Simultaneous measurement of fluorescence, conversion and physical/mechanical properties for monitoring bulk and localized photopolymerization reactions in heterogeneous systems

Article abstract of DOI:10.1039/c6ra06341a

An FT-NIR spectrometer, rheometer and fluorescence spectrophotometer were coupled for the real-time monitoring of polymerization reactions, allowing the simultaneous tracking of polymerization kinetics, storage modulus as well as fluorescence. In this study, a methacrylate functionalized dansyl chromophore (DANSMA) was synthesized and two different nanogels were made from urethane dimethacrylate and isobornyl methacrylate. Two series of resin formulations were prepared using the DANSMA probe, ethoxylated bisphenol A dimethacrylate as the matrix monomer, Irgacure 651 as the initiator and the dispersed, monomer-swollen nanogels to give clear UV-curable resins. Placement of the fluorescent probe either throughout the resin or linked into the nanogel before its dispersion in the matrix provides a tool to study how the nanogel structure affects local network development by means of fluorescence from the DANSMA probe. We demonstrate the potential of this new technique using a composite as the two phase system (resin and polymerizable nanogel) including a dansyl derivative as a polymerizable probe to follow the reactions that are taking places in both phases.

Full text of DOI:10.1039/c6ra06341a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Simultaneous measurement of fluorescence,
conversion and physical/mechanical properties for
monitoring bulk and localized
Cite this: RSC Adv., 2016, 6, 41275
photopolymerization reactions in heterogeneous
systems
ab
b
c
a
a
a
S. Medel, P. Bosch, I. Grabchev, P. K. Shah, J. Liu, A. Aguirre-Soto
ad
and J. W. Stansbury*
An FT-NIR spectrometer, rheometer and fluorescence spectrophotometer were coupled for the real-time
monitoring of polymerization reactions, allowing the simultaneous tracking of polymerization kinetics,
storage modulus as well as fluorescence. In this study,
a methacrylate functionalized dansyl
chromophore (DANSMA) was synthesized and two different nanogels were made from urethane
dimethacrylate and isobornyl methacrylate. Two series of resin formulations were prepared using the
DANSMA probe, ethoxylated bisphenol A dimethacrylate as the matrix monomer, Irgacure® 651 as the
initiator and the dispersed, monomer-swollen nanogels to give clear UV-curable resins. Placement of
the fluorescent probe either throughout the resin or linked into the nanogel before its dispersion in the
matrix provides a tool to study how the nanogel structure affects local network development by means
of fluorescence from the DANSMA probe. We demonstrate the potential of this new technique using
a composite as the two phase system (resin and polymerizable nanogel) including a dansyl derivative as
a polymerizable probe to follow the reactions that are taking places in both phases.
Received 9th March 2016
Accepted 19th April 2016
DOI: 10.1039/c6ra06341a
www.rsc.org/advances
We have prepared polymeric nanogels that can be swollen by
monomers in order to tune nal network structure and bulk
Introduction
The design of improved polymeric materials is facilitated properties. The nanogel additives achieve this control in part by
6
through a better understanding of the network formation inuencing reaction kinetics in their immediate environment.
process. Controlling polymerization kinetics is important to Polymerization kinetics can be followed using NIR, but it is
achieve the desired network structure and physical properties. mainly a bulk analysis technique and cannot probe the local
7
Nanostructured materials, due to their very large surface area, kinetics in and around the nanogels. Fluorescence probe
are ideal additives for tuning networks due to their high degree methods are very useful for monitoring processes in polymers
of interactions. In the case of nanogel additives, the surface because of their high specicity, selectivity and rapid response
effects are compounded with the effects of monomer inltrated time. In addition, probes can be used for in situ measurements.
within the nanoparticles. These polymeric materials reinforced The uorescent response of chromophores has been shown to
with nanogel additives show enhancements in strength as well detect changes in the local environment due to their sensitivity
8
as reductions in polymerization shrinkage and stress, desirable to factors such as polarity and viscosity.
in many applications including dental materials. Previous
1,2
Monomer reaction data from other instruments like photo-
studies have demonstrated the effects of nanogels as additives DSC or FT-IR combined with uorescent probe analysis can
to model dental monomer and composite formulations that can distinguish the different steps involved in a polymerization
3–5
potentially deliver enhanced clinical outcomes.
mechanism at the microscopic level, which cannot be observed
only with calorimetric or spectroscopic measurements. In the
9
same way, physical and mechanical properties can be correlated
Department of Chemical and Biological Engineering, University of Colorado Boulder, with conversion by coupling FT-IR spectroscopy and dynamic
a
3
415 Colorado Ave., Boulder, 80303, CO, USA. E-mail: Jeffrey.stansbury@denver.edu
10
mechanical techniques, such as tensometer,
dynamic
mechanical analyzer or rheometry. This work represents the
rst time that we can examine changes and heterogeneities at
b
Department of Macromolecular Chemistry, Institute of Polymer Science and
Technology, ICTP-CSIC, Juan de la Cierva 3, E-28006, Madrid, Spain
11
12

c
Soa University “St. Kliment Ohridski”, Faculty of Medicine, 1470, Soa, Bulgaria
d
the sub-microscopic level by augmenting an FT-NIR analytical
approach with uorescence spectroscopy.
Department of Craniofacial Biology, School of Dental Medicine, University of
th
Colorado Denver, 12800 East 19 Ave., Aurora, 80045, CO, USA
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 41275–41286 | 41275

View Article Online
RSC Advances
Paper
Herein, we report the coupling of rheology and uorescence
Fourier transform near-infrared (FT-NIR) spectra were
spectroscopy with NIR to study the polymerization behaviour of recorded using a Thermo Scientic Nicolet 6700 spectropho-
ꢁ
1
nanogels dispersed in a resin composite. This provides new tometer with a resolution of 4 cm . The kinetics of the pho-
insights into the time-dependent correlation between micro- topolymerization were monitored using real-time FTIR
scopic viscoelastic properties and polymerization kinetics in spectroscopy equipped with two ber optic cables (Multimode
different areas of the composite materials: inside and outside Fiber Optics) through a NIR ber port (Smart Fiberport, Nicolet
the nanogel additive, in this example.
Instrument). These bers feature pure fused silica ber cores
1 mm diameter), numerical aperture of 0.22 mm, a wide
(
wavelength range (350–2400 nm) and transmission efficiency
higher than 90%.
Experimental section
Materials
The samples were exposed simultaneously to UV light (EXFO
Acticure 4000 mercury arc lamp using a 365 nm narrow band-
pass lter with light intensity on the surface of the sample
For the chromophore synthesis, dansyl chloride (>99%),
2-aminoethyl methacrylate hydrochloride (97%), triethylamine
ꢁ
2
measured in mW cm using an EIT radiometer) which induces
the polymerization, and to a NIR beam, which provides for in
situ analysis of the extent of the reaction. Reaction progress was
followed by monitoring the methacrylate group conversion by
(
>99%), ethyl acetate (99.5%), dichloromethane (99.5%, dried
˚
over 4 A molecular sieves) from Sigma-Aldrich were used as
received unless specied otherwise.
For the nanogel syntheses and resin formulations, isobornyl
methacrylate (IBMA, 98.5%), 2-mercaptoethanol (ME), azobisi-
sobutyronitrile (AIBN, 98%), ethoxylated bisphenol-A dimetha-
crylate (BisEMA, >98%) and ethyl acetate (99.5%) were obtained
from Sigma-Aldrich. Urethane dimethacrylate (UDMA) was ob-
tained from Esstech. 2,2-Dimethoxy-1,2-diphenylethan-1-one
the real-time decrease in the absorbance of the ]CH rst
2
ꢁ
1
overtone absorption band centered at 6165 cm
as the
monomer is converted to polymer. The methacrylate group
conversion is calculated according to the following formula:
Conversion (t) ¼ (1 ꢁ A
t
/A
0
) ꢂ 100
(Irgacure® 651) was purchased from BASF (Ciba-Geigy). The
inhibitor was removed from IBMA by treatment with activated
where A is the initial area of the absorbance peak related to the
0
basic alumina (Brockman I, Sigma-Aldrich). AIBN was puried
methacrylate double bound of the monomer and A
area measured at different irradiation times.
t
is the peak
ꢀ
by recrystallization from methanol (mp 104 C) before use and
dried under vacuum at room temperature. All other materials
were used as received.
Volumetric shrinkage was measured using a linometer
(
ACTA, Academic Center for Dentistry, Amsterdam, The Neth-
Flash column chromatography was performed using silica
13
erlands), as described previously. Briey, a drop of monomer
held between a free aluminum disc and a xed glass slide was
irradiated with the resulting displacement of the disc and the
differences in potential monitored by a Linear Variable Differ-
ential Transformer.
˚
gel (60 A pore size, 230–400 mesh particle size, Merck). For the
synthesis of N-dansyl-2-aminoethyl methacrylate, the reaction
was monitored using thin layer chromatography (TLC) on silica
gel-coated plates (Merck 60 F254). Detection was performed
with UV light.
The linometer was equipped with an EXFO Acticure 4000
mercury arc lamp (l ¼ 365 ꢃ 10 nm) and coupled to FTIR
(Thermo Scientic Nicolet 6700 spectrophotometer equipped
with 100 mm diameter NIR ber optic cables) that provides
measurement of volumetric shrinkage and methacrylate
conversion simultaneously. The NIR source was directed
through the center of the sample and conversion was calculated
using the same approach as in the kinetic analysis. Incident UV
Methods
1
13
H NMR and C NMR spectra were recorded on a Bruker 300
MHz spectrometer with samples dissolved in CDCl at room
3
temperature. Chemical shis were assigned using the residual
undeuterated solvent signal as an internal reference.
Elemental analyses were made with a LECO CHNS-932
apparatus.
Mass spectra were recorded using an HPLC-MSD 1100 mass
spectrometer (MS) with an atmospheric pressure ionization
source or an electrospray ionization (ESI) source.
Triple-detection Gel Permeation Chromatography (GPC;
Viscotek) with differential Refractive Index (RI, Viscotek VE
ꢁ2
light irradiance was 25 mW cm for 10 min in all experiments.
Three replicates were conducted for each system.
Post-gel polymerization stress development was monitored
in real-time under ambient conditions using a cantilever beam-
based tensometer (Volpe Research Center, American Dental
Association Health Foundation, Gaithersburg, MD). As
14
described previously, a monomer sample is placed between
two quartz rods, which are attached to the xed base of the
apparatus and to a deformable cantilever beam. As polymeri-
zation proceeds, the polymer bonds to the silane methacrylate
surface treated rod ends and deects the beam with stress then
calculated based on the cross-sectional area of the specimen
and a calibration curve of the beam constant obtained previ-
3
580), viscosity and light scattering detectors (Viscotek 270 dual
detector) was employed for the analysis of nanogel weight
average molecular weight (M ), number average molecular
weight (M ), polydispersity index (PDI) and average hydrody-
namic radius (R ). The mobile phase was HPLC grade THF/
w
n
h
ꢁ1
ꢀ
acetone (90 : 10 v/v) at a ow rate of 1.0 mL min at 35 C in
a series of four columns spanning molecular weight of 10 to
1
4
7
ously. The rate of stress development (R
) was calculated as the
s
0
Da. Poly(methyl methacrylate) standards were used for

rst derivative of the stress versus time curve.
calibration.
41276 | RSC Adv., 2016, 6, 41275–41286
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
The tensometer was also equipped with an EXFO Acticure equiv.), triethylamine (1.1 mL, 7.65 mmol, 4.5 equiv.) and ethyl
4
000 mercury arc lamp (l ¼ 365 ꢃ 10 nm) that provided UV acetate (20 mL) were stirred at room temperature for 1 h. The
activation of the photopolymerization using one of the quartz white precipitated solid was separated by ltration. Then,
rods as a light guide. The sample was probed by FTIR (Thermo a solution of dansyl chloride (550 mg, 2.04 mmol, 1.2 equiv.)
Scientic Nicolet 6700 spectrophotometer equipped with NIR dissolved in ethyl acetate (20 mL) was added drop-wise to the

ber optic cables), which allows the shrinkage stress data to be reaction mixture. The mixture was stirred at room temperature
coupled directly with methacrylate conversion. The details of for 4 h. The volatiles were removed under vacuum and the crude
this instrument and its operation are described more product was puried by column chromatography on silica gel
15,16
completely in other publications.
All samples were disc- (dichloromethane/ethyl acetate, 12 : 1 v/v) giving DANSMA (462
shaped, with 6 mm diameter and 1 mm thickness. Incident mg, 75%) as a pure yellow-green powder.
ꢁ
2
1
UV light irradiance was 1 mW cm for 30 min in all experi-
ments. Three replicates were conducted for each material.
H NMR (300 MHz, CDCl ) d ppm: 8.58 (d, H, J ¼ 8.5 Hz), 8.26
3
(m, 2H), 7.55 (m, 2H), 7.19 (d, H, J ¼ 7.3 Hz), 5.89 (s, H), 5.47 (s,
–CH
–, J ¼ 5.2
–), 2.91 (s, 6H, N–(CH ), 1.79
For the rheology tests, a rheometer (ARES TA Instruments) H), 5.17 (t, H, J ¼ 6.0 Hz), 4.06 (t, 2H, –NH–CH
was used to measure the initial viscosities and the dynamic Hz), 3.22 (m, 2H, –NH–CH –CH
evolution of storage modulus for nanogel modied resins with 8 (s, 3H, C–(CH )).
3
C NMR (75 MHz, CDCl ) d ppm: 166.8, 151.5, 135.4, 134.2,
2
2
2
2
3 2
)
3
ꢀ
ꢁ1
13
or 20 mm diameter parallel plates at 20 C, 5 rad s frequency
and 50% constant strain (ensuring that the test was carried out 130.6, 129.8, 129.6, 129.5, 128.5, 126.2, 123.3, 118.9, 115.4, 62.8,
within the linear viscoelastic regime). Three replicates were 45.5, 42.4, 18.1.
conducted for each system.
Elemental analysis calcd for C H N O S: C, 59.65; H, 6.12;
18 22 2 4
A chamber was constructed to allow nitrogen purging of all N, 7.73; S, 8.85. Found: C, 59.89; H, 6.31; N, 7.24; S, 8.19.
+
samples. Each sample underwent 30 min of N
analysis with the plates separated 2 mm to remove dissolved H] 363.0.
oxygen and avoid oxygen-inhibited edge effects that otherwise Syntheses of nanogels (Ng and NgDANS). Nanogels were
confound the rheological data. Aerwards the two quartz plates developed by following a slight modication of a literature
2
purge before
22 2 4
ESI-MS (m/z) calcd for C18H N O S [M] 362.0; found [M +
+
18
were brought to a distance of 0.3 mm, creating a cylindrical strategy. Urethane dimethacrylate (UDMA, 2.0 g), isobornyl
sample volume of 0.3 mm thickness and 8 or 20 mm diameter dimethacrylate (IBMA, 3.8 g), ethyl acetate (24 mL, a four-fold
depending on the sample.
excess relative to monomers), 2-mercaptoethanol (ME, 0.13 g,
Coupling rheometer to FTIR and to uorimeter (Scheme 1). 10 mol% relative to monomers) and azobisisobutyronitrile
In this inert atmosphere chamber a uorescence detector USB4000 (AIBN, 28 g, 1 mol% relative to the IBMA monomer content).
ꢀ
Ocean Optics (equipped with a Toshiba 3648 element CCD-charge The solutions were heated in an oil bath for 3 h at 80 C under
coupled device-linear array detector to increase the signal/noise a nitrogen atmosphere with a circulated water condenser in
ratio while detecting photons across the 200–1100 nm range) place.
was coupled through a 50 mm optical ber. It is of critical impor-
tance that this ber is aligned horizontally with the sample and at troscopy, reaching more than 60% conversion. The clear
an optimal 90-degree angle from the incident light source. nanogel reaction mixture was added dropwise to a 10-fold
Aer the reaction, the samples were analysed by FTIR spec-
For following the polymerization reaction, changes in excess of methanol (400 mL), which resulted in the precipita-
emission have been analysed in terms of the ratio between the tion of the polymeric materials. The precipitate was isolated by

2 1
uorescence intensity at two different wavelengths (FI /FI ), as decanting and the residual solvent was removed under high
9,17
1
discussed in previous papers. The wavelengths used for the vacuum affording the nanogel (2.71 g, 72%). H NMR charac-
dansyl probe in this study were 495 nm (lem1) and 545 nm terization of the resulting nanogels demonstrated copolymer
(l
em2), which avoids interaction with the photocuring process.
compositions similar to that of the monomer feed ratios. Since
The rheometer was also equipped with an EXFO Acticure the nanogel synthesis was stopped below full conversion, there
4
000 mercury arc lamp (l ¼ 365 ꢃ 10 nm) that was coupled to an are some pendant reactive groups derived from UDMA that are
in-house designed optical apparatus allowing both UV and NIR located throughout the nanogel structure.
direct transmission access to specimens within the rheometer
Two types of nanogels were synthesized: control nanogel,
so storage modulus and methacrylate conversion are registered designated as Ng and nanogel containing the chromophore,
14
simultaneously. Methacrylate conversion was measured using designated as NgDANS, in which DANSMA was added as
a Thermo Scientic Nicolet 6700 spectrophotometer equipped a comonomer in the nanogel synthesis at a concentration of
with near NIR ber optic cables. The NIR source was directed 0.03 mol% relative to the overall monomer content. The reac-
through the center of the sample and conversion was calculated tion, isolation and characterization procedures were identical
using the same approach as in the kinetic analysis. Incident UV for the Ng and NgDANS materials.
ꢁ2
light irradiance was 1 mW cm for 30 min in all experiments.
Formulation of resins. Two different sets of formulations
were prepared to obtain a broad range of formulated methac-
rylate resin system, which enable us to investigate the poly-
merization behaviour of nanogels dispersed to form a resin
Synthetic procedures
Synthesis of N-dansyl-2-aminoethyl methacrylate (DANSMA). composite. Formulations in Series 1 include the uorescent
-Aminoethyl methacrylate ꢂ HCl (280 mg, 1.70 mmol, 1 label DANSMA, as a polymerizable comonomer in the BisEMA
2
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 41275–41286 | 41277

View Article Online
RSC Advances
Paper
(0.1 wt%, relative to resin) was added to each sample. Samples
were mechanically agitated (2–48 h) until all prepolymers were
fully dispersed to give optically clear monomer/nanogel
compositions.
Final resin compositions and their references are summa-
rized in Table 1.
Results and discussion
Synthesis of N-dansyl-2-aminoethyl methacrylate (DANSMA)
The synthesis of the reactive dansyl probe followed a different
19
synthetic strategy than the method reported recently. Here the
synthesis was performed in a two-step one pot reaction. Meth-
acrylate functionalization of dansyl chloride was carried out by
liberation of 2-aminoethyl methacrylate from its hydrochloride,
under mild conditions, and subsequent in situ acylation of the
amino group with dansyl chloride at room temperature. The
DANSMA chromophore was isolated with an overall yield of
Scheme
1 Experimental system used in the connection among
rheometer (1), IR (2) and fluorimeter (3). Sample (4) is previously purge
under nitrogen inside a chamber (5) and it is irradiated with 365 nm UV
light (6).
75% (Scheme 3).
In this way, total reaction time is reduced from 24 h to 5 h,
one purication step is avoided and the total yield is improved
from 13% (overall two-step procedure as previously described in
the literature) to 75%.
Syntheses of nanogels (Ng and NgDANS)
For the nanogel syntheses, UDMA and IBMA were mixed at
20 : 80 molar ratio and 2-mercaptoethanol was added as a chain
transfer agent. Fine control of the mercaptoethanol concentra-
tion avoids macrogelation along with control of the average
primary chain length as well as nanogel size and molecular
weight. AIBN was added as the free radical thermal initiator.
DANSMA was added at a concentration of 0.03 mol% relative to
the monomer content. The overall process of nanogel synthesis
is shown in Scheme 4.
GPC analysis demonstrated that both series of nanogels
presented relatively broad molecular weight distributions as
expected for the globular, highly branched (M ꢁ H a < 0.5),
multi-chain structures. Both nanogels present similar mean
hydrodynamic radius results (ꢄ7 nm in THF). Specic GPC data
are summarized in Table 2.
Scheme 2 Series 1 and 2 of the prepared formulations.
resin while formulations in Series 2 introduce the uorescent
label covalently linked to the nanogel particles (Scheme 2). In
both series of materials, the dansyl chromophore was included
at very low concentration. Its presence as free comonomer did
not alter the BisEMA reaction kinetics and similarly, no kinetic
differences were noted for nanogel additives with or without the
presence of the tethered probe.
Nanogels were added to the BisEMA resin in mass ratio of 5,
10, 15 and 20 wt%. Unlled BisEMA without any of the dansyl
Ethoxylated bisphenol-A dimethacrylate monomer (BisEMA)
was chosen as the photopolymerizable resin. It is widely used in
probe (NgDANS0) was used as the control. Irgacure 651
Table 1 Formulation of resins. Irgacure 651 (0.1 wt%) was added in each mixture. Unfilled BisEMA (NgDANS0) was used as the control experiment
Reference
BisEMA, wt%
NgDANS, wt%
Ng, wt%
DANSMA, wt%
Series 1
Series 2
Ng0
Ng5
Ng10
Ng15
100
95
90
85
80
100
95
90
85
80
0
0
0
0
0
0
5
10
15
20
0
5
10
15
20
0
0
0
0
0
0.01
0.01
0.01
0.01
0.01
0
0
0
0
0
Ng20
NgDANS0
NgDANS5
NgDANS10
NgDANS15
NgDANS20
41278 | RSC Adv., 2016, 6, 41275–41286
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Scheme 3 Synthesis of DANSMA chromophore.
Scheme 4 Reaction description for the nanogel synthesis of labelled
nanogels.
Scheme 5 Molecular structures for methacrylate monomers.
dental restorative applications as well as in UV cured coatings
due to its resistance to swelling and staining, exibility, low
2
shrinkage, high modulus and strength. Monomer structures
are shown in Scheme 5.
Scheme 6 Graphical representation of resin with nanogel formula-
tions where nanogels are swollen with monomer resin before pho-
topolymerization and “IPN and copolymer formation” is shown in the
Formulation of resins
When nanogels are dispersed in resin of an appropriate solu- transition to the final polymer. Dispersed nanogel is shown repre-
senting low nanogel loading but a continuous nanogel-based phase
would result above the percolation threshold.
bility parameter, the internal free volume associated with the
nanogel structure and solution polymerization process allows
18
for signicant swelling by monomer. Once this secondary
photopolymerization reaction begins, the matrix monomers in
the composite formulation polymerize outside and inside the
nanogel particles including copolymerization with residual
methacrylate groups in the nanogel. This provides both
a physically interpenetrated polymer network as well as a cova-
lent attachment between the nanogel and the matrix polymer
formulation with 20% nanogel content. Without a reactive
diluent comonomer such as TEGDMA, it was difficult to obtain
homogeneous formulations without solvent dilution and
subsequent removal when nanogel content exceeded 20 wt%.
Series 1 and 2 (Ng and NgDANS, respectively) presented
similar viscosity values with the nanogel content due to their
close similarity in terms of nanogel composition and size.
(Scheme 6).
Considering the different formulations listed in Table 1,
these could be represented in the four series as shown in
Scheme 7. As particle–particle interactions become dominant
Photopolymerization kinetics
above the nanogel percolation threshold, which is typically Photopolymerization kinetic proles and polymerization rate
encountered at about 10 wt% loading depending on the nanogel plots for Series 2 (formulations with the uorescent label linked
size and swelling potential, the viscosity of resin formulations to the nanogel particles) are shown in Fig. 2.
greatly increases with increasing nanogel content. In prior
The kinetic proles show the conventional behaviour of
studies, TEGDMA formulations loaded up to 50 wt% with a photocrosslinking curing reaction of a glassy polymer with
18
similar nanogels reach 100 Pa s, while BisGMA monomer autoacceleration in the initial stage of polymerization and nal
1
1
(
specically designed for use in dental materials) has mono- conversion values limited by vitrication. During UV irradia-
20
meric viscosity values between 700 and 1300 Pa s. Viscosity tion, good photostability of the chromophore was observed. The
values of the formulations studied in the present work are presence of the chromophore, whether freely dispersed or
shown in Fig. 1.
localized within the nanogel, does not affect the kinetics of the
Viscosity increases with increasing nanogel content from polymerization reaction relative to the chromophore-free
Pa s for the neat BisEMA resin to almost 600 Pa s for the analogue controls due to its low content in the mixture.
9
n w w n
Table 2 Reaction conversion, number and weight averaged molecular weights (conv., M and M , respectively), polydispersity index (M /M ),
Mark–Houwink ‘a’ value and hydrodynamic radius (R ) for the nanogels
h
ꢁ1
ꢁ1
Conv. (%)
M
n
(g mol
)
M
w
(g mol
)
M
w
/M
n
MH-a
h
R (nm)
Ng
NgDANS
69
60
32 000
28 200
107 900
138 600
3.37
4.91
0.362
0.355
6.59
7.22
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 41275–41286 | 41279

View Article Online
RSC Advances
Paper
Scheme 7 Graphical representation of the different studied systems
depending on the composition. Blue color indicates the resin and
orange color indicates DANSMA fluorescent label which can be
located linked either in the bulk of the resin (Ng5–20) or only inside the
nanogel particles (NgDANS5–20).
Fig. 2 (a) Kinetic profile and (b) instantaneous rate of polymerization
p
(R ) as a function of conversion for Series 2 formulations (NgDANS0–
ꢁ2
2
0). Irradiated with 365 nm UV light at 25 mW cm for 10 min.
It is assumed that the monomer-swollen nanogel environment
restricts monomer mobility and slows the polymerization
reaction relative to the bulk resin. In prior studies with a lower
viscosity monomer matrix such as TEGDMA, addition of
nanogel increased the overall reaction rate with little effect on
Fig. 1 Initial viscosity values for NgDANS0–20 resin formulations
before photopolymerization. Note: error bars are included but are
generally smaller than the symbols.
18

nal conversion.
Volumetric polymerization shrinkage
As expected with the addition of the high glass transition
ꢀ
Volumetric shrinkage was measured in real-time during the
photopolymerization process with a linometer. The degree of
conversion of the shrinkage specimens was simultaneously
monitored by transmission mode NIR spectroscopy via ber
optic cables. Plots obtained by linometer measurements are
shown in Fig. 3 and data are compiled in Table 3. The nanogel
systems generated much lower shrinkage than the unlled
BisEMA. Volumetric polymerization shrinkage decreases in
a relatively linear manner with increasing nanogel content with
some indication that above the percolation threshold of
ꢄ10 wt%, the degree of the shrinkage reduction is somewhat
diminished. A prior study with a different nanogel/resin
formulation showed progressive shrinkage reduction
temperature nanogels (bulk nanogel T
g
ꢄ 90 C) to a moderately
viscous monomer, progressive differences in rate of polymeri-
zation and limiting conversion values were observed as a func-
tion of nanogel content. Final conversion values decreased from
7
0% for BisEMA control to 63% for the formulation with 20%
nanogel content. Also, a similar trend is found for both the
initial and maximum rates of polymerization. In general, the
increase in nanogel loading caused a decrease in the initial rate
of polymerization and a delay in the maximum rate of poly-
merization. These effects are likely caused by the restricted
environment for BisEMA monomer inside and around the
nanogel particles. The viscosity of neat BisEMA has been shown
21
to promote autoacceleration from the onset of polymerization.
22
continuing to 40 wt% nanogel loading levels. Here, at
41280 | RSC Adv., 2016, 6, 41275–41286
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
approximately compared to the unlled resin, with nal
shrinkage values around 3%.
As shown in Fig. 4, the simultaneously acquired conversion/
time and volumetric shrinkage/time plots can be coordinated to
display shrinkage with respect to conversion, which provides
a nonlinear relationship.
Correlation between volumetric shrinkage with conversion
showed three stages: an initially low rate of shrinkage up to 20%
conversion followed by an accelerating progression in
shrinkage, which accounts for the majority of the shrinkage the
material undergoes most of the shrinkage percentage. Finally,
beyond approximately 60% conversion, shrinkage in the vitri-

ed state is signicantly suppressed relative to conversion. In
the intermediate stage, formulations with low nanogel content
5 and 10%) behave similar to the unlled control resin.
(
However, when including 15 and 20% nanogel, where a more
signicant proportion of the resin matrix is located within the
swollen nanogel particles, a delay in the onset of the more rapid
shrinkage phase is observed above 30% conversion. The
predominant contribution to the volumetric polymerization
shrinkage occurs during the deceleration phase of the poly-
merization reaction, which is reached at higher conversion for
formulations with nanogel contents that are above the perco-
lation threshold. This as well as the reduced overall reactive
group concentration associated with the reactive nanogel
additive as compared with the monomer lead to the lower
shrinkage of the nanogel-modied resin.
Nearly all polymerizations involve some degree of shrinkage
during polymer formation and these results in strain. In
compositionally homogeneous materials, the static polymeri-
Fig. 3 Volumetric polymerization shrinkage as a function of (a) irra-
diation time and (b) the nanogel content (at the end of the polymeri-
ꢁ2
zation). Samples irradiated with 365 nm UV light at 25 mW cm for zation shrinkage obtained depends on the initial reactive group
23,24
1
0 min.
concentration and the degree of conversion achieved.
In our
systems, the shrinkage of the formulations was signicantly
reduced with increasing nanogel fraction. Polymerization stress
Table 3 Mean volumetric polymerization shrinkage values (%) of the is related to the storage modulus and to the polymerization-
composite resins after polymerization. Three replicates were con-
ducted for each system
25
induced strain, so as polymerization shrinkage is reduced,
there is potential for reduction in polymerization stress as well.
Sample
Nanogel (wt%)
Volumetric shrinkage (%)
Ng0
NgDANS0
Ng5
0%
4.82 ꢃ 0.17
5.01 ꢃ 0.23
—
5%
NgDANS5
Ng10
NgDANS10
Ng15
NgDANS15
Ng20
NgDANS20
4.38 ꢃ 0.11
3.36 ꢃ 0.28
3.80 ꢃ 0.17
3.34 ꢃ 0.23
3.50 ꢃ 0.31
3.16 ꢃ 0.35
3.24 ꢃ 0.04
10%
15%
20%
equivalent mass fraction loading levels, the formulations with
the chromophore dispersed in the resin (Ng) showed slightly
lower values although not statistically signicant differences as
compared with formulations with the chromophore linked to
the nanogel (NgDANS). In resin loaded with 20 wt% nanogel Fig.
content (NgDANS20), the shrinkage was reduced by 35%
4
Volumetric polymerization shrinkage of control and
NgDANS5–20 resins as a function of conversion degree. Samples
irradiated with 365 nm UV light at 25 mW cm for 10 min.
ꢁ2
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 41275–41286 | 41281

View Article Online
RSC Advances
Paper
Polymerization stress
introduced as an additional detection technique in the system.
To enable this, storage modulus was continuously measured in
real-time during polymerization by conducting photo-
polymerization between the quartz plates of a photorheometer
To examine the polymerization stress behaviour of our labelled
systems, stress evolution during polymerization was monitored
in real time using a tensometer while simultaneously recording
the degree of conversion by NIR. For the BisEMA control, stress
rapidly increased reaching a nal stress value of about 5.8 MPa
(Fig. 6). In this way, the photorheometer was coupled with both
NIR and uorometric detection systems that allowed simulta-
neous irradiation of the sample, measurement of conversion
and monitoring of uorescence emission while also registering
the evolution of the storage modulus. All of the above measured
parameters, except uorescence emission, are “macroscopic”,
meaning they provide information averaged over the entire
sample.
(
Fig. 5). By incorporation of 5% or 10% nanogel content in the
monomer, signicant progressive reductions in stress (12 and
1%, respectively) were achieved compared to the unlled
4
BisEMA control. In these formulations with relatively low
nanogel content, the stress reduction was achieved without
compromise in nal conversion of the resulting polymers.
However, since the uorescence probe is located in different
spatial domains (inside the nanogel or in the bulk of the resin),
Photo-rheology study. Real-time and simultaneous
monitoring of photopolymerization kinetics, storage modulus
and uorescence emission changes
Once conrmed that the photocuring conditions do not affect
the chromophore beyond its copolymerization and that the
macroscopic polymerization parameters do not appreciably vary
when the uorescent label is added to the system, either inside
the nanogel particles or in the bulk of the photopolymerizable
formulation, spectroscopic uorescence monitoring was
Fig. 6 Photographs of the in-house customized design coupling the
photorheometer (1), FT-NIR (2) and fluorescence detection system (3)
Fig. 5 Polymerization stress in BisEMA monomer and resin formula- is shown. Prior to the plates being set to the desired gap, the open
tions with 5 and 10% nanogel content. Irradiated with 365 nm UV light sample (4) is purged under nitrogen inside a chamber (5) followed by
ꢁ
2
at 1 mW cm for 30 min.
irradiation with 365 nm UV light (6).
41282 | RSC Adv., 2016, 6, 41275–41286
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
the measurement of uorescence emission will give informa- (Fig. 8b, d, f and h on the right), these proles are different as
tion only over the microenvironment of the probe. the probe is not sensing the bulk but the variations occurring
Both Series 1 and 2 formulations showed similar develop- due to polymerization in and around the nanogel particles.
ment of mechanical properties during the photopolymerization Thereby, at the beginning of the polymerization, in all cases, the
reaction (Fig. 7). Final storage modulus values were very similar, uorescence emission intensity ratio decreases during the rst
around 0.80–0.85 GPa, independent of nanogel content. In seconds of irradiation, which indicates a temporary decrease in
Fig. 7b the variation of storage modulus as a function of rigidity inside the nanogel particles. This may be related to
conversion is shown only for the Series 2 materials. The slight a physical expansion of the nanogel as contraction of the bulk
initial increase in storage modulus can be better appreciated on matrix begins. This also could be coupled with some degree of
the logarithmic axis, showing the property development prole monomer diffusion in or out of the nanogel driven by the
that includes macrogelation before 5% conversion based on the localized changes in free volume or conversion differential.
crossover between the loss and storage moduli (not shown here During autoacceleration (expanded inset plots), the delay in
12
but previously demonstrated for nanogel/resin systems ). The polymerization inside the nanogel particles is shown clearly, as
equivalent linear data better shows the dramatic increase in well as a slower rate of polymerization based on the diverged
storage modulus at the end of the irradiation process, which signals up to 40–50% conversion depending on nanogel
corresponds with vitrication.
content. At higher conversions, behaviours inside and outside
For all the formulations under study here, Fig. 8 shows the nanogel are similar. These differences found between the
representative polymerization proles recorded by FT-NIR and bulk and the nanogel during the photopolymerization reaction
by uorescence spectroscopy. Resin-labelled formulations are are more notable with the addition of higher nanogel content to
shown on the le whereas the analogous nanogel-labelled the formulation as might be expected since the fraction of
formulations are shown on the right side.
Before irradiation, a higher emission intensity ratio value is nanogel loading.
observed when the probe is linked to the nanogel particles When monitored by FTIR, the kinetics of polymerization are
NgDANS), which indicates an initial higher rigidity around the similar in the systems loaded with the same nanogel content.
monomer infused in the nanogel particles increases with
(
probe located exclusively inside the nanogel. This behaviour is However, kinetic proles obtained from uorescence moni-
explicitly shown in Fig. 9. The initial uorescence value of FI2/ toring showed two different behaviours when the probe is
FI is approximately the same independent of nanogel content linked to the nanogel or dispersed in the resin.
1
in the sample, which is logical as the only variation is the
nanogel content in the system and the uorescence ratio is not observed during the photopolymerization, where the consider-
dependent on the chromophore concentration. able delay in the degree of double bond conversion inside the
Fig. 10 shows, as an example, the two different behaviours
As irradiation takes place, the evolution of the reaction is nanogel and the higher overall rigidity associated with the
perfectly followed by uorescence emission of the probe. When nanogel domain throughout the polymerization are
DANSMA is dispersed in the resin (Fig. 8a, c, e and g on the le), highlighted.
kinetic proles obtained by uorescence are identical to the
The changes in uorescence intensity ratio of the probe have
FTIR conversion plot throughout the entire polymerization been correlated with the storage modulus (Fig. 11). The
range. When the probe is located exclusively inside the nanogel response of the probe shows a parallelism but different
0
Fig. 7 Storage modulus (G ) during photopolymerization reaction of formulations with 0, 5, 10, 15 and 20 wt% nanogel content vs. (a) irradiation
0
time and (b) conversion. Prior to polymerization, the viscosity effect associated with nanogel loading levels is reflected in the higher initial G
ꢁ
2
0
values. Samples were irradiated with 365 nm UV light at 1 mW cm for 30 min. Identical G data are plotted on either log(left axis) or linear (right
axis) scales.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 41275–41286 | 41283

View Article Online
RSC Advances
Paper
Fig. 8 Variation of fluorescence (emission intensity ratio, lem (1) ¼ 495 nm, lem (2) ¼ 545 nm) and reaction conversion with irradiation time
during photopolymerization reaction of formulations with 5, 10, 15 and 20% nanogel content. Samples irradiated with 365 nm UV light at
ꢁ
2
1
mW cm for 30 min.
41284 | RSC Adv., 2016, 6, 41275–41286
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
2 1
Fig. 9 Initial values of FI /FI before irradiation for all the different
nanogel contents.
Fig. 10 Variation of fluorescence intensity ratio with conversion for
the formulation with 5% nanogel content. Samples irradiated with 365
ꢁ2
nm UV light at 1 mW cm for 30 min.
situations inside and outside nanogel particles. As shown
above, a higher emission intensity ratio value is observed
when the probe is linked to the nanogel particles (NgDANS).
At higher conversions, uorescence behaviour is identical
inside and outside the nanogel although shied in terms of
the intensity ratio. However, at lower conversions, a more
2 1
compressed increase in (FI /FI ) is observed for the nanogel-
bound probe. This fact, together with the different kinetic
behaviour observed in this polymerization range, suggests
that the differences between loaded and unloaded systems
are mostly produced at this stage of polymerization.
These preliminary results are very promising. Since
conventional techniques used to study the polymerization
process provide information about the bulk or the macroscopic
system, labelling with uorescent probes in different phases of
a heterogeneous system could allow the detailed study the
potentially complex processes that take place in each phase.
Fig. 11 Variation of storage modulus with fluorescence intensity ratio
Conclusions
(
l
em (1) ¼ 495 nm, lem (2) ¼ 545 nm) during the photopolymerization
of formulation with 5, 10, 15 and 20% nanogel content. Samples irra-
The uorescent probe DANSMA has been included in a resin
composite in two different ways, either chemically attached to
the nanogel or dispersed in the matrix monomer BisEMA. The
ꢁ2
diated with 365 nm UV light at 1 mW cm for 30 min.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 41275–41286 | 41285

View Article Online
RSC Advances
Paper
probe showed intense uorescence emission, which varies in
intensity and in the maximum wavelength of emission during
the photopolymerization reaction. This provides the possibility
to study in situ and in real time how the polymerization kinetics
are locally affected by the nanogel and its structure as the
microstructural differences in the networks formed are
responsible of the changes in the uorescence spectra. These
6 C. Chen, J. Liu, F. Sun and J. W. Stansbury, J. Polym. Sci., Part
A: Polym. Chem., 2014, 52, 2830–2840.
7 J. Liu and J. W. Stansbury, Dent. Mater., 2014, 30, 1252–1262.
8 P. Bosch, F. Catalina, T. Corrales and C. Peinado, Chem.–Eur.
J., 2005, 11, 4314–4325.
9 S. Medel and P. Bosch, React. Funct. Polym., 2015, 93, 101–
110.
changes have been simultaneously monitored and correlated 10 Z. Wang, F. A. Landis, A. A. M. Giuseppetti, S. Lin-Gibson
with photopolymerization kinetics and with the mechanical
data obtained by photo-rheology study.
and M. Y. M. Chiang, Dent. Mater., 2014, 30, 1316–1324.
11 D. A. Abu-elenain, S. H. Lewis and J. W. Stansbury, Dent.
Mater., 2013, 29, 1173–1181.
While conventional techniques to study polymerization
processes give average information about the macroscopic 12 J. Liu, I. Y. Rad, F. Sun and J. W. Stansbury, Polym. Chem.,
system, uorescence is sensitive to the microenvironment of the 2014, 5, 227–233.
probe and this allowed us to distinguished two different 13 A. J. d. Gee, A. J. Feilzer and C. L. Davidson, Dent. Mater.,
behaviours inside the nanogel and in the matrix resin. These 1993, 9, 11–14.
results are promising since the use of uorescent probes in the 14 C. S. Pfeifer, N. D. Wilson, Z. R. Shelton and J. W. Stansbury,
different phases of a composite material could allow study of
Polymer, 2011, 52, 3295–3303.
the processes that take place in each of them.
15 H. Lu, J. W. Stansbury, S. H. Dickens, F. C. Eichmiller and
C. N. Bowman, J. Mater. Sci.: Mater. Med., 2004, 15, 1097–
1103.
Acknowledgements
1
6 H. Lu, J. W. Stansbury, S. H. Dickens, F. C. Eichmiller and
C. N. Bowman, J. Biomed. Mater. Res., Part B, 2004, 71, 206–
213.
Financial support for S. M. provided by the Ministerio de
Econom ´ı a y Competitividad (MINECO) through the Projects
MAT 2009-09671 and MAT 2012-31709 is acknowledged. The 17 C. Peinado, E. F. Salvador, T. Corrales, P. Bosch and
author S. M. thanks MINECO for the EEBB-I-13-064064 grant
and the BES 2010-040128 FPI fellowship. NIH/NIDCR
F. Catalina, J. Polym. Sci., Part A: Polym. Chem., 2004, 42,
1227–1238.
R01DE014227, the NSF Industry/University Cooperative 18 J. Liu, G. D. Howard, S. H. Lewis, M. D. Barros and
Research Center for Fundamentals and Applications of Photo-
J. W. Stansbury, Eur. Polym. J., 2012, 48, 1819–1828.
polymerization, and CONACYT (Consejo Nacional de Ciencia y 19 Y. Hoshino, Y. Arata, Y. Yonamine, S.-H. Lee, A. Yamasaki,
Tecnolog ´ı a) 308269 supported the implementation of ber-
R. Tsuhara, K. Yano, K. J. Shea and Y. Miura, Polym. J.,
optic coupled spectroscopic methods for the kinetic analysis
2015, 47, 220–225.
of photopolymerization reactions. Additional support was 20 M. Trujillo-Lemon, M. S. Jones and J. W. Stansbury, J.
provided by NIH/NIDCR R01DE022348 and the I/UCRC for
Fundamentals and Applications of Photopolymerization.
Biomed. Mater. Res., Part A, 2007, 83, 734–746.
21 S. H. Dickens, J. W. Stansbury, K. M. Choi and C. J. E. Floyd,
Macromolecules, 2003, 36, 6043–6053.
2
2 R. R. Moraes, J. W. Garcia, M. D. Barros, S. H. Lewis,
C. S. Pfeifer, J. Liu and J. W. Stansbury, Dent. Mater., 2011,
Notes and references
1
2
3
J. L. Ferracane, Dent. Mater., 2011, 27, 29–38.
J. W. Stansbury, Dent. Mater., 2012, 28, 13–22.
N. B. Cramer, J. W. Stansbury and C. N. Bowman, J. Dent.
Res., 2011, 90, 402–416.
R. R. Mor ˜a es, J. W. Garcia, N. D. Wilson, S. H. Lewis,
M. D. Barros, B. Yang, C. S. Pfeifer and J. W. Stansbury, J.
Dent. Res., 2012, 91, 179–184.
2
7, 509–519.
2
2
2
3 J. W. Stansbury, M. Trujillo-Lemon, H. Lu, X. Ding, Y. Lin
and J. Ge, Dent. Mater., 2005, 21, 56–67.
4 M. Dewaele, D. Truffier-Boutry, J. Devaux and G. Leloup,
Dent. Mater., 2006, 22, 359–365.
5 R. R. Braga and J. L. Ferracane, Crit. Rev. Oral Biol. Med.,
4
5
2004, 15, 176–184.
P. K. Shah and J. W. Stansbury, Dent. Mater., 2014, 30, 586–
593.
41286 | RSC Adv., 2016, 6, 41275–41286
This journal is © The Royal Society of Chemistry 2016