Methacrylamide-Methacrylate Hybrid Monomers for Dental Applications

Abstract

Objectives –

The susceptibility of methacrylates to hydrolytic and enzymatic degradation may be a contributing factor limiting the clinical lifespan of resin composite restorations. The elimination of labile ester bonds is a potential advantage of methacrylamides, which have been shown to produce more stable restorative interfaces. The rationale of this study is to design hydrolytically and enzymatically stable adhesive monomers, with the added benefit of being able to form crosslinked networks. The objective of this study was to synthesize difunctional, hybrid methacrylate-methacrylamide monomers, and evaluate them as potential monomers for dental adhesives.

Materials and Methods –

HEMA, TEGDMA (controls) or secondary methacrylamides (HEMAM – commercially available, 2EM and 2dMM – newly synthesized) either bearing a hydroxyl group or a methacrylate functionality (Hybrids-Hy), were added at 40 mass% to bisGMA. The photoinitiator system consisted of 2-dimethoxyphenyl acetophenone (DMPA) and diphenyl iodonium hexafluorophosphate (DPI-PF6) at 0.2 and 0.4 mass%, respectively. Polymerization kinetics were followed in real-time by near-IR spectroscopy during light activation at 630 mW/cm² for 300 s. Water sorption and solubility (WS, SL) were measured according to ISO 4049. Storage modulus in shear (G’) for 300s was obtained by oscillatory rheometry. For the microtensile bond strength (μTBS), fully formulated adhesives containing 40 vol% ethanol were used to restore caries-free human third molars. Bonded specimens with 1 mm² cross-sectional area were tested after 48 h and 6 months storage in water at 37 °C. Single bond (SB) was tested as a commercial control. Data were analysed with one-way ANOVA and Tukey’s Test and Student’s t-test (α=0.05).

Results –

In general, hybrid versions showed lower polymerization rate and degree of conversion, whereas the methacrylate controls, HEMA and TEGDMA, showed the highest values. The hybrid versions showed lower values of WS and SL than their monofunctional versions. HEMAM Hy showed the highest values of G’ and TEGDMA, 2EM, and 2dMM-Hy the lowest. The μTBS values between 48 h and 6 months were statistically reduced only for the HEMA and both 2dMM materials. The formulation containing the monofunctional methacrylamide (HEMAM) showed only 9% reduction in μTBS after 6 months of aging, while the other groups showed a decrease ranging between 18 and 33%.

Conclusion –

Overall, hybrid monomers showed lower reactivity than their analogous monofunctional versions, but had markedly lower water sorption. Shear storage modulus was affected differently by the addition of the second functionality. HEMAM-containing systems were able to maintain stable long-term dentin bond strength, which demonstrates that bonding stability is a result of the complex interplay among the factors studied.

Clinical significance:

The novel monomers showed here are potential alternatives to the current methacrylate adhesives, with selected formulations presenting greater bond stability.

1. Introduction

Since resin-based restorative materials were developed over 60 years ago, remarkable improvements have been made in the filler particle systems, photoinitiator effectiveness, and light curing devices [1, 2]. These advances have made adhesive restorative materials the most popular option for direct esthetic dental restorations. However, the limited clinical durability of resin-based materials remains as a challenge, and annually results in costly replacements of dental restorations [3]. The premature degradation of these materials has been ascribed to ester-containing monomer breakdown [4, 5]. Despite the high reactivity and satisfactory mechanical properties of methacrylate-based materials, the ester bonds in these monomers are prone to hydrolysis and enzymatic attack in the hostile oral environment, leading to polymer degradation [6–8]. More recently, it has been demonstrated that esterases produced by microorganisms compound the hydrolytically-mediated methacrylate monomer degradation [7, 9]. In addition, hydrophilic methacrylate monomers, such as HEMA, are more susceptible to water sorption, which further increases the likelihood for mechanical failure [10].

A suitable alternative to overcome this problem relies on the incorporation of ester-free monomers in the formulation of the organic matrix. Methacrylamides have been shown to be promising options for dental adhesive formulations [11, 12]. The absence of labile ester bonds makes them highly resistant to enzymatic and hydrolytic degradation, which enables the potential for maintaining high dentin bond strengths even after long term aging [13]. Our previous systematic evaluation in which acrylamides and methacrylamides were compared as diluent monomers in bisGMA and UDMA-based dental adhesive formulations indicated that, while some secondary methacrylamides showed minor or no reduction in dentin microtensile bond strength after 6-month storage in water, the fully methacrylate-based compounds showed reductions of up to 30% [11]. However, this work also highlighted two major concerns associated with the amide functionality – lower reactivity and higher water absorption. The lone pair of electrons in the nitrogen atom is responsible for resonance stabilization with the carbonyl, which is translated into higher stability of the amide bond in comparison with an ester bond on a methacrylate. This is an advantage from the standpoint of hydrolysis, but also leads to decreased reactivity of the nearby vinyl group [14]. In addition, the amides are highly susceptible to establishing strong hydrogen bonds with water molecules and becoming hydrated [15]. This is due to the fact that amides are hydrogen-bond acceptors via the oxygen atom in the carbonyl, and hydrogen-bond donors via the –NH moiety [15].

Considering that any monomer intended for dental applications would ideally present the reactivity and water absorption resistance of the methacrylates, and the degradation resistance of the methacrylamides, the aim of the present study was to use our previous work with mono-functional methacrylamides [11] as a platform to design, synthesize and evaluate crosslinkable difunctional methacrylamide-methacrylate hybrid moleculesas diluent monomers for HEMA-free dental adhesive formulations. The rationale of the present study relies on the hypothesis that the incorporation of the methacrylate functionality on the methacrylamide chemical structure will increase the reactivity and the final degree of conversion, and the enhanced crosslinking provided by the di-functional species will make the network more resistant to water absorption, ultimately mechanically reinforcing the interfacial bond.

2. Materials and Methods

2.1. Tested monomers and synthesis procedures

The tested monomers are shown in Figure 1. All commercially-available monomers were purchased from Sigma-Aldrich (Milwaukee, WI, USA) at 97% purity and used as received. The chemical structure of the secondary methacrylamide N-hydroxyethyl methacrylamide (HEMAM) was modified with ethyl and methyl substituents on the first (alpha) carbon (2EM and 2dMM, respectively), as described previously [11]. The hybrid versions of these monomers were isolated via chromatography, as described in the supporting information. NMR and IR spectral data can also be found in the supporting information.

Figure 1.

Chemical structure and acronyms of the tested monomers copolymerized at 40 mass% with bisGMA. TEGDMA and HEMA were tested as difunctional and monofunctional methacrylate controls, respectively.

N-hydroxyethyl methacrylate (HEMA) was tested as monofunctional methacrylate control. Triethyleneglycol dimethacrylate (TEGDMA) was tested as difunctional methacrylate control to provide a comparison with the difunctional methacrylamide-methacrylate hybrid monomers. The partition coefficient (log P) for each monomer was calculated using the software package Chem Draw Ultra 14.1 (Perkin Elmer, San Jose, CA, USA).

2.2. Tested formulations and photocuring conditions

The monomers shown in Figure 1 were mixed at 40 mass% with bisphenol A-glycidyl methacrylate (bisGMA). The photoinitiator system consisted of DMPA (2,2-dimethoxy-2-phenylacetophenone, λ_max=365 nm) and DPI-PF6 (diphenyliodonium hexafluorophosphate) at 0.2 mass% and 0.4 mass%, respectively. Butylated hydroxytoluene (BHT) was added at 0.1 mass% to each formulation as a free-radical inhibitor. For the dentin microtensile bond strength test only, experiments were conducted with fully formulated adhesives containing 40 vol% of ethanol to provide appropriate viscosity for proper dentin penetration and subsequent volatilization by air drying. All photocuring procedures were accomplished by a mercury arc lamp (Acticure 4000 UV Cure, Mississauga, Canada) filtered to 320–500 nm at 630 mW/cm² measured directly at the sample surface using a thermopile power meter (Molectron PM100, Portland, OR, USA). The choice of light source was intended to match the initiator system used.

2.3. Kinetics of polymerization

The kinetics of polymerization were assessed in near-IR spectroscopy (Nicolet 6700, Thermo Scientific, USA) in real time during the photopolymerization of disc-shaped samples (10 mm in diameter and 0.8 mm in thickness; measured with a digital caliper to 0.01 mm) for 300 seconds (n = 3). Each spectrum was collected with 2 scans at 4 cm⁻¹ resolution. This resolution allowed for baseline correction without compromising the sampling rate and signal-to-noise ratio. The final carbon-carbon double bond conversion (final DC) was calculate based on the areas of the peaks (obtained with the processing tool in the OMNIC software) at 6165 and 6135 cm⁻¹, which correspond to the vinyl overtone for methacrylates and methacrylamides, respectively. The maximum rate of polymerization (RP_MAX), representing the reactivity of the monomers, was determined as the first derivative of the degree of conversion as a function of the time. The degree of conversion at the maximum rate of polymerization (DC at RP_MAX) was used to estimate the time point in conversion at which diffusional limitations lead to deceleration.

2.4. Water sorption and solubility

Water sorption (WS) and solubility (SL) were measured according to the ISO 4049:2019. Briefly, the same samples obtained in the polymerization kinetics test (n=3), after having their initial mass M1 determined, were immersed in 5 mL of triple distilled water for 7 days. At the end of this period, M2 was measured and the samples were stored in a desiccator containing silica gel and connected to the house vacuum. Sample weights were measured daily until the final mass did not change to the nearest 0.0001 g (M3). WS and SL were calculated in μg/mm³ according the following equations, where V is the volume of the disc in mm³:

$$WS=\frac{M2-M3}{V}$$

$$SL=\frac{M1-M3}{V}$$

2.5. Storage modulus in shear

The storage modulus in shear (G’, n=5) was assessed in an oscillatory rheometer (Discovery HR-1 Hybrid Rheometer, TA Instruments, New Castle, DE, USA), using an 8-mm diameter aluminum plate attached to the upper fixture and an acrylic plate mounted to the UVVis accessory on the bottom. Approximately 0.02 g of each material (the exact mass was recorded for each specimen and used to calculate G’) was placed between the parallel plates, and the light was delivered through the acrylic via the optical apparatus in the UV-Vis accessory. Samples were tested in oscillation mode (sine wave) at 10 Hz and 0.1% strain with a gap of 0.3 mm during the photopolymerization for 300 s (n = 3).

2.6. Dentin microtensile bond strength

Selected formulations with the highest G’ and lowest WS and SL were subjected to dentin microtensile bond strength testing (μTBS). Sound human dentin of extracted caries-free third molars was used as the substrate. The study was approved by the Oregon Health & Science University IRB (IRB00012056). The enamel was removed and the resulting surface was roughened by hand with light pressure and one pass across wet #600 silicon carbide paper to simulate smear layer formation. The dentin surface was etched for 15 s with 37% phosphoric acid (3M ESPE), rinsed and dried with the aid of gentle air stream for about 10 s. Two layers of the adhesive were applied and, after solvent evaporation, the second layer was photocured for 60 s at 630 mW/cm² by the mercury arc lamp. Restorative procedures consisted of a block of Filtek Supreme (shade A2 – 3M ESPE) built in 2 increments of 2 mm each, photoactived with the light guide directly over the surface for 20 s at 1200 mW/cm² with an Elipar™ DeepCure-S LED (3M ESPE). Adper Single Bond (3M ESPE) was tested as a commercial adhesive control, in two consecutive layers, air-dried to remove excess solvent, and photoactivated for 20 s using the same light curing unit settings (n=6).

24 hours after the restorative procedures, teeth were sectioned under water in a slow speed diamond saw (Accutom-50, Struers) to obtain sticks of 1 mm² cross-sectional area (checked with a digital caliper to 0.01 mm). The sticks were tested after 24 h or 6 months water storage at 37 °C. Sticks were glued with cyanoacrylate (Zap-it, Dental Ventures of America, Corona, CA, USA) onto custom-made jigs (Odeme Equipment, Luzerna, SC, Brazil – pictured in Figure SXXX of the supplemental information) attached to a universal testing machine (Criterion MTS, Eden Prairie, MN, USA) and tested in tension until failure (0.5 mm/min).

2.7. Statistical Analysis

Data was statistically analysed by one-way ANOVA and Tukey’s test (α = 0.05), after assessment of normality and homoscedasticity. For μTBS, Student’s t-test was carried out to compare the effect of the storage time (α = 0.05). In the instances where the normality tests failed, the nonparametric Kruskal-Wallis test was carried out (α = 0.05).

3. Results

Kinetics of polymerization curves (average of three curves) are depicted in Figure 2 and results shown in Table 1. RP_MAX ranged between 0.11 and 0.03 %^.s⁻¹, with TEGDMA and HEMAM Hy showing the highest and lowest values, respectively. The other groups were statistically similar (Table 1). A similar trend was found for the DC at RP_MAX results, which ranged between 21.0 and 8.7%, with the methacrylates TEGDMA and HEMA showing the highest values and HEMAM Hy the lowest. In terms of final DC, the monofunctional HEMA and HEMAM showed the highest values (89.0 and 83.2%, respectively) and the hybrid versions HEMAM Hy, 2EM Hy and 2dMM Hy the lowest (63.5, 63.3, and 59.4%, respectively). In general, the alpha-substituted methacrylamides 2EM and 2dMM showed lower values than the monofunctional methacrylate control HEMA (73.6, 76.7 and 89.0%, respectively).

Figure 2.

Degree of conversion (%) as a function of time (%) and rate of polymerization (%^.s-1) as a function of degree of conversion curves for all tested diluent monomers copolymerized at 40 mass% with bisGMA. The polymerization reaction was followed in real-time during polymerization for 300s at 630 mW/cm² by near-IR spectroscopy.

Table 1.

Average (standard deviation) of maximum rate of polymerization (RP_MAX, %.s⁻¹), degree of conversion at maximum rate of polymerization (DC at RP_MAX, %), and final degree of conversion (Final DC) for all tested compositions. Values followed by different letters within the same column indicate significant differences (p<0.05).

Co-monomer	Polymerization Kinetics
	RP MAX(% s−1)	DC at RP MAX (%)	Final DC (%)
TEGDMA	0.11 (0.02) A	19.6 (5.3) A	78.0 (2.7) BC
HEMA	0.07 (0.01) AB	21.0 (3.5) A	89.0 (1.9) A
HEMAM	0.05 (0.006) B	13.0 (1.5) AB	83.2 (3.2) AB
HEMAM Hy	0.03 (0.005) C	8.7 (1.9) B	63.5 (0.8) D
2EM	0.09 (0.01) AB	16.8 (2.3) AB	73.6 (1.4) C
2EM Hy	0.08 (0.02) AB	14.7 (3.8) AB	63.3 (2.0) D
2dMM	0.09 (0.02) AB	15.7 (3.4) AB	76.7 (2.2) C
2dMM Hy	0.07 (0.01) AB	12.5 (2.5) AB	59.4 (1.2) D
p value	<0.001	0.007	<0.001

Regarding water sorption and solubility (Figure 3), the WS values ranged between 33.4±3.2 and 183.0±5.7 μg/mm³ with the highest value being for the methacrylamide HEMAM, followed by 2EM, HEMA and 2dMM (101.3±1.5, 93.9±4.8, and 79.1±0.9 μg/mm³, respectively). TEGDMA and the hybrids were similar (35.5±1.8, 38.7±1.8, 44.0±0.8, and 33.4±3.2 μg/mm³, respectively). In terms of SL, the results ranged between −12.7±1.6 and 6.4±2.2 μg/mm³ for HEMA and 2EM/2dMM, respectively. The hybrids HEMAM Hy, 2EM Hy and 2dMM Hy were statistically similar to TEGDMA (−1.6±0.0, 0.0±0.0, −0.5±2.4, and −4.2±6.0 μg/mm³, respectively).

Figure 3.

Water sorption (WS, μg/mm³) and solubility (SL, μg/mm³) for all tested formulations after 7-day water storage. Different letters indicate significant difference among the groups (p<0.001 for WS and SL).

The shear storage modulus, G’, values ranged between 160.7±8.0 and 115.7±7.0 MPa for the hybrids HEMAM Hy and 2dMM Hy, respectively (Figure 4). In general, the groups were statistically similar and significant difference was only observed between HEMAM Hy versus TEGDMA, 2EM and 2dMM Hy.

Figure 4.

Shear Storage Modulus (G’, MPa) of the polymers tested in oscillatory mode at a frequency of 10 Hz with 0.1% strain, recorded after 5 minutes photoactivation at 630 mW/cm². Different letters indicate significant differences among the groups (p = 0.024).

Dentin μTBS results are shown in Figure 5. Single Bond showed statistically higher values at both 48 h and 6 months (53.4±9.8 and 43.3±5.3 MPa, respectively), while all other groups were statistically similar to each other (ranging between 42.3±9.6 and 27.9±6.0 MPa and between 32.7±3.3 and 19.2±4.5 MPa for 48h and 6 months storage time, respectively). The μTBS decreased for all materials between 48 h and 6 months, and this reduction was statistically significant for the HEMA and two 2dMM compounds. The formulation containing the monofunctional methacrylamide (HEMAM) showed the lowest bond strength reduction (about 9%) after 6 months of aging, while the other groups showed a decrease ranging between 18 and 33%.

Figure 5.

Dentin microtensile bond strength (μTBS, MPa) after 24h (blue) and 6-month (fuchsia) water storage. Different uppercase letters indicate significant differences among the formulations (24h: p<0.001 and 6 months: p<0.0001), and lowercase letters indicate significant differences between the storage times within the same experimental formulation (Single Bond: p=0.0522, HEMA: p=0.0355, HEMAM: p=0.4507, HEMAM Hy: p=0.2915, 2dMM: p=0.050, and 2dMM Hy: p=0.0171).

4. Discussion

The limited clinical durability reported for current esthetic, direct dental resin composite materials highlights the need for the development of alternative monomers to replace the widely-used methacrylates which, despite the high reactivity and reasonable mechanical properties, are highly susceptible to hydrolytic and enzymatic degradation due to the presence of ester bonds [4, 6, 8]. Experimental dental adhesive formulations containing methacrylamides have shown significant long term dentin bonding stability, in spite of their lower reactivity, and of the high hydrophilicity that resulted in reduced values of certain mechanical properties [11, 12]. In an attempt to improve the reactivity of the amides and control their water sorption, in this study hybrid methacrylamide-methacrylate difunctional monomers were designed, synthesized and tested as alternative co-monomers for HEMA-free dental adhesive formulations. The results showed that except for HEMAM Hy, all hybrid versions showed reactivity (RP_MAX) similar to the methacrylate controls (TEGDMA and HEMA). HEMAM was expected to present the highest reactivity due to the absence of bulky substituents. The absence of substituents, in theory, would facilitate the access of the amine radicals to the vinyl groups. Albeit not statistically significant, the opposite was actually observed: the non-substituted HEMAM showed 45% lower RP_MAX than the alpha-substituted versions. Steric interactions of substituents near amide bonds have been shown to cause slight rotation about the amide C-N bond [16], reducing the ability of the nitrogen atom to donate electrons into the conjugated system. This distortion of the amide bond results in a longer amide C-N bond with less double bond character [16]. Compared with the non-substituted HEMAM, the distorted amides of the 2EM and 2dMM versions are not able to stabilize a radical as effectively, which could increase the rate of polymerization (additional information provided with Supplemental Figure 7). One additional explanation is based on the electron-donating nature of the alkyl chains, which may have created a partial negative charge on the alpha-carbon in 2EM and 2dMM [17]. Combined with the negative partial charge inherent in the amide bond, this might have led to a spatial separation from the electron-rich vinyl group and, ultimately, exposed the double bond to free radical propagation. In short, the attachment of a second vinyl functionality to a sterically-hindered chemical structure made the resultant hybrid compound (HEMAM Hy) even less reactive.

In general, all hybrid versions showed numerically or statistically (or both) lower RP_MAX, DC at RP_MAX and Final DC than their OH-bearing versions. This was expected, since the reaction involves co-polymerizations between difunctional and monofunctional monomers, each with distinct individual reactivities. Expectations were that autoacceleration and autodeceleration would be impacted, ultimately leading to structural heterogeneity, unequal functional group reactivity and a delay in volumetric shrinkage rate [18]. The decrease in reactivity in systems containing high ratios of multifunctional molecules is related to polymer crosslinking, which impairs macro-radical diffusion in the reaction environment. Since the mobility of the reactive species is hindered, both the termination and propagation kinetic constants decrease, which explains the lower rates of polymerization. In addition, the propagation becomes diffusion-controlled earlier in the conversion in systems with higher ratios of difunctional monomers, as evidenced by the lower DC at RP_MAX results. At RP_MAX, diffusional limitations reach a threshold beyond which the reaction starts to decelerate, until the network completely vitrifies. DC at RP_MAX results demonstrated that HEMAM Hy showed the lowest conversion at that point, which indicates its network vitrified much sooner in conversion.

One additional factor to be considered is the unequal functional-group reactivity in difunctional monomers. It has been demonstrated that, on average, only one unit of double bonds reacts per monomer independent of the number of functionalities (from one to five - [18]. As one of the functional groups reacts and forms a covalent bond with a growing chain and/or another molecule, a congestion by the physical presence of surrounding ligands is created, which slows down or even prevents reaction at the second functional group [18]. In molecules with short and rigid carbon chains, such as the hybrids tested in the present study, this congestion is even stronger due to the close proximity of the functionalities and the hindrance to molecular stretching or rotation. The presence of aliphatic side chains in some of the molecules compound the steric hindrance effects. Furthermore, for the hybrid monomers, the process is further complicated by the inherent unequal reactivity between methacrylates and methacrylamides. The methacrylamides are markedly less reactive than the methacrylates due to the strong resonance stabilization of the vinyl group provided by the nitrogen atom [14]. In other words, despite having the same number of mesomeric structures, the amide functionality is more stabilized than the ester due to the fact that the nitrogen atom is less electronegative than the oxygen and, consequently, is a better donor of nonbonding electrons [19]. Therefore, it can be postulated that the more reactive methacrylate reacted first, further decreasing the reactivity of the already stable methacrylamide functionality. Even though this falls outside the scope of this study, one future strategy to balance this uneven reactivity between the methacrylate and methacrylamide could be to vary the extender chain length, which could be tailored to modulate molecular degree of freedom, and therefore, make the methacrylamide vinyls more readily available to react [20].

Interestingly, in the kinetics curves profiles of HEMA, HEMAM and HEMAM Hy, two distinct slopes are noticed (Figure 6), which may indicate the presence of two phases with different compositions as the polymerization reactions take place at different rates [21]. The decrease in viscosity promoted by the incorporation of HEMA, HEMAM and HEMAM Hy into the formulations may increase the mobility of the system, which may have caused the polymerization of the more reactive bisGMA to take place more or less independently, at a faster rate and with earlier vitrification compared with the other co-monomers [21]. On the other hand, the polymerization of the diluent-rich phase is hypothesized to have taken place at a slower rate, with delayed gelation and vitrification. It has been shown that during the copolymerization between methacrylates and methacrylamides, a radical is easily formed from a methacrylate molecule and it more likely reacts with a like monomer [18]. Conversely, the amide radical has been shown to more likely react with the more reactive methacrylate rather than with another lower reactivity methacrylamide molecule, thereby enriching the co-polymer with methacrylate units in comparison with the initial comonomer ratio [19]. In short, this differential reactivity may have led to the formation of interpenetrating polymer networks (IPNs) [22]. The lowest reactivities of HEMAM and especially HEMAM Hy are consistent with the possibility of IPN formation. It is interesting to note that the conversion at which the kinetic curve enters the deceleration phase shifts to earlier stages as the monomer reactivity decreases, which indicates a reduction in diffusivity of both polymer and monomer species [18]. Finally, the non-substituted HEMAM showed statistically higher final degree of conversion than the alpha-substituted versions 2dMM and 2EM. The increase in final double bond conversion showed by HEMAM may be associated with the relative lower viscosity of this compound, which likely played a role in preserving sufficient mobility in the system up to much higher levels in conversion [23].

Figure 6.

Rate of polymerization (%^.s-1) as a function of degree of conversion curves for HEMA, HEMAM and HEMAM Hy, co-polymerized at 40 mass% with bisGMA. The degree of conversion at which an inflection of the deceleration curve is observed is marked by the asterisk and highlighted in the insert table. A double-staged kinetic profile was evident in the curves and may indicate the presence of polymerization induced phase separation (PIPS) and interpenetrating polymer network (IPN) formation.

Methacrylamides have hydrogen-bond acceptor (O-H dipole) and hydrogen-bond donor (N-H dipole) capabilities, which favors their interaction with water [15]. Therefore, one additional reason for the incorporation of the methacrylate functionality on the secondary methacrylamides was to reduce the latter’s hydrophilicity. The methacrylate-methacrylamide hybrids (HEMAM Hy, 2EM Hy, and 2dMM Hy) showed dramatic reduction in water sorption in comparison to their methacrylamide versions (HEMAM, 2EM, and 2dMM) (Figure 3), with methacrylate hybrids showing 3 to 6-fold greater log P values. This means they are a lot more hydrophobic than the methacrylamide analogs (Table 2). Several factors contribute to this increased hydrophobic character: increase in molecular weight, substitution of the hydrophilic hydroxyl group and addition of an aliphatic side radical, which all decrease the molecule polarity [17]. The positive results of SL shown by the alpha-substituted methacrylamides (2EM and 2dMM) indicate a higher degree of mass loss due to leaching out of unreacted monomers – the final degree of conversion was 75% on average.

Table 2.

Molecular weight (MW), partition coefficient (log P) and percentage of carbon double bonds from methacrylamide and methacrylate functional groups (% [C=C]), and final degree of conversion (Final C=C) of the tested co-monomers.

Co-Monomer	MW (g/mol)	Log P	% [C=C] methacrylamide	% [C=C] methacrylate	Final C=C conversion (%)
TEGDMA	286.32	1.42	0	100	78
HEMA	130.14	0.45	0	100	89
HEMAM	129.15	−0.21	57	43	83.2
HEMAM HY	196.23	1.06	31.5	68.5	65.3
2EM	157.11	0.60	52	48	73.6
2EM HY	224.48	1.86	30	70	63.3
2dMM	157.21	0.33	52	48	76.7
2dMM HY	224.48	1.59	30	70	59.4

Regarding the mechanical properties, it was expected that the incorporation of difunctional molecules into the formulations would enhance the crosslinking and, ultimately, the shear modulus and μTBS. However, no clear trend was identified in the shear storage modulus results among the difunctional molecules. HEMAM Hy showed the highest values and 2dMM Hy and TEGDMA the lowest ones, which indicates that the molecular packing and the intermolecular interactions are playing key roles. It is known for co-polymerizations between TEGDMA and bisGMA that heterogeneous and poorly-packed polymer networks result, due to TEGDMA’s tendency to cyclization, as well as bisGMA’s rigidity [24]. Cyclization is likely in difunctional molecules with flexible backbones, ultimately leading to the formation of a network with reduced cross-linking density and glass transition temperature, despite the high levels of final degree of conversion [18, 25, 26]. In addition, the flexibility of the pendant groups and crosslinks make the TEGDMA molecule susceptible to rotational motion and with tendency to occupy more space, which compromises the packing efficiency and increases the free volume [24]. And finally, at the relatively high irradiances used in this study, the polymerization reaction takes place at higher rates (as evidenced by the RP_MAX results), leading to the formation of a stiff framework with greater free volume [24]. In the case of 2dMM Hy, the presence of two bulky methyl substituent groups (each with three terminal hydrogens), jeopardizes the molecular packing arrangement. The combination with the second functionality on the rigid backbone makes 2dMM Hy a bulkier and highly sterically hindered molecule, which may have compromised not only the reactivity but also the polymer network structure. Conversely, the structure of HEMAM Hy does not contain any substituents, and its polymerization reaction took place at slow rates which, in tandem with the potential phase separation indicated by the double-staged kinetic profile, may have led to toughening of the material, as previously demonstrated [27]. On a related note, bars were prepared for dynamic mechanical analysis test, but the experiment was not conducted because, after the post-curing heat processing necessary prior to the DMA test (16 hours at 180°C), the bars of HEMAM Hy, 2EM and 2dMM groups became too brittle and showed evidence of significant internal cracking (Figure 7). Though it is not completely clear why this happened for these specific groups, it can be speculated that the aforementioned molecular packing characteristics may have drastically reduced the toughness of the materials, which caused them to break upon thermal contraction.

Figure 7.

Representative bars designed for dynamic mechanical test after post-processing for 16 hours at 180°C. Numerous cracks are noticed on the surface of the methacrylate methacrylamide HEMAM Hy and on the alpha-substituted methacrylamides 2EM and 2dMM bars.

And finally, in respect to the μTBS results, SB showed statistically higher values at both 48 h and 6 months, while all other groups were statistically similar to each other. The μTBS decreased for all materials between 48 h and 6 months, and this reduction was significant for the HEMA and two 2dMM compositions. The formulation containing the monofunctional methacrylamide (HEMAM) showed the lowest bond strength reduction (about 9%) after 6 months of aging, while the other groups showed a decrease ranging between 18 and 33%. The bonding stability of some of these methacrylamides is surprising given their high WS, reduced conversion and mechanical properties compared with the HEMA control. These results follow the same trend reported previously, and it points to the complexity of dentin bonding, which is not directly related to the monomer’s properties [11]. Interestingly, the interaction between the methacrylamides and the dentin substrate was chemical-structure dependent, which makes it difficult to draw general conclusions. Studies have shown that the amides are able to establish hydrogen bonds with specific sites of the collagen, which may have contributed to some form of substrate reinforcement [28].

5. Conclusion

The hybrid strategy resulted in molecules with markedly lower water sorption. The potential increase in reactivity was overshadowed by electronic and steric factors on the tested monomers. Even though the microtensile bond strength was not improved in relation to the control, the stability of the bond observed with selected groups is an encouraging result.