Journal Identification = DENTAL Article Identification = 1822 Date: June 24, 2011 Time: 12:49 pm

Lz

RBa

b

c

a

A

R

R

1

A

K

Z

A

M

S

T

R

1

Ai

D

0d

d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 779–785

av ai lab l e at www.sc iencedi rec t .com

jou rna l h om epa ge: www.int l .e lsev ierhea l th .com/ journa ls /dema

ong-term microtensile bond strength of surface modifiedirconia

obert L. Smitha, Carlos Villanuevab, James K. Rothrocka, Cristina E. Garcia-Godoya,rian R. Stonerc, Jeffrey R. Piascikc, Jeffrey Y. Thompsona,b,∗

Bioscience Research Center, Nova Southeastern University, Fort Lauderdale, FL, USADepartment of Prosthodontics, Nova Southeastern University, Fort Lauderdale, FL, USACenter for Materials and Electronic Technologies, RTI International, Research Triangle Park, NC, USA

r t i c l e i n f o

rticle history:

eceived 2 July 2010

eceived in revised form

1 March 2011

ccepted 28 March 2011

eywords:

irconia

dhesion

icrotensile

ilanation

ribochemical

esin cement

a b s t r a c t

Objective. To compare long-term microtensile bond strength of zirconia, surface-modified via

a novel treatment, to current surface conditioning methods for zirconia, when resin bonded

to dental composite.

Methods. Two ProCAD (porcelain) and 10 sintered ZirCAD (ZrO2) blocks

(18 mm × 14 mm × 12 mm) were obtained from manufacturers. Twelve Herculite XRV

composite blocks were fabricated (18 mm × 14 mm × 12 mm). Bonding surface of blocks was

polished through 1200-grit SiC and air-abraded (50 �m alumina, 0.28 MPa, 20 s). Blocks were

then separated into six groups: (1) porcelain (control), HF-etched/silane-treated, (2) ZrO2,

tribochemical-coated/silane-treated, (3) ZrO2, primer-treated, (4) ZrO2, modified via novel

3.2 nm silica layer/silane-treated, (5) ZrO2, modified via novel 5.8 nm silica layer/silane-

treated, and (6) ZrO2, modified via novel 30.4 nm silica layer/silane-treated. Blocks were

bonded to composite using Clearfil Esthetic cement. Blocks were stored in distilled water

(37 ◦C, 24 h), then cut into microtensile bars (n = 8/gp), then bond strengths were measured

using a universal testing machine at 0, 1, 3, and 6 months. All groups were statistically

analyzed (ANOVA, Tukey’s, p < 0.05).

Results. At 6 months (aging), all silica seed layer specimens displayed microtensile bond

strength similar to CoJet specimens but less than that of silane-modified dental porcelain.

Conclusion. The deposition of a silica layer on zirconia resulted in similar or superior long-

term resin bond strength when compared to traditional silanation and bonding techniques

for zirconia but lower than that for silane-treated dental porcelain.

© 2011 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

. Introduction

dvances in ceramic materials over the years have resultedn the introduction of zirconia (ZrO2) as a viable mate-

∗ Corresponding author at: Nova Southeastern University, College of Der., Fort Lauderdale, FL 33328-2018, USA. Tel.: +1 954 262 7395; fax: +1 9

E-mail address: jeffthom@nova.edu (J.Y. Thompson).109-5641/$ – see front matter © 2011 Academy of Dental Materials. Puoi:10.1016/j.dental.2011.03.018

rial for use in dental prosthetics. Zirconia provides superior

ntal Medicine, Department of Prosthodontics, 3200 S. University54 262 1782.

strength and fracture toughness when compared to porcelainand other silica-based materials while presenting improvedesthetic properties compared to metallic-based prosthetics.Although these properties have led to the use of zirconia in

blished by Elsevier Ltd. All rights reserved.

Journal Identification = DENTAL Article Identification = 1822 Date: June 24, 2011 Time: 12:49 pm

l s 2

780 d e n t a l m a t e r i a

a variety of dental applications, there exists a problem withzirconia.

Bonding of zirconia to tooth structure or other syntheticmaterials is difficult when compared to silica-based materials.The bonding of traditional dental ceramics, i.e., silica-basedceramics, utilizes mechanical and adhesive bonding [1].Mechanical bonding is dependent on the micromechani-cal interlocking between the resin cement and silica-basedceramics caused by surface roughening. Phosphoric acid(H3PO4) or hydrofluoric acid (HF) etching is a common methodused to roughen silica-based ceramic surfaces [1].

Chemical adhesion between resin cement and a silica-based ceramic is achieved with the use of silanes. Silanesare bi-functional compounds that promote chemical bondingbetween dissimilar organic (i.e., resin cements) and inorganic(i.e., silica-based) materials [2]. This is achieved by functionalalkoxy groups on the silane molecule bonding to the silica(SiO2) phase hydroxyl groups (–OH) on the surface of the den-tal ceramic. Organo-silanes also have a degradable functionalgroup that copolymerizes with the organic matrix of resincement [3,4]. These processes create the chemical bonds nec-essary for the successful bonding of resin cement to dentalceramics. Silanes are also responsible for increasing surfaceenergy and wettability of ceramic surfaces, which enhancesboth mechanical and chemical bonding.

These traditional methods of mechanical and adhesivebonding used on silica-based ceramics are not applicable foruse with high-strength ceramics, i.e., zirconia. The absence ofsilica or any substantial glassy phase in the microstructure ofzirconia eliminates the viability of acid etching as a methodto roughen the surface for significant mechanical bonding,and nullifies the use of traditional silanes since there is nosilica present to readily form surface hydroxyls for chemicalbonding [5,6]. The difficulty of bonding to zirconia has resultedin alternative methods of adhesion being developed. Currenttechniques used to facilitate mechanical bonding of zirconiainclude grinding, particle air-abrasion using alumina or otherabrasive particles, and rotary abrasion using diamond burs.However, surface grinding and abrasion can introduce surfaceflaws that can decrease the fracture strength of zirconia [7,8].

Chemical adhesion of resin cements to zirconia has beenaddressed using several techniques. Application of a silicacoating through tribochemical and plasma spray techniqueshas yielded varying results with respect to bond strength. Thetribochemical technique, a commonly used method of silicacoating high strength ceramics, embeds silica on a ceramicsurface by air-abrading the surface with silica-coated aluminaparticles [5,9]. It has been demonstrated that applying a tri-bochemical coating, followed by traditional silanation, doesimprove the resin adhesion of zirconia [2,10–13]. However,bond strengths are not as high as those reported for resinbonded to silane-treated porcelain, and there is still the issueof creating surface flaws from air-abrasion that could decreasethe fracture strength of zirconia.

Improvements in resin cements and silane coupling agentshave been shown to increase bond strength. The combination

of silane primers and resin cements, that contain phospho-ric acid in the form of a phosphate monomer (MDP), havedemonstrated improvements in adhesion to zirconia ceram-ics [5,14–16]. However, the use of these phosphate containing

7 ( 2 0 1 1 ) 779–785

primers and resin cements alone produce lower bond strengthwhen compared to tribochemical coating coupled with silana-tion and resin bonding [2,10,17–20].

Recently, Piascik et al. [21] reported on a method for improv-ing adhesion to zirconia ceramics via a unique vapor-phasedeposition technique, whereby silicon tetrachloride is com-bined with water vapor to form an ultra-thin silicate layer onthe zirconia surface. The study confirmed that this modifica-tion improved adhesion of zirconia substrates to resin cementwhen used in conjunction with traditional silanation andbonding techniques. Application of the chloro-silane pretreat-ment resulted in a resin bond strength comparable to that ofsilane-treated dental porcelain. With the confirmation of theeffectiveness of the chloro-silane pretreatment, the long-termusage of the vapor-phase deposition technique needs to beestablished. Therefore, the present study evaluates the long-term microtensile bond strength of the zirconia-compositeinterface modified using this novel chemical surface treat-ment and the effect of silicate layer thickness on long-termmicrotensile bond strength. It is hypothesizes that the use ofthe chloro-silane surface treatment to zirconia will result in along-term resin bond strength similar to that of silane-treateddental porcelain and that increasing silicate layer thicknesswill result in a decrease in long-term bond strength.

2. Materials and methods

For this study, two porcelain blocks (ProCAD, IvoclarVivadent, Schaan, Liechtenstein) and 10 sintered zirconiablocks (Lava, 3M-ESPE, St. Paul, MN, USA), all measuring12 mm × 14 mm × 18 mm, were obtained from the manufac-turers. Following the procedures implemented by Plascik et al.[21], the bonding surfaces of the blocks were polished through1200 grit using silicon carbide (SiC) abrasive paper (CarbiMet2, Buehler, Lake Bluff, IL) to ensure equal starting surfaceroughness. Additionally, all surfaces were air abraded (100-�malumina abrasive, 0.28 MPa, 20 s) and ultrasonically cleaned(distilled water, 300 s) prior to chemical surface treatments orbonding procedures. Twelve composite blocks (Herculite XRV,Kerr Corporation, Orange, CA, USA) were fabricated by con-densing material into a Teflon mold (12 mm × 14 mm × 18 mm)in 2 mm increments and light-cured (700 mW/cm2 for 40 s perincrement, Optilux 501, Kerr Corporation, Orange, CA). Thebonding surfaces of these blocks were polished through 1200grit using SiC abrasive paper. Specimens for microtensile test-ing were fabricated with minor variations for each surfacetreatment/ceramic substrate group. Two specimen blocks foreach of the six different groups were fabricated as follows,with Groups 2 and 3 representing current clinical proceduresused with zirconia restorations:

• Group 1 (control): Porcelain, acid etched with 5% HF gel(IPS Ceramic Etching Gel, Ivoclar Vivadent, Schaan, Liecht-enstein) for 60 s (rinsed and air-dried), treated with silane(Monobond-S, Ivoclar Vivadent, Schaan, Liechtenstein), and

bonded to corresponding dental composite block usingdual-cure, MDP-containing (phosphoric acid modified) resincement (Clearfil Esthetic Cement and DC Bond, Kuraray,Okayama, Japan).

Journal Identification = DENTAL Article Identification = 1822 Date: June 24, 2011 Time: 12:49 pm

2 7

•

•

•

•

•

cib22iL11sttChssiTa2

w(Cutmt

�

wb

soA

d e n t a l m a t e r i a l s

Group 2: Zirconia (no novel surface treatment), silica coatedwith 30-�m Al2O3 particles modified with salicylic acid(CoJet, 3M-ESPE, St. Paul, MN – 0.28 MPa, 10-mm workingdistance, 15 s), treated with silane, and bonded to dentalcomposite block using dual-cure resin cement.

Group 3: Zirconia (no novel surface treatment), treated withphosphoric acid modified primer (Metal/Zirconia Primer,Ivoclar Vivadent, Schaan, Liechtenstein), and bonded todental composite block using dual-cure resin cement.

Group 4: Zirconia (novel surface treatment yielding silicalayer with thickness of 3.2 nm) treated with silane andbonded to dental composite block using dual-cure resincement.Group 5: Zirconia (novel surface treatment yielding silicalayer with thickness of 5.8 nm) treated with silane andbonded to dental composite block using dual-cure resincement.

Group 6: Zirconia (novel surface treatment yielding silicalayer with thickness of 30.4 nm) treated with silane andbonded to dental composite block using dual-cure resincement.

All ceramic/composite specimens were bonded under aompressive load of 7 N for 10 min using a universal test-ng machine (Instron Model 8841, Canton, MA, USA). Afteronding, the interface was light-cured on all four sides for0 s each and stored in distilled water at 37 ◦C for 24 h. After4 h, the specimens were sectioned using a diamond wafer-ng blade (15 LC IsoMet Diamond Wafering Blade, Buehler,ake Bluff, IL, USA) with a low-speed diamond saw (IsoMet000, Buehler, Lake Bluff, IL, USA) into wafers (cross-section4 mm × 1.5 mm) and then into microtensile bars (cross-ection 1.5 mm × 1.5 mm). The outer material from each ofhe larger blocks was discarded, as these surfaces were con-aminated with the wax (Type I Impression Compound, Kerrorp., Romulus, MI, USA) used to bond to the specimenolder during sectioning. After fabricating the microten-ile specimens, each group was randomly divided into 4ub-groups, 0 (baseline), 1, 3, and 6 months, and storedn distilled water at 37 ◦C for the designated time period.he cohesive tensile strength of the dental composite waslso tested at baseline after storage in water at 37 ◦C for4 h.

For microtensile testing, an outer edge of each specimenas bonded to a microtensile fixture using cyanoacrylate glue

Rocket Heavy and Accelerator, Dental Ventures of America,orona, CA, USA). The microtensile specimens were testedsing a universal testing machine (Instron Model 8841, Can-on, MA, USA) at a crosshead speed of 1.0 mm/min. The

icrotensile bond strength, �, was calculated using the equa-ion:

= P

A

here P is the load at the moment of failure (N) and A is theonding area of the specimen (mm2).

The effect of the various surface treatments on microten-ile bond strength was evaluated using a one-way analysisf variance (ANOVA), with a level of significance set at 0.05.

post hoc Tukey’s test was performed if the microtensile

( 2 0 1 1 ) 779–785 781

bond strengths were determined to be significantly different(p < 0.05).

Fractured surfaces of the samples were analyzed using astereomicroscope (SMZ-140, VWR International, West Chester,PA, USA) and scanning electron microscope (SEM – Quanta200, FEI, Hillsboro, OR, USA) to assess whether the failuremodes were adhesive (partial or complete failure in the adhe-sive or debonding of the adhesive from the surface), cohesive(partial or complete failure in the ceramic or composite), ormixed (combination of adhesive/cohesive failure). All fracturesurfaces were gold sputter coated before SEM analysis. Energy-dispersive X-ray spectroscopy (EDS – INCA x-sight, OxfordInstruments, Tubney Woods, Oxfordshire, UK) was used todetermine the elemental composition of the features on thefracture surface of the specimens and to aid in determiningthe failure mode.

3. Results

The mean microtensile bond strength values and standarddeviations are reported in Table 1. Microtensile specimensfrom the air abrasion and the three novel silica layer groupsdebonded during the cutting process, resulting in smallerspecimen groups.

One-way ANOVA and Tukey’s post hoc test showed thatthere was a significant difference (p < 0.05) between the spec-imen groups for each time period. For the baseline, thecohesive tensile strength of the composite group was signifi-cantly different than the bond strength of any other specimengroup. Porcelain, CoJet, and the 3.2 nm groups had signifi-cantly greater bond strengths than the other groups at t = 0(baseline). After 1 month, the porcelain group displayed sig-nificantly greater mean bond strength than any of the otherspecimen groups. After 3 months, the mean strengths for theporcelain, CoJet, and 3.2 nm groups were found to be signifi-cantly greater than those of the other groups. There was nosignificant difference (p > 0.05) between mean strengths of airabrasion, 5.8 nm, and 30.4 nm groups. After 6 months, onlyCoJet exhibited a bond strength that was statistically equiva-lent to porcelain. The mean bond strengths of the silica seedlayer groups (3.2, 5.8, and 30.4 nm) were only statistically sim-ilar to CoJet and there was no statistical difference betweenthe silica-seed layer groups.

Over the four time periods, every group showed a signif-icant difference in bond strength except for the air abrasionand 5.8 nm groups. Porcelain showed a significant increase inbond strength after 1 and 3 months but showed a decrease inbond strength after 6 months that was statistically similar tobaseline. CoJet showed a significant increase in bond strengthafter 3 months followed by a decrease after 6 months that wassimilar to baseline. After 3 months, the 3.2 nm group exhibiteda significant increase in bond strength followed by a signifi-cant decrease in bond strength. The 5.8 nm group only showeda significant decrease in bond strength after 6 months. The

30.4 nm group showed a significant increase in bond strengthfrom baseline to 1 month. However, after 3 months, the bondstrength was not significantly different from the baseline or 1month data.

Journal Identification = DENTAL Article Identification = 1822 Date: June 24, 2011 Time: 12:49 pm

782 d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 779–785

Table 1 – Microtensile bond strength (MPa) mean values and standard deviations for specimen groups.

Material n/group Baseline 1 Month 3 Months 6 Months

Composite (cohesive strength) 8 60.3 (17.6)d* - - -Porcelain 8 25.6 (5.6)a1 38.9 (8.1)a2 40.2 (8.7)a2 27.2 (5.2)a1

Air abrasion 6 15.0 (6.8)bc1 18.1 (5.8)bd1 14.3 (0.8)b1 0.0 (0.0)CoJet 8 24.0 (11.1)ac1 25.7 (10.0)bc1 44.9 (12.5)a2 22.1 (12.5)ab1

3.2 nm 6 25.2 (4.3)a1 27.5 (5.9)c1 35.0 (5.2)a2 13.3 (1.6)b3

5.8 nm 5 18.2 (5.8)bc1 13.2 (2.2)d12 16.5 (10.1)b12 9.8 (4.6)b2

30.4 nm 5 12.1 (3.4)b1 20.4 (5.6)bc2 15.8 (5.1)b12 16.6 (5.5)b12

) betn dif

∗ Different superscript letters indicate significant difference (p < 0.05indicate significant difference (p < 0.05) for specimen groups betwee

Optical and scanning electron microscopy revealed that allbond failures were adhesive in nature (partial or complete fail-ure of the adhesive). Fig. 1A shows a typical adhesive failure

that was observed for the microtensile specimens. Elemen-tal mapping from energy-disperse X-ray spectroscopy (EDS) ofthe specimen fracture surface (Fig. 1B and C) showed the loca-

Fig. 1 – SEM and EDS micrographs of 3.2 nm zirconia specimen szirconia fracture surface. (B) EDS analysis showing presence of zpresence of resin cement (SiO2, green areas). (D) EDS spectrum ilfor the resin cement section. (For interpretation of the referencesweb version of the article.)

ween groups for same time interval. Different superscript numbersferent time intervals.

tion of zirconia and resin cement (silica [SiO2]). EDS spectra(Fig. 1D) showed the presence of only resin cement (identifiedby the distinguishable compounds of barium oxide [BaO] and

SiO2) on the zirconia fracture surface. This observation wasobserved for all other specimen groups, confirming that thefailures were of an adhesive nature.

howing example of partial adhesive failure. (A) SEM ofirconia (ZrO2, white areas). (C) EDS analysis showinglustrating only the presence of silicon (Si) and barium (Ba)

to color in this figure legend, the reader is referred to the

Journal Identification = DENTAL Article Identification = 1822 Date: June 24, 2011 Time: 12:49 pm

2 7

4

Ipcisottfgtlfo

ol[wlsisadsaTnnn

aboaisipcptHdidcTitsfsobi

d e n t a l m a t e r i a l s

. Discussion

mprovement in the retention of zirconia to resin cements hasrogressed in recent years. The long-term stability of resinements is important due to the environment and time-spann which that dental materials will be used. Long-term watertorage or thermal cycling have been the most common meth-ds used to artificially age resin bonds for mechanical testingo determine hydrolytic stability. Research has shown thathe method of artificial aging does not affect bond strengthor MDP-based resin cements to zirconia [11,14]. There is noeneral agreement on the preferred method of artificial agingo test resin bond strengths. Therefore, the authors selectedong-term water storage in order to remove the affects that dif-erences in the coefficient of thermal expansion would haven the retention of the bonding materials.

Particle air abrasion of zirconia ceramics, with aluminar other ceramic particles, has been shown to produce

ow bond strengths relative to that of porcelain materials5,10,12,13,17,22]. Our research confirmed this observation,

ith the air-abrasion specimens (Group 3) having consistentlyow bond strengths throughout the study. The air-abrasionpecimens were also the most susceptible to debonding dur-ng specimen fabrication and testing, resulting in the loss ofpecimens and the absence of data during the 6-month evalu-tion. Premature debonding could be due to the forces exerteduring specimen fabrication either exceeding the initial bondtrength of the specimens or weakening the bonds enough tollow for hydrolytic degradation of the cement after 6 months.his reinforces the concept that mechanical adhesion alone isot suitable for providing the maximum resin bond strengtheeded for dental prosthetics, and that chemical adhesion isecessary for achieving optimal bond strength.

The use of a tribochemical silica coating, followed by silane coupling agent, is another method used for resinonding of zirconia dental materials [23]. The combinationf mechanical bonding (surface roughening) and chemicaldhesion (bonding of silane to silica) has been effective inncreasing the retention of zirconia to resin cements. Ourtudy has shown that the application of a tribochemical sil-ca coating, followed by the application of silane (Group 2),rovided equivalent long-term bond strength to resin cementompared to silane-treated dental porcelain over the 6-montheriod. This is in agreement with previous research studieshat showed the benefits of tribochemical coating [10–12,17].owever, the tribochemical group exhibited the greatest stan-ard deviation out of all the groups for each time period. It

s possible that the tribochemical coating process could pro-uce a non-uniform silica layer on the bonding surface, whichould be responsible for the large variation in bond strength.he variation in bond strength could be alleviated by increas-

ng specimen group size. Another possibility could be thathe hardness of zirconia could make it more difficult for theilica coating from the alumina particles to bond to the sur-ace, resulting in a low concentration of silica on the bondingurface [5,17]. This problem could be addressed by increasing

perating pressure, which has been shown to increasing resinonding [24]. Also, as previously mentioned, there is still an

ssue with the possibility of decreased fracture strength of zir-

( 2 0 1 1 ) 779–785 783

conia due to the creation of micro-flaws from particle abrasion.However, this is less of a concern, as it has been shown thatthe application of resin luting agents can “heal” surface micro-flaws [25,26] and there was no zirconia present on the surfaceof the composites showing cohesive failure of the ceramic.

The application of a vapor-deposited, silica seed layer onthe bonding surface of zirconia has been shown to providesimilar resin bond strength to that of silane-treated dentalporcelain [21]. As with the air-abraded specimens, some speci-mens debonded during fabrication and water storage. Baselineevaluation of the silica seed layer specimens (Groups 4–6)showed that only the 3.2 nm group (Group 4) exhibited similarresin bond strength to that of silane-treated dental porcelain,which is in agreement with the results reported by Piascik et al.As mentioned in that previous study, it is possible that addi-tional layers, which are added to increase the silica seed layerthickness, are partially bonded together by secondary chem-ical bonds, e.g. van der Waals bonds. This could result in adecrease in bond strength between zirconia and the compos-ite.

After 6 months, the silica seed layer specimens exhibitedresin bond strength significantly less than that for silane-treated dental porcelain but equivalent to the tribochemicalsilica specimens. It is possible that, after 6 months, the resincement became susceptible to hydrolytic degradation whenbonded to the silica seed layer. Another possible reason forthe decrease in bond strength of the silica seed layer speci-mens could be the bonding between the silica seed layer andthe zirconia surface. The silica seed layer may only be bondedto zirconia through mechanical and/or secondary chemicalbonding, which would not be as strong as the covalent bond-ing formed between silane and resin cement. Over time, waterabsorption at the interface could decrease the bond strengthof the silica seed layer. Energy-dispersive X-ray spectroscopy(EDS) analysis showed the partial or complete absence of sil-icon from the fracture surface of zirconia specimens after6 months (Fig. 2). Although EDS is a bulk characterizationmethod and not a surface analysis technique, EDS was used asa supplemental analysis tool to illustrate the elemental com-position of the fracture surfaces. The information obtainedfrom both SEM and EDS would indicate that the compositeresin was not present on the surface of zirconia. Further workis necessary, such as using X-ray photoelectron spectroscopyor Auger electron spectroscopy, to determine bonding states,chemical composition of the fracture surface of microtensilespecimens, and to determine the nature of the chemical bond-ing between the silica seed layer and zirconia and how itaffects bond strength.

A characteristic of the novel modification process thatcould influence behavior, and must be analyzed further, isthe possibility that the porous, under-dense nature of thevapor deposited silica layer would make it more suscep-tible to stress-enhanced hydrolytic degradation. While thehigh surface area silica that forms as a result of mixingsilicon-tetrachloride and water might improve the adhesiveproperties of the surface, the defective nature of the films

may make them more susceptible to chemical attack [27,28].Improving the homogeneous structure of these depositedfilms could improve resistance to this form of time and envi-ronment dependent degradation. Research is currently being

Journal Identification = DENTAL Article Identification = 1822 Date: June 24, 2011 Time: 12:49 pm

784 d e n t a l m a t e r i a l s 2 7 ( 2 0 1 1 ) 779–785

Fig. 2 – EDS and SEM micrographs of 5.8 nm zirconia specimen showing example of partial debonding of silica seed layerfrom zirconia surface. (A) EDS spectrum illustrating the absence of silica seed layer on surface of zirconia where cement isnot present. (B) SEM of zirconia fracture surface showing location of cement on fracture surface.

r

conducted to evaluate the effect of post-deposition densifica-tion on the stability of the silica layer in hopes of improvingthe lifetime behavior of this treatment technique.

There was no correlation between the silica seed layerthickness and long-term resin bond strength after 6 months.Only during the 3-month evaluation did the 3.2 nm specimensexhibit resin bond strength similar to that of silane-treated dental porcelain. It is possible that the complexsilica seed layer/silane/MDP-containing resin cement regiontakes several months to completely polymerize and becomehydrolytically stable. Piwowarczyk et al. [12] showed anincrease in the shear bond strength of tribochemical coatedzirconia using MDP and phosphoric acid (meth)acrylate con-taining resin cements after 14 days of water storage followedby thermocycling. This increase in bond strength, after artifi-cial aging, could also be occurring for the tribochemical silicaand 3.2 nm groups between 1 and 3 months.

All of the failures observed in this study were adhesive innature (partial or complete failure of the adhesive). This isin disagreement with the results reported by Piascik et al., inwhich all groups exhibited a combination of adhesive, cohe-sive (partial or complete failure of composite or ceramic),and mixed mode (combination of adhesive/cohesive) fail-ure. The difference in failure modes could be due to thegreater cohesive strength of the composite used in our study(60.3 ± 17.6 MPa) compared to 29.2 ± 3.7 MPa in the Piascik et al.study, resulting in the absence of cohesive failures. Energy-dispersive X-ray analysis confirmed the absence of any of thedistinguishable compounds present in the dental composite

(titanium oxide [TiO2] and zinc oxide [ZnO]) (Fig. 1D) on thefracture surface of the ceramic and the absence of zirconia onthe surface of the composite, reinforcing that all failures wereadhesive.

5. Conclusion

The long-term resin microtensile bond strength to zirconia,modified with a vapor-deposited silica seed layer, was eval-uated. After 6 months, the application of the silica seedlayer provided similar or improved resin bond strength whencompared to current conventional bonding techniques usedfor zirconia. However, after 6 months, the use of the silicaseed layer on zirconia resulted in lower bond strength whencompared to silane-treated dental porcelain. Further work isnecessary to optimize the silica seed layer to enhance theadhesive strength of zirconia and other high strength ceram-ics to resin-based materials for clinical applications and todetermine the effect that thermocycling has on the long-termadhesive strength of the silica seed layer bonding system.

Acknowledgements

This research was supported in part by NIH/NIDCRR56DE020142 and through a grant provided by the RTIInternational Research and Development Fund.

e f e r e n c e s

[1] Blatz MB, Sadan A, Kern M. Bonding to silica-basedceramics: clinical and laboratory guidelines. QuintessenceDent Technol 2002;25:54–62.

[2] Akgungor G, Deniz S, Murat A. Influence of different surfacetreatments on the short-term bond strength and durabilitybetween a zirconia post and a composite resin corematerial. J Prost Dent 2008;99(5):388–99.

Journal Identification = DENTAL Article Identification = 1822 Date: June 24, 2011 Time: 12:49 pm

2 7

d e n t a l m a t e r i a l s

[3] Soderholm KJ, Shang SW. Molecular orientation of silane atthe surface of colloidal silica. J Dent Res 1993;72:1050–4.

[4] Barghi N. To silanate or not to silanate: making a clinicaldecision. Compend Contin Educ Dent 2000;21:659–62.

[5] Kern M, Wegner SM. Bonding to zirconia ceramic: adhesionmethods and their durability. Dent Mater 1998;14:64–71.

[6] Blatz MB, Sadan A, Kern M. Resin-ceramic bonding: a reviewof the literature. J Prosthet Dent 2003;89:268–74.

[7] Kosmac T, Oblak C, Jevnikar P, Funduk N, Marion L. Theeffect of surface grinding and sandblasting on flexuralstrength and reliability of Y-TPZ zirconia ceramic. DentMater 1999;15:426–33.

[8] Kosmac T, Oblak C, Jevnikar P, Funduk N, Marion L. Strengthand reliability of surface treated Y-TZP dental ceramics. JBiomed Mater Res Part B 2000;53:304–13.

[9] Kern M, Thompson VP. Sandblasting and silica coating of aglass-infiltrated alumina ceramic: volume loss, morphology,and changes in the surface composition. J Prosthet Dent1994;71:453–61.

[10] Atsu SS, Kilicarslan MA, Kucukesmen HC, Aka PS. Effect ofzirconium-oxide ceramic surface treatments on the bondstrength to adhesive resin. J Prost Dent 2006;95(6):430–6.

[11] Kumbuloglu O, Lassila LVJ, User A, Vallittu PK. Bonding ofresin composite luting cements to zirconium oxide by twoair-particle abrasion methods. Oper Dent 2006;31(2):248–55.

[12] Piwowarczyk A, Lauer HC, Sorensen JA. The shear bondstrength between luting cements and zirconia ceramicsafter two pre-treatments. Oper Dent 2005;30(3):382–8.

[13] Wegner SM, Kern M. Long-term resin bond strength tozirconia ceramic. J Adhes Dent 2000;2:139–47.

[14] Wegner SM, Gerdes W, Kern M. Effect of different artificialaging conditions on ceramic-composite bond strength. Int JProsthodont 2002;15:267–72.

[15] Wolfart M, Lehmann F, Wolfart S, Kern M. Durability of theresin bond strength to zirconia ceramic after using different

surface conditioning methods. Dent Mater 2007;23(1):45–50.

[16] Luthy H, Loeffel O, Hammerle CHF. Effect of thermocyclingon bond strength of luting cements to zirconia ceramic.Dent Mater 2006;22(2):195–200.

( 2 0 1 1 ) 779–785 785

[17] Blatz MB, Chiche G, Holst S, Sadan A. Influence of surfacetreatment and simulated aging on bond strengths of lutingagents to zirconia. Quintessence Int 2007;38(9):745–53.

[18] Tanaka R, Fujishima A, Shibata Y, Manabe A, Miyazaki T.Cooperation of phosphate monomer and silica modificationon zirconia. J Dent Res 2008;87(7):666–70.

[19] Nothdurft FP, Motter PJ, Pospiech PR. Effect of surfacetreatment on the initial bond strength of different lutingcements to zirconium oxide ceramic. Clin Oral Invest2009;13(2):229–35.

[20] Sahafi A, Peutzfeldt A, Asmussen E, Gotfredsen K. Bondstrength of resin cement to dentin and to surface-treatedposts of titanium alloy, glass fiber, and zirconia. J AdhesDent 2003;5(2):153–62.

[21] Piascik JR, Swift EJ, Thompson JY, Grego S, Stoner BR. Surfacemodification for enhanced silanation of zirconia ceramics.Dent Mater 2009;25(9):1116–21.

[22] Derand P, Derand T. Bond strength of luting cements tozirconium oxide ceramics. Int J Prosthodont 2000;13:131–5.

[23] Komine F, Tomic M. A single-retainer zirconium dioxideceramic resin-bonded fixed partial denture for single toothreplacement: a clinical report. J Oral Sci 2005;47(3):139–42.

[24] Heikkinen TT, Lassila LVJ, Matinlinna JP, Vallittu PK. Effect ofoperating air pressure on tribochemical silica-coating. ActaOdontol Scand 2007;65:241–8.

[25] Burke FJ, Fleming GJ, Nathanson D, Marquis PM. Areadhesive technologies needed to support ceramics? Anassessment of the current evidence. J Adhes Dent2002;4:7–22.

[26] Blatz MB. Long-term success of all-ceramic posteriorrestorations. Quintessence Int 2002;33:415–26.

[27] Isobe H, Kanoko K. Porous silica particles prepared fromsilicon tetrachloride using ultrasonic spray method. J ColloidInterf Sci 1999;212(2):234–41.

[28] Naono H, Hakuman M, Tanaka T, Tamura N, Nakai K. Porous

texture and surface character of dehydroxylated andrehydroxylated MCM-41 mesoporous silicas—analysis ofadsorption isotherms of nitrogen gas and water vapor. JColloid Interf Sci 2000;225(2):411–20.