Abstract
Objectives
Tetrahydrofurfuryl-methacrylate (THFM) and hydroxypropyl-methacrylate (HPM) were used to partially or fully replace HEMA in experimental RMGICs. The experimental materials were compared with home and commercial products in terms of degree of conversion, polymerization shrinkage and exotherm.
Methods
Two commercial RMGICs used were Fuji-Plus (FP, GC, Japan) and RelyX-Luting (RX, 3M-ESPE, USA). Two additional in-house liquids were prepared based on the commercial materials liquids. Eight experimental liquid compositions (F1, F2, F3 and F4 based on Fuji-Plus; R1, R2, R3 and R4 based on RelyX-Luting) were prepared replacing 100% HEMA with HPM in F1 and R1 or 70%/30% HPM/THFM in F2 and R2. 50% HEMA was replaced with THFM in F3 and R3 compared to 30% in F4 and R4. All liquids were mixed with the corresponding commercial powder. Degree of conversion was determined using Fourier-transform infrared spectroscopy (FT-IR). Polymerization shrinkage and exotherm were measured using the bonded-disk method.
Results
Compositions containing HPM (F1 and R1) showed lower degree of conversion compared to experimental compositions containing THFM, home and commercial materials ( p < 0.0001). FP-commercial showed significantly higher polymerization shrinkage and exotherm compared to all other materials in both groups ( p < 0.0001). FP-commercial showed higher degree of polymerization shrinkage and exotherm at 5 min compared to all materials due to the incorporation of an additional cross-linker (glycerol-dimethacrylate). In general, compositions containing THFM, presented polymerization shrinkage and degree of conversion values similar to their corresponding commercial products.
Significance
THFM monomer showed promising results and could be potentially useful in the development of new RMGICs with improved properties.
1
Introduction
Dimensional stability of dental cements is an important feature that can affect the longevity of the restoration. Shrinkage that occurs following application of the material can lead to marginal gaps and leakage, which could lead to secondary caries, and contribute to the failure of the restoration . Shrinkage of resin-based dental materials is directly associated with the percentage degree of conversion of the carbon-carbon double bonds to single bonds following polymerization . The occurrence of polymerization shrinkage is a result of the change in the molecular bonds (from van-der Waals forces to covalent single bonds) and in the distances between atoms . This reaction is also exothermic and, in the case of light activated products, there can be extra heat generated from the light curing device .
Resin modified glass ionomer cements (RMGICs) set by both an acid–base reaction, similar to the conventional glass ionomer cements (GICs), and polymerization of the monomer (commonly used is hydroxyethyl methacrylate; HEMA). GICs do not undergo polymerization shrinkage and have a ‘very low to non-existent reaction exotherm’ . However, due to the fact that RMGICs contain HEMA and undergo polymerization (light and/or chemically), a degree of polymerization shrinkage and a rise in temperature on setting, are commonly associated with this type of cement . Although it is favorable to increase the degree of conversion (polymerization) of dental materials in order to enhance their mechanical and biological properties , a high degree of conversion is less desirable due to the possibility of increasing polymerization shrinkage, exotherm and brittleness of the material .
The larger the molar volume of the monomer, the fewer monomer units are required to be converted for the same volume of material . Therefore, in theory, in order to achieve RMGICs with a higher degree of conversion, but with clinically acceptable polymerization shrinkage, monomers with higher molecular weights and molar volume could be used . Tetrahydrofurfuryl methacrylate (THFM) has been studied for use in RMGICs, where it partially replaced HEMA and this resulted in improved properties (water uptake and dimensional change) . Another monomer that could be used to replace HEMA is hydroxyl propyl methacrylate (HPM), which is the next member in the homologous series, just below HEMA, which contains an extra CH 2 group between the hydroxyl and methacrylate groups. It has a lower setting exotherm and reduced water uptake compared to HEMA . Both HPM and THFM have higher molecular weights and higher molar volumes compared to HEMA. Thus it can be assumed that they may demonstrate lower shrinkage and higher degree of conversion values .
The polymerization process can be characterized by Fourier Transform Infra-Red Spectroscopy (FT-IR) as this technique is based on the absorption of radiation of the functional groups in the polymer chain, and their molecular vibration in the infrared frequency range . Moreover, this technique (FT-IR) can be used to calculate the degree of conversion of resin composites by measuring the ratio of absorbance intensities of the aliphatic carbon carbon double bond (C C), at ∼1638 cm −1 , and aromatic double bond (C C) at ∼1609 cm −1 . However, a study by Young et al. (2002) showed that the peak at 1638 cm −1 could not be used to quantify the degree of conversion in RMGICs since the formulation contains water, which absorbs at this wavenumber, thus making it difficult to determine the peak base at this point. Therefore, a different peak was used (at 1320 cm −1 ) and the degree of conversion results obtained using this peak was in agreement with those obtained from Raman studies .
The polymerization shrinkage strain can be measured using the bonded-disk method, which was first mentioned by Walls et al. and Bausch et al. and further developed by Watts and Cash . This technique was subsequently improved and modified by Watts and Hindi . The shrinkage strain of the material is measured using the bonded-disk method after applying the mixed material onto a rigid glass plate (3 mm thickness), which consequently bonds to one surface of the glass maintaining a constant dimension of the circumference at 8 mm diameter . Watts and Cash demonstrated that no change occurred in the circumference of the tested material as a result of bonding to the glass plate; hence shrinkage occurred only in the vertical plane, in the unbound surface . Therefore, the measurement of the strain ( ε ) in the axial plane corresponds to a close approximation to the volumetric strain of the material. Another advantage of this technique is that the polymerization exotherm can be measured simultaneously, since a thermocouple can be inserted in the material during the measurement of shrinkage.
Therefore, the objectives of this study were to:
- –
Analyse the composition and setting reactions of commercial and experimental RMGICs using FT-IR, and moreover to measure and compare the degree of monomer(s) conversion by comparing the absorbance peak immediately and 30 min after mixing.
- –
Measure and compare the percentage polymerization shrinkage and exotherm simultaneously of commercial and experimental RMGICs in order to determine the effect of replacing HEMA, fully or partially, with higher molar volume monomers.
2
Materials and methods
2.1
Materials
Two commercial chemically cured resin modified glass-ionomer cements (RMGICs) were included in this study, namely Fuji Plus (FP, GC Corporation, Tokyo, Japan) and RelyX Luting (RX, 3M ESPE, St Paul, MN, USA). Two additional control liquids were prepared in-house (FP home and RX home), based on the corresponding commercial liquid composition, and these were to be mixed with the corresponding commercial powders.
Eight experimental liquid formulations were prepared for this study, four for each of the commercial products (RX, FP), where HEMA was replaced with either 100% HPM (F1 and R1), 70%/30% HPM/THFM (F2 and R2), 50%/50% THFM/HEMA (F3 and R3) and 30%/70% THFM/HEMA (F4 and R4). All experimental and home liquids were mixed with the corresponding commercial powder.
3
Methods
3.1
FT-IR analysis technique
All commercial, home and experimental materials (unset liquids and set materials) were tested using FT-IR spectrometer (Perkin Elmer series 880) supplied with a Golden Gate Single Reflection Diamond attenuated total reflectance (ATR) accessory (Graseby Specac Ltd, Kent, UK). A total of 144 FT-IR spectra were obtained divided as four spectra taken at four time points (at 0, 5, 10 and 30 min), from the start of mixing for all commercial, home and experimental cements ( n = 3 per material). Furthermore, the corresponding FT-IR spectra were compared and analysed with spectra obtained from known monomers ( n = 3).
Spectra of absorbance versus wavenumber were obtained at a resolution of 4 cm −1 , and 10 scans in the range of 4000–500 cm −1 , were used to determine the absorbance peaks of the materials before and after curing.
For identification of degree of polymerization, materials were mixed at room temperature (23 ± 1 °C) according to the manufacturers recommended powder: liquid mixing ratio (2:1 g:g for FP group and 1.6:1 g:g for RX group). In order to standardize the amount of RMGIC investigated, a wax ring of 8 mm diameter and 1.5 mm thickness was prepared using a profile wax stick (Berg Dental, Engen, Germany) and placed on the ATR accessory with the ATR crystal in the middle of the wax ring. The mixture was then placed inside the wax ring; the top surface was covered with an acetate sheet and a scan was run as zero time. After 10 min from the start of mixing, pressure was applied to the top surface of the sample using the compression head anvil in the ATR, in order to maintain good contact with the crystal. Subsequent spectra were generated at 0, 5, 15 and 30 min from the start of mixing thereafter. The analysis was repeated 3 times for all materials at each time point.
3.2
Treatment of FT-IR data
The peak height at 1320 cm −1 was used to quantify the degree of conversion (DC) of the monomer by measuring the net peak height difference between the absorption from 0 time ( A 0 ) to 30 min later ( A t ). Eq. (1) was used to calculate the degree of conversion.
DC(%)=(A0−At)/A0×100.
Degree of conversion values calculated at each time point (5, 10 and 30 min) were compared between all materials for a significant difference using one-way ANOVA followed by post-hoc Tukey test at significance level of p = 0.05.
3.3
Polymerization shrinkage strain and exotherm
Polymerization shrinkage strain was measured using the bonded-disk instrument, which was calibrated prior to starting the shrinkage experiment. The experiment was conducted in a temperature-controlled instrument at both 23 ± 1 °C and at 37 ± 1 °C, the latter to mimic the oral environment temperature. The instrument was switched on and left to equilibrate in temperature for 2 h prior to starting the testing.
The instrument consists of a linear variable displacement transducer (LVDT – sensitivity greater than 0.1 μm), fixed to an aluminium stand above a brass anvil (40 mm diameter). The experiment assembly (16 ± 0.1 mm diameter and 1 ± 0.1 mm thickness square cross sectioned), comprised a brass ring bonded centrally on a glass slide (74 mm length, 25 mm width and 3 mm thickness. The glass slide and brass ring was covered with a square microscope cover slip (22 × 22 mm area and 0.13 mm thickness; Number.0, VWR International, Radnor, PA, USA). The whole assembly was placed centrally on a temperature controlled metal plate and specimen platform, so that the LVDT was lightly in contact with the centre of the flexible cover slip. The movement of the cover slip was detected by the LVDT. The latter was connected to a signal-controlling unit (E309, RDP Electronics, Wolverhampton, UK) that transferred the data to a computer through a data logging system hardware and software (ADC-20 multichannel unit and Pico-log software, Picotech, Cambridge, UK).
In order to measure the polymerization exotherm simultaneously, a thermocouple was passed through a carefully drilled groove in the centre of the glass slide reaching the centre of the brass ring. The end of the thermocouple was placed centrally in the test material. This thermocouple was attached to a type-K thermocouple amplifier (TCK-4, Audon Electronics, Nottingham, UK) which transferred the exotherm data to the ADC-20 unit.
3.4
Treatment of shrinkage strain data
A total of 60 samples divided into 12 groups ( n = 5) of different compositions (commercial, home and experimental) were investigated for polymerization shrinkage strain and exotherm using bonded-disk technique.
The final specimen thickness ( L ) was measured using a digital micrometre accurate to 10 μm. The calibration coefficient of displacement/voltage was used to calculate the displacement of the cover slip during the experiment ( dL ), which was then used to determine the original specimen thickness ( L 0 ) in μm, following Eq. (2) .
L0=dL+L.
The percentage shrinkage strain was calculated using Eq. (3) . These values were then plotted against time (s) to originate the shrinkage strain kinetic curve from the start to the end of the experiment.
3.5
Treatment of exotherm data
The maximum temperature reached was recorded and compared for significant differences between the materials using one-way ANOVA followed by post-hoc Tukey test at a significant level of p = 0.05. Moreover, the time to reach the peak temperature was also recorded as the period from the start of mixing to the time required to reach the maximum temperature.
4
Results
4.1
Composition analysis
Table 1 and Fig. 1 summarize FT-IR wavenumbers of PAA, water and all monomers used in the commercial, experimental and home liquids. The only difference between the spectra of RX and FP liquids was the additional peaks at 1059 cm −1 (C N stretch), 1250 and 1535 cm −1 (N H bond), representing UDMA in FP liquid compositions (∼4%). Peaks at 1631 cm −1 and at 1701 cm −1 represent C C and C O group respectively, and peaks at 1301 cm −1 and 1324 cm −1 refer to the C O stretch in monomers . Tartaric acid (peak at 1086 cm −1 ) was present at 5–10%, according to the manufacturer’s material’s safety data sheet (MSDS) in commercial materials, and at 5% in the home and experimental liquids. HPM showed an additional peak at 1059 cm −1 corresponding to the C OH group in its structure, while HEMA and THFM showed additional peaks at ∼1020 and 1086 cm −1 , corresponding to the C O H group in HEMA’s structure and the ring stretch in THFM .
| FTIR wavenumber (cm −1 ) | Assignment | Component |
|---|---|---|
| 1020 | OH Ring stretch |
HEMA THFM |
| 1059 | OH C N |
HPM UDMA |
| 1086 | OH Ring stretch |
HEMA THFM |
| 1250 | N H | UDMA |
| 1300 | C O | Monomer (all three) |
| 1320 | C O | Monomer (all three) |
| 1379 | C H | PAA and monomer (all three) |
| 1405 | C H | PAA and monomer (all three) |
| 1450 | C H | PAA and monomer (all three) |
| 1535 | N H | UDMA |
| 1630 | C C | Monomer (all three) |
| 1630 | O H | Water |
| 1700 | C O | Monomer (all three) |
Fig. 1 c and d show the FTIR spectra of experimental liquid compositions in the FP and RX groups respectively (between wavenumbers of 800 cm −1 and 1800 cm −1 ). Compositions of cements containing HPM, from both groups (F1, F2, R1 and R2), showed an additional peak at ∼1059 cm −1 representing the C OH stretch. Moreover, peaks at ∼1084 cm −1 were sharper in compositions containing THFM and HEMA (F3, F4, R3, R4).
4.2
Analysis of reaction
According to the literature (discussed in the introduction), regarding the measurement of degree of conversion concerning RMGIC, the peak at 1320 cm −1 was used instead of 1638 cm −1 . The degree of conversion results obtained using this peak, were in agreement with those obtained from Raman studies .
Fig. 2 a and b is representative FT-IR spectra taken during the setting of Fuji Plus commercial and home respectively, as soon as the cement was applied to the ATR crystal (0 min) and at 5, 10 and 30 min, from the start of mixing. As can be seen in Fig. 2 a, changes in absorbance occurred in FP (commercial) in the 5 min’ spectrum when compared to that at 0 min. These changes included loss of C C at 1630 cm −1 , 1320 and 1300 cm −1 and a shift of the latter two wavenumbers to 1275 and 1253 cm −1 . Small changes also occurred at 1700 and 975 cm −1 representing C O and glass Si O stretch respectively . These changes also occurred in FP Home ( Fig. 2 b) but at later time points, thus indicating a slower reaction.
