d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 304–311

Available online at www.sciencedirect.com

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

Use of NIR light and upconversion phosphors in

light-curable polymers

Alexander Stepuk a , Dirk Mohn a,b , Robert N. Grass a , Matthias Zehnder b ,

Karl W. Krämer c , Fabienne Pellé d , Alban Ferrier d , Wendelin J. Stark a,∗
a Department of Chemistry and Applied Biosciences, Institute for Chemical and Bioengineering, ETH Zurich, 8093 Zurich, Switzerland
b Department of Preventive Dentistry, Periodontology, and Cardiology, University of Zurich, Center of Dental Medicine, 8032 Zurich,
Switzerland
c Department of Chemistry and Biochemistry, University of Bern, 3012 Bern, Switzerland
d Laboratoire de Chimie de la Matière Condensée de Paris (LCMCP), UMR7574 CNRS/UPMC/Chimie ParisTech, Chimie ParisTech, 75231

Paris 05, France

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

Article history: Objective. Light-curable polymers are commonly used in restorative surgery, prosthodontics
Received 12 October 2011 and surgical procedures. Despite the fact of wide application, there are clinical problems
Accepted 22 November 2011 due to limitations of blue light penetration: application is restricted to defects exposed to
the light source, layered filling of defect is required.
Methods. Combining photo-activation and up conversion allows efficient polymer hardening
Keywords: by deep penetrating near-infrared (NIR) light. The prerequisite 450 nm blue light to poly-
Near-infrared merize dental resins could be achieved by filler particles, which absorb the incident NIR
Polymers irradiation and convert it into visible light.
Composites Results. The on spot generated blue light results in uniform polymer hardening. Composite
Upconversion samples of 5 mm thickness were cured two times faster than pure polymer cured by blue
Luminescence light (30 and 60 s, respectively). Overall degree of monomer conversion resulted in higher
Dental materials values of more than 40%. The enhanced transmission of NIR light was confirmed by optical
analysis of dentin and enamel. The NIR transmittance surge in the 800–1200 nm window
could improve sealing of complex and deep caries lesions.
Significance. We demonstrate faster curing and an improved degree of polymerization by
using upconversion filler particles as multiple light emission centers. This study represents
an alternative approach in curing dental resins by NIR source.
© 2011 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

viscous bisphenol-A glycidyl dimethacrylate (BisGMA), a reac-

1. Introduction tive diluent (triethyleneglycol dimethacrylate, TEGDMA) and a
photo-initiator such as camphorquinone [2,3]. The advantage
Light-curing resins have been widely used in restorative den- of light-curing over chemical curing in dental applications is
tistry for several decades [1]. These resins are acrylate based that the process is easily controllable and fast. Furthermore,
and can be polymerized by blue light. They consist of a light-curing resins are frequently used in adhesive dentistry to

∗
Corresponding author at: Institute for Chemical and Bioengineering, Wolfgang-Pauli-Str. 10, ETH Zurich, HCI E 107, CH-8093 Zurich,
Switzerland. Tel.: +41 44 632 0980; fax: +41 44 633 1571.
E-mail address: [email protected] (W.J. Stark).
0109-5641/$ – see front matter © 2011 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.dental.2011.11.018
 d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 304–311 305

Fig. 1 – Treatment of caries lesions (dark area, I) using blue light induced curing (II) demands a time consuming layer by
layer application and hardening with possible shrinkage and reinfection. Generating blue light inside the polymer through
irradiation by near infrared (NIR) allows polymerization in one single step. Since tooth tissue is quite transparent to NIR,
irradiation of larger and complex defects proceeds more reliable and can be initiated through irradiation through the tooth
sides.

bond ceramic restorations such as inlays, overlays, or crowns TmF3 and YbF3 ) were prepared from corresponding oxides by
to dental hard tissues. However, despite widespread applica- treating them with HF. Sintering the stoichiometric mixtures
tion, these resins suffer from shrinkage [4] and incomplete of derived salts at 550 ◦ C in HF/Ar gas stream resulted in
curing of the resin [5]. The application thickness per treatment hexagonal NaYF4 host matrix doped with Yb and Tm. The
is limited by the blue light penetration through dental hard tis- phase composition was analyzed by X-ray diffraction (XRD)
sue and the polymer itself. Current clinical practice, therefore, recorded on an X’Pert PRO-MPD (PANalytical Netherlands, Cu
relies on layer-by-layer filling of the restorative polymers into K␣ radiation, X’Celerator linear detector system, step size
a cavity (Fig. 1). The multistep layer-by-layer treatment com- 0.033◦ , under ambient conditions). The derived UC particles
pels advanced skills in restoration procedures from a dentist were afterwards grounded and analyzed according to protocol
and involves long treatment time at considerable costs. Layer- (Fig. 3a and b).
by-layer technique often requires destruction of healthy tooth
tissues to provide complete exposure of the restorative toward
blue light. 2.2. Preparation of Heliobond® /NaYF4 :Yb,Tm
In the present work, we therefore investigate a new way composites
to avoid problems inherent to conventional light-curing den-
tal resins. Deep tissue penetrating near infrared light (NIR) and Polymeric matrix consisted of clinically approved ingredi-
upconversion (UC) phosphors (a specific class of materials that ents: adhesive mixture of TEGDMA and BisGMA (40/60%)
allows to combine several NIR photons and emit a blue light with camphorquinone as an initiator in Heliobond® (Ivoclar,
photon, for review see [6]) can generate the necessary light Vivadent, Liechtenstein) [10]. Powders of UC phosphors were
inside the polymer to homogeneously harden a restorative premixed with Heliobond® under different ratios (0/100,
resin in the tooth. Different energy of the incident NIR irradi- 10/90 or 20/80) using a “speed mixer” (Hauschild Eng. DAC
ation avoids the blue light absorbance loss problem of current 150 FVZ Germany, 3000 rpm – rates per minute) during
dental treatments (Fig. 2a and b). 30 s.
NIR has recently attracted tremendous interest when Then the samples were molded in cylinders (Ø1 mm) of
dealing with low radiation penetration depth in living tis- different thicknesses (1–10 mm). Finally, the premixed and
sue [7]. Non-invasive excitation with NIR light has a higher molded specimens were exposed to a NIR laser ( = 980 ± 5 nm,
transmittance in the so-called “therapeutic window” at continuous wave, CNI, China) for intervals from 30 s to
800–1200 nm [8]. Following such approach, we demonstrate 5 min. For comparison, commercial non-filled polymeric sam-
that a clinically relevant dental resin can be cured using ples were cured with blue light ( = 450 nm, 550 mW cm−2 ,
NIR light in one step behind various dental hard tissue Optilux 500, Kerr, USA) for 60 s, following the manufacturer’s
barriers. specification.
To evaluate the curing time of pure polymer and
composite samples, a needle was manually pressed
2. Materials and methods through the surface of the sample. The absence of a
visible needle penetration proved that the sample was
2.1. Preparation of NaYF4 :Yb,Tm particles hardened.
Scanning electron microscope (SEM; LEO 1530 Gem-
The blue emitting hexagonal UC phosphors were prepared ini, Zeiss, Germany) images characterized the morphology
by solid state synthesis [9] in the group of Karl Krämer at and particles’ distribution in the UC phosphor/polymer
the University of Bern. Briefly, the rare earth fluorides (YF3 , composites.
306 d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 304–311

Fig. 2 – (a) Light activated polymerization in a bulk sample requires light to reach all parts of the material (top) which limits
sample size as most polymers and photo-initiators strongly absorb the required short-wavelength blue light. Generation of
the blue light on spot, throughout the sample (bottom) can be realized through upconversion of several NIR photons to yield
one blue light photon. This approach combines deep sample NIR penetration with reliable generation of blue light
throughout the material (b, bottom). The blue light emission intensity strongly depends on the NIR excitation power (c).

2.3. Absolute UC luminescence spectroscopy centrifuge particle size analyzer (Brookhaven Instruments,
USA).
Absolute UC luminescence spectra were measured using an
integrating sphere (Fig. 3c) following a method described 2.5. Polymer conversion rate
by Ivanova et al. [11]. NIR excitation radiation was deliv-
ered through a Ti:sapphire laser pumped by a continuous Cured samples (n = 10) were ground into fine powders by
wave argon ion laser (Coherent Innova 90, USA) and 200 ␮m corn sandpaper and then mixed with KBr powder
focused on an integrating sphere containing the sample. (2 wt.%). The absorbance peaks were measured using diffuse
The excitation power was controlled by a photodiode and reflection in a Fourier transform infrared spectrometer (FTIR,
adjusted using neutral density filters. The total UC lumi- Tensor 27, Bruker Optics, 500–4000 cm−1 wavelength range,
nescence was collected in a SpectralonTM sphere (Labsphere resolution 2 cm−1 , 64 scans per sample). Non-cured polymer
Inc., USA) with 5 cm diameter (AvaSphere, Avantes BV, samples were examined with a Bruker Optics Vertex 70v FTIR
Netherlands). The 1 cm diameter sample port was covered spectrometer (500–4000 cm−1 wavelength range, resolution
with SpectralonTM . Emitted light was transmitted by an optical 2 cm−1 , 64 scans per sample).
fiber to a spectrometer (AVASpec-1024TEC). The setup (inte- The degree of conversion of each composite (n = 3) was
grating sphere and spectrometer) was calibrated by a halogen determined from the ratio of the absorbance intensities of
tungsten lamp (10 W, fan-cooled Avalight-HAL, Avantes BV, C C (peak at 1638 cm−1 ) to one of the C· · ·C (peak at 1608 cm−1 )
Netherlands). which measures the monomer content before and after curing
[12]:
2.4. Upconversion phosphors preparation and  
characterization [(C = C)/(C· · ·C)] after curing
DC(%) = 1− × 100% (1)
[(C = C)/(C· · ·C)] before curing
The UC phosphors were grounded in ball-mill (Pulverisette 7,
Tracomme, Switzerland) (10 mm ZrO2 balls; 1:1 weight ratio) 2.6. NIR and blue light transmittance of teeth
in isopropyl alcohol (technical grade) during 1–5 cycles for
30 min. The hydrodynamic particle size distribution was mea- Bovine enamel and dentin cross-sections of 0.5 mm and
sured (2 wt.% UC powder in ethylene glycol) using an X-ray disk 1 mm thickness were analyzed for light transmission. The

Fig. 3 – NaYF4 :Yb,Tm particles characterization. (a) XRD spectrum; (b) particle size distribution; (c) spectral irradiance of
upconversion phosphors under different excitation powers.
 d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 304–311 307

Fig. 4 – (a) Two thermocouples were placed at 1 and 2 mm

distances from the tooth cavity filled by composite resin
and cured by NIR laser. (b) Temperature increase upon Fig. 5 – SEM micrographs of polymer/UC phosphor
curing by NIR laser and blue light (*) [14]. composite: (a) cross-section and (b) magnified view of UC
particles incorporated in polymer.

specimens (n = 5) were stored in water at 5 ◦ C prior to anal-

ysis. Samples were examined under dry and wet conditions 3. Results
using an UV–Vis–NIR spectrophotometer (Cary 6000i, Varian,
USA). Hexagonal sodium yttrium fluoride codoped with ytterbium
and thulium [6], i.e. ␤-NaYF4 :25%Yb3+ ,0.3% Tm3+ , was used
2.7. Temperature rise as UC phosphor. Microcrystalline particles of the pure hexag-
onal phase (Fig. 3a) were prepared by solid state synthesis
Tooth chamber temperature rise upon curing with NIR [9]. To prevent particles from fast sedimentation in the poly-
laser was measured by K-type thermocouple (Thermocoax, mer before curing, ball-milling of the phosphors reduced their
Suresnes, France), which was placed in the prepared cylindri- mean particle size to 2.8 ␮m (Fig. 3b). Non-milled powders
cal cavities (1 mm × 3 mm) at ambient conditions (T = 25 ◦ C). showed much faster sedimentation in ethylene glycol and Bis-
The thermocouple was placed at 1 and 2 mm distance from GMA/TEGDMA polymer leading to inhomogeneous samples.
the cavity filled with resin samples (n = 10) (Fig. 4a). Since homogeneous hardening of a larger sample requires
a good distribution of the light-emitting UC phosphor particles
2.8. Data analysis within the polymer matrix, we investigated the even distribu-
tion of the UC particles (Fig. 5a). SEM images presented the
All data are stated as mean values ± standard deviation. Sta- UC particles of hexagonal morphology (Fig. 5b) in agreement
tistical significance (p < 0.05) between pure polymer and UC with earlier reports [9] with a size ranging from 2 to 3 ␮m.
phosphors/polymer composite was evaluated by Student’s t- This optically determined size range correlates well with the
test for curing performance experiments. Results were taken findings from X-ray diffraction (Fig. 3a) and X-ray disk centrifu-
as significant at p < 0.05 for optical parameters for dentin and gation (Fig. 3b). The latter method gives an integrated (average,
enamel. mass-based) hydrodynamic particle size diameter [13], which
308 d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 304–311

is the most relevant measure for sedimentation in a suspen-

sion.
The UC luminescence spectrum on 980 nm excitation
shows mainly the targeted blue emissions with maxima at
457 nm and 480 nm together with a weaker red emission at
652 nm (Fig. 3c). Absolute luminescence spectra revealed max-
imum spectral irradiance of 50 ␮W cm−2 nm−1 measured at
480 nm in the blue spectral range under 160 W cm−2 NIR exci-
tation power. Integrating over the peaks of useful radiation
between 400 and 500 nm resulted in a blue light intensity of
1 mW cm−2 for 90 W cm−2 excitation power (Fig. 2c).
The hardening of a 7 mm thick doped polymer slab in a
single step was about 3 times faster, than by curing it with a
dental lamp (Fig. 6a), even if only the irradiation time is consid-
ered. Furthermore, a 0.5 mm enamel plate installed in front of
the laser source increased overall curing time for NIR laser and
dental lamp cured samples (thickness 2 mm). The UC particles
doped composites were set in 45 ± 3 s, and blue light resulted
in almost two times slower curing rate (Fig. 6b). Upon curing
the samples with NIR laser, the tooth tissue revealed tempera-
ture increase similar to reported conventional blue light curing
[14]. The temperature rose by 12 ± 0.35 ◦ C and 10.91 ± 1.6 ◦ C
as measured by the thermocouple placed at 1 and 2 mm dis-
tances from the cavity, correspondingly (Fig. 4b).
The maximum degree of monomer conversion (DC) of 42%
was achieved for a NIR cured samples of 2 mm thickness,
which is 10% higher than the DC of the blue light-cured resin
(Fig. 6c). The visible absorption spectrum of the light-cured
dental resin has a maximum at 470 nm (Fig. 7a), caused by the
presence of the polymerization initiator (camphorquinone).
NIR transmission studies revealed increasing transmission
for enamel (∼3 ) to longer wavelengths from visible to NIR
(Fig. 7b). Calculated attenuation coefficients gave values of
˛ = 50 ± 4 cm−1 (n = 5) for dentin and ˛ = 35 ± 2 cm−1 (n = 5) for
enamel (Table 1).

4. Discussion

Present clinical application of dental acrylates uses a tech-

nique where the light-curable resin is applied as multiple
layers of 1–2 mm thickness with intermittent blue light irradi-
ation (Fig. 1). This technique at least partially avoids otherwise
inhomogeneous curing and shrinkage of the bulk sealants.
The viscosity of clinically used acrylates (0.5–30 Pa s [15,16]) Fig. 6 – (a) Polymer curing times to harden samples of
generally depends on BisGMA to TEGDMA ratio, filler con- 1–10 mm thickness when using blue light (squares) and NIR
tent and assures negligible particle settling (less than 0.1 mm irradiation (spheres). (b) Dental polymer curing through a
sedimentation per day [17]) (loss to the bottom of handling side wall of a tooth, modeled by putting an enamel platelet
vials or during application at the patient) for UC phosphors between sample and radiation source. (c) The quality of a
of a few micron size in spite of their relatively high density polymerization can be quantified through the degree of
(4.5–5.3 g cm−3 [17]). monomer conversion (DC) and can be significantly
Shrinkage is of particular relevance as detachment of the improved when generating blue light throughout the
polymer from the tooth surface provides an attractive niche sample using upconversion and NIR irradiation.
for caries re-infection, a frequently observed clinical incident
[18]. Since the here proposed improvement in curing would
allow single processing of a tooth lesion, shrinkage is even The currently used dental light-curing restorative materi-
more critical for good clinical performance. Adjusting the min- als contain mineral additives in the range from 20 to 80 wt.%
eral particle content in the composites reduces shrinkage to [19]. The effect of UC phosphor doping was investigated
less than 1%, which is comparable to commercial filled dental at lower levels of loadings. At 20 wt.% UC phosphor filling,
resins (Table 2). the resulting composite material displayed fast and efficient
 d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 304–311 309

Table 1 – Optical properties of enamel and dentin.

Attenuation coef.a /cm−1 Anisotropy factorb Optical penetration depth/mm

 = 980 nmc  = 450 nmc

Enamel 35 (2) 0.96 6.8 (0.5) 2.6
Dentin 50 (4) 0.93 3.8 (1.8) 0.7
a
Calculated from NIR transmission spectra.
b
From [26].
c
Calculated according to [24,25,27].

Table 2 – Polymerization shrinkage (n = 5).

Polymerization shrinkage
Blue light cured resin 5.5 (0.5)
NIR cured resin with 20 wt.% UC particle filler 0.5 (0.3)
BisGMA/TEGDMA = 70/30 polymer 7.04 (0.8)a
BisGMA/TEGDMA = 70/30 polymer with 75 wt.% particle filler 1.94 (0.23)a
a
Shrinkage data according to [15].

setting (rapid and homogeneous polymerization throughout blue light) requires quantification. A so-called SpectralonTM
the sample). Robustness of the method could be demonstrated sphere was used to quantify the luminescence irradiation
by NIR curing over the full range of clinically relevant sam- power of pure, compacted NaYF4 :Yb,Tm pills. UC emission
ple thicknesses from 1 mm (small surface defect in a tooth) to intensity increases non-linearly with NIR excitation power,
7 mm. as observed in more fundamental, previous studies using
A fair comparison of the “on spot generated” blue light similar phosphors [9]. For comparison, the widely used clin-
(from NIR UC) to the currently used dental lamps (also emitting ical Optilux 500 blue lamp for light-curing dental materials
provides a much higher irradiance of 200–300 mW cm−2 [20].
Despite its much weaker intensity of 1 mW cm−2 , the blue light
emitted from the UC particles successfully cures the acrylate
polymer.
NIR-assisted curing could significantly reduce treatment
time in dental practices. In a clinical setting, one has to add
the time to re-apply additional layers. The curing time of 1 mm
and 2 mm thick samples was not significantly different for
the dental lamp and NIR laser cured samples (Fig. 3a). That
is expected from the typical penetration depth of blue light
(2–3 mm) allowing the light to reach the whole sample.
The exothermic effect of the light-curing resin polymer-
ization and significant energy absorption during irradiation
lead to temperature increase of the restorative and heat up
the surrounding tissue, which is unfavorable from a clinical
prospective. Curing by NIR laser the composite resin did not
rise the chamber temperature significantly higher than the
reported values for curing by a conventional dental lamp [14].
The elevated temperatures upon curing still remain within
physiologically acceptable temperatures (Fig. 4b).
The curing time and sample thickness impacts the degree
of monomer conversion (DC) in polymerization and was inves-
tigated for samples with a fixed loading of UC phosphors.
Since monomer curing happens in the bulk of the mate-
rial, the homogenous light emission from the UC phosphor
particles, distributed throughout the sample, provides higher
curing rates. For better illustration, one may consider the
mean distance the blue light has to travel before polymeriza-
tion is initiated (i.e. activating the photo-labile initiator): the
blue light emitted by the UC phosphor particles only needs
Fig. 7 – (a) Visible absorption spectrum of BisGMA/TEGDMA to cross the acrylate sample over several microns, i.e. the
resin prior to polymerization. (b) NIR light transmission for average particle distance. For curing the resins with blue irra-
enamel and dentin samples (n = 5). diation, the light travels the path from the surface to the
310 d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 304–311

interior of the material (i.e. about the sample thickness of where s denotes the scattering coefficient and g is an
several millimeters). The visible spectrum of the light-cured anisotropy factor.
dental resin has an absorption maximum at 470 nm (Fig. 6a), The penetration depth depends on the reduced scatter-
caused by the presence of the polymerization initiator (cam- ing and absorption coefficients; therefore, it also correlates
phorquinone). The reported depth of blue light penetration with the attenuation coefficient. Consequently, for enamel,
normally does not exceed 2–3 mm [21] which limits the depth the maximum light penetration depth was 6.8 ± 0.5 mm
of cure. However, for the NIR region, experiments did not show (Table 1), whereas dentin showed a shorter penetration depth
any strong absorption peaks, therefore, allowing efficient pen- of 3.8 ± 1.8 mm. Those values exceed more than twice the
etration of the light through the sample. Each upconversion numbers reported for the visible range: 2.6 mm for dentin [25]
particle serves as an emitting center. For efficient curing of the and only 0.7 mm for enamel [26,27].
polymer, generated blue light travels the path equal to the dis-
tance between the phosphors, therefore the absorption effect
is significantly weaker compared to the conventional curing of 5. Conclusion
the resin by the dental lamp. Note that any classical strategy
to solve this problem of sample transparency is bound to fail As a result of these studies, we suggest a novel technique
since any photo-activator must absorb light to actually trig- for the polymerization of light-curing resins: blending the
ger the polymerization. That means, making a non-absorbing resins with UC phosphor particles and curing the material
photo-activator is per se impossible, as no absorption would with blue UC emission upon NIR excitation. This appears to
induce no reaction. The presently used clinical material is a be a promising approach for single stage dental restorative
delicate balance between high power irradiation (to drive at treatment in spots that are not accessible to direct light. The
least some light down to the lower parts of the 1–2 mm sample) composites containing UC particles showed shorter curing
and keeping the photo-initiator loading low (i.e. higher trans- times and higher degrees of monomer conversion in compari-
parency) while maintaining a clinically reliable and robust son to the polymer curing by blue light. The depth of NIR light
degree of polymerization (i.e. mechanical stability, acceptable transmission through teeth tissues is higher than that of blue
leaching of non-cured monomers, shrinkage). light (450 nm). For this reason, the prepared composites could
Treating complex shaped caries lesions with light cured restore cavities up to 7–10 mm depth in one step, even posi-
dental resin often requires cavity opening for light beam pen- tioned behind thin enamel or dentin obstacles. Accordingly,
etration. This voluntary destruction of otherwise still healthy these materials may substitute commonly used light-curing
tissue stays in contrast to the clinical philosophy of mini- polymers, providing additional advantages.
mizing treatment damage [22]. Ideally, complex caries lesions
would only be cleaned and directly filled with polymer, then
irradiated through the tooth wall itself. We therefore per-
Acknowledgements
formed preliminary curing tests on the composite material
behind enamel and dentin obstacles as an accurate way to
The authors acknowledge the financial support by ETH Zurich.
demonstrate a broad range of clinical situations. The curing
We are grateful to Prof. Albert Mehl (ZZM) for valuable discus-
time of such composites is significantly shorter, because of
sions. The generous provision of teeth samples by Beatrice
the improved NIR light penetration through the dental tissues.
Sener (ZZM) is kindly acknowledged.
A 0.5 mm enamel obstacle placed between the light source
and the sample decreased the light intensity and therefore
resulted in longer curing times. The curing time is significantly
references
shorter for the UC phosphor composites than for samples
cured by a blue light source (Fig. 3b).
NIR irradiation has a better penetration through the hard
[1] Moszner N, Salz U. Recent developments of new
tissues compared to visible light, which results in a so-called components for dental adhesives and composites.
therapeutic window around 800–1400 nm [23]. The absorption Macromol Mater Eng 2007;292:245–71.
coefficient of dentin in the visible light range is a = 4 cm−1 [24]. [2] Craig RG. Restorative dental materials. 10th ed. St. Louis:
Following previous studies on the optical properties of hard Mosby; 1997.
tissues, the penetration depth could be calculated as: [3] Wang K, Ma G, Qin X, Xiao M, Nie J. Cyclic acetals as
coinitiators in CQ-induced photopolymerizations. J Polym
2010;42:450–5.
1 [4] Ferracane JL. Placing dental composites – a stressful
ı=  (2)
experience. Oper Dent 2008;33:247–57.
3 · a · (a + s )
[5] Tarle Z, Meniga A, Ristic M, Sutalo J, Pichler G, Davidson CL.
The effect of the photopolymerization method on the
where ı denotes the penetration depth of light, and a and quality of composite resin samples. J Oral Rehabil
1998;25:436–42.
s denote the absorption and reduced scattering coefficients,
[6] Auzel F. Upconversion and anti-stokes processes with f and
respectively. The reduced scattering coefficient depends on
d ions in solids. Chem Rev 2004;104:139–73.
the anisotropy and is defined as: [7] Sershen SR, Westcott SL, Halas NJ, West JL.
Temperature-sensitive polymer-nanoshell composites for
photothermally modulated drug delivery. J Biomed Mater
s = (1 − g) · s (3) Res 2000;51:293–8.
 d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 304–311 311

[8] Sommerdijk JL. Influence of the host lattice on the [17] Dolez PI, Goff A, Love BJ. Settling behavior of specific acrylic
infrared-excited blue luminescence of Yb3+ , Tm3+ -doped particles in bis-GMA based reactive slurries. Sep Sci Technol
compounds. J Lumin 1973;8:126–30. 2002;37:2007–19.
[9] Krämer KW, Biner D, Frei G, Gudel HU, Hehlen MP, Lüthi SR. [18] Cenci MS, Pereira-Cenci T, Cury JA, ten Cate JM. Relationship
Hexagonal sodium yttrium fluoride based green and blue between gap size and dentine secondary caries formation
emitting upconversion phosphors. Chem Mater assessed in a microcosm biofilm model. Caries Res
2004;16:1244–51. 2009;43:97–102.
[10] Schneider OD, Stepuk A, Mohn D, Luechinger NA, Feldman [19] Kim K-H, Ong JL, Okuno O. The effect of filler loading and
K, Stark WJ. Light-curable polymer/calcium phosphate morphology on the mechanical properties of contemporary
nanocomposite glue for bone defect treatment. Acta composites. J Prosth Dent 2002;87:642–9.
Biomater 2010;6:2704–10. [20] Hirose T, Wakasa K, Yamaki M. Spectral distribution of
[11] Ivanova S, Pelle F. Strong 1.53 ␮m to NIR–Vis–UV visible-light curing units – effect on the depth of cure in a
upconversion in Er-doped fluoride glass for high-efficiency dental composite resin. J Mater Sci 1990;25:1209–13.
solar cells. J Opt Soc Am B 2009;26:1930–8. [21] Taira M, Okazaki M, Takahashi J. Studies on optical
[12] Tarumi H, Imazato S, Ehara A, Kato S, Ebi N, Ebisu S. properties of two commercial visible-light-cured composite
Post-irradiation polymerization of composites containing resins by diffuse reflectance measurements. J Oral Rehabil
bis-GMA and TEGDMA. Dent Mater 1999;15: 1999;26:329–37.
238–42. [22] Banerjee A, Watson TF, Kidd EAM. Conservative dentistry:
[13] Limbach LK, Li Y, Grass RN, Brunner TJ, Hintermann MA, dentine caries excavation: a review of current clinical
Muller M, et al. Oxide nanoparticle uptake in human lung techniques. Br Dent J 2000;188:476–82.
fibroblasts: effects of particle size, agglomeration, and [23] Fraqkowiak D, Planner A, Wiktorowicz K. Near-infrared
diffusion at low concentrations. Environ Sci Technol applications in biotechnology. New York: CRC Press; 2001.
2005;39:9370–6. [24] Zijp JR, ten Bosch JJ. Theoretical model for the scattering of
[14] Bouillaguet S, Caillot G, Forchelet J, Cattani-Lorente M, light by dentin and comparison with measurements. Appl
Wataha JC, Krejci I. Thermal risks from LED- and Opt 1993;32:411–5.
high-intensity QTH-curing units during polymerization of [25] ten Bosch JJ, Zijp JR. Dentine and dentine reactions in the
dental resins. J Biomed Mater Res B 2005;72B:260–7. oral cavity. Oxford: IRL Press Ltd.; 1987.
[15] Kim JW, Kim LU, Kim CK, Cho BH, Kim OY. Characteristics of [26] Fried D, Glena RE, Featherstone JDB, Seka W. Nature of light
novel dental composites containing 2,2-bis scattering in dental enamel and dentin at visible and
4-(2-methoxy-3-methacryloyloxy propoxy) phenyl propane near-infrared wavelengths. Appl Opt 1995;34:
as a base resin. Biomacromolecules 2006;7:154–60. 1278–85.
[16] Ling L, Xu X, Choi G-Y, Billodeaux D, Guo G, Diwan RM. Novel [27] Spitzer D, ten Bosch J. The absorption and scattering of light
F-releasing composite with improved mechanical in bovine and human dental enamel. Calcified Tissue Int
properties. J Dent Res 2009;88:83–8. 1975;17:129–37.