view, whereas the color

Mustafa et al

Table II. Material properties (Young modulus and Poisson ratio) of simulated3D models

3D ModelYoung Modulus

(GPa) Poisson Ratio Reference

Cancellous bone 5.5 0.30 30

Cortical bone 13.7 0.30 31

Mucosa 0.003 0.40 34

PDL 0.05 0.45 32,35

Pulp 0.02 0.45 36

Enamel 40 0.30 37

Dentin 15 0.31 37,38

CL 12.5 0.30 39

AL 2.5 0.28 39

HL 38,40-42

Primer with no HEMA 0.16 0.28

Primer with 5% HEMA 0.8 0.28

Primer with 25% HEMA 4 0.28

Primer with 50% HEMA 8 0.28

PDL, periodontal ligament; CL, composite resin layer; AL, adhesive layer; HL, hybrid layer;HEMA, hydroxyethyl methacrylate.

- 2014 5

contour of the dentin was presentedfrom the lingual view. High stresseswere concentrated at the center supe-rior region of the CL because of theplacement of the axial and buccolin-gual loads. The stresses graduallydecreased from the concentrated re-gions outward. No significant differ-ence was found between the 5 CLgroups, indicating that the differentmaterial properties of the HL did notaffect the stress distribution in the CLregion. The loading forces were thentransferred to the AL, HL, and dentin

Table III. Descriptive and Dunnett T3

Group Mean ±SD

Primer with no HEMA 11.78 �4.77

Primer with 5% HEMA 14.32 �5.84

Primer with 25% HEMA 16.96 �3.00

Primer with 50% HEMA 14.36 �2.46

Control 8.30 �4.61

HEMA, hydroxyethyl methacrylate.*P<.05

Mustafa et al

regions situated below the CL. Similarstress distribution trends were noticedbecause these bodies adhered to oneanother.

The mean value of the EQVM inTable IV shows that the average stresslevels gradually increased from themodel of primer with no HEMA to themodel of primer with 50% HEMA forthe AL and HL. The results for the CLand dentin show the opposite effect.The maximum values of the EQVMincreased as the percentage of HEMAincreased from the primer with no

analyses of bond strength data (MPa) by

MaximumValue

Primer With5% HEMA

Prime25% H

16.75 .050 .00

20.83 - .5

22.98 -

18.50 -

17.34 -

HEMA to the primer with 50% HEMAfor the CL and AL, whereas the HLshowed the opposite effect. SimilarEQVM maximum values wereobserved for the dentin in all of thegroups.

The statistical analysis found a sig-nificant difference between each groupfor the AL, HL, and dentin, with Pvalues of .001, <.001, and <.001,respectively. The results are in agree-ment with the color contour plot, wherehigher stresses were generated for themodels with a higher percentage ofHEMA. The model of a primer with50% HEMA that was assigned thehighest Young modulus (8000 MPa)produced the greatest stress value in theHL region, which was greater than 15MPa. The HL region for the controlgroup was not modeled in this study,because the HL did not appear in theSEM image.

DISCUSSION

In this study, the null hypothesiswas rejected. Different compositionsof experimental silane primers haveaffected the strength properties andmorphology of the bonding HL, thusproducing different patterns of stressdistribution because of variations inthe material properties. The HL be-tween the dentin and the compositeresin is susceptible to slow hydrolyticdegradation because of incompleteinfiltration of the composite resin intothe demineralized collagen fibril

using 1-way ANOVA

Significance

r WithEMA

Primer With50% HEMA Control

3* .032* .209

44 1.000 .009*

- .046* <.001*

- - <.001*

- - -

3 Morphology of hybrid layer formation. A, Primer with no HEMA (original magnification �300). B, primer with 5%HEMA (original magnification �300). C, primer with 25% HEMA (original magnification �700). D, primer with 25%HEMA (original magnification �2000). E, primer with 50% HEMA (original magnification �300). F, control specimens(original magnification �300). (HEMA, hydroxyethyl methacrylate.)

6 Volume - Issue -

network.4 This condition might lead tolow bond strength and less durabilityfor the HL.4 In this study, the hydro-phobic materials treated with ethanolwere incorporated into the dentininterface to form the hydrophobic HLby replacing the rinse water withethanol on the acid-etched dentinmatrices.30 The formation of the

The Journal of Prosthetic Dentis

hydrophobic HL increased the resis-tance to HL degradation, thusimproving its strength and durability.30

Several factors contribute to thestrength of the HL formation. The studyfound a nonsignificant relationship be-tween the bond strength and the HLthickness. The thickness varied between1.17 mm and 8.37 mm, but these

try

values were not correlated with theapplication of different compositions ofthe experimental silane primers. Yosh-iyama et al3 stated that bonding wasmore dependent on the number of resintags that formed in the dentinal tubulesthan on the HL thickness. Barceleiroet al14 mentioned that the thinner andirregular HL might have a negative

Mustafa et al

4 Color contour of equivalent von Mises stress at 5 differentgroups of composite resin layer, adhesive layer, hybrid layer,and dentin.

- 2014 7

effect on the bonding strength anddurability of the HL formation. Tyasand Burrow18 verified that the forma-tion of a thin HL from a self-etchingprimer solution produced greater

Table IV. Mean and maximum FEA EQ

Region

Mea

Primer Withno HEMA

Primer Wit5% HEMA

CL 39.61 �85.90 39.46 �85.9

AL 6.17 �2.95 6.23 �3.00

HL 5.63 �2.97 6.40 �3.15

Dentin 12.01 �5.25 12.01 �5.22

CL 1462.25

AL 23.95

HL 45.28

Dentin 78.75

FEA, finite element analysis; EQVM, equivalent voAL, adhesive layer; HL, hybrid layer.*P<.05.

Mustafa et al

bond strength than did many othersystems with a thicker HL.

In this study, several techniques wereimplemented during the restorationprocess to form a good-quality HL with

VM (MPa) values for CL, AL, HL, and de

n of EQVM ±SD (MPa)

h Primer With25% HEMA

Primer With50% HEMA C

1 39.43 �85.92 39.42 �85.92 39.4

6.26 �2.98 6.27 �2.98 6.4

7.96 �3.19 9.76 �3.29

11.95 �5.20 11.91 �5.19 12.

Maximum Value of EQVM

1462.29 1462.30

24.21 24.26

43.27 37.77

78.75 78.75

n Mises stress; HEMA, hydroxyethyl methacrylate; CL

high mechanical strength. The dentinspecimens were etched with phosphoricacid to form a demineralized dentin.The demineralized dentins were thenrinsed with water and dried with air,leaving some water to hydrate thecollagen fibrils.29 The hydrated collagenfiber network provides a large volume ofmicro spaces for the diffusion affinity ofhydrophilic monomers to form the ho-mogeneous HL.30 This environmentimproves the bond strength and dura-bility of the bonding performance.

The addition of HEMA to theexperimental primer solution apparentlyhas a role in enhancing the mechanicalproperties of the HL. The main bondingeffect of the HEMA-based primers isattributed to the polar solvents used todissolve the HEMA.30 These solventswould expand the collagen matrix30 forthe formation of the HL. An excessiveamount of HEMA significantly reducedthe bond strength of the HL, as pre-sented in Table III. The threshold per-centage of the HEMA composition inthe primers was found to be 25%,because the bond became weaker afterthe application of 50% HEMA. Thiscould be attributed to the possibility ofcontamination in the areas in which thedentinal fluid is more likely and the

ntin

df F Pontrol

2 �85.92 2930 0.00 1.000

0 �2.91 31 060 4.83* .001

- 24 557 2066.41* <.001

74 �5.21 30 700 26.63* <.001

(MPa)

1462.30 1462.31

24.27 24.49

38.41 -

78.75 76.42

, composite resin layer;

8 Volume - Issue -

HEMA is partially solubilized.29 Thecontaminated areas might exhibit lowersurface energy, leading to an incohesivedispersion of the adhesive resin that isadded subsequently.29

For optimum performance, theadhesive system should have a surfacetension equal to or slightly less thanthe critical surface tension of the solidsubstrate.43 In general, primers have asurface tension that is less than thesurface free energy of the dentinetchant conditioners.44 The incorpo-ration of the hydrophilic HEMA intothe primer formula provides affinitywith the dentin surface,45 whereas thepresence of condensed bridged orga-nosiloxane groups in the molecularbackbone (1,2-bis(triethoxysilyl)ethane)might provide affinity with the resinadhesive.18,26,46

In addition to the bond strengthanalyses, FEAs were conducted to vali-date the experimental results. This en-gineering technique has been widelyused in dentistry to analyze the effect ofmasticatory occlusal transfer to theteeth and the surrounding bodies.32,33

In this study, the technique wasapplied to simulate and predict theoutcome of the composite resin resto-ration of a physical model of a premolartooth. In Figure 3, the model of anexperimental primer with 5% and 25%HEMA showed evenly distributed stresson the HL because of the low range ofthe mean and maximum EQVM valuescompared with those of the model withexperimental primer with no HEMA andthe model with experimental primerwith 50% HEMA. Such high stressvalues might initiate microfractures andcracks, causing detachment from thenearest layer or body. These results aresimilar to those of the laboratoryexperimental data, where the specimensof an experimental primer with 25%HEMA were found to provide the bestadhesion between the dentin and com-posite resin. The highest value of themean EQVM values was noticed on all 4bodies (CL, AL, and dentin) of thecontrol models, which suggested thatthe HL formation after a restoration isimportant and that reducing stress

The Journal of Prosthetic Dentis

transfer is necessary to reduce micro-fractures and bond detachment.

Several limitations were noted in thelaboratory experimental study and theFEA simulation analyses. The weakerbond strength of the experimentalsilane primer with 50% HEMA wasattributed to the partially solubilizedHEMA, which is caused by contamina-tion by a high saturation constituent. Atechnique that could avoid contami-nation should be investigated, becausea high percentage of HEMA in theprimer solution is essential to generatea strong bond with the HL. In addition,the process of artificially speeded agingand thermocycling for the bondstrength test could be compared,because both artificial aging processesaffect the bond strength between thedentin and the restorative material. Inthe simulation analyses, the Youngmoduli of CL, AL, HL, and dentin wereassumed from the literature becausetheir actual values were not determinedexperimentally. The FEA simulation mayachieve its clinically genuine state if thevalue of the Young modulus is deter-mined from an actual laboratoryexperiment.33

CONCLUSIONS

The addition of 25% HEMA in anexperimental silane primer generated avisible HL barrier between the restor-ative material and the dentin. The 25%HEMA addition was significantly foundto be the optimum percentage thatproduces the greatest HL bond strengthcompared with compositions withHEMA at 0%, 5%, and 50%. This layerwas responsible for increasing thedurability of the composite resin resto-ration by improving the bond strengthand reducing stress transfer.

REFERENCES

1. Swift EJ, Perdigão J, Heymann HO. Bondingto enamel and dentin: a brief history andstate of the art, 1995. Quintessence Int1995;26:95-110.

2. Meerbeek BV, Munck JD, Yoshida Y, Inoue S,Vargas M, Vijay P, et al. Adhesion to enameland dentin: current status and future chal-lenges. Oper Dent 2003;28:215-35.

try

3. Yoshiyama M, Carvalho R, Sano H, Horner J,Brewer PD, Pashley DH. Interfacialmorphology and strength of bonds made tosuperficial versus deep dentin. Am J Dent1995;8:297-302.

4. Hashimoto M, Ohno H, Kaga M, Endo K,Sano H, Oguchi H. In vivo degradation ofresin-dentin bonds in humans over 1 to 3years. J Dent Res 2000;79:1385-91.

5. Nakabayashi N, Ashizawa M, Nakamura M.Identification of a resin-dentin hybrid layer invital human dentin created in vivo: durablebonding to vital dentin. Quintessence Int1992;23:135-41.

6. Perdigao J, Swift EJ, Denehy GE, Wefel JS,Donly KJ. In vitro bond strengths and SEMevaluation of dentin bonding systems todifferent dentin substrates. J Dent Res1994;73:44-55.

7. Nakabayashi N, Nakamura M, Yasuda N.Hybrid layer as a dentin-bonding mecha-nism. J Esthet Restor Dent 1991;3:133-8.

8. Spencer P, Wang Y. Adhesive phase separa-tion at the dentin interface under wetbonding conditions. J Biomed Mater Res2002;62:447-56.

9. Nakabayashi N, Saimi Y. Bonding to intactdentin. J Dent Res 1996;75:1706-15.

10. Krejci I, Schüpbach P, Balmelli F, Lutz F. Theultrastructure of a compomer adhesiveinterface in enamel and dentin, and its mar-ginal adaptation under dentinal fluid ascompared to that of a composite. DentMater 1999;15:349-58.

11. Ferrari M, Cagidiaco CM, Mason PN.Morphologic aspects of the resin-dentininterdiffusion zone with five different dentinadhesive systems tested in vivo. J ProsthetDent 1994;71:404-8.

12. Ferrari M, Mannocci F, Kugel G, Garcia-Godoy F. Standardized microscopic evalua-tion of the bonding mechanism of NRC/Prime & Bond NT. Am J Dent 1999;12:77-83.

13. Nakajima M, Sano H, Urabe I, Tagami J,Pashley DH. Bond strengths of single-bottledentin adhesives to caries-affected dentin.Oper Dent 2000;25:2-10.

14. Barceleiro MO, Dias KRHC, Sales HX,Silva BC, Barceleiro CG. SEM evaluation ofthe hybrid layer after cavity preparation withEr:YAG laser. Oper Dent 2008;33:294-304.

15. Pisani-Proença J, Erhardt MCG, Amaral R,Valandro LF, Bottino MA, Del Castillo-Salmerón R. Influence of different surfaceconditioning protocols on microtensilebond strength of self-adhesive resin ce-ments to dentin. J Prosthet Dent 2011;105:227-35.

16. Aguiar TR, André CB, Correr-Sobrinho L,Arrais CAG, Ambrosano GMB, Giannini M.Effect of storage times and mechanical loadcycling on dentin bond strength of conven-tional and self-adhesive resin luting cements[published online December 16, 2013]. JProsthet Dent.

17. Kenshima S, Francci C, Reis A, Loguercio AD,Filho LER. Conditioning effect on dentin,resin tags and hybrid layer of different acidityself-etch adhesives applied to thick and thinsmear layer. J Dent 2006;34:775-83.

Mustafa et al

- 2014 9

18. Tyas MJ, Burrow MF. Adhesive restorative ma-terials: a review. Aust Dent J 2004;49:112-21.

19. McGuckin RS, Powers JM, Li L. Bondstrengths of dentinal bonding systems toenamel and dentin. Quintessence Int1994;25:791-6.

20. Chen L, Shen H, Suh BI. Effect of incorpo-rating BisGMA resin on the bonding proper-ties of silane and zirconia primers. J ProsthetDent 2013;110:402-7.

21. Lung YK, Matinlinna JP. Silanes for adhesionpromotion and surface modification. In:Moriguchi K, Utagawa S, editors. Silane:chemistry, applications and performance.Hauppauge: Nova Science Publishers; 2013.p. 87-110.

22. Mustafa AA, Matinlinna JP, Choi AH,Razak AAA. Fracture strength and fracto-graphic analysis of zirconia copings treatedwith four experimental silane primers. J AdhesSci Technol 2012;27:68-80.

23. Lung CYK, Matinlinna JP. Aspects of silanecoupling agents and surface conditioning indentistry: an overview. Dent Mater 2012;28:467-77.

24. Matinlinna JP, Tsoi JKH, de Vries J,Busscher HJ. Characterization of novel silanesystems on titanium implant surfaces. ClinOral Implants Res 2013;24:688-97.

25. Matinlinna JP, Choi AH, Tsoi JKH. Bondingpromotion of resin-composite to silica-coated zirconia implant surface using a novelsilane system. Clin Oral Implants Res2013;24:290-6.

26. Lung CYK, Matinlinna JP. Resin bonding tosilicatized zirconia with two iso-cyanatosilanes and cross-linking silane, partI: experimental. Silicon 2010;2:153-61.

27. Matinlinna JP, Lassila LVJ, Vallittu PK. Theeffect of a novel silane blend system on resinbond strength to silica-coated Ti substrate.J Dent 2006;34:436-43.

28. Munck JD, Landuyt KV, Coutinho E,Poitevin A, Peumans M, Lambrechts P, et al.Micro-tensile bond strength of adhesivesbonded to class-I cavity-bottom dentin afterthermo-cycling. Dent Mater 2005;21:999-1007.

29. Eick JD, Gwinnett AJ, Pashley DH,Robinson SJ. Current concepts on adhesionto dentin. Crit Rev Oral Biol Med 1997;8:306-35.

Mustafa et al

30. Pashley DH, Tay FR, Carvalho RM,Rueggeberg FA, Agee KA, Carrilho M, et al.From dry bonding to water-wet bonding toethanol-wet bonding: a review of the in-teractions between dentin matrix and sol-vated resins using a macromodel of thehybrid layer. Am J Dent 2007;20:7-20.

31. Müller H-P, Schaller N, Eger T. Ultrasonicdetermination of thickness of masticatorymucosa: a methodologic study. Oral SurgOral Med Oral Pathol Oral Radiol Endod1999;88:248-53.

32. Saidin S, Abdul Kadir MR, Sulaiman E, AbuKasim NH. Effects of different implant-abutment connections on micromotion andstress distribution: prediction of microgapformation. J Dent 2012;40:467-74.

33. Eskitascioglu G, Usumez A, Sevimay M,Soykan E, Unsal E. The influence ofocclusal loading location on stresses trans-ferred to implant-supported prostheses andsupporting bone: a three-dimensional finiteelement study. J Prosthet Dent 2004;91:144-50.

34. Rho JY, Ashman RB, Turner CH. Young’smodulus of trabecular and corticalbone material: ultrasonic and microtensilemeasurements. J Biomech 1993;26:111-9.

35. Eliguzeloglu E, Eraslan OO, Huma,Eskitascıoglu GB, Sema. Effect of hybrid layerand thickness on stress distribution of cervi-cal wedge-shaped restorations. Eur J Dent2010;4:160-5.

36. Mahmoudi M, Saidi A, GandjalikhanNassab SA, Hashemipour MA. A three-dimensional finite element analysis of theeffects of restorative materials and post ge-ometry on stress distribution in mandibularmolar tooth restored with post-core crown.Dent Mater J 2012;31:171-9.

37. Rees JS, Jacobsen PH. Elastic modulus of theperiodontal ligament. Biomaterials 1997;18:995-9.

38. Hasegawa A, Shinya A, Nakasone Y,Lassila LVJ, Vallittu PK, Shinya A. Develop-ment of 3D CAD/FEM analysis system fornatural teeth and jaw bone constructed fromx-ray CT images. Int J Biomater 2010; ArticleID 659802:1-7.

39. Rees JS, Jacobsen PH. The elastic moduli ofenamel and dentine. Clin Mater 1993;14:35-9.

40. Yasuda G, Inage H, Kawamoto R,Shimamura Y, Takubo C, Tamura Y, et al.Changes in elastic modulus of adhesive andadhesive-infiltrated dentin during storage inwater. J Oral Sci 2008;50:481-6.

41. Ausiello P, Apicella A, Davidson CL. Effect ofadhesive layer properties on stress distribu-tion in composite restorationsea 3D finiteelement analysis. Dent Mater 2002;18:295-303.

42. Martini AP, Anchieta RB, Rocha EP, FreitasJunior AC, Almeida EOd, Sundfeld RH,et al. Influence of voids in the hybrid layerbased on self-etching adhesive systems: a3-D FE analysis. J Appl Oral Sci 2009;17:19-26.

43. Baier RE. Principles of adhesion. Oper Dent1992;5:1-9.

44. Attal JP, Asmussen E, Degrange M. Effectsof surface treatment on the free surfaceenergy of dentin. Dent Mater 1994;10:259-64.

45. Wang Y, Spencer P. Hybridization efficiencyof the adhesive/dentin interface with wetbonding. J Dent Res 2003;82:141-5.

46. George LA, Ramzy MS, Bradley FC. Func-tionalization of mesostructured inorganic-organic and porous inorganic materials.Curr Opin Colloid Interface Sci 2009;14:281-92.

Corresponding author:Dr Ammar A. MustafaDental Materials Science DepartmentFaculty of DentistryInternational Islamic University MalaysiaKuantan Campus, 25200 PahangMALAYSIAE-mail: drammar71@gmail.com

AcknowledgmentsThe authors thank the InternationalIslamic University Malaysia (IIUM) and theUniversity of Hong Kong (HKU) for the prepara-tion of silane primers, and the Medical DevicesTechnology Group (MEDITEG), UniversitiTeknologi, Malaysia (UTM), for the high-performance workstation.

Copyright ª 2014 by the Editorial Council forThe Journal of Prosthetic Dentistry.