d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1531–1541

Available online at www.sciencedirect.com

ScienceDirect

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

Effect of high-intensity curing lights on the

polymerization of bulk-fill composites

Mindy M. Daugherty a,1 , Wen Lien b , Michael R. Mansell a ,

Douglas L. Risk b,2 , Daniel A. Savett b , Kraig S. Vandewalle c,∗
a United States Army, 36014 Wratten Loop, Ft. Hood, TX 76544, USA
b United States Air Force, 3650 Chambers Pass, Ft. Sam Houston, TX 78023, USA
c United States Air Force, 1615 Truemper St., Joint Base San Antonio — Lackland, TX 78236, USA

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

Article history: Objectives. The purpose of this study was to investigate the effect of high-irradiance light-
Received 2 April 2018 curing-units (LCUs) on the depth-of-cure (DoC) and degree-of-polymerization (DoP) of bulk-
Received in revised form fill composites (BFCs).
25 May 2018 Methods. The DoC of composites (Beautifil-Bulk, SHOFU; Filtek-Bulk-Fill, 3 M ESPE; Tetric-
Accepted 7 June 2018 EvoCeram-Bulk-Fill, Ivoclar; Sonic-Fill-2, Kerr; Venus-Bulk-Fill, Heraeus; Z250, 3M-ESPE)
were measured according to ISO-4049 using high-irradiance LCUs (FlashMax-P3, CMS-
Dental; SPEC3, Coltene) and conventional LCU (Paradigm, 3M-ESPE) for exposure times: 3/9-s,
Keywords: 3/20-s, and 10/20-s respectively. Using FTIR, the DoP per composite was measured at the bot-
High-intensity curing lights tom surface as a function of post-curing times for the LCUs at the same exposure times. Data
Depth of cure was analyzed with nonlinear regression and ANOVA/Tukey.
Degree of polymerization Results. Significant differences in DoC were found amongst the LCUs for the various expo-
Bulk-fill composites sure times. All BFCs failed to meet the DoC claimed by manufacturers and failed to satisfy
ISO-4049 with the high-irradiance LCUs with 3-s exposures. Standard irradiance and 20-s
exposures outperformed all other irradiance-exposure combinations for maximizing the
DoC and DoP of BFCs. A minimum of 15.3 J/cm2 radiant exposure was required to achieve
an adequate maximum polymerization rate. Venus Bulk exhibited the highest DoC and DoP
for any LCU-exposure-time combination.
Significance. Among the different combinations of BFCs and LCUs, DoC and DoP were always
increased with longer exposure time, but there exists a theoretical radiant-exposure limit
beyond which DoP or DoC remains unchanged. However, high DoC or DoP are not always
associated with one another. Thus, the exposure-reciprocity law must be approached
thoughtfully since irradiance and exposure can independently affect DoP and DoC.
© 2018 The Academy of Dental Materials. Published by Elsevier Inc. All rights reserved.

∗
Corresponding author.
E-mail addresses: mindy.m.daugherty.mil@mail.mil (M.M. Daugherty), wen.lien.mil@mail.mil
(W. Lien), michael.r.mansell.mil@mail.mil (M.R. Mansell), drisk@amdpi.com (D.L. Risk),
daniel.a.savett.mil@mail.mil (D.A. Savett), kraig.s.vandewalle.civ@mail.mil (K.S. Vandewalle).
1
Present address: United States Army, 10590 Enduring Freedrom Dr., Ft. Drum, NY 13602, USA.
2
Present address: Private Practice, 2992 Waldorf Marketplace, Waldorf, MD 20603, USA.
https://doi.org/10.1016/j.dental.2018.06.005
0109-5641/© 2018 The Academy of Dental Materials. Published by Elsevier Inc. All rights reserved.
1532 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1531–1541

LCU can properly initiate polymerization within the recom-

1. Introduction mended curing time. Today, the standard in most dental
practices is the LED (light-emitting diode) LCU. The LED LCU
Dentistry has witnessed significant innovation of resin-based
exploits the electronic nature between two dissimilar semi-
restorative materials in the last three decades. The success
conducting materials to produce light. They differ from the
of this ongoing evolution appears to have been promul-
quartz-tungsten-halogen (QTH) LCUs in which light produc-
gated from new treatment strategies based on scientific
tion is based on the scientific principle of electroluminescence
evidences and practical constraints imposed by our clinical
rather than incandescence [6]. In comparison to QTH LCUs, the
practices, patient demands, and empirical research. Espe-
advantages of LED LCUs are high photonic output in relatively
cially for restorative dentistry, a paradigm shift toward light
short exposure time, energy efficiency, and narrow spectral
cured materials and away from amalgams is driven partially
output [7,8].
by the post-handling of mercury and its environmental con-
The recent introduction of LED LCUs with “high power
cerns, partly by the gradual clinical success of RBCs, and by
and short exposure” has raised concerns regarding their
the increasing patient demand for esthetics. RBC is presently
role, effectiveness, and safety in operative dentistry [9,10].
a routine armamentarium for dental operative procedures
Questions have arisen among dental professionals as to
with approximately 146 million RBCs placed annually in the
whether the Bunsen–Roscoe concept of exposure reciprocity
United States [1,2]. Placing these RBCs is an intricate process
[11] can be applied to aid in the acceleration of dental photo-
that often poses many challenges such as adequate isola-
polymerization when using these new LCUs. There has been a
tion, etch-and-bond treatment of dentin and enamel surfaces,
discursive controversy around extrapolating the total energy
incremental or bulk placement, and adequate light curing
principle to curing of the dental RBCs [12–15]. The assump-
(photo-polymerization). Proper execution of these steps can
tion is that the photo-response of a photo-sensitive RBC only
minimize iatrogenic errors and can enhance the longevity of
depends on the total absorbed energy dose (radiant exposure
the restoration. However, the role of the dental LCU in photo-
or H, unit: J/cm2 ) rather than the two components, irradiance
polymerization has often been overlooked, perhaps because
and exposure time, which their product determines radiant
of its presumed simplicity in application relative to all of the
exposure [11]. Furthermore, these extrinsic factors can be
other challenging requirements when restoring a tooth.
complicated by the different manufacturers’ claims on the
Photo-polymerization is an important process because it
bulk-fill RBC depth of cure. Most are publicized to be bulk
serves as a “rate-limiting” factor, controlling the outcome of
placed and cured in 4 mm increments, and few are cured in
a RBC restoration. For example, a RBC requires a synergis-
5 mm increments [16]. Although reducing the time required
tic combination of intrinsic and extrinsic factors to achieve
to cure these materials has short-term, cost-saving implica-
photo-polymerization efficiency [3]. The former includes the
tions, the quality, integrity, and performance of the restoration
filler content/morphology, monomer ratios/chemistry, or pho-
remains the primary concern. Improperly polymerized RBCs
toinitiator concentrations/types, and the latter is involved
can result in insufficient monomer-to-polymer conversion,
with curing parameters such as exposure times, source-
a predisposing factor to clinical problems such as marginal
to-object distance, or standard-versus-high irradiance mode
discrepancy, secondary caries, and fracture, whereas a prop-
[2]. Although controlling, manipulation, and formulation of
erly cured composite will exhibit good physical properties in
intrinsic factors can have a direct and significant impact on the
strength, wear, and toughness [17].
final physical properties of a RBC, they are very much depen-
The present study aims to investigate the effect of
dent on the extrinsic attributes. It has been shown that the
high-irradiance LCUs on the depth-of-cure (DoC) and degree-
energy and spectral wavelengths of a LCU must match that of
of-polymerization (DoP) of bulk-fill RBCs. The null hypotheses
the RBC’s manufacturer photo-polymerization requirements
were:
[4,5]. Without proper delivery of the required photon ener-
gies by the LCU, the predictive course of photo-polymerization
reaction can be jeopardized in which suboptimal quantum (1) There are no significant differences in DoC and DoP
yields and monomer-to-polymer conversions are incurred and amongst various commercial bulk-fill RBC brands cured
cascaded across each of the photopolymerization kinetic steps by a LCU having a high irradiance (i.e., ≥2000 mW/cm2 )
such as free-radical initiation, chain-propagation, and chain- with either a short (≤10 s) or an ultra-short exposure
termination mechanisms. The aberrant products from these time (≤5 s) versus a LCU having a conventional irradiance
inchoate steps are ultimately reflected in the final outcomes (<2000 mW/cm2 ) and standard exposure time (20 s).
of the RBC’s physical properties and clinical performance [3]. (2) Given the same radiant exposure (energy dose), there is
Furthermore, of these two factors, the former is governed by no significant difference in DoC and DoP amongst various
the proprietary manufacturing formulations which are neither commercial bulk-fill RBCs.
changeable nor controllable by the end user. In contrast, the
latter is most susceptible to operator error, particularly involv-
2. Materials and methods
ing their light-delivery techniques. There are other extrinsic
factors like LCU’s beam profile, spectral emission, and irra-
2.1. LCU and bulk-fill RBC characteristics
diance, which are strictly set forth by the manufacturer as
well.
Three LED LCUs, having different irradiances, were studied:
When a curing light tip is placed over a resin-based
FlashMax-P3 (CMS Dental, Copenhagen, Denmark), Paradigm
composite, the dentist must be confident that the chosen
(3M ESPE, St. Paul, MN, USA), and SPEC3 (Coltene, Cuyahoga
 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1531–1541 1533

Falls, OH, USA). Table 1a shows the manufacturer salient char-

Standard = 1600 ± 10%

1200 ± −10% to +20%
acteristics for each of the three LCUs. These LCUs were used

and 3 K = 3000–3500
5000–6000
to polymerize five bulk-fill RBCs and one non-bulk-fill RBC in
different combinations. This study follows the ISO-4049 [18]

(mW/cm2 )
irradiance
Claimed
in defining a non-bulk-fill or conventional RBC as a RBC with
a manufacture’s claimed depth of cure of less than or equal
to 3 mm. Table 1b shows the manufacturer salient character-
istics for each of the RBCs. The standard or main mode of
FlashMax-P3 and of Paradigm was selected for exposure times
of 3/9 s and of 10/20 s respectively. For SPEC3, both the main

wavelength
(nm) peak
and 3K modes were applied to cure RBCs for the exposure

Claimed

455 ± 10

455–465
times of 3 s and 20 s respectively. To validate the manufacturer
claimed irradiance, an integrating sphere (sphere Ø = 15 cm

460
and entry port Ø = 19 mm; Lapsphere, North Sutton, NH, USA),
coupled to a calibrated spectrophotometer (USB4000-UV–VIS,

wavelength
(nm) range
Ocean Optics, Dunedin, FL, USA), was used to measure the

450–470
Claimed

430–480

430–490
irradiances of the three LCUs. The measured irradiance was
determined by dividing the integration of the spectral radiant
power (mW/nm) as a function of wavelength (nm) over the
entire wavelength range by the active area of the LCU tip. The

Time to fully
diameter of the LCU tip was measured with a digital microme-
ter (±0.1 mm). Respectively, the light guide diameter were 0.72,

2h

2h
charge
0.78, and 0.91 cm for SPEC3, FlashMax-P3, and Paradigm; for

2–3 h
Paradigm and SPEC3, the outer rim thickness of the light guide
(0.08 cm) was negated. Also, irradiance measurements were
made using a 4 mm diameter aperture at the entrance of the
integrating sphere to determine the estimated amount of irra- Battery type
diance delivered to a 4 mm diameter cylindrical specimen. The

Li-Ion

Li-Ion

Li-Ion
outputs of the 3 LCUs were measured 3 times each, after which
the mean and standard deviation were calculated. For each
combination of LCU’s irradiance mode and exposure time, the
Continuous 1 s or 3 s

between each 1 s or
radiant exposure (J/cm2 ), a measure of photon dose, was deter-

3 K: 1 s, 2 s, 3 s, and
ortho = continuous

between each 3 s
mode = 120 s and
with 0.5 s pause

mined from multiplying the measured irradiance (mW/cm2 )

mode options

tack-cure mode
Other curing

3 s with pause
3 s exposure

by the exposure time.

Continuous

exposure
2.2. Depth of cure (DoC)

The DoC was measured according to the ISO-4049 [18]. Sam-

ples were made for each composite brand using stainless steel
5 s, 10 s, 15 s, or 20 s
Standard (main)

(SS) molds. All SS molds were 4 mm in diameter but varied in

exposure time

5 s, 10 s, or 15 s
curing mode

height. The determination for which mold height to be used

was based on each RBC manufacturer’s claimed DoC and on
1 s or 3 s

the formula, mold height = 2 × (manufacturer’s claim DoC) + 2,

as specified by the ISO-4049 [18]. Six RBC brands (5 bulk-fills
and 1 conventional) and three LCU brands were evaluated.
Cordless or corded

Table 1b showed the manufacturer claimed DoC for each of

Power option

the RBC brands. Once a proper mold was selected and placed
Cordless
Table 1a – LCU brands and features.

onto a Mylar strip on a glass microscope slide, it was filled

Cordless

with the test RBCs, prepared in accordance with the man-

ufacturer’s instructions. Care was taken to exclude any air
bubble. Next, another Mylar strip was placed over the sample,
followed by a second microscope slide. After finger pressure
LED

LED

LED

was used to displace excess materials, the second microscope

slide was removed. Then, the exit aperture of the LCU light
(manufacturer)

tip was positioned concentrically to the mold opening and

SPEC 3 (Coltene)
& light source

(CMS Dental)

Paradigm (3M

directly against the Mylar surface. Immediately after curing,

FlashMax P3

the sample was removed from the mold. Any remaining soft
Product

ESPE)

uncured RBC was manually scraped away using a plastic spat-

ula. The maximum height of the remaining hard RBC was
measured three times with a digital caliper (±0.1 mm). The
1534 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1531–1541

three measurements were averaged and divided by two to

Universal

Bis-EMA (bisphenol-A polyethylene glycol diether dimethacrylate), Bis-GMA (bisphenol-A diglycidyl ether dimethacrylate), Bis-MPEPP (bisphenol-A polyethoxy-dimethacrylate), DMA (dimethacrylate),
Universal
yield DoC. Based on the aforementioned ISO-4049 method [18],

A2
Shade

the DoC was measured in triplicate (n = 3) for each of the vari-

IV A
A2

A2
ous RBC + LCU combinations. For each combination, a new and
un-polymerized RBC was cured by using a LCU that was either
of the two high-irradiance LCUs (FlashMax-P3 or SPEC3) or a

40–3000 nm (mean
conventional LCU (Paradigm) for the exposure times of 3/9 s,

20–5000 nm
10–3500 nm
(20 nm/4–11 nm)

EBADMA (ethoxylated bisphenol-A-dimethacrylate), S-PRG (surface modified pre-reacted glass), TEGDMA (triethylene glycol dimethacrylate), UDMA (urethane dimethacrylate).
3/20 s, or 10/20 s respectively.
Cluster/non-
aggregated
Filler size

25–400 nm

550 nm)
2.3. Degree of polymerization (DoP)
N/A

A Fourier transform infrared attenuated total reflection (FTIR-

ATR) spectrometer (Spotlight 400, PerkinElmer, Waltham, MA,

Ba-Al-F-Si glass, YbF3, SiO2

USA) was used to measure the DoP per RBC brand at the
Zirconia/silica bottom surface (specimen thickness = 4 mm, Ø = 10 mm) as a
Zirconia/silica, YbF3

(monomer, glass, &

function of post-curing times for Paradigm, FlashMax-P3, and

pre-polymer filler
Filler

YbF3), spherical
(F-B-Al-Si glass)

SPEC3. RBC was placed following manufacturer’s instruction

Ba-Al-Si glass,
SiO2 , Ba-glass

mixed oxide

inside a SS mold (height = 4 mm, Ø = 10 mm) with direct con-

S-PRG filler

tact to the ATR crystal. Care was taken to exclude any air
bubbles. Next, a Mylar strip was placed over the sample, fol-
lowed by a microscope slide. After finger pressure was used to
displace excess materials, the microscope slide was removed.
Then, the exit aperture of the LCU light tip was positioned con-
wavelength

centrically to the mold opening and directly against the Mylar

440–490

400–500

400–520

400–500

400–500
Curing

surface. Depending on which LCU brand was applied, each

(nm)

N/A

new un-polymerized RBC sample received an exposure time

of either 10 or 20 s for Paradigm, either 3 or 9 s for FlashMax-
P3, or either 3 or 20 s for SPEC3, which produced 6 different
Bis-MPEPP,TEGDMA

Bis-GMA, TEGDMA,

combinations of RBC brand, LCU brand, and exposure time.

12-dodecane-DMA
Bis-GMA, UDMA,

Bis-GMA, UDMA,
UDMA, EBADMA

For each combination, real-time recording of the infrared (IR)

Bis-GMA, DMA

spectrum (2500–600 cm−1 ) began immediately after curing and

UDMA and 1,
chemistry

continued at steady intervals for 7 min (0, 15, 35, 55, 70, 90,
Bis-EMA

Bis-EMA
Matrix

105, 120, 180, 240, 300, 420 s), then again after 24 h. At each of
the collecting intervals, 4 scans (8 cm−1 resolution per scan)
were averaged to produce an IR spectrum, and 14 IR spectra
(including the baseline uncured spectrum) per each combi-
nation were recorded. The DoP (%) was determined by the
53–54
74.5

58.4

68.5

38.0
60.0
Filler wt% &

following equation:

  A1636  
vol.%

76–77
Table 1b – Resin-based composite brands and features.

87.0

76.5

83.5

65.0
82.0

A1609 polymer
DoP = 100 1 −  A1636 
A1609 monomer
4.0

5.0

5.0

4.0

4.0
2.5

where A1636 and A1609 represented the heights of the IR absorp-

tion intensities at 1636 cm−1 (C C of aliphatic bond stretch)
Ivoclar Vivadent
Manufacturer

Heraeus Kulzer

and 1609 cm−1 (C C of aromatic bond stretch)

and claimed

 respectively
 A1636   A1636 
DOC (mm)

and where / represented the

3M-ESPE

3M ESPE

A1609 polymer A1609 monomer

SHOFU

absorption-peak ratio of the aliphatic to aromatic stretch

Kerr

of the polymer normalized by the absorption-peak ratio of

aliphatic to aromatic stretch of the monomer.
Resin-based composite

Tetric EvoCeram Bulk Fill

2.4. Statistics and data analysis

Z250TM (Control)

Data were analyzed with nonlinear regression

  and
Beautifil Bulk

Sonic FillTM 2
FiltekTM Bulk

d(DoP)
Venus-Bulk

ANOVA/Tukey (˛ = 0.05). The maximum DoP rate, dt

,
®

max
defined as the maximum rate at which DoP changed with
respect to post-curing times, was calculated from the
following [19–22]:
 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1531–1541 1535

Fig. 1 – Effect of three commercial LCUs’ radiant exposures (J/cm2 ) on six RBC brands’ DoCs. Each bar represents an average
DoC of three samples per RBC brand and per radiant exposure (n = 3) from which for each of the radiant exposures, a mean
DoC of the six RBC brands’ DoCs (n = 18) was calculated and depicted as black squares on the figure.

(1) Each DoP-post curing time curve per RBC brand & radi- gorized: standard, short, and ultra-short exposure times to
ant exposure (n = 36) was fitted with a sigmoidal function, be 20 s, 9–10 s, and 3 s respectively. Using these 3 different
LCU brands and their various curing combinations, 6 radiant
a
DoP (t) = −(t−t0 )
exposures (7.1, 9.0, 12.3, 21.4, 24.5, and 36.5 J/cm2 ) were cal-
1+e ⁄b
culated from the 6 combinations of irradiances and exposure
where t = time, t0 = time at which sigmoidal reached maximum times (2.378 × 3, 3.001 × 3, 1.226 × 10, 2.378 × 9, 1.226 × 20, and
derivative, and a & b were the fitting constants. 1.827 × 20 J/cm2 ) respectively.
d(DoP(t))
(2) Once a, b, and t0 were found, the term, dt
, repre-
senting the rate of polymerization (Rp ), was used to obtain
  3.2. Influence of radiant exposure on DoC and DoP
d(DoP) d(DoP(t))
the maximum Rp or dt
value by solving dt
at t0 ,
a
  max
Figs. 1 and 2 show how successive radiant exposures can
which is 4b .
Then, for each set of 6 RBCs’ maximum Rp values that impact the DoC and DoP measurements of the 6 RBC brands.
were attained from exposing to the same radiant exposure, a Comparing amongst radiant exposures (black dashed line),
mean maximum Rp value per radiant exposure was calculated. the same upper-case letter per radiant exposure is not signifi-
These mean maximum Rp values were plotted as a function cantly different than each other (p > 0.05). Comparing amongst
of radiant exposure (J/cm2 ). RBC brands within each radiant exposure, the same lower-
case letter per RBC brand is not significantly different than
each other (p > 0.05). These differences indicated that the DoC
3. Results and DoP of bulk-fill RBCs were dependent more on material
brand and less on radiant exposure for high values of radiant
3.1. Irradiance exposure conditions.
Comparison between the insets on Figs. 1 and 2 showed the
Significant differences (p < 0.05) in mean irradiances mean irradiance values to conduct DoC measurements were
were found amongst the 3 LCU brands. LCU rankings of much higher than those of DoP measurements. These differ-
mean irradiances from highest to lowest were: SPEC3 (3 s ences in mean irradiance values can be best clarified from
mode = 3001 ± 8 mW/cm2 ), FlashMax-P3 (2378 ± 22 mW/cm2 ), the mathematical construct of irradiance, which is the rate
SPEC3 (main mode = 1827 ± 5 mW/cm2 ), and Paradigm of energy passage per unit area, orthogonal to the direction of
(1226 ± 0.3 mW/cm2 ). For this study, we categorized: (1) energy transport. This mathematical ratio demonstrates irra-
SPEC3 (3 s mode) and FlashMax-P3 as high irradiance diance is inversely proportional to the energy transport area.
(≥2000 mW/cm2 ) group and (2) SPEC3 (main mode) and Consequently, the smaller the active area, the larger the irra-
Paradigm as the conventional irradiance (<2000 mW/cm2 ) diance value. For example, the mean irradiance delivered to
group. The consensus is any LCU, having >1500 mW/cm2 each DoC specimen (see Fig. 1 inset) was calculated via divid-
irradiance, is considered high intensity [23]. We also cate- ing the total spectral radiant power output by the specimen’s
1536 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1531–1541

cross sectional area, which was 0.13 cm2 for Ø = 0.40 cm. Here,
the specimen’s cross sectional area was smaller than the exit
aperture of the LCU’s light tip and was limiting the amount
of photo-energy that could be transported from the light tip.
Therefore, the specimen’s cross sectional area was used as
the “normalizing” active area to calculate irradiance. In con-
trast, the mean irradiance delivered to each DoP specimen (see
Fig. 2 inset) was calculated via dividing the total spectral radi-
ant power output by the cross sectional, active area per LCU’s
light tip. Here, the specimen’s cross sectional area (Ø = 1 cm)
was slightly larger than the LCU’s light tip. Hence, the area of
energy transport was adopted from the cross sectional active
area of the LCU’s light tip. Because the total active area for the
DoC specimen was smaller than total active area of the DoP
specimen, given the same amount of exposure time, the irradi-
ance (total spectral radiant power per unit area) received by the
DoC specimens were relatively more than the DoP specimens.
Even though there are large differences between the insets
on Figs. 1 and 2, Fig. 3 shows radiant exposure, a product of
irradiance and exposure time, is more correlative to DoC and
DoP than the irradiance parameter by itself. When different
irradiances in combination with various exposure times are
converted to radiant exposures, Fig. 3 illustrates that in gen-
eral, the higher the radiant exposures ascended, the faster the
rate of change in DoC or DoP detected, but then the growth
leveled off at high radiant exposures.
Fig. 4 illustrates the mean maximum Rp plotted against
its respective radiant exposure (here, the time of dark-cure
was immediately measured beginning from the cessation of
radiant exposure to post one day). Each point in Fig. 4 rep-
resents the mean of 6 RBCs’ maximum Rp (5 bulk-fill and 1
conventional) per radiant exposure. The relationship between
mean maximum Rp and radiant exposure was best fitted using
a sigmoidal nonlinear regression. The curve was forced to
go through the origin to ensure the assumption that at zero
deliverance of radiant exposure, the mean maximum Rp was
also zero. To estimate the minimum radiant exposure, we
used Ilie and Durner approach [24], where lines of best fit
were inserted through the first and last points of the approxi-
mate function. The point, obtained from the projection of the
best fit lines’ intersection, was determined to be the mini-
mum radiant exposure that is required to achieve an adequate
maximum Rp . The intersection of these lines is considered
to estimate the inflection point of the curve, approximat-
ing the glass transition state, which describes the motions
of monomeric segments and polymeric chains as they tran-
sit from a viscoelastic matrix to an elastoplastic network [24].
A minimum of 15.3 J/cm2 is required to lengthen the RBCs’
glass-transition kinetics after which Rp ascends to its asymp-
totic limit, where it remains steady and independent of radiant
exposure when approaching infinity. Alternatively, the x-axis
value of the inflection point could also be determined by taking
the first derivative of the sigmoidal function, which resulted a
minimum radiant exposure of 9.0 J/cm2 . Comparing the two
minimum radiant exposure values, the method of Ilie and
Durner is more conservative.
As shown in Fig. 5, irrespective of various radiant exposures
supplied, Venus-Bulk showed the highest DoC (p < 0.05), while
Sonic Fill 2 exhibited the lowest DoC (p < 0.05) when compared
to all other bulk-fill RBC brands in this study. This further
demonstrates material brand dependence. However, DoP val-
ues of Venus-Bulk and Sonic Fill 2, measured at the bottom
surface of a 4-mm specimen thickness, are not significantly
different than each other and are ranked the highest (p < 0.05)
in comparison to all other bulk-fill RBC brands, whose DoP val-
ues were also measured at an equivalent depth of 4 mm (Fig. 5).
Interestingly, even though Sonic-Fill-2 has a high DoPs, its DoC

Fig. 2 – Effect of three commercial LCUs’ radiant exposures (J/cm2 ) on six RBC brands’ DoPs. Each bar represents an average
DoP of one day of dark-cure measurements per RBC brand and per radiant exposure (n = 12) from which for each of the
radiant exposures, a mean DoP of the six RBC brands’ DoPs (n = 72) was calculated and depicted as black squares on the
figure.
 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1531–1541 1537

Fig. 3 – DoP or DoC in relationship with radiant exposure.

is one of the lowest significant values in comparison with the
other five RBC brands.
The ISO-4049 states that DoC measurement should not be
below 0.5 mm from the values claimed by the manufacturer
of the tested product using the recommended curing times.
In this study, Beautifil Bulk, Venus Bulk, Tetric EvoCeram Bulk,
and Z250 all met the standard and were within the range of
the manufacturers’ claimed DoC when using the control LCU,
Paradigm (1226 mW/cm2 ) for 20 s. Filtek Bulk and SonicFill 2
each claim 5 mm DoC, and both fell below the ISO-4049 stan-
dard at 3.6 mm and 3.2 mm respectively.
RBCs. The DoP of bulk-fill RBCs under the curing combina-
tion of standard irradiance & exposure is not significantly
different than the high irradiance & short exposure combina-
tion. However, the curing combination of standard irradiance
& exposure did significantly outperform the other two com-
binations (standard irradiance & short exposure and high
irradiance & ultra-short exposure).
4. Discussion

Within the limits of this study, our null hypotheses were

3.3. Influence of irradiance and exposure combination
on DoC and DoP
Fig. 6 demonstrates that light curing protocol with low irradi-
ance & long exposure combination significantly outperforms
all other combinations for maximizing the DoC of bulk-fill
rejected since: (1) significant differences were found for the
DoC and DoP amongst the various commercial bulk-fill RBC
brands cured by an LCU having a high irradiance and either
a short or an ultra-short exposure time versus an LCU having
a conventional irradiance and standard exposure time; and,
(2) there were significant differences in DoC and DoP amongst

Fig. 4 – Relationship between mean maximum Rp and radiant exposure.

1538 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1531–1541

Fig. 5 – Comparison amongst various bulk-fill RBC brands against themselves and a non-bulk-fill RBC (Z250, Control). Each
black bar represents a mean DoP, which was calculated from averaging DoPs (n = 72) of the six radiant exposures per RBC
brand; and, each gray bar represents a mean DoC, which was calculated from averaging DoCs (n = 54) of the six radiant
exposures per RBC brand. The same case letters across columns are not significantly different than each other (p > 0.05).

various commercial bulk-fill RBCs when using the same radi-
ant exposure (energy dose).
4.1. Influence of radiant exposure on bulk-fill RBCs
There is still much to learn about the complex interactions
between photopolymerization and light curing parameters
(e.g., radiant exposure, irradiance, and exposure duration).
Today, there is no adoption of a standard that recommends a
minimal radiant exposure beyond which this threshold should
be adequate to meet the energetic demands of any photo-
cured RBC, resulting in clinically relevant properties. However,
the current consensus [25] is that there exists a minimum
amount of radiant exposure and exposure duration that are
resin-based material specific and that must be given along
with the correct wavelengths to attain acceptable outcomes.
Furthermore, “not all LCUs are created equal” [25].
Historically, clinical practice suggests a minimum radiant
exposure of 12–24 J/cm2 to adequately cure a 2 mm thick RBC
[26–28]. This evidence is based on surveying the QTH LCUs
and conventional RBCs since LED technology and bulk-fill
RBC were not available at the time. For example, Rueggeberg
et al. found that a combination of at least 400 mW/cm2 and
60 s (equating to a radiant exposure of 24 J/cm2 ) was required
to adequately achieve a 2 mm DoC [26]. Comparatively, the
ISO 10650:1999 recommended the measured irradiance of the
QTH LCU to be above 300 mW/cm2 , which under an exposure
duration of 40 s, this equated to a 12 J/cm2 radiant exposure
[27]. Similarly, Fan et al. was able to show that for most con-
ventional RBCs, a radiant exposure of one half smaller than
24 J/cm2 (or 12 J/cm2 ) was capable to satisfy the minimum,
1.5 mm DoC requirements, in accordance to the ISO-4049 [28].
As LED technology evolved over the years, data have shown
that a QTH LCU having a comparable irradiance output of a LED
LCU requires at least twice the amount of exposure duration
to achieve comparable biaxial flexure strength [29]. In general,
these studies are dependent on one crucial premise that deliv-
ering a minimum radiant exposure not only can adequately
cure but also can generate clinically relevant strength.
For comparison, this study has found that as a “rule of
thumb”, based on the mean maximum polymerization rate
as a function of radiant exposure (Fig. 4), a minimum radiant
exposure of 14 J/cm2 is required to adequately cure a bulk-fill
RBC. Furthermore, our findings provided supportive evidence
to the claims [14,15,30] that the Bunsen–Roscoe reciprocity law
does not apply under conditions involving high irradiance and
short exposure duration. In practical terms, we have observed
that the range of irradiance-and-exposure combination during
which the greatest sensitivity (i.e., maximum polymerization
rate) exists varies according to the brand of bulk-fill RBC. For
example, two different RBC brands in this study (Venus-Bulk
and Beautifil-Bulk at 24.5 J/cm2 ), given equal amounts of radi-
ant exposure, produced different DoP (see Fig. 2).
Although bulk-fill RBCs are claimed to have improved DoC
(on average up to 4 mm), saved time, and reduced the poten-
tial for entrapment of air and containments, they are like
any RBCs, not perfect [31]. Their photo-energetic demands
for polymerization are complex. Caution must be exercised to
ensure that when restoring with these bulk-fill RBCs, an ade-
quate amount of radiant exposure is delivered. The radiant
exitance, if not controlled properly, can produce deleterious
events, consequently leading to increased risk of premature
RBC failures. Furthermore, having an improved DoC neither
buffers for a reduction in curing times nor provides a sense of
complacency for the clinicians to ignore standard curing pro-
tocols by which the false assumption is that RBC with greater
DoC are impervious to radiant exposure effects or equates to
improved DoP. For instance, this is best observed in Fig. 5,
where there is no observable relationship between DoC and
DoP.
Additionally, our findings support the views of a previous
study by Ilie and Stark, which were: (1) short exposure and
 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1531–1541 1539

Fig. 6 – Effect of irradiances and exposures on the mean DoP and DoC of five bulk-fill RBCs and one non-bulk-fill RBC (Z250,
control). The same case letter across the columns is not significantly different than each other (p > 0.05). The blue and red
bars refer to the DoP and DoC axes respectively.

high irradiance resulted in significantly lower DoC for bulk-fill
RBCs as compared to long exposure and low irradiance; (2) an
increase in radiant exposure resulted in an increase in DoC;
and (3) the minimum radiant exposure to achieve a 4 mm DoC
was in the range of 6–24 J/cm2 [32].
4.2. Fitting function
Past studies have employed several fitting approaches
to describe the kinetic behavior of photopolymerization
[15,19–22,24]. Of those approaches, there are two fitting func-
tions, reflecting two different points of view, of which each
argues in favor of its empirical evidence on providing clarifi-
cation to the onset of RBC network formation. Both of these
fitting functions involve the summation of two reactions and
claim that a function, composed of a double component, can
better approximate the physical features of network forma-
tion than the traditional, one component fitting function. The
first viewpoint is based on the assumption that RBC reacts to
light curing in a viscoelastic manner during which the network
formation of a RBC can be modeled by a superposition of two
exponentials [24]. This double exponentials function describes
the transitional kinetics from a gel phase to a glassy phase
[24]. The second viewpoint is based on the curing process and
its categorization into two stages, primary cure and post-cure
[15]. And respectively, the fitting function accounts for these
stages by separating them into two time-dependent factors:
exponential and logarithmic [15]. The difference between the
two viewpoints is the latter ensures that at time zero, when
photo-irradiation is absent, DoP values are best fitted by start-
ing at the origin.
In this study, a sigmoidal function was used to fit the
kinetic curve of DoP versus post-cure time and also to cor-
relate the mean maximum Rp against radiant exposure. The
significance of the sigmoidal curve is threefold [19]: (1) at the
immediate onset of post-cure time, the slow-to-rise, nonlinear
1540 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1531–1541
response of the kinetic curve describes an induction period at
which oxygen is known strongly to inhibit free-radical photo-
polymerization; (2) once the oxygen consumption has reached
its peak as post-curing time extends, free radicals cascade
with less impediment, culminating in exponential growth of
polymerization; and, (3) when post-curing time further length-
ens, the overall reaction slows as the result of several possible
reaction kinetics: (a) progressive depletion of free radicals and
reactive double bonds and (b) rising rate of increasing molec-
ular weight of the chains that have not yet been terminated,
causing restrictions to the chains’ mobility (Trommsdorff or
gel effect) [19].
Naturally, there are pros and cons to each of the afore-
mentioned fitting approaches. It may suffice to point out here
that the application of a sigmoidal fitting function has a dis-
tinctive mathematical advantage. Unlike an exponential or
logarithmic fitting function, the sigmoidal fitting function for
the standard curve of DoP versus time will always yield a
derivative with an absolute maximum from which the rate
of polymerization per time (Rp ), as well as the maximum Rp ,
can be calculated and can then be plotted as a function of
DoP. Another option is to use, the fitting function consisted
of the two time-dependency factors, to model our data, but
our measurements were solely based on the post-cure phe-
nomenon. We did consider the fitting approach that uses only
the logarithmic time-dependency factor, which was intended
for modeling the post-cure stage, to approximate our results.
However, the disadvantage of the logarithmic fitting function
is that the Rp cannot be approximated from the first derivative
of a logarithmic function. Further study on which of the fitting
approaches are best to characterize the photopolymerization
kinetics may be necessary.
4.3. Limitations
This study has several limitations. First, the DoP and DoC
measurements do not yield information about the mechanical
properties of the bulk-fill RBCs in response to high irradiance
and short exposure. However, based on past studies, which
have demonstrated that a direct correlation existed between
hardness and DoP [33], this study approximates that as the
extent of DoP advances during post-cure, bulk-fill RBC hard-
ness must also increase. Second, more studies are needed to
explain the phenomena of contractional stress during and
after polymerization under the influence of high-irradiance
and short exposure. Earlier studies have found that greatly
increased radiant exposures are linked to significant rises in
polymerization stress but do not affect DoP significantly after
a certain threshold has been met [34,35]. Third, the rise of pul-
pal temperature when teeth are exposed to prolonged high
irradiance is an important concern, which this study did not
assess. A recent study has shown that consecutive curing
cycles under various irradiance protocols have the potential
to generate enough heat to surpass the 5.5 ◦ C temperature
threshold, inducing irreversible pulpal damage [36]. Based on
our data, we have found that there exists a threshold after
which the amount of radiant exposure plays less of a signifi-
cant role in DoP amplification. Since supplying more radiant
exposures than the threshold augments no additional benefits
and serve only to increase unnecessary pulpal temperature,
perhaps, the curing protocol may need to consider having an
appropriate amount of radiant exposure such that its radiant
temperature is negligible to affect pulpal irritation. Further
study on the effect of radiant exposure and pulpal temperature
rise is required.
5. Conclusion
The clinical applications of this study emphasize the impor-
tance of material and equipment selection. It is imperative
that clinicians understand the properties of composite mate-
rials chosen, whether standard or bulk-fill, and modify the
placement procedure as necessary. Just as importantly, clin-
icians should be familiar with the chosen LCU and its features
and realize that each brand of LED LCU is designed to be
used differently. There is not a “one size fits all” dental LCU
available that one can operate the same way for all material
types.
The DoC and DoP are governed by three interrelated but
independent fundamental light-curing parameters: radiant
exposure, irradiance, and exposure duration. Bulk-fill compos-
ites cured with high-irradiance and short/ultra-short exposure
time may not provide adequate DoC and DoP, which can lead
to undesirable clinical properties. Irrespective of bulk-fill RBC
brands and LCUs, long exposure time always produced highest
DoC and DoP. However, there exists a theoretical radiant expo-
sure limit above which DoP or DoC plateaus. High DoC does
not always associate with high DoP. Cautions must be taken
when exercising exposure reciprocity law since irradiance and
exposure can independently affect DoP and DoC.
references
[1] American Dental Association. ADA 2005-06 survey of dental
services rendered; 2007. Chicago, IL.
[2] Price RB, Shortall AC, Palin WM. Contemporary issues in
light curing. Oper Dent 2014;39(1):4–14.
[3] Leprince JG, Palin WM, Hadis MA, Devaux J, Leloup G.
Progress in dimethacrylate-based dental composite
technology and curing efficiency. Dent Mater
2013;29(2):139–56.
[4] Price RB, Felix CA. Effect of delivering light in specific narrow
bandwidths from 394 to 515 nm on the micro-hardness of
resin composites. Dent Mater 2009;25(7):899–908.
[5] Nomoto R. Effect of light wavelength on polymerization of
light-cured resins. Dent Mater J 1997;16(1):60–73.
[6] Jandt KD, Mills RW. A brief history of LED
photopolymerization. Dent Mater 2013;29(6):605–17.
[7] Leonard DL, Charlton DG, Roberts HW, Cohen ME.
Polymerization efficiency of LED curing lights. J Esthet
Restor Dent 2002;14(5):286–95.
[8] Vandewalle KS, Roberts HW, Andrus JL, Dunn WJ. Effect of
light dispersion of LED curing lights on resin composite
polymerization. J Esthet Restor Dent 2005;17(4):244–54.
[9] Kutuk ZB, Gurgan S, Hickel R, Ilie N. Influence of extremely
high irradiances on the micromechanical properties of a
nano hybrid resin based composite. Am J Dent
2017;30(1):9–15.
[10] Feng L, Carvalho R, Suh BI. Insufficient cure under the
condition of high irradiance and short irradiation time. Dent
Mater 2009;25(3):283–9.
 d e n t a l m a t e r i a l s 3 4 ( 2 0 1 8 ) 1531–1541
[11] Bunsen RW, Roscoe HE. Photochemische untersuchungen.
Ann Phys 1859;108(2):193.
[12] Halvorson RH, Erickson RL, Davidson CL. Energy dependent
polymerization of resin-based composite. Dent Mater
2002;18(6):463–9.
[13] Peutzfeldt A, Asmussen E. Resin composite properties and
energy density of light cure. J Dent Res 2005;84(7):659–62.
[14] Hadis M, Leprince JG, Shortall AC, Devaux J, Leloup G, Palin
WM. High irradiance curing and anomalies of exposure
reciprocity law in resin-based materials. J Dent
2011;39(8):549–57.
[15] Selig D, Haenel T, Hausnerová B, Moeginger B, Labrie D,
Sullivan B, et al. Examining exposure reciprocity in a resin
based composite using high irradiance levels and real-time
degree of conversion values. Dent Mater 2015;31(5):
583–93.
[16] Van Ende A, De Munck J, Lise DP, Van Meerbeek B. Bulk-fill
composites a review of the current literature. J Adhes Dent
2017;19(2):95–109.
[17] Demarco FF, Corrêa MB, Cenci MS, Moraes RR, Opdam NJ.
Longevity of posterior composite restorations: not only a
matter of materials. Dent Mater 2012;28(1):87–101.
[18] ISO-4049. Dentistry—polymer-based filling, restorative and
luting materials. Geneva, Switzerland: International
Organization for Standardization; 2009.
[19] Decker C, Decker D. Photoinitiated radical polymerization of
vinyl ether-maleate systems. Polymer 1997;38(9):2229–37.
[20] Trujillo M, Newman SM, Stansbury JW. Use of near-IR to
monitor the influence of external heating on dental
composite photopolymerization. Dent Mater
2004;20(8):766–77.
[21] Pereira SG, Telo JP, Nunes TG. Towards a controlled
photopolymerization of dental dimethacrylate monomers:
EPR studies on effects of dilution, filler loading, storage and
aging. J Mater Sci: Mater Med 2008;19(9):3135–44.
[22] Nunes TG, Ceballos L, Osorio R, Toledano M. Spatially
resolved photopolymerization kinetics and oxygen
inhibition in dental adhesives. Biomaterials
2005;26(14):1809–17.
[23] Shortall AC, Felix CJ, Watts DC. Robust spectrometer-based
methods for characterizing radiant exitance of dental LED
light curing units. Dent Mater 2015;31(4):339–50.
[24] Ilie N, Durner J. Polymerization kinetic calculations in dental
composites: a method comparison analysis. Clin Oral
Investig 2014;18(6):1587–96.
1541
[25] Roulet JF, Price R. Light curing – guidelines for practitioners –
a consensus statement from the 2014 symposium on light
curing in dentistry held at Dalhousie University, Halifax,
Canada. J Adhes Dent 2014;16(4):303–4.
[26] Rueggeberg FA, Caughman WF, Curtis Jr JW. Effect of light
intensity and exposure duration on cure of resin composite.
Oper Dent 1994;19(1):26–32.
[27] ISO 10650. Dental equipment—powered polymerization
activators. Geneva, Switzerland: International Organization
for Standardization; 1999.
[28] Fan PL, Schumacher RM, Azzolin K, Geary R, Eichmiller FC.
Curing-light intensity and depth of cure of resin-based
composites tested according to international standards. J
Am Dent Assoc 2002;133(4):429–34.
[29] Rueggeberg FA, Cole MA, Looney SW, Vickers A, Swift EJ.
Comparison of manufacturer-recommended exposure
durations with those determined using biaxial flexure
strength and scraped composite thickness among a variety
of light-curing units. J Esthet Restor Dent 2009;21(1):43–61.
[30] Leprince JG, Hadis M, Shortall AC, Ferracane JL, Devaux J,
Leloup G, et al. Photoinitiator type and applicability of
exposure reciprocity law in filled and unfilled photoactive
resins. Dent Mater 2011;27(2):157–64.
[31] Leprince JG, Palin WM, Vanacker J, Sabbagh J, Devaux J,
Leloup G. Physico-mechanical characteristics of
commercially available bulk-fill composites. J Dent
2014;42(8):993–1000.
[32] Ilie N, Stark K. Curing behaviour of high-viscosity bulk-fill
composites. J Dent 2014;42(8):977–85.
[33] Ferracane JL. Correlation between hardness and degree of
conversion during the setting reaction of unfilled dental
restorative resins. Dent Mater 1985;1(1):11–4.
[34] Calheiros FC, Braga RR, Kawano Y, Ballester RY. Relationship
between contraction stress and degree of conversion in
restorative composites. Dent Mater 2004;20(10):939–46.
[35] Asmussen E, Peutzfeldt A. Polymerization contraction of
resin composite vs. energy and power density of light-cure.
Eur J Oral Sci 2005;113(5):417–21.
[36] Runnacles P, Arrais CA, Pochapski MT, Dos Santos FA, Coelho
U, Gomes JC, et al. In vivo temperature rise in anesthetized
human pulp during exposure to a polywave LED light curing
unit. Dent Mater 2015;31(5):505–13.