- Email: [email protected]
Dynamic monitoring of curing photoactive resins: A methods comparison
d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 565–570
available at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/dema
Dynamic monitoring of curing photoactive resins: A methods comparison M. Rosentritt a,∗ , A.C. Shortall b , W.M. Palin b a
Regensburg University Medical Center, Department of Prosthetic Dentistry, Franz Josef Strauss Allee 11, D-93042 Regensburg, Germany Biomaterials Unit, University of Birmingham, School of Dentistry, College of Medical and Dental Sciences, St Chad’s Queesway, Birmingham B4 6NN, United Kingdom b
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objective. The aim of this investigation was to determine reaction enthalpy, ion viscosity
Received 28 July 2009
and curing light transmission changes of unfilled methacrylate-based systems in order to
Received in revised form
compare methods that monitor photoactive resin polymerization.
26 November 2009
Methods. Photoinitiator (0.2%, w/v, camphoroquinone), accelerator (0.3%, w/v, amine) and
Accepted 12 February 2010
inhibitor (ranging from 0 to 1%, w/v, butylated hydroxytoluene, BHT) were incorporated in an experimental BisGMA/TEGDMA co-monomer mixture (50/50, w/v). The concentration of BHT was varied from 0.00, 0.01, 0.05, 0.10, 0.50 to 1.00% (w/v). Light transmission (LT),
Keywords:
reaction enthalpy (UV-differential scanning calorimetry, DSC), and ion viscosity (dielectrical
Inhibitor
analysis, DEA) were determined during irradiation of the resins (40 s; halogen light curing-
BHT
unit). Statistical analysis was performed using two-way ANOVA followed by post hoc tests
Real-time reaction
(˛ = 0.05). Curve fitting and regression calculation were done.
Photoactive resin
Results. There was no significant change in the time to reach the maximum rate of poly-
DSC
merization (reaction time) in the individual systems up to a BHT concentration of 0.05%
DEA
(P > 0.05). Starting at a concentration of 0.10% BHT an increase in time of reaction could be
Light transmission
found from 4.0 s (LT), 4.07 s (DEA) and 4.9 s (DSC) to a maximum of 7.4 s (DSC), 9.43 s (DEA) and 9.67 s (LT). Linear increase (y = 5.588 × x) in time to the maximum speed of reaction could be found with a correlation of R2 = 0.992. Conclusions. The speed of polymerization reaction is strongly influenced by BHT concentration. The linear relationship should allow for the prediction of the speed of reaction during blending of a methacrylate-based resin. The three test systems allow for monitoring the complex polymerization kinetics of unfilled methacrylate-based systems. © 2010 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
The majority of dental composites consist of dimethacrylate resins reinforced with inorganic filler particles. The properties of the composite can be optimized by the manufacturer to suit a particular application (e.g. Class IV anterior or Class II pos-
∗
terior cavities, filled adhesives) by varying the filler (volume percentage, filler size and particle distribution) and polymer matrix composition. Smaller, more flexible monomer chains, such as those in triethylene glycol dimethacrylate (TEGDMA) and hydroxyethyl methacrylate (HEMA) in larger volumes are used to create less viscous but highly cross-linked systems.
Corresponding author. Tel.: +49 941 944 6054; fax: +49 941 944 6171. E-mail address: [email protected] (M. Rosentritt). 0109-5641/$ – see front matter © 2010 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dental.2010.02.006
566
d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 565–570
Complex, bulky monomers with phenolic rings (bisphenol glycol dimethacrylate, BisGMA) or urethane components (urethane dimethacrylate, UDMA) with higher molecular weight allow for the fabrication of higher strength, more viscous resins. Other components may be modified to alter reactivity of the system and influence network formation (by varying initiator, co-initiator types and concentrations, dye and pigment inclusions) and improve shelf-life (UV-stabilizers and inhibitors). Polymerization retarders or inhibitors are added to improve shelf-life and operatory light stability during in vivo placement and contouring but may also influence polymerization kinetics and degree of cure of the polymer [1]. For monitoring curing characteristics and efficiency of polymerization, differential scanning calorimetry (DSC) [2], Fourier transform infrared spectroscopy (FTIRS) [3], rheology, temperature measurements, simple micro-hardness tests [4] or dielectric analysis (DEA) [5] are used. For characterization of photopolymerization reactions in real-time a fast testing frequency and a high data acquisition rate of measurement is essential. In DEA, the specimen is exposed to an alternating sinusoidal voltage. The dielectric specimen affects the amplitude and phase of the input signal depending on ionic mobility and the alignment of the dipoles in the alternating current field [6]. It is considered that ion mobility is reduced during the polymerization reaction and finally stopped, so the change of ion viscosity is regarded as an indicator for the reaction and degree of cure. This allows for monitoring the chemical process of the resin from monomer with different viscosities over gelation to an insoluble polymer [7]. The first studies of dynamic measurements of light transmission (LT) during polymerization of light-activated composites identified that variations in light unit irradiance over time and changes induced by the polymerization reaction process itself influenced transmission and also proposed LT monitoring as a simple technique for determining appropriate radiation time for any LCU/composite combination [8,9]. Since then additional work has led to an improved understanding of the effects of photoinitiator absorption [10] and filler/resin refractive index mismatch changes on the cure behavior of unfilled resins and resin composites [11]. DSC investigations on the reaction of resin polymers have been described [12] and measurements are based on enthalpy changes during the reaction of the system. It is assumed that following repeated irradiation of a photoreactive system the resin is converted to a maximum degree. The calculation of the enthalpy (area below the heat flow over time of the first polymerization peak) gives the energy of the thermal and chemical reaction (including absorption/reflection). The curve provides information about the speed and enthalpy of reaction. The hypothesis of this investigation was that the measurement of reaction enthalpy, ion viscosity and curing light transmission changes of unfilled methacrylate-based systems allow for a combined interpretation of polymerization characteristics.
w/v, camphoroquinone), accelerator (0.3%, w/v, amine) and inhibitor (ranging from 0 to 1%, w/v, butylated hydroxytoluene, BHT) were incorporated. The concentration of BHT varied from 0.00, 0.01, 0.05, 0.10, 0.50 to 1.00% (w/v). All specimens (n = 3) for each monomer formulation were irradiated for 40 s with a quartz tungsten halogen, QTH light curing-unit, LCU (Elipar Trilight, 3M EPSE, Seefeld, Germany;) exit window operating at 900 ± 50 mW/cm2 throughout the present work. The QTH LCU was fitted with a 12 mm diameter light guide exit window.
2.
2.3.
Materials and methods
The investigated resin composed of a BisGMA/TEGDMA comonomer mixture (50/50, w/v). The photoinitiator (0.2%,
2.1.
Light transmission
Experimental resins were transferred from a lightproof dropper bottle into the cavity (4 mm diameter) of a black nylon disc mould (24 mm diameter, 2 mm thickness) taking care not to incorporate air bubbles. Opposing surfaces were covered with mylar matrix strips. Changes in light transmission (LT) throughout polymerization of specimens were recorded with an established photodiode technique [9,10,11], where the light curing-unit was held concentrically in an alignment ring for reproducible curing of the resin specimens. Light passes through the specimen above a diffuser to homogenize irradiation and eliminate effects of light dispersion. Light is detected through a 4 mm diameter aperture by a calibrated photodiode at the base of the lightproof box and the voltage output processed by a computer (Fig. 1a). LT traces of the polymerizing monomers were normalized against light output since the operation efficiency decreased as the unit warmed up and irradiance fell by approximately 5% over the 40 s radiation time.
2.2.
UV-differential scanning calorimetry (DSC)
Special modified UV-DSC (DSC 204 F1, Netzsch, Selb, Germany) was used in combination with the light curing-unit, which was positioned directly above the specimen pan. The reference pan was kept empty. All measurements were performed isothermally at 37 ◦ C. N2 -atmosphere (20 ml/min) was used to avoid an oxygen inhibition layer. Each specimen (15 ± 0.04 mg) was exposed to five 40 s curing cycles, with a 2 min pause in between each cycle (Fig. 1b). Data acquisition rate was 10 s−1 . Control measurements without resin and with resin but without light activation were performed. The speed of reaction (height of the peak) and enthalpy of reaction (peak area) curves (f = conversion) were evaluated, as well as time to maximum speed and maximum peak (Netzsch complex evaluation program). The heat of reaction (HR ) curve was calculated according to Eq. (1): HR = Q1 −
Q2 + Q3 + Q4 + Q5 4
(1)
where Q1 equals the energy of the thermal and chemical reaction (including absorption/reflection) and Q2 − Q5 is the absorption/reflection of the cup.
Dielectrical analysis (DEA)
The light curing performance of dental resins was investigated with a dielectric analyzer (DEA 231/1 Epsilon,
d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 565–570
567
Fig. 1 – Examples of the measured curves for DSC, DEA and LT (BHT concentration 10%).
Netzsch-Gerätebau, Selb, Germany). All measurements were performed with an interdigitated electrode IDEX sensor (frequency 1000 Hz, data acquisition rate 0.055 s). The electrodes of the sensor were arranged in a fringed design with comb electrodes (65S, A/D Ratio: 80, distance between electrodes: 115 m, sensing area: 1.2 mm × 2.5 mm). The mode of action is described in detail by Zahouily et al. [13]. A perforated (5 mm diameter, 1 mm thick) polyethylene mask film was fixed on the sensor and the resins were applied with a dental spatula and pressed flat with a microscope slide. The glass slide also served to reduce oxygen inhibition at the specimen surface throughout irradiation. A retaining jig was used to assemble the LCU directly above the material (Fig. 1c). Temperature was controlled with a thermocouple on a second input of the DEA electronics. Control measurements with mask film without resin and with resin but without light activation were performed. A logarithmical ionic viscosity/time graph was recorded. All analysis was performed with Proteus Analysis software (Netzsch, Selb, Germany). Statistical analysis was performed using two-way ANOVA followed by post hoc Scheffe pair-wise comparisons between means with Bonferroni adjusting (˛ = 0.05). Curve fitting and regression calculation were done using SPSS 16.0.
Fig. 2 – The change in energy, ion viscosity and light transmission throughout curing of resin mixtures containing various inhibitor concentrations.
3.
Results
Fig. 2 shows the time-resolved data obtained by each test method for resins containing increasing concentration of inhibitor. Fig. 2a demonstrates the changes in light transmission through the sample throughout polymerization and features distinct inflections, which occur at longer times throughout cure dependent upon inhibitor concentration. For each test method, there was no apparent change in the polymerization kinetics and transmission patterns up to a BHT concentration of 0.05% (P > 0.05). Proceeding light activation (t = 0) the DSC signal showed a steep exothermal increase (>10 mW/mg), reaching maximum energy (maximum polymerization velocity) following approximately 5 s. With further progression of the polymerization, energy performance reversed to a steep endothermal reaction (Fig. 2b). Ion viscosity in DEA initially decreases following polymerization by half of a logarithmic decade, and after ∼2 s increases by about 1–2 decades. 5 s after light activation, ion viscos-
568
d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 565–570
Fig. 3 – Time of reaction of DSC, DEA and LTC tests with different BHT concentrations.
Fig. 4 – Regression coefficient (R2 ) and gradient (m) of the individual test methods and linear curve fitting.
ity increases further on, but with a smaller slope, indicating reduced ion mobility (Fig. 2c). Starting at a concentration of 0.10% BHT an increase in time of reaction could be found in all tests. The time increased from 4.0 s (LT), 4.07 s (DEA) and 4.9 s (DSC) to a maximum of 7.4 s (DSC), 9.43 s (DEA) and 9.67 s (LT) (Fig. 3). A mathematical curve fitting analysis showed a linear increase in the time to the maximum speed of reaction. Curve fitting parameters and correlations vary for the individual tests methods. The individual linear regressions and detailed information of the curve fits are displayed in Fig. 4. An overall equation for the three test methods is expressed by y = 5.588 × x with a correlation of R2 = 0.992.
4.
Discussion
The hypothesis of this investigation that the three different test systems allow for monitoring the complex reaction of unfilled methacrylate-based systems can be corroborated. The change in light transmission (LT), reaction enthalpy (DSC) and ion viscosity (DEA) provided similar steps of polymerization, but small differences in the time to reach these individual steps. All test methods allow for the investigation of the light
polymerization of an unfilled resin system. In principle the measurement of filled composite resins is possible, however, filled resins complicate LT tests since the change in optical properties of the resin matrix throughout cure may increase and/or decrease LT due to changes in extent of interfacial scattering between resin and filler [11]. Furthermore constraint induced by surrounding cavity walls will heighten the effects of polymerization contraction stress leading to filler to resin matrix micro-gap formation and additional effects on LT [14]. In contrast to other test methods, which may use low irradiance (∼50 mWcm−2 ) to slow reaction speeds and allow for suitable data acquisition, the described systems work with appropriate irradiance levels associated with dental light sources and therefore remain relevant to the clinical situation. While with matching radiant exposure the polymerization as assessed by degree of conversion of an unfilled dimethacrylate resin using low irradiance over a prolonged period may equal that of “standard” polymerization over a clinically relevant time, it has been argued that slower reaction will lead to less heterogeneity of the polymer network and a reduction in the flexural modulus of the cured resin [15]. Even if irradiance regulates the rate of polymerization it is unlikely that final polymer conversion will be significantly affected unless resin viscosity varies greatly since such rapid reactions result in a diffusion-controlled process and a monomolecular termination pathway [16]. Additionally, slower polymerization occurs in resin composites at greater distances from the irradiated surface as irradiance falls with increasing depth. While no significant differences in Wallace hardness at depths between 0.5 mm and 3.5 mm from the specimen surface were found for a light-cured composite after dry storage, hardness declined to a greater extent with increasing depth following subsequent ethanol immersion and this has been attributed to slower polymerization resulting in a polymer with reduced cross-link density [17]. The chain polymerization is initiated by free-radical polymerization, including elementary reaction steps (production of radical, initiation of chain radicals, chain propagation and radical removal [18]). Under certain reaction conditions, branched or cross-linked polymers form (network formation). Autoacceleration (gel effect) and vitrification together with a progressive slowing of the polymerization [13,19,20] may take place during the reaction [21]. High-molecular-weight polymer is formed immediately in the chain polymerization [21]. Structure and properties [22–24] of a polymer may be dependent on parameters of the reaction, such as polymerization temperature [25] curing irradiance [26] and monomer viscosity [16]. The three methods employed in the present study show similar data to describe the reaction with fast autoacceleration properties and reduced speed after approximately 5 s. Immediately following light activation, UV-DSC curves provide a strong exothermal signal > 10 mg/mW. The acceleration of the reaction can be evaluated via derivation of energy versus time and shows a maximum for all resins after ∼2 s. The integral of the curve is related to the conversion of the resin during the experiment and the whole area below the curve shows the polymerization energy. At the point of maximum polymerization energy (∼5 s), autoacceleration stops and speed of reaction is dramatically reduced and autodeceleration occurs,
d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 565–570
which is in accordance with earlier investigations using similar resin formulations [5]. New aspects of the reaction arise regarding DEA results. The expected increase of ion viscosity, which should arise because of the hindered mobility of conductive components of the resin during polymerization, occurred with a delay of ∼2 s. This short decrease of ion viscosity in the range of half a decade on the scale of the ion viscosity can be explained with a decrease of viscosity of the material due to heating from the light source. At the point of lowest ion viscosity, polymerization energy (compare UV-DSC) provided highest acceleration of the reaction. The steep increase in ion viscosity after 2 s (Fig. 1) could be interpreted as a result of autoacceleration progressively hindering ion mobility. The mathematical conversion point of the curve indicated a distinct change in rate of ion viscosity after ∼5 s that correlates with the reduced speed of reaction. While UV-DSC showed only a principle slowdown of the speed of reactivity, DEA and LT may provide detailed information about the speed of reactivity by analyzing the gradient of the particular curve. Up until the distinct inflection points in the light transmission traces, which occur at longer times as inhibitor concentration increases, the similar LT patterns (before 4.3 s) can be attributed to consumption of the inhibitor and retardation of the curing process. The slight decrease in LT observed prior to the inflection point might be associated with heating of the resins by the curing lamp which decreases viscosity. According to Beer–Lambert’s Law, as polymerization proceeds light transmission through the curing resin will increase as the photoinitiator is consumed and as a decrease in volume shortens the optical path length. Indeed, light transmission will be significantly affected by shrinkage phenomena, where specimen constraint will alter shrinkage vectors and patterns of light transmission [27]. Here, using a low aspect ratio (4 mm diameter, 2 mm thickness), specimen constraint and subsequent change in specimen geometry following light irradiation would be expected once polymer mobility rapidly decreases. Subsequently, the restricted specimens exhibited a distinct change in rate of light transmission giving the observed inflection on the LT trace. The exothermic polymerization reaction can be characterized by gelation (liquid-rubber) and vitrification (rubber to glass transition) and directly measured using UV-DSC techniques (Fig. 1). The gel-point corresponds to formation of an infinite cross-linked network and is accompanied by a dramatic physical change. Gel-point occurs at early stages of conversion exemplified here in LT experiments by the marked change in transmission following inhibitor consumption. Gelation results in a distinct increase in density, which would be expected to increase light scattering and decrease LT. However, we assume the sudden increase in LT, which may have resulted from shrinkage strain of the constrained sample and compliant acetate strip, may have outweighed any decrease in LT from increased density. Further confirmation of the effects of sample viscosity changes on LT throughout cure requires rheological characterization. As polymerization proceeds, contraction increases throughout the gradual vitrification process demonstrated by an increase in LT as optical path length and photoinitiator concentration decrease.
569
DEA and LT experiments were performed with data acquisition rates of 16.7 and 10 Hz, respectively, which appears to follow the speed of polymerization reaction effectively. The processing speed of DSC (10 Hz) also appears sufficient for the determination of the reaction and allows a beneficial estimation of the degree of conversion of the resin via integration of the curve. However, UV-DSC data lagged behind in comparison to DEA and LT experiments, which might be due to inertia (time constant and resolution) of the DSC cell. Although different aspects and properties of the resins were investigated, reaction enthalpy, ion viscosity and light transmission show comparable results of the speed of reaction. All test systems revealed no influence of the BHT concentration up to 0.10%. From BHT concentrations between 0.25% and 1.00%, the time of reaction increases from approximately 4–5 s up to 7.5 s (DSC) and 9.5 s (LT). We assume that some BHT had been added during formulation of the resins to improve storability and shelf-life of the resin, and in this work, monomers were used without purification. We believe that a possible explanation for the non-significant influence of inhibitor on reaction kinetics up to a concentration of 0.10% may possibly relate to trace impurities of the delivered resins rather than insensitivity of the analytical techniques employed. Further research is required to confirm this hypothesis. Above the threshold value of 0.10% BHT, the time to the highest speed of the reaction increased by about 80–100%. For achieving elaborated information on the speed of polymerization, further investigations are necessary for example at different temperatures or light intensity. It should be considered that sources of errors for all methods are adjunct to the polymerization: shrinkage of the resin may cause stress on the sensor or even (partly) de-bonding of the resin. Furthermore, the results may be influenced by variations in film thickness or polymerization temperature [19,28], among others. Trace impurities of BHT in the delivered resins, which were used without purification, may influence the results. Even a shielded DEA inductive field is very sensitive to environmental influences. All three test methods used here for characterizing the polymerization process of photoactive resins assessed the degree of polymerization indirectly. Fourier transform infrared spectroscopy techniques can be used measure conversion directly by analyzing the extent of C C saturation in the midIR [29] or C–H saturation in the near-IR region [30]. Although such techniques provide a powerful analytical tool for measuring degree of conversion (especially, the near-IR methods, which allow dynamic transmission measurements in realtime), complications arise comparing filled resin systems, since differences in silane concentrations between materials may significantly affect the final conversion value [31]. In summary, a significant linear correlation (R2 = 0.992) was established between BHT concentration and time/speed of reaction for each method of characterization. This infers that the speed of polymerization reaction can be strongly influenced by BHT concentration. The three different methods showed in accordance that this linear relationship should allow for the prediction of the speed of reaction for photoactive resin systems and, if correlated to physical properties, may allow forecasting material properties on a fine scale. The three different methods, which determine the change in light
570
d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 565–570
transmission (LT), reaction enthalpy (DSC) and ion viscosity (DEA) allow for monitoring the complex reaction of unfilled methacrylate-based systems.
5.
Conclusions
Improvements in light-activated resins and adhesive restorative techniques are providing better, less invasive methods for replacing lost tooth tissues. Adequate polymerization in depth is fundamental to the clinical success of these restorative materials. By better understanding the dynamics of the polymerization process and the factors governing the nature and extent of polymerization it may ultimately be possible to tailor new material formulations to better suit the diverse needs of restorative clinical practice. Traditional methods of measuring resin conversion only offer limited insight into how final mechanical properties of a restorative resin ensue. In this work three methods, which complement and offer good alternatives to classical conversion monitoring techniques, were compared in regard to their ability to monitor the photopolymerization process of a model unfilled resin mixture. A highly significant correlation was established reaction kinetics of model resin mixtures as influenced by level of inhibitor concentration and polymerization reaction rate as indirectly monitored by light transmission changes, reaction enthalpy (UV-DSC) and ion viscosity (DEA).
references
[1] Braga RR, Ferracane JL. Contraction stress related to degree of conversion and reaction kinetics. J Dent Res 2002;81: 114–8. [2] Rosentritt M, Behr M, Leibrock A, Handel G. Veneering composites—a thermoanalytical examination. J Mater Sci Mater Med 1999;10:91–8. [3] Lovell LG, Berchtold KA, Elliott JE, Lu H, Bowman CN. Understanding the kinetics and network formation of dimethacrylate dental resins. Polym Adv Technol 2001;12:335–45. [4] Jung H, Friedl KH, Hiller KA, Haller A, Schmalz G. Curing efficiency of different polymerization methods through ceramic restorations. Clin Oral Invest 2001;5:156–61. [5] Rosentritt M, Behr M, Knappe S, Handel G. Dielectric analysis of light-curing dental restorative materials—a pilot study. J Mater Sci 2006;41:2805–10. [6] Kim H, Char K. Dielectric changes during the curing of epoxy resin based on the diglycidyl ether of bisphenol A (DGEBA) with diamine. Bull Korean Chem Soc 1999;20: 1329–34. [7] Kranbuehl D, Delos S, Yi E, Mayer J, Jarvie T. Dynamic dielectric analysis: non-destructive material evaluation and cure cycle monitoring. Polym Eng Sci 1986;26:338–45. [8] Komatsu K, Nemoto K, Horie K. Transmittance of light-cured composite resins. 1. Measurement of transmittance with time. Shika Zairyo Kikai 1990;9:102–11. [9] Harrington E, Wilson H, Shortall A. Light-activated restorative materials: a method of determining effective radiation times. J Oral Rehabil 1996;23:210–8.
[10] Ogunyinka A, Palin WM, Shortall AC, Marquis PM. Photoinitiation chemistry affects light transmission and degree of conversion of curing experimental dental resin composites. Dent Mater 2007;23:807–13. [11] Shortall AC, Palin WM, Burtscher P. Refractive index mismatch and monomer reactivity influence composite curing depth. J Dent Res 2008;87:84–8. [12] Breschi L, Cadenaro M, Antoniolli F, Sauro S, Biasotto M, Prati C, et al. Polymerization kinetics of dental adhesives cured with LED: correlation between extent of conversion and permeability. Dent Mater 2007;23:1066–72. [13] Zahouily K, Decker C, Kaisersberger E, Gruener M. Real-time UV cure monitoring. Eur Coat J 2003;11:245–9. [14] Feng L, Suh BI. Gaps between filler and resin induced by polymerization shrinkage. J Dent Res 2009 [87th general Session IADR, Miami, FL., USA. Abstr #1016]. [15] Feng L, Suh BI. A mechanism on why slower polymerization of a dental composite produces lower contraction stress. J Biomed Mater Res Part B: Appl Biomater 2006;78B:63–9. [16] Feng L, Suh BI. Exposure reciprocity law in photopolymerization of multi-functional acrylates and methacrylates. Macromol Chem Phys 2007;208:295–306. [17] Asmussen E, Peutzfeldt A. Polymer structure of a resin composite in relation to distance from the surface. Eur J Oral Sci 2003;111:277–9. [18] Oster G, Yang N. Photopolymerisation of vinyl monomers. Chem Rev 1968;68:125–51. [19] Decker C, Moussa K. A new class of highly reactive acrylic monomers. 1. Light-induced polymerisation. Makramol Chem Rapid Commun 1990;11:159–67. [20] Navarrete M, Pineda J, Vera-Graziano R. Multifractality in the copolymerization of Bis-GMA/TEGDMA by pulsed photoacoustics. J Appl Polym Sci 2009;111:1199–208. [21] Odian G. Principles of polymerization. 4th ed. New York: Wiley; 2004. p. 465–533. [22] Floyd CJE, Dickens SH. Network structure of Bis-GMA- and UDMA-based resin. Dent Mater 2006;22:1143–9. [23] Loshaek S, Fox TG. Cross-linked polymers. I. Factors influencing the efficiency of cross-linking in copolymers of methyl methacrylate and glycol dimethacrylates. J Am Chem Soc 1953;75:3544–50. [24] Watts DC. Reaction kinetics and mechanics in photo-polymerised networks. Dent Mater 2005;21:27–35. [25] Trujillo M, Newmann SM, Stansbury JW. Use of near-IR to monitor the influence of external heating on dental composite polymerisation. Dent Mater 2004;20:766–77. [26] Lovell LG, Newman SM, Donaldson MM, Bowman CN. The effect of light intensity on double bond conversion and flexural strength of a model, unfilled dental resin. Dent Mater 2003;19:458–65. [27] Palin WM, Senyilmaz DP, Marquis PM, Shortall AC. Cure width potential for MOD resin composite molar restorations. Dent Mater 2008;24:1083–94. [28] Decker C. Kinetic study of light-induced polymerization by real-time UV and IR spectroscopy. J Pol Sci Part A: Pol Chem 1992;30:913–28. [29] Palin WM, Fleming GJ, Burke FJ, Marquis PM, Randall RC. Monomer conversion versus flexure strength of a novel dental composite. J Dent 2003;31:341–51. [30] Stansbury JW, Dickens SH. Determination of double bond conversion in dental resins by near infrared spectroscopy. Dent Mater 2001;17:71–9. [31] Halvorson RH, Erickson RL, Davidson CL. The effect of silane content on conversion of resin-based composite. Dent Mater 2003;19:327–33.
available at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/dema
Dynamic monitoring of curing photoactive resins: A methods comparison M. Rosentritt a,∗ , A.C. Shortall b , W.M. Palin b a
Regensburg University Medical Center, Department of Prosthetic Dentistry, Franz Josef Strauss Allee 11, D-93042 Regensburg, Germany Biomaterials Unit, University of Birmingham, School of Dentistry, College of Medical and Dental Sciences, St Chad’s Queesway, Birmingham B4 6NN, United Kingdom b
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objective. The aim of this investigation was to determine reaction enthalpy, ion viscosity
Received 28 July 2009
and curing light transmission changes of unfilled methacrylate-based systems in order to
Received in revised form
compare methods that monitor photoactive resin polymerization.
26 November 2009
Methods. Photoinitiator (0.2%, w/v, camphoroquinone), accelerator (0.3%, w/v, amine) and
Accepted 12 February 2010
inhibitor (ranging from 0 to 1%, w/v, butylated hydroxytoluene, BHT) were incorporated in an experimental BisGMA/TEGDMA co-monomer mixture (50/50, w/v). The concentration of BHT was varied from 0.00, 0.01, 0.05, 0.10, 0.50 to 1.00% (w/v). Light transmission (LT),
Keywords:
reaction enthalpy (UV-differential scanning calorimetry, DSC), and ion viscosity (dielectrical
Inhibitor
analysis, DEA) were determined during irradiation of the resins (40 s; halogen light curing-
BHT
unit). Statistical analysis was performed using two-way ANOVA followed by post hoc tests
Real-time reaction
(˛ = 0.05). Curve fitting and regression calculation were done.
Photoactive resin
Results. There was no significant change in the time to reach the maximum rate of poly-
DSC
merization (reaction time) in the individual systems up to a BHT concentration of 0.05%
DEA
(P > 0.05). Starting at a concentration of 0.10% BHT an increase in time of reaction could be
Light transmission
found from 4.0 s (LT), 4.07 s (DEA) and 4.9 s (DSC) to a maximum of 7.4 s (DSC), 9.43 s (DEA) and 9.67 s (LT). Linear increase (y = 5.588 × x) in time to the maximum speed of reaction could be found with a correlation of R2 = 0.992. Conclusions. The speed of polymerization reaction is strongly influenced by BHT concentration. The linear relationship should allow for the prediction of the speed of reaction during blending of a methacrylate-based resin. The three test systems allow for monitoring the complex polymerization kinetics of unfilled methacrylate-based systems. © 2010 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
The majority of dental composites consist of dimethacrylate resins reinforced with inorganic filler particles. The properties of the composite can be optimized by the manufacturer to suit a particular application (e.g. Class IV anterior or Class II pos-
∗
terior cavities, filled adhesives) by varying the filler (volume percentage, filler size and particle distribution) and polymer matrix composition. Smaller, more flexible monomer chains, such as those in triethylene glycol dimethacrylate (TEGDMA) and hydroxyethyl methacrylate (HEMA) in larger volumes are used to create less viscous but highly cross-linked systems.
Corresponding author. Tel.: +49 941 944 6054; fax: +49 941 944 6171. E-mail address: [email protected] (M. Rosentritt). 0109-5641/$ – see front matter © 2010 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dental.2010.02.006
566
d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 565–570
Complex, bulky monomers with phenolic rings (bisphenol glycol dimethacrylate, BisGMA) or urethane components (urethane dimethacrylate, UDMA) with higher molecular weight allow for the fabrication of higher strength, more viscous resins. Other components may be modified to alter reactivity of the system and influence network formation (by varying initiator, co-initiator types and concentrations, dye and pigment inclusions) and improve shelf-life (UV-stabilizers and inhibitors). Polymerization retarders or inhibitors are added to improve shelf-life and operatory light stability during in vivo placement and contouring but may also influence polymerization kinetics and degree of cure of the polymer [1]. For monitoring curing characteristics and efficiency of polymerization, differential scanning calorimetry (DSC) [2], Fourier transform infrared spectroscopy (FTIRS) [3], rheology, temperature measurements, simple micro-hardness tests [4] or dielectric analysis (DEA) [5] are used. For characterization of photopolymerization reactions in real-time a fast testing frequency and a high data acquisition rate of measurement is essential. In DEA, the specimen is exposed to an alternating sinusoidal voltage. The dielectric specimen affects the amplitude and phase of the input signal depending on ionic mobility and the alignment of the dipoles in the alternating current field [6]. It is considered that ion mobility is reduced during the polymerization reaction and finally stopped, so the change of ion viscosity is regarded as an indicator for the reaction and degree of cure. This allows for monitoring the chemical process of the resin from monomer with different viscosities over gelation to an insoluble polymer [7]. The first studies of dynamic measurements of light transmission (LT) during polymerization of light-activated composites identified that variations in light unit irradiance over time and changes induced by the polymerization reaction process itself influenced transmission and also proposed LT monitoring as a simple technique for determining appropriate radiation time for any LCU/composite combination [8,9]. Since then additional work has led to an improved understanding of the effects of photoinitiator absorption [10] and filler/resin refractive index mismatch changes on the cure behavior of unfilled resins and resin composites [11]. DSC investigations on the reaction of resin polymers have been described [12] and measurements are based on enthalpy changes during the reaction of the system. It is assumed that following repeated irradiation of a photoreactive system the resin is converted to a maximum degree. The calculation of the enthalpy (area below the heat flow over time of the first polymerization peak) gives the energy of the thermal and chemical reaction (including absorption/reflection). The curve provides information about the speed and enthalpy of reaction. The hypothesis of this investigation was that the measurement of reaction enthalpy, ion viscosity and curing light transmission changes of unfilled methacrylate-based systems allow for a combined interpretation of polymerization characteristics.
w/v, camphoroquinone), accelerator (0.3%, w/v, amine) and inhibitor (ranging from 0 to 1%, w/v, butylated hydroxytoluene, BHT) were incorporated. The concentration of BHT varied from 0.00, 0.01, 0.05, 0.10, 0.50 to 1.00% (w/v). All specimens (n = 3) for each monomer formulation were irradiated for 40 s with a quartz tungsten halogen, QTH light curing-unit, LCU (Elipar Trilight, 3M EPSE, Seefeld, Germany;) exit window operating at 900 ± 50 mW/cm2 throughout the present work. The QTH LCU was fitted with a 12 mm diameter light guide exit window.
2.
2.3.
Materials and methods
The investigated resin composed of a BisGMA/TEGDMA comonomer mixture (50/50, w/v). The photoinitiator (0.2%,
2.1.
Light transmission
Experimental resins were transferred from a lightproof dropper bottle into the cavity (4 mm diameter) of a black nylon disc mould (24 mm diameter, 2 mm thickness) taking care not to incorporate air bubbles. Opposing surfaces were covered with mylar matrix strips. Changes in light transmission (LT) throughout polymerization of specimens were recorded with an established photodiode technique [9,10,11], where the light curing-unit was held concentrically in an alignment ring for reproducible curing of the resin specimens. Light passes through the specimen above a diffuser to homogenize irradiation and eliminate effects of light dispersion. Light is detected through a 4 mm diameter aperture by a calibrated photodiode at the base of the lightproof box and the voltage output processed by a computer (Fig. 1a). LT traces of the polymerizing monomers were normalized against light output since the operation efficiency decreased as the unit warmed up and irradiance fell by approximately 5% over the 40 s radiation time.
2.2.
UV-differential scanning calorimetry (DSC)
Special modified UV-DSC (DSC 204 F1, Netzsch, Selb, Germany) was used in combination with the light curing-unit, which was positioned directly above the specimen pan. The reference pan was kept empty. All measurements were performed isothermally at 37 ◦ C. N2 -atmosphere (20 ml/min) was used to avoid an oxygen inhibition layer. Each specimen (15 ± 0.04 mg) was exposed to five 40 s curing cycles, with a 2 min pause in between each cycle (Fig. 1b). Data acquisition rate was 10 s−1 . Control measurements without resin and with resin but without light activation were performed. The speed of reaction (height of the peak) and enthalpy of reaction (peak area) curves (f = conversion) were evaluated, as well as time to maximum speed and maximum peak (Netzsch complex evaluation program). The heat of reaction (HR ) curve was calculated according to Eq. (1): HR = Q1 −
Q2 + Q3 + Q4 + Q5 4
(1)
where Q1 equals the energy of the thermal and chemical reaction (including absorption/reflection) and Q2 − Q5 is the absorption/reflection of the cup.
Dielectrical analysis (DEA)
The light curing performance of dental resins was investigated with a dielectric analyzer (DEA 231/1 Epsilon,
d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 565–570
567
Fig. 1 – Examples of the measured curves for DSC, DEA and LT (BHT concentration 10%).
Netzsch-Gerätebau, Selb, Germany). All measurements were performed with an interdigitated electrode IDEX sensor (frequency 1000 Hz, data acquisition rate 0.055 s). The electrodes of the sensor were arranged in a fringed design with comb electrodes (65S, A/D Ratio: 80, distance between electrodes: 115 m, sensing area: 1.2 mm × 2.5 mm). The mode of action is described in detail by Zahouily et al. [13]. A perforated (5 mm diameter, 1 mm thick) polyethylene mask film was fixed on the sensor and the resins were applied with a dental spatula and pressed flat with a microscope slide. The glass slide also served to reduce oxygen inhibition at the specimen surface throughout irradiation. A retaining jig was used to assemble the LCU directly above the material (Fig. 1c). Temperature was controlled with a thermocouple on a second input of the DEA electronics. Control measurements with mask film without resin and with resin but without light activation were performed. A logarithmical ionic viscosity/time graph was recorded. All analysis was performed with Proteus Analysis software (Netzsch, Selb, Germany). Statistical analysis was performed using two-way ANOVA followed by post hoc Scheffe pair-wise comparisons between means with Bonferroni adjusting (˛ = 0.05). Curve fitting and regression calculation were done using SPSS 16.0.
Fig. 2 – The change in energy, ion viscosity and light transmission throughout curing of resin mixtures containing various inhibitor concentrations.
3.
Results
Fig. 2 shows the time-resolved data obtained by each test method for resins containing increasing concentration of inhibitor. Fig. 2a demonstrates the changes in light transmission through the sample throughout polymerization and features distinct inflections, which occur at longer times throughout cure dependent upon inhibitor concentration. For each test method, there was no apparent change in the polymerization kinetics and transmission patterns up to a BHT concentration of 0.05% (P > 0.05). Proceeding light activation (t = 0) the DSC signal showed a steep exothermal increase (>10 mW/mg), reaching maximum energy (maximum polymerization velocity) following approximately 5 s. With further progression of the polymerization, energy performance reversed to a steep endothermal reaction (Fig. 2b). Ion viscosity in DEA initially decreases following polymerization by half of a logarithmic decade, and after ∼2 s increases by about 1–2 decades. 5 s after light activation, ion viscos-
568
d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 565–570
Fig. 3 – Time of reaction of DSC, DEA and LTC tests with different BHT concentrations.
Fig. 4 – Regression coefficient (R2 ) and gradient (m) of the individual test methods and linear curve fitting.
ity increases further on, but with a smaller slope, indicating reduced ion mobility (Fig. 2c). Starting at a concentration of 0.10% BHT an increase in time of reaction could be found in all tests. The time increased from 4.0 s (LT), 4.07 s (DEA) and 4.9 s (DSC) to a maximum of 7.4 s (DSC), 9.43 s (DEA) and 9.67 s (LT) (Fig. 3). A mathematical curve fitting analysis showed a linear increase in the time to the maximum speed of reaction. Curve fitting parameters and correlations vary for the individual tests methods. The individual linear regressions and detailed information of the curve fits are displayed in Fig. 4. An overall equation for the three test methods is expressed by y = 5.588 × x with a correlation of R2 = 0.992.
4.
Discussion
The hypothesis of this investigation that the three different test systems allow for monitoring the complex reaction of unfilled methacrylate-based systems can be corroborated. The change in light transmission (LT), reaction enthalpy (DSC) and ion viscosity (DEA) provided similar steps of polymerization, but small differences in the time to reach these individual steps. All test methods allow for the investigation of the light
polymerization of an unfilled resin system. In principle the measurement of filled composite resins is possible, however, filled resins complicate LT tests since the change in optical properties of the resin matrix throughout cure may increase and/or decrease LT due to changes in extent of interfacial scattering between resin and filler [11]. Furthermore constraint induced by surrounding cavity walls will heighten the effects of polymerization contraction stress leading to filler to resin matrix micro-gap formation and additional effects on LT [14]. In contrast to other test methods, which may use low irradiance (∼50 mWcm−2 ) to slow reaction speeds and allow for suitable data acquisition, the described systems work with appropriate irradiance levels associated with dental light sources and therefore remain relevant to the clinical situation. While with matching radiant exposure the polymerization as assessed by degree of conversion of an unfilled dimethacrylate resin using low irradiance over a prolonged period may equal that of “standard” polymerization over a clinically relevant time, it has been argued that slower reaction will lead to less heterogeneity of the polymer network and a reduction in the flexural modulus of the cured resin [15]. Even if irradiance regulates the rate of polymerization it is unlikely that final polymer conversion will be significantly affected unless resin viscosity varies greatly since such rapid reactions result in a diffusion-controlled process and a monomolecular termination pathway [16]. Additionally, slower polymerization occurs in resin composites at greater distances from the irradiated surface as irradiance falls with increasing depth. While no significant differences in Wallace hardness at depths between 0.5 mm and 3.5 mm from the specimen surface were found for a light-cured composite after dry storage, hardness declined to a greater extent with increasing depth following subsequent ethanol immersion and this has been attributed to slower polymerization resulting in a polymer with reduced cross-link density [17]. The chain polymerization is initiated by free-radical polymerization, including elementary reaction steps (production of radical, initiation of chain radicals, chain propagation and radical removal [18]). Under certain reaction conditions, branched or cross-linked polymers form (network formation). Autoacceleration (gel effect) and vitrification together with a progressive slowing of the polymerization [13,19,20] may take place during the reaction [21]. High-molecular-weight polymer is formed immediately in the chain polymerization [21]. Structure and properties [22–24] of a polymer may be dependent on parameters of the reaction, such as polymerization temperature [25] curing irradiance [26] and monomer viscosity [16]. The three methods employed in the present study show similar data to describe the reaction with fast autoacceleration properties and reduced speed after approximately 5 s. Immediately following light activation, UV-DSC curves provide a strong exothermal signal > 10 mg/mW. The acceleration of the reaction can be evaluated via derivation of energy versus time and shows a maximum for all resins after ∼2 s. The integral of the curve is related to the conversion of the resin during the experiment and the whole area below the curve shows the polymerization energy. At the point of maximum polymerization energy (∼5 s), autoacceleration stops and speed of reaction is dramatically reduced and autodeceleration occurs,
d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 565–570
which is in accordance with earlier investigations using similar resin formulations [5]. New aspects of the reaction arise regarding DEA results. The expected increase of ion viscosity, which should arise because of the hindered mobility of conductive components of the resin during polymerization, occurred with a delay of ∼2 s. This short decrease of ion viscosity in the range of half a decade on the scale of the ion viscosity can be explained with a decrease of viscosity of the material due to heating from the light source. At the point of lowest ion viscosity, polymerization energy (compare UV-DSC) provided highest acceleration of the reaction. The steep increase in ion viscosity after 2 s (Fig. 1) could be interpreted as a result of autoacceleration progressively hindering ion mobility. The mathematical conversion point of the curve indicated a distinct change in rate of ion viscosity after ∼5 s that correlates with the reduced speed of reaction. While UV-DSC showed only a principle slowdown of the speed of reactivity, DEA and LT may provide detailed information about the speed of reactivity by analyzing the gradient of the particular curve. Up until the distinct inflection points in the light transmission traces, which occur at longer times as inhibitor concentration increases, the similar LT patterns (before 4.3 s) can be attributed to consumption of the inhibitor and retardation of the curing process. The slight decrease in LT observed prior to the inflection point might be associated with heating of the resins by the curing lamp which decreases viscosity. According to Beer–Lambert’s Law, as polymerization proceeds light transmission through the curing resin will increase as the photoinitiator is consumed and as a decrease in volume shortens the optical path length. Indeed, light transmission will be significantly affected by shrinkage phenomena, where specimen constraint will alter shrinkage vectors and patterns of light transmission [27]. Here, using a low aspect ratio (4 mm diameter, 2 mm thickness), specimen constraint and subsequent change in specimen geometry following light irradiation would be expected once polymer mobility rapidly decreases. Subsequently, the restricted specimens exhibited a distinct change in rate of light transmission giving the observed inflection on the LT trace. The exothermic polymerization reaction can be characterized by gelation (liquid-rubber) and vitrification (rubber to glass transition) and directly measured using UV-DSC techniques (Fig. 1). The gel-point corresponds to formation of an infinite cross-linked network and is accompanied by a dramatic physical change. Gel-point occurs at early stages of conversion exemplified here in LT experiments by the marked change in transmission following inhibitor consumption. Gelation results in a distinct increase in density, which would be expected to increase light scattering and decrease LT. However, we assume the sudden increase in LT, which may have resulted from shrinkage strain of the constrained sample and compliant acetate strip, may have outweighed any decrease in LT from increased density. Further confirmation of the effects of sample viscosity changes on LT throughout cure requires rheological characterization. As polymerization proceeds, contraction increases throughout the gradual vitrification process demonstrated by an increase in LT as optical path length and photoinitiator concentration decrease.
569
DEA and LT experiments were performed with data acquisition rates of 16.7 and 10 Hz, respectively, which appears to follow the speed of polymerization reaction effectively. The processing speed of DSC (10 Hz) also appears sufficient for the determination of the reaction and allows a beneficial estimation of the degree of conversion of the resin via integration of the curve. However, UV-DSC data lagged behind in comparison to DEA and LT experiments, which might be due to inertia (time constant and resolution) of the DSC cell. Although different aspects and properties of the resins were investigated, reaction enthalpy, ion viscosity and light transmission show comparable results of the speed of reaction. All test systems revealed no influence of the BHT concentration up to 0.10%. From BHT concentrations between 0.25% and 1.00%, the time of reaction increases from approximately 4–5 s up to 7.5 s (DSC) and 9.5 s (LT). We assume that some BHT had been added during formulation of the resins to improve storability and shelf-life of the resin, and in this work, monomers were used without purification. We believe that a possible explanation for the non-significant influence of inhibitor on reaction kinetics up to a concentration of 0.10% may possibly relate to trace impurities of the delivered resins rather than insensitivity of the analytical techniques employed. Further research is required to confirm this hypothesis. Above the threshold value of 0.10% BHT, the time to the highest speed of the reaction increased by about 80–100%. For achieving elaborated information on the speed of polymerization, further investigations are necessary for example at different temperatures or light intensity. It should be considered that sources of errors for all methods are adjunct to the polymerization: shrinkage of the resin may cause stress on the sensor or even (partly) de-bonding of the resin. Furthermore, the results may be influenced by variations in film thickness or polymerization temperature [19,28], among others. Trace impurities of BHT in the delivered resins, which were used without purification, may influence the results. Even a shielded DEA inductive field is very sensitive to environmental influences. All three test methods used here for characterizing the polymerization process of photoactive resins assessed the degree of polymerization indirectly. Fourier transform infrared spectroscopy techniques can be used measure conversion directly by analyzing the extent of C C saturation in the midIR [29] or C–H saturation in the near-IR region [30]. Although such techniques provide a powerful analytical tool for measuring degree of conversion (especially, the near-IR methods, which allow dynamic transmission measurements in realtime), complications arise comparing filled resin systems, since differences in silane concentrations between materials may significantly affect the final conversion value [31]. In summary, a significant linear correlation (R2 = 0.992) was established between BHT concentration and time/speed of reaction for each method of characterization. This infers that the speed of polymerization reaction can be strongly influenced by BHT concentration. The three different methods showed in accordance that this linear relationship should allow for the prediction of the speed of reaction for photoactive resin systems and, if correlated to physical properties, may allow forecasting material properties on a fine scale. The three different methods, which determine the change in light
570
d e n t a l m a t e r i a l s 2 6 ( 2 0 1 0 ) 565–570
transmission (LT), reaction enthalpy (DSC) and ion viscosity (DEA) allow for monitoring the complex reaction of unfilled methacrylate-based systems.
5.
Conclusions
Improvements in light-activated resins and adhesive restorative techniques are providing better, less invasive methods for replacing lost tooth tissues. Adequate polymerization in depth is fundamental to the clinical success of these restorative materials. By better understanding the dynamics of the polymerization process and the factors governing the nature and extent of polymerization it may ultimately be possible to tailor new material formulations to better suit the diverse needs of restorative clinical practice. Traditional methods of measuring resin conversion only offer limited insight into how final mechanical properties of a restorative resin ensue. In this work three methods, which complement and offer good alternatives to classical conversion monitoring techniques, were compared in regard to their ability to monitor the photopolymerization process of a model unfilled resin mixture. A highly significant correlation was established reaction kinetics of model resin mixtures as influenced by level of inhibitor concentration and polymerization reaction rate as indirectly monitored by light transmission changes, reaction enthalpy (UV-DSC) and ion viscosity (DEA).
references
[1] Braga RR, Ferracane JL. Contraction stress related to degree of conversion and reaction kinetics. J Dent Res 2002;81: 114–8. [2] Rosentritt M, Behr M, Leibrock A, Handel G. Veneering composites—a thermoanalytical examination. J Mater Sci Mater Med 1999;10:91–8. [3] Lovell LG, Berchtold KA, Elliott JE, Lu H, Bowman CN. Understanding the kinetics and network formation of dimethacrylate dental resins. Polym Adv Technol 2001;12:335–45. [4] Jung H, Friedl KH, Hiller KA, Haller A, Schmalz G. Curing efficiency of different polymerization methods through ceramic restorations. Clin Oral Invest 2001;5:156–61. [5] Rosentritt M, Behr M, Knappe S, Handel G. Dielectric analysis of light-curing dental restorative materials—a pilot study. J Mater Sci 2006;41:2805–10. [6] Kim H, Char K. Dielectric changes during the curing of epoxy resin based on the diglycidyl ether of bisphenol A (DGEBA) with diamine. Bull Korean Chem Soc 1999;20: 1329–34. [7] Kranbuehl D, Delos S, Yi E, Mayer J, Jarvie T. Dynamic dielectric analysis: non-destructive material evaluation and cure cycle monitoring. Polym Eng Sci 1986;26:338–45. [8] Komatsu K, Nemoto K, Horie K. Transmittance of light-cured composite resins. 1. Measurement of transmittance with time. Shika Zairyo Kikai 1990;9:102–11. [9] Harrington E, Wilson H, Shortall A. Light-activated restorative materials: a method of determining effective radiation times. J Oral Rehabil 1996;23:210–8.
[10] Ogunyinka A, Palin WM, Shortall AC, Marquis PM. Photoinitiation chemistry affects light transmission and degree of conversion of curing experimental dental resin composites. Dent Mater 2007;23:807–13. [11] Shortall AC, Palin WM, Burtscher P. Refractive index mismatch and monomer reactivity influence composite curing depth. J Dent Res 2008;87:84–8. [12] Breschi L, Cadenaro M, Antoniolli F, Sauro S, Biasotto M, Prati C, et al. Polymerization kinetics of dental adhesives cured with LED: correlation between extent of conversion and permeability. Dent Mater 2007;23:1066–72. [13] Zahouily K, Decker C, Kaisersberger E, Gruener M. Real-time UV cure monitoring. Eur Coat J 2003;11:245–9. [14] Feng L, Suh BI. Gaps between filler and resin induced by polymerization shrinkage. J Dent Res 2009 [87th general Session IADR, Miami, FL., USA. Abstr #1016]. [15] Feng L, Suh BI. A mechanism on why slower polymerization of a dental composite produces lower contraction stress. J Biomed Mater Res Part B: Appl Biomater 2006;78B:63–9. [16] Feng L, Suh BI. Exposure reciprocity law in photopolymerization of multi-functional acrylates and methacrylates. Macromol Chem Phys 2007;208:295–306. [17] Asmussen E, Peutzfeldt A. Polymer structure of a resin composite in relation to distance from the surface. Eur J Oral Sci 2003;111:277–9. [18] Oster G, Yang N. Photopolymerisation of vinyl monomers. Chem Rev 1968;68:125–51. [19] Decker C, Moussa K. A new class of highly reactive acrylic monomers. 1. Light-induced polymerisation. Makramol Chem Rapid Commun 1990;11:159–67. [20] Navarrete M, Pineda J, Vera-Graziano R. Multifractality in the copolymerization of Bis-GMA/TEGDMA by pulsed photoacoustics. J Appl Polym Sci 2009;111:1199–208. [21] Odian G. Principles of polymerization. 4th ed. New York: Wiley; 2004. p. 465–533. [22] Floyd CJE, Dickens SH. Network structure of Bis-GMA- and UDMA-based resin. Dent Mater 2006;22:1143–9. [23] Loshaek S, Fox TG. Cross-linked polymers. I. Factors influencing the efficiency of cross-linking in copolymers of methyl methacrylate and glycol dimethacrylates. J Am Chem Soc 1953;75:3544–50. [24] Watts DC. Reaction kinetics and mechanics in photo-polymerised networks. Dent Mater 2005;21:27–35. [25] Trujillo M, Newmann SM, Stansbury JW. Use of near-IR to monitor the influence of external heating on dental composite polymerisation. Dent Mater 2004;20:766–77. [26] Lovell LG, Newman SM, Donaldson MM, Bowman CN. The effect of light intensity on double bond conversion and flexural strength of a model, unfilled dental resin. Dent Mater 2003;19:458–65. [27] Palin WM, Senyilmaz DP, Marquis PM, Shortall AC. Cure width potential for MOD resin composite molar restorations. Dent Mater 2008;24:1083–94. [28] Decker C. Kinetic study of light-induced polymerization by real-time UV and IR spectroscopy. J Pol Sci Part A: Pol Chem 1992;30:913–28. [29] Palin WM, Fleming GJ, Burke FJ, Marquis PM, Randall RC. Monomer conversion versus flexure strength of a novel dental composite. J Dent 2003;31:341–51. [30] Stansbury JW, Dickens SH. Determination of double bond conversion in dental resins by near infrared spectroscopy. Dent Mater 2001;17:71–9. [31] Halvorson RH, Erickson RL, Davidson CL. The effect of silane content on conversion of resin-based composite. Dent Mater 2003;19:327–33.