d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1211–1221

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.intl.elsevierhealth.com/journals/dema

Effect of light curing units on the polymerization of

bulk fill resin-based composites

Carlos Alberto Kenji Shimokawa a , Míriam Lacalle Turbino a ,

Marcelo Giannini b , Roberto Ruggiero Braga c , Richard Bengt Price d,∗
a University of São Paulo, School of Dentistry, Restorative Dentistry, São Paulo, São Paulo, Brazil
b University of Campinas, Piracicaba Dental School, Restorative Dentistry, Piracicaba, São Paulo, Brazil
c University of São Paulo, School of Dentistry, Biomaterials and Oral Biology, São Paulo, São Paulo, Brazil
d Dalhousie University, Faculty of Dentistry, Dental Clinical Sciences, 5981 University Ave., Halifax, Nova Scotia,

B3H 4R2, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Objective. To determine the potential effect of four different light curing units (LCUs) on the
Received 14 February 2018 curing profile of two bulk fill resin-based composites (RBCs).
Received in revised form Methods. Four LCUs (Bluephase 20i, Celalux 3, Elipar DeepCure-S and Valo Grand) were used
15 April 2018 to light cure two RBCs (Filtek Bulk Fill Posterior Restorative and Tetric EvoCeram Bulk Fill).
Accepted 2 May 2018 The effective tip diameter, radiant power, radiant emittance, emission spectrum and light
beam profile of the LCUs were measured. Knoop microhardness was measured at the top
and bottom surfaces of RBC specimens that were 12-mm in diameter and 4-mm deep (n = 5).
Keywords: The distribution of the spectral radiant power that was delivered to the surface of the spec-
Dental curing lights imen and the light transmission through the 4-mm thick specimens was measured using
Composite resins an integrating sphere. Two-way ANOVA and Tukey tests (˛ = 0.05) were applied.
Hardness tests Results. The Valo Grand produced the most homogeneous microhardness across the surfaces
Polymerization of the RBCs (p > 0.05). When the Celalux 3, Bluephase 20i and Elipar DeepCure-S lights were
used, the center of the specimens achieved greater hardness values compared to their outer
regions (p < 0.05). Approximately 10% of the radiant power delivered to the top reached the
bottom of the specimen, although almost no violet light passed through 4 mm of either RBC.
A positive correlation was observed between the radiant exposure and microhardness.
Significance. The characteristics of the LCUs influenced the photoactivation of the RBCs. The
use of a wide tip with a homogeneous light distribution is preferred when light curing RBCs
using a bulk curing technique.
© 2018 Published by Elsevier Inc. on behalf of The Academy of Dental Materials.

1. Introduction

The manufacturers of bulk fill resin-based composites (RBCs)

claim that it possible to successfully photo-cure 4 or even

∗
Corresponding author.
E-mail addresses: [email protected] (C.A.K. Shimokawa), [email protected] (M.L. Turbino), [email protected] (M. Gian-
nini), [email protected] (R.R. Braga), [email protected] (R.B. Price).
https://doi.org/10.1016/j.dental.2018.05.002
0109-5641/© 2018 Published by Elsevier Inc. on behalf of The Academy of Dental Materials.
1212 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1211–1221

Table 1 – Information provided by the manufacturers about their light curing units.
Light curing unit Manufacturer Stated irradiance Emission Stated tip
(mW/cm2 ) spectrum (nm) diameter (mm)
Bluephase 20i (sn: P626170S548780) Ivoclar Vivadent, Amherst, NY, USA 1200 (±10%) 385–515 8
Celalux 3 (sn: 1637091) VOCO, Cuxhaven, Germany approx. 1300 450–480 8
Elipar DeepCure-S (sn: 933112-001111) 3M Oral Care, St. Paul, MN, USA 1470 (−10%/+20%) 430–480 10
Valo Grand (sn: MFG3277-5) Ultradent, South Jordan, UT, USA 1000 (±10%) 395–480 12

5-mm increments of RBC instead of the customary 2 mm 4. There will be no correlation between the radiant exposure
increment. While this approach may introduce less voids delivered and the microhardness results.
between each increment and reduce overall treatment times,
the ability to adequately photo-cure such a large volume of
2. Methods
RBC in one exposure is a concern.
Many types of light curing units (LCUs) are available for
The details of the four light emitting diode (LED) LCUs
dentists to purchase. The spectral radiant power, light tip
(Bluephase 20i, Celalux 3, DeepCure-s, Valo Grand) are
diameter, and radiant emittance (tip irradiance) from these
reported in Table 1. These LCUs were used to photo-cure two
LCUs are often different and this may have a negative impact
commonly used bulk fill resin-based composites: Filtek Bulk
on the ability of these units to photo-cure RBCs [1–3]. Although
Fill Posterior Restorative – A2 (3M Oral Care, St. Paul, MN, USA)
camphorquinone is currently used in all RBCs, some RBCs
and Tetric EvoCeram Bulk Fill – IVA (Ivoclar Vivadent, Amherst,
also include alternative photoinitiators requiring activation by
NY, USA).
lower wavelengths of light. These RBCs benefit from the use
of a broad spectrum light source that delivers violet light as
well as blue light [4–6]. However, since the lower wavelengths 2.1. Characterization of the LCUs
of light (in the violet range) do not penetrate as far into the
RBC as the longer wavelengths of blue light, the alternative The light tip diameter, radiant power, radiant emittance, emis-
photoinitiators in the deeper regions may receive an insuffi- sion spectrum and light beam profile of the four LCUs were
cient amount of the lower wavelengths of light and the RBC analyzed. Both the external and the internal fiber to fiber
may remain undercured in the deeper parts of the RBC [7,8]. tip diameters were measured with a digital caliper (Mitutoyo,
The homogeneity, or the lack of thereof, of the emitted light Kawasaki, Kanagawa, Japan). The effective tip diameter from
beam with respect to the radiant emittance and spectrum of where light could be emitted was taken to be the maximum
light across the RBC surface, can affect the polymerization of fiber to fiber distance. For the Valo Grand, the diameter of the
the RBC. Additionally, the effective light tips of some LCUs lens was used as the effective tip diameter (Fig. 1).
are 7 mm or less in diameter, and multiple light exposures Five measurements of the total radiant power (mW) emit-
are required to fully cover the entire restoration with light ted between 350 and 550 nm and spectral radiant power
[3,9]. This is not a concern if an incremental RBC filling tech- (mW/nm) from each LCU were measured using a 6 integrating
nique is used, since even small tips are able to cover the entire sphere (Labsphere, North Sutton, NH, USA) that was attached
increment. However, with the introduction of bulk fill RBCs to a fiber-optic spectrometer USB4000 (Ocean Optics, Dunedin,
manufacturers have promoted filling the entire cavity with a IL, USA). An internal traceable light source inside the SCL-600
single portion of RBC and then curing the restorations with sphere (Labsphere) was used to calibrate the system before the
one light exposure. When LCUs with small diameter light tips measurements were made. When measuring the LCUs, the tip
are used to light cure large restorations in molar teeth, any of the LCU was positioned 2-mm from the 16-mm aperture
RBC that is not covered by the light tip may be inadequately into the sphere and all the light that was emitted from the
polymerized. LCU was captured by the sphere. The radiant emittance for
Several studies have evaluated the depth of cure of bulk fill each LCU was calculated as the quotient of the radiant power
RBCs at the center of the specimens [10,11], but none has ver- and the effective tip diameter values shown in Fig. 1.
ified the adequacy of both the depth and the width of cure of The light beam profiles from the LCUs were evaluated using
the bulk fill RBCs when using different LCUs. Therefore, this a Laser Beam Profiler (Ophir Spiricon, Logan, UT, USA). This
study determined the light output characteristics of four dif- device uses a digital camera with a 50-mm focal distance lens
ferent LCUs and investigated their ability to light cure two bulk (USB-L070, Ophir Spiricon) positioned at a fixed distance from
fill RBCs in one exposure. The null hypotheses were: a 40-degree holographic screen (Edmund Industrial Optics,
Barrington, NJ, USA). A custom-made blue filter (International
Light Technologies, Peabody, MA, USA) was used to flatten the
spectral response of the camera and BeamGage v.6.6 software
1. There will be no difference between the light emitting char- (Ophir Spiricon) was used to determine the photonic count
acteristics of the four LCUs; received by each camera pixel. With the 40◦ holographic screen
2. The effective tip diameters and light beam homogeneity of positioned 2-mm away from the light tip, the distribution of
the LCUs will not influence the curing profile of the two the radiant power was recorded for each LCU. In addition, to
bulk fill RBCs; evaluate the distribution of the violet light (∼400 nm) only, the
3. The emission spectra of the LCUs will not influence the beam profile was also recorded through a 400 nm narrow band-
curing profile of the two bulk fill RBCs; pass filter (Edmund Industrial Optics) with a full width half
 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1211–1221 1213

Fig. 1 – Measured effective tip diameters of the four light curing units.

maximum range of ±5 nm. Prior to making any recordings, ness testing, the RBCs were covered with a 0.14 mm thick
the effect of the ambient light was accounted for by normaliz- microscope cover glass before light curing. This microscope
ing every pixel to a similar level using the UltraCal feature in cover glass had no measurable effect on the light transmis-
the BeamGage software. The radiant power (mW) results pre- sion. Each LCU was used for 10 s to light cure the RBCs, with
viously collected with the integrating sphere together with the the tip positioned at the center of the specimen, 2-mm away
effective tip diameters were used to produce calibrated values from the top surface.
of the irradiance (mW/cm2 ) that would be received by the spec- The real-time light transmission through the specimens
imens positioned 2-mm away from the light tip. This data was was measured during photocuring using the information col-
exported to OriginPro 2017 software (OriginLab, Northampton, lected by a smaller FOIS-1 integrating sphere (Ocean Optics)
MA, USA) to make two and three dimensional images of the that was attached to a fiber-optic spectrometer USB-4000
light beam profile. (Ocean Optics). A LS-1-CAL-INT (Ocean Optics) traceable light
source was used to calibrate the system before starting the
measurements. By using a 10-mm aperture into this sphere,
2.2. Light delivered across the top surface of the RBC
just the light that passed through the 10 mm diameter of
specimens
RBC was captured, rather than including the light that also
passed through the 12-mm diameter semi-opaque Delrin ring.
To determine the distribution of the spectral radiant power
Real-time data collection was started 5 s before light activa-
that was delivered to the top surface of the specimens, a 1 mm2
tion and continued until 5 s after the end of photoactivation,
square aperture was placed over the entrance into the 6 inte-
using OceanView software (Ocean Optics). Using a scan time
grating sphere (Labsphere). Each LCU tip was placed 2-mm
of 200 ms, the real-time radiant spectral power and the total
away from this 1 mm2 aperture. Starting with the LCU cen-
radiant power (between 350 and 550 nm) that exited from the
tered over the aperture, the light entering the aperture was
bottom of two specimens from each combination of LCUs and
measured. The LCU tip was then moved in 1-mm increments
RBCs was recorded and averaged. After the RBC had been pho-
in all directions (north, south, east and west) using an X–Y
toactivated, the specimen was stored in air, in the dark, and
Axis Rack & Pinion Stage (#62-041, Edmund Optics) to record 97
at 37 ◦ C for 24 h.
individual spectral radiant power measurements, 1 mm apart,
across the light beam from each LCU. For the LCUs with narrow
tips, these measurements went beyond the edge of the light
2.4. Microhardness measurements
beam. Thus, all the light that would reach the specimens that
were 2-mm away from the tip, was captured. These measure-
After 24 h of storage, the Knoop microhardness was mea-
ments were made in duplicate for each LCU, and averaged. The
sured at the top and bottom surfaces with a microhardness
mean values were later matched to the same regions where
tester (HMV 123, Mitutoyo), applying a 15 gf for 8 s. The indents
the microhardness was measured (Fig. 2).
were made at locations corresponding to where the radiant
power had been measured through the 1 mm aperture into
2.3. Specimen preparation and real time light the integrating sphere, for a total of 97 indents across each of
transmission through the RBC the surfaces of the RBC specimens (Fig. 2). The results were
exported into the OriginPro 2017 software and converted to
For each of the four LCUs, five resin specimens were made a percentage scale that was relative to the maximum micro-
®
using each RBC in semi-opaque white Delrin molds (40 spec- hardness value achieved by any of the four tested LCUs on that
imens in total). The molds were 4-mm deep and had a 12-mm RBC. Hardness maps were made using OriginPro software with
internal diameter. To produce a flat surface suitable for hard- a small smoothing factor of 0.001.
1214 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1211–1221

Fig. 2 – Microhardness maps. (a) An example of the indentation scheme and locations where the light emitted by the LCUs
was measured. The squares represent the 1 mm2 aperture into the sphere and where one indent was made at the center of
each square. The percent of maximum hardness means are represented in the center of the squares. The circles illustrate
where the microhardness data was grouped into three regions: center, middle and outer. (b) A 3D map made from the
percent of maximum hardness at each indent location. (c) A smoothed hardness map made from the percent of maximum
hardness.

The microhardness data recorded at both top and bot- had the widest (11.6 mm). Although the Valo Grand emitted
tom surfaces of the 12 mm diameter specimens were grouped the highest radiant power, due to its large effective tip area
into three regions: center (2 mm radius, containing 9 hard- (105.7 mm2 ), it delivered the lowest radiant emittance.
ness values), middle (2 mm intermediate radius, containing Three-dimensional (3-D) beam profile results with the
28 hardness values) and outer (2 mm external radius, contain- LCU tip positioned 2-mm away from the 40◦ holographic
ing 60 hardness values), thus the microhardness values were screen, made with and without the 400 nm bandpass filter are
reported in three regions across the RBC specimens. For each reported in Fig. 3. The Celalux 3 delivered the narrowest and
specimen, the mean microhardness values of each region were the least homogeneous light beam profile. When measured
calculated to allow the center, middle and outer regions to be through the 400 nm bandpass filter, no light could be mea-
compared with each other. sured from DeepCure-S and Celalux 3. Comparing the Valo
Grand and Bluephase 20i multi-peak lights, the distribution
2.5. Statistical analysis of violet light was more homogeneous for Valo Grand (Fig. 3).
The means and standard deviations of the Knoop micro-
The top and bottom surfaces and the two RBCs were analyzed hardness results measured across the top and bottom surfaces
separately. Two-way repeated measures analysis of variance of the Filtek Bulk Fill Posterior Restorative and Tetric EvoCeram
(ANOVA) tests followed by Tukey post-hoc multiple compar- Bulk Fill specimens are reported in Tables 3 and 4, respectively.
ison tests (˛ = 0.05) were applied to the microhardness data, The smoothed hardness maps for the composites are shown
considering these factors: the three regions (center, middle or in Fig. 4.
outer) and the LCU (Bluephase 20i, Celalux 3, Elipar DeepCure- For both RBCs, Valo Grand produced a more homoge-
S and Valo Grand). Correlation tests were applied to verify neous hardness across both the top and bottom surfaces
if there was correlation between the light distribution over and the hardness values measured in the central, middle
the specimens and the microhardness data. The mean of the and outer regions at the top surfaces of the specimen were
two repeats of the radiant exposures delivered to each 1 mm not significantly different (p ≥ 0.05). This is illustrated by the
square from each LCU (on a semi-log scale) and the mean of homogeneity of the color across the hardness map for the Valo
the 5 repeats of the microhardness measurements made at Grand. In contrast, when Celalux 3, Bluephase 20i and Elipar
each surface of each RBC were used for the correlation analy- DeepCure-S lights were used, the hardness values from the
sis. central, middle and outer regions across the RBC specimens
were significantly different (p < 0.05).
At the bottom of the 4-mm thick specimens, the microhard-
3. Results
ness values depended on the location. On average, it reached
about 70% of the maximum hardness value, represented by the
The measured properties of each LCU are shown in Table 2.
green color in Fig. 4. Comparing the bottom central regions,
Two LCUs emitted a single peak of blue light (Celalux 3 and Eli-
it can be seen that the use of multi-peak LCUs resulted in
par DeepCure-S) and two LCUs (Bluephase 20i and Valo Grand)
a lower percentage of the maximum hardness values, com-
emitted multiple-peak wavelengths of light that included
pared to the specimens that had been photoactivated with
violet light below 410 nm. The Celalux 3 had the smallest effec-
single-peak blue LCUs. This effect was more evident for the
tive tip diameter (7.1 mm, area 39.6 mm2 ), while Valo Grand
 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1211–1221 1215

Table 2 – Characteristics of the light curing units used in this study.

Light curing unit Measured Measured Calculated Measured Calculated radiant
external tip effective tip effective tip radiant power emittance (mW/cm2 )
diameter (mm) diameter (mm) area (mm2 ) (mW)
Bluephase 20i 8.0 7.2 40.7 570 ± 2 1400 ± 6
Celalux 3 8.0 7.1 39.6 432 ± 4 1092 ± 11
Elipar DeepCure-S 9.8 9.0 63.6 792 ± 4 1246 ± 7
Valo Grand 15.0 11.6 105.7 1007 ± 13 953 ± 12

Fig. 3 – Three-dimensional representations of the beam profile captured with the LCU tips 2-mm away from the holographic
screen. The Bluephase 20i and Valo were also taken through the 400 nm narrow bandpass filter.

Table 3 – Means ± the standard deviations for the Knoop microhardness values measured in the central, middle and
outer regions of the specimens made with the Filtek Bulk Fill Posterior Restorative.
Top Bottom

Central Middle Outer Central Middle Outer

Bluephase 20i 70.9 ± 4.3
Aa
69.4 ± 4.2 Ab
65.7 ± 5.3 Ac
60.1 ± 2.3
ABa
58.3 ± 2.0
Bb
53.7 ± 2.5Bc
Celalux 3 69.2 ± 1.9Aa 67.3 ± 1.7Ab 62.6 ± 2.0Ac 57.9 ± 3.4Ba 56.3 ± 2.7Bb 52.4 ± 2.5Bc
DeepCure-S 70.5 ± 3.2Aa 68.9 ± 3.1Ab 66.0 ± 3.2Ac 64.3 ± 1.4Aa 62.7 ± 1.7Ab 59.7 ± 1.6Ac
Valo Grand 67.3 ± 1.5Aa 67.3 ± 1.2Aa 66.5 ± 1.1Aa 58.3 ± 3.0Ba 58.2 ± 3.0Ba 56.1 ± 2.5ABb

Different superscript uppercase letters represent statistical differences within columns (lights) and different superscript lowercase letters
represent statistical differences within rows (p < 0.05).
Center 2 mm radius; middle 2 mm intermediate radius; and outer 2 mm external radius.

Tetric EvoCeram Bulk Fill specimens, where the Celalux 3 and specimens. Of note, the ratios of violet to blue light at the
Elipar DeepCure-S specimens produced greater hardness val- top was markedly different to the ratio at the bottom of the
ues than the Bluephase 20i (p < 0.05), while the Valo Grand specimens because very little of the lower wavelengths of vio-
produced intermediate microhardness results. let light passed through the 4 mm thick RBC. Only 5.5 and
The spectral radiant power and the total radiant power of 3.4 mW of violet light passed through Filtek Bulk Fill Posterior
the light arriving at the top of the specimens and at the bot- Restorative when Bluephase 20i and Valo Grand were used,
tom of specimens after passing through the RBCs are reported respectively, while 3.1 and 2.1 mW of violet light reached the
in Fig. 5. Approximately 10 per cent of the total radiant power bottom of Tetric EvoCeram Bulk Fill for the same LCUs. The
delivered to the top of the RBCs reached the bottom of the spectral radiant power that reached the bottom of the speci-
1216 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1211–1221

Table 4 – Means ± the standard deviations for the microhardness values measured in the three different regions of the
specimens made with the Tetric EvoCeram Bulk Fill.
Top Bottom

Central Middle Outer Central Middle Outer

Bluephase 20i 62.4 ± 1.9
Aa
61.0 ± 1.7Ab
53.9 ± 2.1 Cc
45.0 ± 3.7 Ba
42.7 ± 3.4
Bb
37.2 ± 2.7Bc
Celalux 3 62.1 ± 2.6Aa 59.6 ± 1.7Ab 52.4 ± 1.5Cc 52.3 ± 1.6Aa 49.1 ± 1.1Ab 42.0 ± 0.9Ac
DeepCure-S 62.2 ± 1.9Aa 60.6 ± 2.2Ab 57.1 ± 2.2Bc 51.7 ± 3.2Aa 50.0 ± 3.6Ab 46.3 ± 3.8Ac
Valo Grand 62.5 ± 0.5Aa 62.4 ± 0.3Aa 61.8 ± 0.5Aa 48.9 ± 1.9ABa 47.9 ± 1.9Aa 45.8 ± 1.8Ab

Different superscript uppercase letters represent statistical differences within columns (lights) and different superscript lowercase letters
represent statistical differences within rows (p < 0.05).
Center 2 mm radius; middle 2 mm intermediate radius; and outer 2 mm external radius.

Fig. 4 – Beam profiles (center image) of the four LCUs and the percent of maximum microhardness achieved by the two RBCs
at the top and bottom. The circles illustrate the locations of the central, middle and outer regions across each RBC. (For
interpretation of the references to color in the text, the reader is referred to the web version of this article.)
 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1211–1221 1217

Fig. 5 – Total radiant power (mW) and spectral radiant power (mW/nm) that reached the bottom of the specimens compared
to what was delivered to the top (note the different Y-axis scales for top and bottom).

Fig. 6 – Spectral radiant power (mW/nm) that reached the bottom of the specimens at different times during the 10 s
exposure. (For interpretation of the colors, the reader is referred to the web version of this article.).

mens at different times during the 10 s exposure is shown in are shown in the scatter plot graphs (Fig. 7). A strong pos-
Fig. 6. It can be seen that for these two RBCs, the transmitted itive correlation could be observed in every case. The least
power increased during the course of the exposure. scatter in the radiant exposures delivered across the surfaces
The correlation between the radiant exposure measured and in the microhardness values was observed from the Valo
at each indentation location and the microhardness values Grand.
1218 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1211–1221

Fig. 7 – Scatter plots of the Knoop microhardness vs. radiant exposure (J/cm2 ) on a logarithmic scale. A positive linear
correlation was verified for all the conditions. (For interpretation of the colors, the reader is referred to the web version of
this article.).

polymerization stresses within the RBC [13,14]. This is thought

4. Discussion to occur because faster polymerization may not allow suf-
ficient time to relieve the stresses generated during the
The four contemporary LCUs used in this study are com-
polymerization reaction. High irradiance values will also pro-
mercially available and widely used however, they emitted
duce more radicals very quickly and this may cause more and
different quantities and qualities of light (Fig. 5). Thus, the
earlier bi-radical termination of the polymer chains, resulting
first null hypothesis was rejected. This result was anticipated
in regions where there is a lower degree of conversion of the
based on previous studies [2,3,12]. There was a difference
RBC [15].
between the measured properties and the information pro-
It is important to know the emission spectrum and the
vided by the manufacturers, which may be partly explained by
beam profile from the LCU. In the present study both the
the fact that the LCUs were measured at a distance of 2-mm
Celalux 3 and Elipar DeepCure-S emitted single-peak blue
from the sensor. Since irradiance is the quotient of the radi-
light, while the Bluephase 20i and Valo Grand emitted two
ant power received by a surface per unit area and the beam
or three emission peaks of light. For a photochemical reac-
width is affected by distance from the tip, the irradiance from
tion to take place, the first law of photochemistry, the
a curing light is also affected by distance. At the light tip, the
Grotthuss–Draper law, states that light must be absorbed by
irradiance will be the same as the radiant emittance.
the chemical substance. The second law of photochemistry,
The effective tip diameters varied among the four LCUs.
the Stark–Einstein law, states that for each photon of light
The Celalux 3 had the smallest tip diameter, 7.1 mm, and the
absorbed, only one molecule within the chemical system
Valo Grand had the largest at 11.6 mm (Table 2). Thus, the emit-
can be activated. In addition, the Planck–Einstein relation-
ting area of the Valo Grand was 2.6 times greater than the
ship means that the lower the wavelength of the photon, the
Celalux 3. As a consequence, although Valo Grand delivered
greater the energy carried by the photon [16]. These three
the highest radiant power, its radiant emittance was the low-
factors explain why two curing lights that deliver the same
est. This is because radiant emittance is dependent on the area
radiant power, but have different emission spectra, differ-
and even small changes in the effective diameter of the light
ent irradiance or spectral beam profiles will produce different
beam will lead to large differences in the calculated radiant
results. At lower wavelengths, there will be fewer photons for
emittance.
the same radiant power and, additionally, the wavelengths
Localized regions of high irradiance values across a light
emitted from the LCU may not be absorbed by the photoini-
beam may not be ideal as they can lead to regions of increased
tiator used within the RBC. The beam profile can show the
 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1211–1221 1219

distribution of both the number and the wavelengths of the specimens was only about 10% of the light that was deliv-
photons from the LCU and identify regions that receive many ered to the top. This occurs in part because some of the light
and those regions that receive few photons in the wavelength is used during the polymerization reaction, part is absorbed
ranges of interest. within the RBC, and part of the light is reflected from the
There was a difference in the irradiance homogeneity of specimen [19]. During photocuring, the power that reaches
the light emitted by these four LCUs. The Celalux 3 emitted the bottom of the specimen increases (Fig. 6) as the refrac-
light with a high irradiance that was concentrated in a 4-mm tive index of the organic matrix of the RBCs becomes similar
diameter central area of the specimens. It delivered a relatively to the refractive index of its filler, thus reducing refraction of
low irradiance and thus fewer photons to the outer regions of light [20,21]. Due to the absorption, refraction and scattering of
the specimens. Regarding the distribution of the photons of light, a longer exposure time may be required to achieve ade-
violet light across the tip of the Bluephase 20i and the Valo quate polymerization in the deeper regions of restorations,
Grand, although the distribution was not completely homo- especially when using multiple peak LCUs that deliver light
geneous, violet light was emitted across the whole tip of both at and below 410 nm. However, it must be remembered that
LCUs. the photons at these lower wavelengths of light contain more
Since distance will affect the amount of light that reaches energy than photons at the longer wavelengths. This, when
the RBC and thus the properties of the RBC [17], the specimens coupled with the increase in the exposure time, may cause
were all photoactivated with the tip of the LCU at a 2-mm dis- an unwanted temperature rise and damage the oral tissues
tance from the surface of the specimen. This distance was [22,23].
considered to reflect the clinical situation, where the light tip The two different RBCs used in this study were specifi-
is rarely in contact with the RBC and 2-mm represents the dis- cally chosen to verify the interaction between the different
tance from a molar cusp tip to the central fossa. Accordingly, photoinitiators present in their composition and the differ-
the irradiance beam profiles of the light that reached the spec- ent emission spectra of the LCUs. The manufacturer of Tetric
imen were also determined with the LCU tip 2-mm distance EvoCeram Bulk Fill states that it contains the alternative
®
from the sensor. Thus, the light that the specimens received photoinitiator Ivocerin that is most sensitive to lower wave-
was reported, rather than the light that was emitted at the tip lengths of violet light around 410 nm. In contrast, Filtek Bulk
of the LCU. Fill Posterior Restorative does not include any additional pho-
The second null hypothesis that tip diameter and beam toinitiator besides camphorquinone. This may explain the
homogeneity would have no effect on the microhardness greater microhardness results (Table 4) at the top surface of
was rejected. Both the tip diameter and the homogeneity of the Tetric EvoCeram Bulk Fill specimens when they were pho-
the light emitted from the LCUs affected the results. The toactivated by the Valo Grand light. Even though this LCU
ANOVA results reported less variation in the surface micro- delivered a lower overall irradiance value, it delivered more
hardness when the Valo Grand was used. When Celalux power in the violet range compared to the other multi-peak
3, Bluephase 20i and Elipar DeepCure-S were used, there light (Bluephase 20i) with 26% of the emitted light from the
were statistically significant differences between microhard- Valo Grand delivered in the violet range in a relatively homo-
ness values at the central, middle and outer regions of geneous manner (Fig. 3). This additional violet light would
®
the specimens. The narrow beam profile of the Celalux 3 efficiently activate the alternative photoinitiator Ivocerin in
meant that only the central 4 mm diameter region emit- Tetric EvoCeram Bulk Fill, thus producing a greater microhard-
ted light that was above 300 mW/cm2 . With the dimensions ness at the top.
of a mandibular molar crown being approximately 11.0 mm Some manufacturers claim an increased depth of cure for
mesiodistally and 10.5 mm buccolingually [18], if the user their bulk fill RBCs compared to conventional RBCs, but the
wishes to photoactivate large bulk fill RBC restorations, mul- reduced penetration of the violet light may lead to inade-
tiple exposures will be required when using this narrow tip quate polymerization in the deepest regions of restorations.
LCU. Although other photoinitiators are more reactive than cam-
Since the use of multiple-peak LCUs resulted in lower phorquinone [24,25] and have a greater ability to absorb violet
hardness values at the bottom of the RBCs compared to the light, there will be little violet light that can reach these
microhardness when they were photoactivated with single- photoinitiators at the bottom of 4 mm of RBC. Thus it is sug-
peak blue LCUs for the same exposure time, the third null gested that the use of photoinitiators that are only activated
hypothesis that the emission spectra of the LCUs will not influ- by violet light, such as Irgacure TPO, is not advisable for bulk
ence the microhardness of the two bulk fill RBCs was also fill RBCs, even though the top of the restoration, which is
rejected. This outcome probably occurred because very little subjected to occlusal forces, should exhibit greater polymer-
of the violet light (<410 nm) reaches the bottom of the speci- ization. Instead it is recommended that a broader spectrum
®
men, as illustrated in Fig. 5. When the spectral radiant power photoinitiator, such as Ivocerin , be included. This photoini-
delivered to the top was compared to the amount that reached tiator is activated by blue light up to 460 nm and so can also
the bottom of the RBC specimen, the ratio of violet to blue light be activated by single-peak blue LCUs that emit light below
dropped from 26% (top) to only 2% at the bottom, when mea- 460 nm.
sured through Tetric EvoCeram Bulk Fill photoactivated with The fourth null hypothesis was also rejected. The micro-
the Valo Grand. This is because the lower wavelengths of vio- hardness maps reflect the beam profiles from the four
let light are more highly scattered within the RBC compared different lights where the polymerization across the sur-
to the longer wavelengths of blue light. Additionally, the total face of bulk fill RBC restorations was dependent on the light
amount of light that reached the bottom of the 4-mm thick beam profile and distribution of radiant power (photons) that
1220 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1211–1221

reached the surface (Fig. 4). Also, when plotted on a semi- [3] Shimokawa CA, Turbino ML, Harlow JE, Price HL, Price RB.
logarithmic scale, as shown in Fig. 7, there was a positive Light output from six battery operated dental curing lights.
linear correlation between the microhardness and the radi- Mater Sci Eng C Mater Biol Appl 2016;69:1036–42.
[4] Santini A, Miletic V, Swift MD, Bradley M. Degree of
ant power measured at each hardness point. This correlation
conversion and microhardness of TPO-containing
was exponential (or linear in a semi-log scale) because the resin-based composites cured by polywave and monowave
relationship between the radiant exposure (dose) and the LED units. J Dent 2012;40:577–84.
polymerization is also exponential. The polymerization reac- [5] Cardoso KA, Zarpellon DC, Madruga CF, Rodrigues JA, Arrais
tion reaches a point of saturation, at which point there are CA. Effects of radiant exposure values using second and
no more radicals to be formed and the non-reacted radicals third generation light curing units on the degree of
conversion of a lucirin-based resin composite. J Appl Oral
are trapped. Thus, delivering more light into the material
Sci 2017;25:140–6.
will not change, or only have a very small effect on, its
[6] ALShaafi MM, Haenel T, Sullivan B, Labrie D, Alqahtani MQ,
degree of conversion [25,26]. Since a positive correlation was Price RB. Effect of a broad-spectrum LED curing light on the
detected, it is recommended that the entire surface of the Knoop microhardness of four posterior resin based
restoration should receive uniform coverage of light that is composites at 2, 4 and 6-mm depths. J Dent 2016;45:
at the same irradiance and at the same wavelengths. This 14–8.
will result in a more uniform polymerization across the sur- [7] Harlow JE, Rueggeberg FA, Labrie D, Sullivan B, Price RB.
Transmission of violet and blue light through conventional
face.
(layered) and bulk cured resin-based composites. J Dent
Considering the results obtained in the present study, the
2016;53:44–50.
spectral radiant power, active tip diameter and beam profile [8] Shimokawa C, Sullivan B, Turbino ML, Soares CJ, Price RB.
characteristics of LCUs should be reported by manufactur- Influence of emission spectrum and irradiance on light
ers and researchers alike. As very little violet light reaches curing of resin-based composites. Oper Dent 2017;42:537–47.
the bottom of 4-mm thick restorations, the polymerization [9] Price RBT. Light curing in dentistry. Dent Clin North Am
of 4-mm deep restorations should not depend solely on pho- 2017;61:751–78.
[10] Benetti AR, Havndrup-Pedersen C, Honore D, Pedersen MK,
toinitiators that are activated by the lower wavelengths of
Pallesen U. Bulk-fill resin composites: polymerization
violet light. Also, dentists should be educated so that they contraction, depth of cure, and gap formation. Oper Dent
purchase and use an appropriate LCU that has a wide light 2015;40:190–200.
tip delivering uniform irradiance and wavelengths of light to [11] Fronza BM, Rueggeberg FA, Braga RR, Mogilevych B, Soares
the RBC. LE, Martin AA, et al. Monomer conversion, microhardness,
internal marginal adaptation, and shrinkage stress of
bulk-fill resin composites. Dent Mater 2015;31:1542–51.
[12] Harlow JE, Sullivan B, Shortall AC, Labrie D, Price RB.
5. Conclusions Characterizing the output settings of dental curing lights. J
Dent 2016;44:20–6.
Within the limitations of this in vitro study using four LCUs [13] Taubock TT, Feilzer AJ, Buchalla W, Kleverlaan CJ, Krejci I,
and two RBCs, the different tip diameters, beam profile and Attin T. Effect of modulated photo-activation on
emission spectra all influenced the photoactivation of the two polymerization shrinkage behavior of dental restorative
resin composites. Eur J Oral Sci 2014;122:293–302.
bulk fill RBCs tested. The use of an LCU that has a wide light
[14] Cunha LG, Alonso RC, Pfeifer CS, de Goes MF, Ferracane JL,
tip delivering uniform irradiance and wavelengths of light to
Sinhoreti MA. Effect of irradiance and light source on
the entire surface of the RBC is preferable to those with narrow contraction stress, degree of conversion and push-out bond
light tips or those that deliver an inhomogeneous distribution strength of composite restoratives. Am J Dent
of irradiance and wavelengths of light. 2009;22:165–70.
[15] Feng L, Carvalho R, Suh BI. Insufficient cure under the
condition of high irradiance and short irradiation time. Dent
Mater 2009;25:283–9.
Acknowledgements [16] Wayne RP. Photochemistry. London: Butterworths; 1970.
[17] Rode KM, Kawano Y, Turbino ML. Evaluation of curing light
The authors wish to acknowledge the technical assistance distance on resin composite microhardness and
of Mr. B. Sullivan and the manufacturers who donated the polymerization. Oper Dent 2007;32:571–8.
resin composites and the curing lights for this study. They [18] Ash MM, Nelson SJ, Ash MM. Dental anatomy, physiology,
and occlusion. 8th ed. Philadelphia: W.B. Saunders; 2003.
also wish to thank the Dalhousie Faculty of Dentistry Research
[19] Musanje L, Darvell BW. Curing-light attenuation in
Fund, University of São Paulo, CAPES and the National Coun- filled-resin restorative materials. Dent Mater 2006;22:804–17.
cil for Scientific and Technological Development (CNPq # [20] Shortall AC, Palin WM, Burtscher P. Refractive index
307217/2014-0), Brazil for supporting this study. mismatch and monomer reactivity influence composite
curing depth. J Dent Res 2008;87:84–8.
[21] Par M, Repusic I, Skenderovic H, Klaric Sever E, Marovic D,
references Tarle Z. Real-time light transmittance monitoring for
determining polymerization completeness of conventional
and bulk fill dental composites. Oper Dent 2018;43:19–31.
[22] Mouhat M, Mercer J, Stangvaltaite L, Ortengren U.
[1] Rueggeberg FA. State-of-the-art: dental photocuring—a Light-curing units used in dentistry: factors associated with
review. Dent Mater 2011;27:39–52. heat development-potential risk for patients. Clin Oral
[2] Price RB, Ferracane JL, Shortall AC. Light-curing units: a Investig 2017;21:1687–96.
review of what we need to know. J Dent Res 2015;94:1179–86.
 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1211–1221 1221

[23] Runnacles P, Arrais CA, Pochapski MT, Dos Santos FA, Coelho [25] Rueggeberg FA, Giannini M, Arrais CAG, Price RBT. Light
U, Gomes JC, et al. In vivo temperature rise in anesthetized curing in dentistry and clinical implications: a literature
human pulp during exposure to a polywave LED light curing review. Braz Oral Res 2017;31:e61.
unit. Dent Mater 2015;31:505–13. [26] Selig D, Haenel T, Hausnerova B, Moeginger B, Labrie D,
[24] Neumann MG, Miranda Jr WG, Schmitt CC, Rueggeberg FA, Sullivan B, et al. Examining exposure reciprocity in a resin
Correa IC. Molar extinction coefficients and the photon based composite using high irradiance levels and real-time
absorption efficiency of dental photoinitiators and light degree of conversion values. Dent Mater 2015;31:583–93.
curing units. J Dent 2005;33:525–32.