The optimal polymerization of resin-based luting cements plays a critical role in the long-term clinical success of dental prostheses and indirect restorations. This study investigated a mutual action between the conformational changes and mechanical properties of a dimethacrylate resin-based luting cement with and without pre-application of the acidic functional monomer 10-methacryloxydecyl dihydrogen phosphate.
Degree of conversion in the luting cement was measured using conventional infrared spectrophotometry. Mechanical properties of the luting cements were also evaluated by quasi-static and dynamic nanoindentation tests.
The results of infrared spectrophotometry and nanoindentation testing were proportional in samples without functional monomer pretreatment. When considerable residual monomer remains within the final products, the mechanical properties of the resin-based luting cements could possibly be impaired. Although the apparent degree of conversion increased with a mixture of functional monomer, a reduction in the cross-linking polymer network may have resulted in the highest viscoelastic creep behavior of the luting cement. The time-dependent behaviors found in the nanoindentation tests likely resulted from linear polymerization chains of the functional monomer.
The application of an acidic functional monomer may affect the viscosity of resin-based luting cements. Quasi-static or dynamic nanoindentation is a useful tool for assessing the polymerization qualities of resin composites.
The integrity of luting cements is an important determinant of the long-term success of dental prostheses and indirect restorations . With increasing demand for esthetic properties and adequate marginal seal, the use of resin-based luting cement has become dominant.
Despite their superior esthetic and mechanical properties, conventional resin-based cements require pre-application of total etch adhesives or a self-etching primer containing an acid-functionalized monomer such as 10-methacryloxydecyl dihydrogen phosphate (MDP), because the resin substrate lacks chemical adhesive properties . Acidic monomers may compromise the curing mechanism of resin-based cements and hence optimal polymerization of the cements with MDP application is uncertain. Inferior polymerization of resin-based materials may negatively affect their mechanical properties, causing significant deterioration of clinical performance .
Infrared (IR) spectrophotometry has been used to evaluate the polymerization of dental composites . In this technique, the relative intensities of the C C double bond peak at 1637 cm −1 and of the phenyl group peak at 1608 cm −1 indicate the degree of conversion. However, IR spectrophotometry also detects MDP vibration modes, which theoretically overlap with the cement spectra, so that isolation of polymerization is unlikely . Thus, an evaluation technique that combines IR spectrophotometry and direct mechanical characterization is needed.
IR spectrophotometry using two-beam transmission devices allows visualization of conformational changes over time in thin film preparations of luting cements. Concurrent microscale mechanical testing on the thin film is desirable. Nanoindentation is depth-sensing mechanical testing that continuously measures hardness and elastic modulus by quasi-static load displacement in the thin film preparation . A drawback in mechanical testing of resin-based luting cements is the unavoidable viscoelastic response, which increases with decreased polymerization because of delayed fluid movement of the residual monomer . However, nanoindentation systems currently have high placement precision and can capture a material’s responses over a range of imposed frequencies using dynamic force and displacement amplitudes , enabling measurement of the viscoelastic properties of resin-based materials.
This study evaluates the polymerization of dimethacrylate resin-based luting cement with and without MDP pre-application by measuring conformational changes over time with quasi-static and dynamic nanomechanical testing.
Materials and methods
This study used dual-polymerizing resin-based luting cement (Rely-X TM ARC, 3M ESPE, Tokyo, Japan) and MDP monomer (Epricode, Kuraray, Tokyo, Japan).
Table 1 shows sample preparation protocols. Mixed resin-based luting cement was placed directly between CaF 2 discs (control) or was photo-irradiated for 30 s with a light cure unit (LC) (DP-075, Morita, Tokyo, Japan). The intensity of the halogen lamp was assessed (470 nm and >500 mW/cm 2 ) with a photoelectric sensor before testing and its intensity was maintained throughout the study. In the MDP-pretreated samples, CaF 2 discs were pretreated with MDP monomer, and then resin-based luting cement was pressed between a pretreated and an untreated disc (MDP-pretreated). Pretreatment was performed according to manufacturer’s instructions. For the mixed samples, the same volume of MDP monomer and resin-based luting cement were mixed and then immediately placed between CaF 2 discs (MDP-mixed). CaF 2 discs were used because of their IR translucency and their distinctive mechanical properties compared with the samples.
|Control||Mixed resin-based luting cement (Rely-X TM ARC, 3M ESPE).|
|LC||Mixed resin-based luting cement was photo-irradiated for 30 s with a light cure unit.|
|MDP-pretreated||Mixed resin-based luting cement was placed on a CaF 2 disc pretreated with MDP monomer (Epricode, Kuraray).|
|MDP-mixed||MDP monomer and resin-based luting cement were mixed and then immediately placed on a CaF 2 disc.|
The samples were subjected to a Fourier Transform Infrared (FTIR) analyzer (FT/IR-660, JASCO, Tokyo, Japan). Conformational changes of each composite over time were monitored for 24 h. At a resolution of 4 cm −1 , we performed 200 iterations within the range from 400 to 4000 cm −1 to characterize the various functional groups. The degree of conversion was measured by the intensities of the C C peak at 1638 cm −1 and the C- – -C reference peak at 1608 cm −1 , using a standard baseline technique . The peak at 1608 cm −1 originated from aromatic rings, whose intensity remains unchanged during polymerization. Therefore, the ratio of the absorbance intensities of C C/C- – -C reveals the polymerization ratio of the samples using following formula:
Degree of conversion ( % ) = 100 − cured peak intensities of C = C/C – – – C uncured peak intensities of C = C/C – – – C × 100
Phase detection by scanning probe microscope
One CaF 2 disc was removed to expose a bare LC sample surface, which was subjected to phase detection. Scanning probe microscopy (SPM) (SPM-9700; Shimadzu, Kyoto, Japan) was performed in the phase mode using rectangular silicon cantilevers with a spring constant of ∼40 Nm −1 and typical resonance frequencies between 250 and 300 kHz. Imaging was accomplished in the attractive tip-sample interaction regime, recording height and phase images, which indicate the distribution of mechanical properties of the sample surface .
The bare sample surfaces were subjected to nanomechanical testing using a quantitative nanomechanical test instrument (TS70 TriboScope; Hysitron, Inc., MN, USA) interfaced with a scanning probe microscope (SPM-9700; Shimadzu) with a diamond Berkovich indenter probe (Hysitron). Fused quartz acted as the calibration material to determine the indenter tip area function and the machine compliance . To minimize errors arising from surface roughness, appropriately smooth regions were chosen with a scanning range of 50 μm × 50 μm and then 1 μm × 1 μm, so that the surface roughness was less than 10% of the minimum measurement range of the indenter penetration depth , assuming a perfect relationship between contact depth and elastic deformation. The distance between indentations was more than 10 μm to avoid any influence of residual stresses from adjacent indentations. A constant thermal drift was monitored with 2 μN of pre-load for 40 s before indentation. The second 20 s of drift rate was subtracted from the overall contact depth.
Indentation tests were performed within the estimated oxygen inhibition region at the center of the sample surface so that superior mechanical properties associated with the degree of conversion could be expected. The tests were performed perpendicular to the selected regions using a loading/partial unloading technique with a load function comprising a total of 33 loading and unloading portions with hold time. Segment time was 1 s. Typical nanoindentation method is based on the assumption of isotropic elastic–plastic materials . However, synthetic polymers often exhibit viscoelastic or time-dependent behavior. The most readily observed effect of viscoelasticity on indentation is creep under constant load. When unloading follows loading without a hold time at peak load, displacement increases slightly in the initial portion of the unloading phase, because the creep rate of the materials is initially higher than the imposed unloading rate . This phenomenon results in a negative, changing slope in the initial unloading phase, making accurate assessment of the elastic modulus impossible. To eliminate this negative influence, each loading portion was followed by a hold time, so that the unloading portion was assumed to be purely elastic rather than viscoelastic. Moreover, an incremental loading rate was applied to a maximum loading force of 200 μN during a single test. The elastic moduli of polymers may vary with different loading rates because viscoelasticity causes a varied unloading slope. A larger number of loading/unloading portions enables multiple loading rates resulting in the final loading force. Therefore, this technique also evaluates the load-dependent behavior of resin-based luting cements by performing multiple loading rates followed by partial unloading during a single test .
The hardness and elastic moduli at indenter contact depths between 10 and 100 nm were noted, based on knowledge of the effective range of the Berkovich tip area function. An applied closed-loop (load) control algorithm was used throughout the tests, unless the dynamic indentations were superimposed in a load function. For the closed-loop control, the indenter tip was first withdrawn from the surface to a set distance (lift height) and then re-captured the sample surface, because the 2 μN pre-load during thermal drift monitoring may influence the ultra-low loading portion of this partial unloading test. The elastic moduli were calculated from force–displacement curves using the standard unloading analysis within the TS-70 proprietary software (Hysitron). The ideal curves fit to each displacement curve were corrected manually. The effective measurement range and constant elastic moduli were confirmed prior to each nanoindentation experiment.
Dynamic nanoindentation test
Force–displacement curves with a loading/partial unloading portion were recorded using 50 μN/s to a maximum load of 200 μN. The initial loading portion was followed by a 10 s holding time and partial unloading to 100 μN. The second loading portion regained the maximum load of 200 μN to enable nearly pure elastic deformation. The second loading portion was followed by a 20 s holding time. Oscillations and five sinusoidal indentations were superimposed after the 20 s holding time on the second force–displacement curve of the loading/partial unloading tests ( Fig. 1 ). The applied amplitudes were set at 20 μN with frequencies at 0 (creep), 1, 2, and 4 Hz. The initial oscillations mimicked the subsequent sinusoidal indentation profiles with appropriate amplitudes and instantaneous holdings, so that material amplitudes by dynamic indentations became constant prior to the sinusoidal frequency. The above load functions were determined in consideration of effective area functions and measurement range, according to suggestions of depth-dependent loading/partial unloading tests.