Abstract
Objectives
Optical properties of an object are determined visually or instrumentally. Although instrumental measurement provides objective and quantitative color coordinates, these values vary by the measurement method such as specimen and background conditions, instrument settings and illuminant. The objective of this study was to review the influence of the measurement method on the instrumental color coordinates of esthetic dental restorative materials.
Methods
Published reports on the measurement method dependent color variations of esthetic restorative materials were reviewed.
Results
Surface roughness influences the color coordinates differently by the surface roughness range and the measurement geometry. Specimen thickness and the kind of illuminant influence the color coordinates, and the influence of background varied by specimen thickness. Therefore, the specular component excluded (SCE) geometry that reflects the surface condition of specimens is suggested as the correct measurement geometry. Surface roughness, thickness and layering of specimens, and the kind of illuminant should be stipulated in each measurement. There should be a standard for the color and gloss of the background.
Significance
Variables in instrumental color measurements should be stipulated to obtain consistent and comparable color coordinates, and a general guideline for instrumental color measurement of dental materials should be established.
1
Introduction
Understanding the optical behavior of esthetic restorative materials is essential to obtain high quality esthetics . Optical properties such as color coordinates and other related parameters of teeth and esthetic restorations are determined visually or instrumentally, and it is generally agreed that visual evaluation is inconsistent and unreliable, while instrumental measurement provides objective and quantified values . However, instrumental color coordinates varied by specimen and background conditions , instrument settings and illuminant . Kind of color measurement instrument such as colorimeter (CM), spectrophotometer (SP) or spectroradiometer (SR) also influenced color coordinates .
Compilation of a data pool of optical parameters that would enable calculation models to be used as a basis for the optimization of the optical approximation of the natural teeth was initiated based on instrumental color measurements . To make a reliable data pool, instrumental color measurement methods should be standardized to compile consistent data regardless of testing laboratories. Although it is generally agreed that instrumental color coordinates vary by measurement method, and many studies reported the influence of measurement methods on color variations, there have been no identified studies on the general review for this issue.
For instrumental color measurement, the CIELAB system is generally used . In this system, the CIE L* coordinate ranges from 0 to 100 and represents the lightness; the CIE a* coordinate ranges from −90 to 70 and represents greenness (positive a* ) and redness (negative a* ); and the CIE b* coordinate ranges from −80 to 100 and represents yellowness (positive b* ) and blueness (negative b* ) . The advantage of this system over the Munsell system (hue, chroma and value color attributes) is that the CIE color coordinates are evenly spaced in terms of visual perception, so that the spectral readings can be correlated with subjective observations . However, it was reported that the instrumental color coordinates were inconsistent with a commonly used visual shade guide, although general trends were similar in many cases . Therefore, correlations between the visual and instrumental colors still need further studies .
The objective of this study was to review the influences of specimen conditions, instrument settings and illuminant on the instrumental color coordinates of esthetic restorative materials. Although the measurement methods influence the color coordinates in complicated and interrelated ways, influence of each variable would be evaluated separately for straightforward understanding. Translucency, fluorescence and opalescence of an object also influence the color perception, and the same variables encountered in color measurement would influence theses parameters, but those need another review.
This review consists of three subtopics: (1) Influence of specimen conditions such as surface roughness/thickness of specimen and that of background condition; (2) Influence of instrument settings such as measurement area and measurement geometries of the specular component excluded (SCE) and the specular component included (SCI); (3) Influence of the kind of illuminant (metamerism).
2
Influence of specimen condition
2.1
Surface roughness
Surface roughness of a specimen influences the instrumental color coordinates. Resin composites with a rough surface appeared lighter and less chromatic than those with a smooth surface under the condition of diffuse reflectance measured by SP , and the inverse of the contrast-gloss ratio was related to the surface roughness of resin composites . Based on a study of the effects of polishing on color, color differences among five resin composites and four polishing methods were found to be 0.2–1.1 <SPAN role=presentation tabIndex=0 id=MathJax-Element-1-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEb’>ΔEbΔEb
Δ E b
units, which were lower than a perceptible limit <SPAN role=presentation tabIndex=0 id=MathJax-Element-2-Frame class=MathJax style="POSITION: relative" data-mathml='(ΔEab≤1)’>(ΔEab≤1)(ΔEab≤1)
( Δ E a b ≤ 1 )
. In another study , color differences between polished and non-polished resin composite specimens were determined by CM; as results, they were 1.3–1.8 <SPAN role=presentation tabIndex=0 id=MathJax-Element-3-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units (Ra – average surface roughness value: 0.47–0.96 μm) for nanohybrid, 1.4–2.0 <SPAN role=presentation tabIndex=0 id=MathJax-Element-4-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units (Ra: 0.52–1.30 μm) for microhybrid and 1.5–2.4 <SPAN role=presentation tabIndex=0 id=MathJax-Element-5-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units (Ra: 0.58–1.83 μm) for hybrid; all of which were within an acceptable range <SPAN role=presentation tabIndex=0 id=MathJax-Element-6-Frame class=MathJax style="POSITION: relative" data-mathml='(ΔEab<3.7)’>(ΔEab<3.7)(ΔEab<3.7)
( Δ E a b < 3.7 )
. Based on these studies, it was confirmed that surface roughness range of resin composites, after polishing or other treatments, significantly influences the instrumental color.
Influence of the surface roughness of resin composites on color was different by the measurement geometry. Specular reflected light at the specimen is excluded in the SCE geometry and is included in the SCI geometry . Color differences between resin composites with non-polished (Ra: 0.15–0.24 μm) and polished (Ra: 0.16–0.21 μm) surfaces were 1.7–3.3 and 0.4–2.6 <SPAN role=presentation tabIndex=0 id=MathJax-Element-7-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units, respectively, under the SCE and the SCI geometry; meanwhile, color differences between the polished specimens measured under the SCE and the SCI geometries were 0.5–2.5 <SPAN role=presentation tabIndex=0 id=MathJax-Element-8-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units . In another study , color differences between non-polished (Ra: 0.05–0.10 μm) and 600-grit SiC paper polished (Ra: 0.20–0.28 μm) resin composites were 4.2–6.0 and 0.3–1.5 <SPAN role=presentation tabIndex=0 id=MathJax-Element-9-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units, respectively, under the SCE and the SCI geometry. Based on these studies, it was confirmed that color differences between the specimens with different surface roughness values measured under the SCE geometry were significantly higher than those under the SCI geometry. Fig. 1 shows the differences in color coordinates of a marketed resin composite by the surface roughness under the SCE and the SCI geometries . Before polishing, the CIE L* value under the SCE geometry was significantly lower than that under the SCI geometry. The CIE L* values increased after polishing under the SCE geometry; however, they did not change under the SCI geometry. Fig. 2 shows the color differences of resin composites by the measurement geometry. Before polishing, the color differences were high (3.8–5.9 <SPAN role=presentation tabIndex=0 id=MathJax-Element-10-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units); however, those after polishing were lower than 1.6 <SPAN role=presentation tabIndex=0 id=MathJax-Element-11-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units. Based on these studies, it was confirmed that color measurement under the SCE geometry reflects the surface roughness of specimens.
As to ceramics, polishing treatments of feldspathic porcelains significantly affected the Ra values (0.82–1.17 μm), but had no significant effect on the color (color difference compared with the control group: 0.6–0.9 <SPAN role=presentation tabIndex=0 id=MathJax-Element-12-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units) . As to the influence of polishing in a feldspathic porcelain , color difference determined by CM under the SCI geometry ranged from 1.0 to 3.4 <SPAN role=presentation tabIndex=0 id=MathJax-Element-13-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units (Ra value of control group: approximately 1.5 μm; polished groups: 1.4–3.1 μm); besides, surface roughness of specimens and color difference with the control group had a linear correlation. Therefore, the same conclusion to that for resin composites was confirmed for ceramics.
As to the combined influence of the surface roughness and the measurement geometry on the color of ceramics, color differences by these factors were determined . Specimens were not-polished (Ra: 5.40 μm), polished with 200, 400, 1000, and 1500-grit silicon carbide papers (Ra: 0.61, 0.29, 0.21, and 0.17 μm, respectively) or glazed (Ra: 0.30 μm). Fig. 3 shows the differences in color coordinates by the surface roughness under the SCE and the SCI geometries . Under the SCE geometry, the CIE L* value after glazing was significantly lower than those after polishings. Fig. 4 shows the color differences between glazed surface and other surface conditions. Color differences measured under the SCE geometry were higher than those under the SCI geometry (2.6–4.7 vs. 0.9–1.6 <SPAN role=presentation tabIndex=0 id=MathJax-Element-14-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units).
In conclusion, the surface roughness of esthetic restorative materials influences the instrumental color coordinates differently by the surface roughness range and the measurement geometry. Generally, the SCE geometry, which reflects the surface conditions of specimens, is suggested as the correct measurement geometry. Besides, the surface roughness or condition of specimens should be stipulated in each measurement.
2.2
Specimen thickness
Several studies were performed to determine the influence of specimen thickness of resin composites on color. This variable is associated with the infinite optical thickness ; when the thickness of a specimen is thinner than this value, measured color would be influenced by background condition. As to the influence of specimen thickness on the SP-based color, lightness and chroma increased over a black background but decreased over a white background as the thickness of resin composites increased from 1.3 to 3.9 mm . Color of resin composites of different thicknesses on a highly chromatic background was successfully predicted by applying the Kubelka–Munk theory, and the infinite optical thickness of this material increased as wavelength increased . Color difference between different thickness-based (0.5–3.0 mm) dentin shade resin composites was as high as 5.1 <SPAN role=presentation tabIndex=0 id=MathJax-Element-15-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units . The infinite optical thickness of resin composites ranged from 4.2 to 6.7 mm and that of an unfilled resin was 5.7 mm , and those for varied shades of resin composites ranged from 3.0 to 5.9 mm .
Color accuracy of prosthetic resin composites was determined by CM to evaluate the relationship between specimen thickness variation (0.5–3.0 mm) and color reproducibility . As results, the CIE L* value consistently decreased as the thickness increased over a white background, although there was no obvious correlation between the thickness of the specimen and either of the CIE a* or b* values. Since the color difference between the 2.5 and 3.0 mm thick specimens were the lowest <SPAN role=presentation tabIndex=0 id=MathJax-Element-16-Frame class=MathJax style="POSITION: relative" data-mathml='(ΔEab<2.0)’>(ΔEab<2.0)(ΔEab<2.0)
( Δ E a b < 2.0 )
, a thickness of at least 2.5 mm was found to be required for the acceptable color reproduction of composite materials. Although the concept determined in this study was different from that of the infinite optical thickness, this thickness threshold would be used as a recommended thickness for color specimens. However, the influence of translucency of an object on the thickness-dependent color variations should be considered.
Color reproduction of esthetic restorative materials by thickness would be evaluated based on four visual categories for color perception such as perfect reproduction, not-perceptible reproduction, acceptable reproduction and not-acceptable reproduction. If the thickness is thicker than the infinite optical thickness, perfect reproduction of the assigned color would be possible. If thinner than this value, perceptibility or acceptability in color reproduction should be evaluated. Considering these, a minimum thickness of 2.5 mm could be a guideline for the acceptable color reproduction.
As to the influence of the specimen thickness of ceramics, opaque and body porcelain specimens were made in different thicknesses and the color was compared with the shade guide tabs . Since the thickness of the opaque and body porcelains to produce optimum color varied by the shade, manufacturers’ recommendations for each shade should include the optimum required thickness for each porcelain. It was also reported that a change of enamel porcelain thickness from 0.6 to 0.3 mm resulted in three units changes in the CIE L* and the metric chroma and 4° change in the hue angle .
Influence of the thickness ratio of opaque and translucent porcelains as well as that of core and veneer porcelains on the color of all-ceramic systems were determined . The results indicated that small changes in the thickness ratio perceivably influenced the final color <SPAN role=presentation tabIndex=0 id=MathJax-Element-17-Frame class=MathJax style="POSITION: relative" data-mathml='(ΔEab>1)’>(ΔEab>1)(ΔEab>1)
( Δ E a b > 1 )
, and the redness coordinate (CIE a* ) correlated more strongly with the thickness than the yellowness coordinate ( b* ) . Besides, core thickness, veneer thickness and their interaction exhibited significant influence on several color coordinates . For the influence of dentin ceramic thickness (0.5–1.5 mm) on the color of all-ceramics, the CIE L* and a* values were affected by the thickness; however, the CIE b* value was not affected by the thickness . Influence of dentin ceramic thickness on the color of two all-ceramics was also evaluated by SP, and it was concluded that the color coordinates were influenced by the thickness differently . Depending on the kind of ceramic and the measurement method, the influence of ceramic thickness on color was different. Generally, ceramic thickness influenced the CIE L* and a* values, but not the CIE b* value. Although these previous studies were based on clinical scenarios, the color of ceramics should be evaluated considering the influences of thickness, thickness ratios of component layers and other variables such as translucency.
2.3
Background condition
Black and white backgrounds are generally used to assess the influence of background on the final color of restorative materials . Color of resin composites before and after polishing was measured over a white background (reflectance = 91.57%) and a light trap (reflectance = 0.01%) by SP . As results, the color differences by the background were 2.4–11.6 <SPAN role=presentation tabIndex=0 id=MathJax-Element-18-Frame class=MathJax style="POSITION: relative" data-mathml='ΔEab’>ΔEabΔEab
Δ E a b
units, and the background influenced three color coordinates differently depending on the material and the specimen condition. Therefore, background significantly influenced the color coordinates and the color differences between different surface conditions. Fig. 5 shows the color difference of resin composites by surface condition over two backgrounds . Color differences between different surface conditions varied by the background, and those over a white background were significantly higher than those over a light trap except a few cases. Since the light trap can eliminate the influence of variations at the background, it was suggested that measured color over the light trap might be the color of a material itself .
