A Novel Combined Silica-Coating and Etching Protocol for Titanium for Improved Adhesion

Durable adhesion between resin composite cement and Ti prostheses is critical for a long clinical life in dentistry. A rough surface is a prerequisite for mechanical interlocking and a special coating method, silica-coating must be followed by silanation for chemical adhesion. Commercially pure grade-2 Ti specimens were modified by silica-coating and silanized with an experimental silane blend primer or two commercially available silane primers. Specimens were assigned randomly: Group-A (control; polished only Ti), Group-B (polished Ti + silica-coating + primers), Group-C (polished Ti + etching with HNO3 + with or without primers), Group-D (polished Ti + silica-coating + etching with HNO3 + with or without primers), Group-E (polished Ti + etching with HCl+H3PO4 + with or without primers), and Group-F (polished Ti + silica-coating + etching with HCl+H3PO4 + with or without primers). Next, Ti specimens were analyzed for their atomic concentration by electron dispersive spectroscopy (EDX). BisCem™ resin cement was used to build up enclosed mold stubs onto the Ti specimens. Enclosed mold micro-shear bond strength (EM-µSBS) was measured after storage in distilled water for 1 day, 1 week, 4 weeks, and 8 weeks. A combined treatment employing silica-coating + the blend HCl+H3PO4 had a more substantial effect on Ti surface roughness when compared to other surface pretreatments. Whereas, the highest EM-µSBS values were observed in Group-D from week 1 to week 8. A novel combined dual surface modification creates higher surface roughness on Ti, and this might significantly improve the durability of resin-Ti adhesion.


Introduction
Boosted by the consolidation of CAD/CAM systems, there has been a continuous call for the use of titanium as a dental prosthesis material for crowns, bridges, implant fixtures, abutment, fixed partial dentures, and removable partial dentures.The handling of titanium and titanium alloys is much less cumbersome with CAD/CAM systems which have made it even more accessible to the dentist and dental laboratories.Moreover, 3D printing will likely change the game in the very near future.Recent research has indicated that laboratory experiments on a modified surface treatment of titanium might lead to sustainable adhesive bond between the resin cement and titanium [1,2].For durable adhesion, titanium surfaces are first silica-coated, then silanated (a.k.a.silanization) with a silane coupling agent primer, providing mechanical and chemical adhesion.This is vital for durable and successful resin Ti bonding in dental applications.On the other hand, SiO 2 -based ceramics need mandatory etching with HF followed by rinsing, drying, silanation and luting for durable adhesion.There are some key publications on surface treatment of indirect restorative dental materials presenting the evolution of adhesion promotion [3][4][5][6].
We have recently introduced a new, dual surface modification method for Ti to promote adhesion [1,2] in which the surface of Ti specimen is silica-coated followed (or preceded) by etching with strong mineral acids, rinsed, dried, Silicon (2023) 15:7705-7715 1 3 and then silanated with trialkoxysilanes.Silica-coating of Ti followed by silanation has been intensively studied leading to, e.g., a promising concept for durable, hydrolytically stable adhesion, called 'a novel silane system' [7][8][9][10].In the current research, we are utilizing the said silane system and further developing the dual surface modification concept to Ti.The significance is that this novel dual surface modification concept is being applied on titanium, contributing to a greater surface area and different surface chemistry which might improve the resin-titanium adhesion.Albeit, gritblasting is being employed in dental practice, but the harsh oral environment easily makes this resin-titanium adhesion somewhat degradable.
Adhesion strength between surface treated titanium and resin cements is being routinely evaluated by the microshear bond strength test (µSBS) in research laboratories.The µSBS test has enabled researchers to assess the bond strength of not only for smaller specimen sizes but it also makes the specimens (at least in principle) defect free.It is noteworthy that such adhesion results would exhibit numerically some 1-2 times higher values than the conventional shear bond strength, SBS, results.However, more feasible, accurate, and reliable enclosed mold based EM-µSBS is being replacing the traditional SBS and µSBS tests [18].It is well known that stress accumulation at the tip of the resin specimen mold during the mold removal can cause an invisible micro-crack which could propagate and lead to failure initiation on the adhesive bond between the resin-titanium.Researchers have developed an enclosed mold micro-shear bond strength test which does not need the mold to be removed at all, and this mode reduces the chances of micro-crack formation at the tip of resin stubs [19][20][21][22].
The aim of this laboratory study was to further improve and assess the dual surface modification method, i.e., silicacoating with etching, which could create enough retention and upon silanation would provide the basis for stronger chemical adhesion of resin composite cement to titanium.The objectives of the study were: a) evaluation of the difference of surface roughness between the conventional acid-etching method to the dual surface modification method, and b) evaluation of the effect of dual surface modification technique with the silane blend and 2 commercially available silanes on resin-titanium adhesion.We hypothesize that the novel dual surface modification method would create higher surface roughness which with silanation would enhance the resin-Ti adhesion.

Materials and Methods
A total of 252 pure grade 2 Ti (Baoji Xinlian Titanium Industry, Baoji city, Shaanxi Province, China) specimens of the size 10.0 mm x 10.0 mm x 1.0 mm were laser-cut.The Ti specimens were polished manually using SiC abrasive papers in the grit sequence of sizes 220, 320, 500, and 1000 (Lunn Major, Struers™, Denmark), under running water for 5 min after each.Specimens were rinsed individually with distilled water and cleansed ultrasonically (Decon FS200 Frequency sweep, Decon Ultrasonics, Hove, Sussex, UK) in 70% ethanol for 10 min each, followed by ultrasonic cleaning in DI water for 10 min.The specimens were air-dried overnight at room temperature in a desiccator.The Ti-specimens were randomly assigned to four study groups according to their aging times, in a group of twenty of each, plus one control specimen of polished titanium in each group.This totals 21 groups x 4 (i.e., aging periods) x 3 (i.e., each group contained 3 specimens), i.e., all together 252.In this pilot study we used only 3 specimens per one study group.The specimens were etched, silica-coated, or coated with a primer (for shear bond strength testing) depending on the type of surface modification as listed in Table 1 below.
One side (the upper half) of Ti-specimens were silicacoated using a 110 µm special silica-coated alumina powder (Rocatec™ plus, 3M ESPE, Seefeld, Germany) at a constant operational pressure of 3.4 bar with a Pen-blaster (Shofu Dental, Kyoto, Japan) held 10 mm away perpendicularly and for a rotating movement for 15 s each.Next, specimens were rinsed with DI water, ultrasonically cleansed in 70% ethanol and DI water followed by drying them overnight at room temperature in a desiccator.
A blend of an experimental silane coupling agent primer consisting of 1% 3-acryloyloxypropyltrimethoxy silane and 0.3% 1,2-bis-(triethoxysilyl)ethane was prepared [1,8,11,13,14,17], and also commercially available Monobond Plus™ and Z-Prime Plus™ were applied with a fine applicator brush onto the specimens (one coat) which were allotted for primer coating.The remaining Ti specimens which were etched after silica-coating by placing a specimen individually into plastic test tubes with pre-heated concentrated mineral acids: nitric acid HNO 3 (69 vol.%,VWR chemicals™) and a blend of (35 vol.%) hydrochloric acid (Sigma Aldrich™) and (85 vol.%), phosphoric acid (Peking Chemical Works™), HCl+H 3 PO 4 in test tubes, which were immersed in a water bath at 80˚C for 1 h.
A surface roughness profilometer (Surtronic 3+, Taylor-Hobson pneumo, Leicester, UK) was used to measure the surface topography of all the Ti specimens, including the control, using a diamond stylus tip (5 µm) moved across the Ti specimen, measuring the surface profile: three readings were randomly taken on each specimen to give an average surface roughness of each specimen.The average of the three readings was noted as R a , i.e., the arithmetic means of the departures of the profile from the mean line.
Next, Ti specimens were affixed with a cold cure acrylic polymer (Ivoclar Vivadent™) which as mixed and poured into a plastic mold.BisCem™, a self-adhesive resin composite cement was added in increments into enclosed polyethylene molds obtained by cutting clean polyethylene tubes to a length of appr.3.0 mm each, with an inner diameter of 1.0 mm [1].Rough edges of the molds were smoothened by a plain cut fissure bur.The molds were held firmly on the silanated or non-silanated Ti surface followed by filling to the brim with the resin cement.These specimens were light-cured (Elipar™ 2500,3M ESPE) from all the sides for 40 s each.1All the specimens were aged in DI water for 1 day, 1 week, 4 weeks or 8 weeks depending on their respective groups.

Scanning Electron Microscopy and EDX
Surface morphology of all the Ti surfaces was analyzed by using a scanning electron microscope (SEM, SU 1510, Hitachi, Tokyo, Japan).The specimens were observed under SEM under the high vacuum mode at 15 kV with magnifications of 250X, 500X and 1000X.The EDX analysis was conducted to analyze the composition of different elements on the Ti specimens (EDX, 550i XRF systems, Austin, TX, USA).A 2.00 mm area was analyzed in three random spots of the specimen maintaining a constant pressure of 1×10 -9  Torr in the specimen chamber.Non-monochromated Mg-Kα radiation with excitation energy of hʋ = 1253.6eV was used.

Adhesion Strength Testing
The enclosed mold micro-shear bond strength test was conducted with a universal testing machine (Electropuls™ E3000, Instron, MA, USA) using a constant cross-head speed of 1.0 mm/min until fracture, the force was applied 0.6 mm away from the interfacial surface between the resin cement and the Ti specimen.The enclosed mold micro-shear bond strength is calculated in MPa by the following formula: Shear bond strength (MPa) = maximum force at adhesion failure∕area of resin stub where the maximum force at fracture = N, and area of stub = mm 2

Failure Mode Analysis
Optical light microscopy was conducted to assess the failure mode: a) the failure was adhesive, i.e., less than 33% of the resin stub was remaining on the Ti surface after adhesion strength testing, b) mixed (between 33% and 66% left on the Ti surface), or c) cohesive, i.e., more than 66% of the cement left [19].A magnification of x20 was used to visually analyze the failure modes [1,8].

Statistical Analysis
The obtained data were statistically analysed with statistics software SPSS™ Version 24.0 (IBM, USA) [1].One-way ANOVA was followed by the Turkey post hoc test used to compare the mean values of surface roughness and EM-µSBS between all groups.Statistical analysis using two-way ANOVA test revealed that both surface treatment and chemical primers significantly affect the adhesion strength (α) on different aging intervals.This was followed by the Bonferroni post hoc multiple comparisons.The statistical significance level was set to be α = 0.05.

Results
The control polished Ti exhibited the lowest surface roughness (R a ) (Table 2) values: group A: R a = 0.24 µm ± 0.01 µm.This value of surface roughness was not significantly different from the surface roughness evaluated of the Ti specimens surface treated with HNO 3 , for group C: R a = 0.24 µm ± 0.00 µm -0.28 µm ± 0.01 µm, with and without silanation.Significantly higher surface roughness was observed in the Ti specimens treated with silica-coating + HCl+H 3 PO 4 and for this group F: R a = 1.18 µm ± 0.34 µm -1.22 µm ± 0.03 µm, with and without silanation when compared to all the specimen groups.The surface roughness of this group was significantly different from all other group of specimens in the study.The surface roughness of titanium specimens treated with only etching with HCl+H 3 PO 4 , group E: R a = 1.03 µm ± 0.03 µm -1.07 µm ± 0.02 µm, was found to be significantly different from all the specimens evaluated for their surface roughness.No significant difference was found between the Ti specimens which were only silica-coated, i.e., group B: R a = 0.77 µm ± 0.01 µm -0.80 µm ± 0.03 µm, and Ti specimens treated with silica-coating + etched with HNO 3 , group D: R a = 0.74 µm ± 0.02 µm -0.86 µm ± 0.02 µm.However, other than from each other, their surface roughness was significantly different from all other specimens in the study (Table 2).Ti specimens only silica coated and the Ti specimens silica-coated and then acid-etched with HNO3 were found to significantly not different to each other when compared on week 8 (Table 3).Whereas Ti specimens treated with HNO 3 followed by silanation, Ti specimens treated with HCl+H 3 PO 4 and silanated, and Ti specimens treated with HCl+H 3 PO 4 and not silanated showed significantly lower EM-µSBS on week 8 of ageing when compared to other surface modified Ti specimens (Table 3).
SEM results of the current study also showed the same trend as the surface roughness measurement.Surface roughness created by silica-coating and etching could be clearly seen in the specimens treated with the dual surface modification method of silica-coating + HCl+H 3 PO 4 .The control Ti and Ti etched with HNO 3 showed a relatively smoother surface compared to the Ti surfaces treated only with silicacoating or etched with HCl+H 3 PO 4 , treatment with silicacoating + etching with HNO 3 .

Discussion
The hypothesis was that the novel dual surface modification method would create higher surface roughness which with silanation would enhance the resin-Ti adhesion.The hypothesis is accepted.The highest average EM-µSBS of enclosed-mold resin bonded to Ti specimens at the end of 8 weeks of aging in DI water was exhibited by the specimens treated only with silica-coating and with specimens treated with silica-coating + etched with HNO 3 with or without silanation.The EM-µSBS difference for these two sets of specimens was not significantly different from each other after the ageing of 8 weeks (Table 3).The lowest EM-µSBS exhibited on week 8 was for the specimens etched only by HNO 3 and it was even lower than the ISO standard for SBS in dental materials lowest threshold of 5 MPa [1].Specimens only etched with a blend of HCl+H 3 PO 4 (almost at the lowest threshold recommended by ISO standard 10477 amendment) with or without silanation.Control polished Ti and Ti etched only by HNO 3 on day 1 had EM-µSBS of 12.31 MPa ± 0.06 MPa and 7.09 MPa ± 0.06 MPa, respectively.Still, after aging for 1 week, 4 weeks, and 8 weeks the enclosed mold had debonded from the Ti specimen.This said, the Ti specimens etched with HNO 3 and silanated, Ti specimens etched with HCl+H 3 PO 4 silanated, and Ti specimens etched with HCl+H 3 PO 4 non-silanated did not debond but exhibited significantly lower EM-µSBS on 8 weeks aging compared to the other surface-treated Ti specimens as seen in Table 3 where the mean values of EM-µSBS can be seen.This could be attributed to HNO 3 being a weaker acid when compared with the other acid blend used in this study.The mean EM-µSBS (MPa) with SD for multiple comparisons of Ti specimens shows that at week 8, for specimens 2 and 10, P = 0.057.
The SEM analysis revealed that Ti specimens treated only with HNO 3 were quite similar in surface roughness with the control, i.e., untreated polished Ti.This was also confirmed by the surface profilometry analysis, which showed similar surface roughness ranging from R a = 0.24 µm ± 0.00 µm to 0.28 µm ± 0.01 µm.No change was observed in surface roughness of the specimens treated with HNO 3 (with or without silanation after etching).That might be attributed to the passivating effect of HNO 3 on Ti, in contrast against oxidization caused by chromic acid, perchloric acid, hypochlorous acid, and chloric acids [23].
Similar surface roughness values provided by profilometry can be seen in Table 2.In contrast, the dual etching method with a blend of HCl+H 3 PO 4 caused significant surface roughness as seen on the SEM images.This is also supported by the surface roughness profilometry results.Surface roughness was visible with clearly marked edges of etched Ti surface (Fig. 1a-d).A similar pattern of surface roughness was also observed in a research report with etching + grit-blasting with alumina particles [24].Now, it might be suggested that the dual surface modification method of creating surface roughness would create greater mechanical interlocking resulting in enhanced adhesion of resin composite cement to Ti as suggested in previous research [1,25].
All the specimen surfaces apparently consisted mostly of Ti and oxygen: this is because titanium is the substrate material and the reason for oxygen being present in the formation of the passivating titanium dioxide layer (TiO 2 ) formed just in ns after cleansing.Another reason for oxygen to appear would be its signals captured from the Ti specimen itself which phenomenon is widely agreed [6,26,27].
The presence of Si could be attributed to polishing titanium with SiC paper resulting in some Si getting attached to the titanium surface [1].The second reason for the presence of Si and the presence of Al is that the titanium specimens are grit-blasted with silica-coated alumina powder [9].These particles are blasted on the surface with high kinetic energy and embedded onto the surface (i.e., they are fusing onto the surface) and they cannot be removed with ultrasonic cleansing, and they may appear on the EDX spectra.However, the coverage of Si was not evenly distributed on all the surfaces treated with silica-coated alumina powder.It is well known that Si plays a more critical role in chemical adhesion than Al, as the ≡Si-O-SI≡ bond has much higher hydrolytic stability compared to ≡Si-O-Al= [10,26,27].
Based on the enclosed mold micro-shear bond strength data, we can observe that the novel combined surface modification method (silica-coating + etching with HCl+H 3 PO 4 ) has a substantial effect on the surface roughness.The EM-µSBS values were the highest with only silica-coating or silica-coating + HNO 3 etching.The effect of HNO 3 here needs more research to explain.
In the current study for the EM-µSBS test, on day 1 all the Ti specimens with or without surface modification and with or without silanation exhibited the value higher than 5.0 MPa, which is the lowest threshold for SBS for dental materials by ISO standard 10477 Amendment [1,4,27].Nevertheless, by the end of week 8 of aging the specimens, control polish Ti, and Ti etched with HNO 3 only (without silanation) could only be tested for their EM-µSBS on day 1 of aging.This is because the resin cement had spontaneously de-bonded from the Ti substrate on storing it in DI water.This trend of a significant decrease in the EM-µSBS was observed in all kinds of surface-modified Ti specimens, with and without silanation.In addition, optical The low coefficient of variance (RSD), indicated by the results, suggests that the acquired data were consistent and not scattered and closely packed with each other.This might suggest that the smaller size of the specimen brings minor variation and could be due to the fabrication of (almost) defect-free specimens [28][29][30].In the future research we intend to apply some more surface analysis methods, such as AFM, XPS, and ToF-SIMS, to observe the enhanced effects of this novel surface modification method in its details and the changes it creates on the Ti surface.

Conclusion
A novel combined surface modification method of silicacoating + etching with HCl+H 3 PO 4 has a substantial effect on the surface roughness.The EM-µSBS values were the highest with only silica-coating or silica-coating + HNO 3 etching.This said method might present a novel Ti surface treatment for significantly more durable resin-Ti adhesion in dentistry.

Fig. 2
Fig. 2 Pictographic mode of failure analyses of some Ti specimens after adhesion strength testing suggesting adhesive failure modes: (a) polished Ti plus HNO 3 etching, (b) polished Ti silica-coated and then etched with HNO 3 , (c) polished Ti silica-coated followed by HCl+H 3 PO 4 blend for etching

Table 1
Materials used in the current study

Table 2
Ti specimens and their surface treatments with mean surface roughness values (with standard deviation, SD) and atomic concentration (%) of various elements after modification of Ti specimens.Abbreviations: see key below Key: PT = polished Ti, SC = silica-coating, ES = experimental silane blend, MB = Monobond™ Plus, ZPP = Z-Prime Plus™, HNO 3 = nitric acid, HCl+H 3 PO 4 = blend of hydrochloric acid and phosphoric acid * The groups A-F show a single type of surface modification method with different silanes application or without any silanation

Table 3
Mean EM-µSBS (MPa) with SD for multiple comparisons of all the Ti specimens with or without silanation after 1 day, 1 week, 4 weeks and 8 weeks, with the P values *