Elsevier

Dental Materials

Volume 21, Issue 8, August 2005, Pages 721-730
Dental Materials

How filler properties, filler fraction, sample thickness and light source affect light attenuation in particulate filled resin composites

https://doi.org/10.1016/j.dental.2005.01.002Get rights and content

Summary

Objective

It was hypothesized that by standardizing variables such as light sources, filler types and filler surface treatment, it should be possible to use Beer–Lambert's law to predict light absorption in visible light-cured dental composites.

Methods

Mixture of 50 wt% bisGMA and 50 wt% TEGDMA to which a photo-initiator (0.35 wt% champhorquinone) and a co-initiator (0.7 wt% of dimethylaminoethylmethacrylate) was prepared. Three different filler types (HBB, SBB and KU) were added to that mixture in eight different volume percentage. Filler particles were either silane surface treated or not. Specimens were made with thicknesses of 1–5 mm. Total number of 1200 specimens were made for this study. Light transmission was obtained for halogen source and laser lights, which made the number of observations 2400. The absorbance values of the different materials were analyzed in Matlab with respect to the differences in filler fraction and sample thickness.

Results

The obtained results revealed that of the two light sources, more light was absorbed by the composite when the laser light was used. Among different filler types, the HBB filler absorbed most light and the KU filler the least. There were significant differences (p<0.05) in light absorption between all three filler types.

Significance

By comparing the modeled surfaces generated by Matlab for different materials it was possible to determine how different variables such as filler type, filler surface treatment and light source affect light attenuation. The characteristic of incident light affected the light absorbance, meaning that not only the composite's composition needs to be considered in light absorption studies of dental composites.

Introduction

Visible light-cured dental composites (VLCDC) are widely used as direct filling materials because of their ‘snap-on-command’ curing mechanism [1]. VLCDC are partly translucent and scatter light. Light penetration decreases with increased material thickness [2]. This is due to the absorption and scattering of light by fillers and other additives [3]. Unfortunately, light scattering leads to limited depth of cure [4], [5]. Limited depth of cure has been pointed out by several researchers [6], [7] as a major clinical drawback with VLCDC, because unpolymerized or partially polymerized material can lead to poor mechanical properties, poor dentin bonding and eventually initiate pulp reactions [8].

By measuring of light transmission through experimental and commercial VLCD composites some investigators have developed mathematical relationships between different factors affecting light transmission [2], [5], [9], [10], [11], [12]. Further studies have shown that light transmission affects depth of cure of the composite [2], [4], [5], [13], [14], [15]. In 1980, Cook claimed that the depth of cure is mainly depended on the composition of the composite, the irradiance light characteristics and the exposure time [9]. Later, O'Keefe et al. [16] showed that there was a wavelength dependency of light transmission through light cured composites.

Because of the above aspects, light transmission of the composite and irradiance of the light source are important factors to consider when one wants to improve depth of cure. In practice different commercial VLCDC containing different matrices and filler particles are used and different light sources are available to cure these composites. Due to these variations, it is important to understand how different material components as well as different light sources affect light penetration through VLCDC.

One approach to enhance our understanding of light penetration in dental composites is to use Beer–Lambert's law. According to this law, if a beam of monochromatic radiation passes through a medium its power attenuates based on the following equationP=P0(1RF)exp(αd),where P0 is the initial optical power, RF is the total Fresnel reflectance coefficient, α is the attenuation coefficient and d is the thickness of the sample [17]. Forming the quotient P/P0 and taking the natural logarithm result in the expressionln(P/P0)=αd+ln(1RF),describing the loss in optical power due to reflections, absorption and scattering upon transmission through the medium, in our case the VLCDC. As a consequence of Beer–Lambert's law, Eq. (2) becomes a linear function where the constant factor results in the efficient Fresnel reflectance coefficient and the inclination give by the attenuation coefficient. The attenuation coefficient, α, should further be divided into an absorption coefficient, αa, and scattering coefficient, αs, which express byα=αa(Vr)+αs(Vf,Δn,r,S),where the dependence of resin volume fraction, Vr, the filler volume fraction, Vf, relative refractive index, Δn, particle radius, r, and surface treatment, S, are variables affecting the scattering coefficient.

The photo-initiator, e.g. camphoroquinon (CQ), has an absorption peak at about 467.5 nm which is shown in Fig. 1. No other absorption bands are present in either the resin or the filler particles within the spectrum of interest wherefore the absorption coefficient αa is assumed to depend only on the matrix volume fraction. This dependency will be predominantly linear wherefore the absorption coefficient is modeled by the first two terms in its Taylor expansion that gives,αa(Vr)αa+αaVr+,where αa is a higher order term and other higher order terms are assumed negligible. Eq. (4) can be written in the form of Eq. (4′) since the relation between Vr and Vf represents as: Vr=1−Vf,αa(Vr)αa+αaαaVf+The scattering coefficient, αs, of the filler will vary even more due to all the variables indicated in Eq. (1). In the following, however, we will treat each type of filler particle separately and only consider the variation due to filler volume fraction and sample thickness for each specific material combination. By further approximation, it is assumed that the scattering coefficient is relatively consistent among the volume fractions being tested. The scattering coefficient may therefore be approximated by its Taylor expansion around the point Vf=0, giving the expressionαs(Vf)αs+αsVf+,where αs and αs are two new constants. The higher order terms of the Taylor expansion are assumed negligible. Accordingly, Eq. (2) can be rewritten asln(P/P0)=(αa+αaαaVf+αs+αsVf)d+ln(1RF)which is rewritten asln(P/P0)=(αa+αa+αs)d+(αaαs)Vfd+ln(1RF).

Eq. (6) can be simplified asZ=A+Bd+CdVf,where Z=ln(P/P0), A=ln(1−Rf) or reflection term, B=(αa+αa+αs) or absorption plus scattering factor and C=αaαs or a factor showing the difference between higher-order absorption and scattering terms. The term CdVf depends on filler volume.

Based on the above relationship, we hypothesize that by standardizing variables such as light sources, filler types and filler surface treatment, it should be possible to employ Eqs. (6), (7) for predicting light absorption in VLCD composites. Accordingly, the objective with this study is to test whether the light absorption in VLCDC can be mathematically modeled by using Eqs. (6), (7) or not.

Section snippets

Composition of the experimental materials

The resin system used in this study consisted of a mixture of 50 wt% bisGMA (2,2-bis[4-(2-hydroxy-3-methacrylyloxy-propoxy)-phenyl] propane) and 50 wt% TEGDMA (triethyleneglycol dimethacrylate) to which a photo-initiator (0.35 wt% champhorquinone) and a co-initiator (0.7 wt% of dimethylaminoethylmethacrylate [DMAEMA]) had been added. Different filler types at different volume fractions were added to that mixture. Two of the filler particle types (SBB and HBB) consisted of spherical Ba–Al–B–Si glass

Refractive index determinations of matrix and filler systems

The obtained results from refractive index measurements for both the resins and the filler particles were: bisGMA/TEGDMA=1.5020 (0.0002), γ-methacryloxypropyl-trimethoxysilane=1.4297 (0.0002), SBB=1.5509 (0.0002), HBB=1.5481 (0.0001), and KU=1.5454 (0.0002). It can be confirmed from above results that all filler particles had significantly higher refractive indices than the matrix. Of the filler particles, however SBB shows biggest difference with respect to the matrix. While KU presents the

Discussion

Based on our approach with respect to Lambert–Beer's law to develop a model and the outcomes of our regression analysis of the experimental results, it seems reasonable to conclude that theoretical predictions and experimental findings coincide. That finding, by itself, is not surprising because Beer-Lambert's law is well established for solutions absorbing radiation, where absorption depends on concentration and path-length of light [5], [18]. Watts and Cash [5] also emphasized that surface

Acknowledgements

The curing unit was generously donated by Dentsply (Spectrum™ 800, Dentsply, De-Trey, Germany). Dr Göran Manneberg, Department for Physics, Royal Institute of Technology, Stockholm, Sweden is highly acknowledged for his help with refractive index measurements for glass particles. This investigation was supported by the Swedish Engineering Research Council.

References (22)

  • F.A. Rueggeberg et al.

    Correlation of parameters used to estimate monomer conversion in a light-cured composite

    J Dent Res

    (1988)
  • Cited by (135)

    View all citing articles on Scopus
    View full text