The influence of vinyl activating groups on β-allyl sulfone-based chain transfer agents for tough methacrylate networks

Article abstract of DOI:10.1002/pola.27993

In radical polymerization of monofunctional monomers, addition fragmentation chain transfer (AFCT) agents are well known to regulate polymerization and yield polymers with lower molecular weights and narrower molecular weight distributions. Papers concerning bulk photopolymerization of monomer mixtures with AFCT agents are rarely found in literature. In this article, AFCT reagents based on β-allyl sulfones with different vinyl activating groups were synthesized and compared. The compounds were tested in mono- and difunctional monomer systems providing information about the influence on photoreactivity, molecular weight, as well as thermal and mechanical properties of the resultant polymers. Where more potent activating groups (-Ph, -CN) markedly influenced polymerization at lower concentrations, the AFCT reagent with an ester activating group reacted at a similar rate to the methacrylate monomer (C_T ≈ 1) and provided the best overall performance.

Full text of DOI:10.1002/pola.27993

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
The Influence of Vinyl Activating Groups on b-Allyl Sulfone-Based Chain
Transfer Agents for Tough Methacrylate Networks
Paul Gauss,1,2 Samuel Clark Ligon-Auer,1,2 Markus Griesser,1,2 Christian Gorsche,1,2
Helena Svajdlenkova,4 Thomas Koch,5 Norbert Moszner,2,3 Robert Liska1,2
1Institute of Applied Synthetic Chemistry, TU Wien, Getreidemarkt 9/163/MC, 1060 Vienna, Austria
2Christian Doppler Laboratory for Photopolymers in Digital and Restorative Dentistry, Getreidemarkt 9/163/MC, 1060 Vienna,
Austria
3Ivoclar Vivadent AG, Bendererstrasse 2, 9494 Schaan, Liechtenstein
4
ꢀ
ꢀ
Polymer Institute of the Slovakian Academy of Science, Dubravska Cesta 9, 845 41 Bratislava, Slovakia
5Institute of Materials Science and Technology, TU Wien, Getreidemarkt 9, 1060 Vienna, Austria
Correspondence to: R. Liska (E-mail: robert.liska@tuwien.ac.at)
Received 27 September 2015; accepted 7 November 2015; published online 17 December 2015
DOI: 10.1002/pola.27993
ABSTRACT: In radical polymerization of monofunctional mono-
mers, addition fragmentation chain transfer (AFCT) agents are
well known to regulate polymerization and yield polymers with
lower molecular weights and narrower molecular weight distri-
butions. Papers concerning bulk photopolymerization of mono-
mer mixtures with AFCT agents are rarely found in literature.
In this article, AFCT reagents based on b-allyl sulfones with dif-
ferent vinyl activating groups were synthesized and compared.
The compounds were tested in mono- and difunctional mono-
mer systems providing information about the influence on
photoreactivity, molecular weight, as well as thermal and
mechanical properties of the resultant polymers. Where more
potent activating groups (-Ph, -CN) markedly influenced poly-
merization at lower concentrations, the AFCT reagent with an
ester activating group reacted at a similar rate to the methacry-
late monomer (CT ꢀ 1) and provided the best overall perform-
C
V
ance.
2015 Wiley Periodicals, Inc. J. Polym. Sci., Part A:
Polym. Chem. 2016, 54, 1417–1427
KEYWORDS: addition-fragmentation chain transfer; mechanical
properties; methacrylates; networks; photopolymerization;
shrinkage stress
and oligomeric methacrylates. Ring opening monomers are
also investigated although poor reactivity has prevented
their more widespread application.8
INTRODUCTION Although the photopolymerization of (meth)-
acrylates has been known for more than six decades, this
technique is still finding more and more fields of advanced
application in packaging, 3D-printing, coatings, and biomedi-
cal fields.1,2 Important benefits of photo curing over thermal
methods are that the reaction tends to be quicker and
requires less energy. In addition, low molecular weight
monomers are normally used to lower viscosity for process-
ing, making the technology potentially free of volatile organic
compounds.3 While acrylates are more reactive and thus of
greater interest in most industrial fields, methacrylate mono-
mers are much less irritating to human skin and are thus
preferred for dental and emerging tissue engineering appli-
cations.4,5 Within these fields, shrinkage and associated
shrinkage stress and high brittleness of the photopolymer
networks remain as persistent problems.6,7 A number of
approaches have been taken to reduce shrinkage and shrink-
age stress including the use of methacrylate monomers,
which shrink less such as bisphenol-A glycidyl methacrylate
Chain transfer is yet another approach for modifying the
cure kinetics and network architecture of a methacrylate-
based polymer. Alkyl thiols are certainly one of the better
studied classes of chain transfer agents, finding industrial
application for the regulation of radical chain growth.9 In the
presence of thiols and a radical initiator, acrylates can
undergo both propagation and chain transfer.10 The propaga-
tion occurs to a greater extent so that a stoichiometric mix-
ture of thiols and acrylates will still contain a substantial
concentration of unreacted thiol after polymerization.11 The
reaction of alkyl thiols with methacrylates is by comparison
substantially slower, which can also limit its utility. While
thiols can be quite effective in improving the impact strength
of acrylate and methacrylate-based resins, the Young’s modu-
lus of such polymers is significantly reduced due to flexible
thio-ether linkages.12 Other problems with thiols are odor
Additional Supporting Information may be found in the online version of this article.
C
V
2015 Wiley Periodicals, Inc.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1417–1427
1417

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 1 Mechanism of irreversible AFCT with b-allyl sulfone reagents.
and poor storage stability due to thermally induced Michael
addition.13 Where thiols function via transfer of a hydrogen
atom, an alternate mode of chain transfer is that of
addition-fragmentation chain transfer (AFCT).14 AFCT may
be classified as either reversible or irreversible. The first
case, reversible AFCT (commonly shortened to RAFT) is a
form of living polymerization and provides polymers with
polydispersity index (PDI) values approaching unity.15,16
More recently, photo-RAFT has been developed, however
high conversions can take hours or days instead of seconds
as is the usual case with (meth)acrylate polymerization.17,18
Irreversible AFCT can also be used to regulate molecular
weight. Although the PDI will not be as low as that achieved
by RAFT (values of 1.5–2.0 are common),19 polymerization
can proceed at a rate comparable to that without chain
transfer.
the CTA and possibly allow better control of the network
structure. Furthermore, the fairly high concentration of ASEE
required to illicit an effect (typically 10–20 wt %) may limit
its use in real world formulations.23 Therefore, we concluded
to also investigate the newly synthesized compounds at
lower concentrations.
EXPERIMENTAL
Materials and Methods
All chemicals were purchased from Sigma Aldrich or Fluka
and used as received unless otherwise noted. Dichlorome-
thane (CH2Cl2) used for synthesis was of high purity grade.
Urethane dimethacrylate (UDMA, CAS: 72869-86-4), 1,10-
decandiol dimethacrylate (D3MA, CAS: 72829-09-5) and
24
R
V
bis(4-methoxybenzoyl)diethyl germanium (Ivocerin ) were
obtained from Ivoclar Vivadent and used as received.
Bis(2,6-dimethoxybenzoyl)-2,4,4-trimethylpentylphosphine
oxide (CGI403) was provided by Ciba. Melting points were
measured on a Zeiss axioscope microscope with a Leitz
heating block. GC-MS measurements were carried out on a
Thermo Fisher Scientific DSQ II employing a Fused Silica
capillary column (30 m 3 0.25 mm). All experiments were
performed at least in triplicate. Results are given as an
average of three measurements with standard deviation.
One of the more commonly used classes of irreversible AFCT
agent (shortened throughout to CTA) is based on an a-olefin
with a vinyl activating group and a labile bond near but not
directly adjacent to the olefin.14 The propagating radical will
thus add to the CTA vinyl group and the resulting radical
may then continue to propagate or fragment to form a new
initiating radical. Although other relaxation paths are possi-
ble, the majority of CTAs are designed to fragment via b-
scission. This gives rise to a terminal alkene and a new radi-
cal. The efficiency of chain transfer is then dependent on
kinetics of addition and fragmentation, and on the initiation
capacity of the newly formed radical. The labile bond of an
irreversible CTA is normally a single bond between carbon
and a heteroatom such as sulfur or phosphorus. In the first
case, allyl sulfides are one of the better studied families of
CTAs where the thiyl radical formed after fragmentation has
good reactivity towards remaining vinyl monomers.20 We
have recently found that allyl sulfones are particularly effi-
cient CTAs for regulating kinetic chain length in methacrylate-
based networks (see mechanism depicted in Scheme 1).21
Such b-allyl sulfone CTAs are effective reagents in
dimethacrylate-based resins for lowering shrinkage stress and
sharpening the glass transition of the network. More recently,
difunctional b-allyl sulfone CTAs were found to improve
impact strength while maintaining a high modulus.22
Synthesis
2-Ethyl-2-(tosylmethyl)acrylate (ASEE)
Ethyl-2-(bromomethyl)acrylate (1 equiv, 4.83 g, 25 mmol),
sodium p-toluenesulfinate (1.1 equiv, 4.9 g, 27.5 mmol) and
polyethylene glycol 400 (0.025 equiv, 0.24 g, 0.6 mmol) were
added to 35 mL THF. The resulting solution was refluxed for
3 h. Afterward, the mixture was extracted with 100 mL
water and 3 3 100 mL diethyl ether. The organic phase was
washed with brine (100 mL) and dried over Na2SO4. After
In this article, we investigate a series of b-allyl sulfone CTAs
(see Fig. 1) with different vinyl activating groups. In compar-
ison to the previously investigated ester 2-ethyl-2-(tosylme-
thyl)acrylate (ASEE), the other CTAs all have activating
groups with different capability of stabilizing the intermedi-
ate radical, which should influence the rate of addition to
FIGURE 1 b-allyl sulfone CTAs with different activating groups.
1418
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1417–1427

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
evaporation of the solvent, the resulting oil was crystallized
overnight. Recrystallization from diethyl ether yielded pure
ASEE as colorless crystals (5.3 g, 79% yield).
water (50 mL). Recrystallization from water gave 7.9 g (66%
yield) of intermediate 2-(tosylmethyl)acrylic acid. Mp: 189–
190 8C. 1H NMR (200 MHz, CDCl3, d, ppm): 8.81 (bs, 1H,
COOH), 7.67 (d, J 5 8.6 Hz, 2H, Ar-H), 7.27 (d, J 5 8.6 Hz,
2H; Ar-H), 6.55 (s, 1H; 5CH2), 5.94 (s, 1H; 5CH2), 4.04 (s,
2H; -SO2-CH2-), 2.36 (s, 3H; Ar-CH3).
Mp: 44.5 8C. TLC (Rf): 0.5 (PE/EtOAc51:1). 1H NMR (200
MHz, CDCl3, d, ppm): 7.72 (d, J 5 8.22 Hz, 2H; Ar-H), 7.31
(d, J 5 8.22 Hz, 2H; Ar-H), 6.50 (s, 1H; 5CH2), 5.91 (s, 1H;
5CH2), 4.14 (s, 2H; -SO2-CH2-), 4.01 (q, J 5 7.24 Hz, 2H; -O-
CH2-CH3), 2.44 (s, 3H; Ar-CH3), 1.18 (t, J 5 7.04 Hz, 3H; -O-
CH2-CH3).
2-(Tosylmethyl) acrylic acid (3.00 g, 12.5 mmol) was
refluxed in 30 mL thionyl chloride for 2 hours. Afterwards,
excess thionyl chloride was removed by distillation under
vacuum. The acid chloride was then dissolved in 100 mL
anhydrous CH2Cl2 and cooled to 0 8C. To this solution N,N-
methylpropylamine (75 mmol) was added dropwise. After
12 h at room temperature the solvent was removed. The
crude product was dissolved in 50 mL CH2Cl2 and washed
with 1 N HCl (2 3 20 mL) and brine (1 3 20 mL). After
drying with Na2SO4, further purification of N-methyl-N-
propyl-2-(tosylmethyl)acrylamide (ASA) was achieved by col-
umn chromatography. About 700 mg (19% yield) product
was collected as a clear liquid.
2-(Tosylmethyl)acrylonitrile (ASN)
Tosyliodide (4.88 g, 17 mmol) and methacrylonitrile (1.16 g,
17 mmol) were dissolved in CCl4 (100 mL) and stirred at
room temperature for 6 h. The solvent and any formed
iodine were removed under vacuum. Another 100 mL CCl4
was added along with triethylamine (TEA, 6.88 g, 68 mmol).
The reaction was heated to reflux and after 12 h cooled to
room temperature. The brown solution was then washed
with a 5% solution of sodium dithionite (2 3 20 mL), 1N
HCl (1 3 20 mL) and brine (1 3 20 mL). The organic phase
was collected, dried with Na2SO4, filtered, and concentrated
on a rotary evaporator. The product was further purified by
column chromatography. About 790 mg 2-(tosylmethyl)acry-
lonitrile (ASN) was isolated as fine white needles (21%
yield).
TLC (Rf): 0.25 (PE/EtOAc 1:1 1 0.5% AcOH). 1H NMR (200
MHz, CDCl3, d, ppm): 7.78 (d, J 5 8.1 Hz, 2H; Ar-H), 7.34 (d,
J 5 8.1 Hz, 2H; Ar-H), 5.51 (bs, 1H; 5CH2), 5.41 (s, 1H;
5CH2), 4.09 (s, 2H; -SO2-CH2-), 3.5–2.7 (m, 5H; N-CH3, N-
CH2-CH2-CH3), 2.38 (s, 3H; Ar-CH3), 1.7–1.3 (m, 2H; N-CH2-
CH2-CH3), 0.85 (t, J 5 7.5 Hz, 3H; N-CH2-CH2-CH3). 13C NMR
(50 MHz, CDCl3, d, ppm): 168.9 (C5O), 144.8 (C4), 136.5
(C4), 131.9 (C4), 129.8 (C3), 128.3 (C3), 124.24 (C2), 60.71
(C2), 50.5 (C2), 34.9 (C1), 21.6 (C1), 19.9 (C2), 11.1 (C1).
Mp: 109–110 8C. TLC (Rf): 0.5 (CH2Cl2). 1H NMR (200 MHz,
CDCl3, d, ppm): 7.73 (d, J 5 8.45 Hz, 2H; Ar-H), 7.33 (d, J 5
8.45 Hz, 2H; Ar-H), 6.15 (s, 1H; 5CH2), 5.94 (s, 1H; 5CH2),
3.84 (s, 2 H; -SO2-CH2-), 2.41 (s, 3H; Ar-CH3).
EIMS (70 eV) m/z (%): 295.05(2) [M]1, 222.99(40) [M-
N(C3H7)(CH3)]1, 154.99(60) [TosSO2]1, 140.09(100) [M-
TosSO2]1, 91(48) [C6H5(CH3)]1.
2-(Tosylmethyl)styrene (BAS)
a-Methyl styrene (2.95 g, 25 mmol), tosyl chloride (4.77 g,
25 mmol), Cu(I)Cl (2.48 g, 25 mmol), and triethylamine
(2.53 g, 25 mmol) were added to 70 mL dry acetonitrile and
heated under argon to reflux. After 3 h, the reaction was
cooled and diluted with 50 mL 0.1 N HCl. The mixture was
extracted with CH2Cl2, the collected organic phases dried
over Na2SO4, and the solvent was evaporated giving a brown
crystalline raw product. Recrystallization from ethanol gave
colorless crystals of poor purity. The product was further
purified by column chromatography giving 3.60 g (53%
yield) 2-(tosylmethyl)styrene (BAS) as a colorless solid.
Elementary analysis: Anal. Calcd. for C15H21NO3S: C 60.99; H
7.17; N 4.74; S 10.85; found: C 60.18; H 7.30; N 4.76; S
10.74.
Sample Preparation for Analysis Monofunctional Test
Systems
For the tests on the monofunctional system, benzyl methac-
rylate (BMA) with 1.0 wt % Ivocerin as photoinitiator was
used as reference mixture. Samples were prepared with 5
double bond percent (db%) to 20 db% of CTA, depending on
the solubility of the CTA. The db% refers to the percentage
of all C5C double bonds in the system including monomer
and CTA. To prepare each sample, monomer, the desired
amount of CTA, and the photoinitiator were weighed into a
brown glass vial, put into an ultra-sonic bath at 40 8C until
the solids dissolved and then mixed with a vortex mixer.
Mp: 96–99 8C. TLC (Rf): 0.3 (CH2Cl2). 1H NMR (200 MHz,
CDCl3, d, ppm): 7.59 (d, J 5 7.64 Hz, 2H; Ar-H), 7.5–6.9 (m,
7H; Ar-H), 5.51 (s, 1H; 5CH2), 5.14 (s, 1H; 5CH2), 4.18 (s,
2H; -SO2-CH2-), 2.32 (s, 3H; Ar-CH3).
N-Methyl-N-propyl-2-(tosylmethyl)acrylamide (ASA)
(2-Bromomethyl)acrylic acid (8.25 g, 50 mmol) was dis-
solved in hot methanol (250 mL) and neutralized with NaOH
(2 g, 50 mmol). Sodium p-toluenesulfinate (8.91 g, 50 mmol)
was added in portions. The reaction mixture was then
heated to reflux for 2 h and concentrated on the rotary evap-
orator. The concentrate was taken up with 500 mL water
and acidified with 1 M HCl to give a precipitate. After cooling
overnight the precipitate was filtered and washed with cold
Difunctional Test Systems
Samples for real-time (RT)-near-infrared (NIR) photorheol-
ogy, dynamic mechanical thermal analysis (DMTA), and Dyn-
stat impact resistance measurements were based on two
dimethacrylate monomers (UDMA and D3MA) as an equimo-
lar mixture with 1.0 wt % Ivocerin as photoinitiator for the
DMTA and Dynstat experiments and 0.3 wt % for the
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1417–1427
1419

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
rheology experiments. In both cases, samples were prepared
with 5–20 db% of CTA, depending on the solubility of the
CTA.
stants for the addition of the phosphinoyl radicals to the
monomer double bonds, kadd, were determined in pseudo-
first-order
according
to
the
following
equation
k
obs 5 k0 1 kadd ꢂ kquencher
.
Photo-DSC
Photo-DSC was conducted on a Netzsch DSC 204 F1 with
auto-sampler. The measurements were carried out at 25 8C
under nitrogen atmosphere. Photoreactivity of the monomers
was tested by weighing accurately 12 6 0.5 mg of the sample
into an aluminum DSC pan, which was subsequently placed
in the DSC chamber after closing it with a glass lid. The sam-
RT-NIR Photorheology
RT-NIR photorheology experiments were performed on an
Anton Paar MCR 302 WESP with a P-PTD 200/GL Peltier
glass plate and a disposable PP25 measuring system. The
rheometer is additionally coupled with a Bruker Vertex 80
FTIR spectrometer with external mirrors to guide the IR
beam though the sample being analysed by rheology. In a
typical experiment, 130 mL of sample was placed at the
centre of the glass plate with the head of the rheometer held
at a measuring position of 200 mm. The formulations were
subjected to an oscillatory sheer with an angular strain of
1% and a frequency of 1 Hz. Polymerization was induced by
UV-irradiation (300 s, 400–500 nm, 0.5 W cm22) projected
via a waveguide from the underside of the glass plate using
a filtered Exfo OmniCure S2000 broadband Hg-lamp. The IR
beam also passes through the glass plate holding the sample
and is reflected by the flat plate rheology head before
returning to an external MCT detector. During the measure-
ments, change in normal force along with the storage modu-
lus and loss modulus were recorded. Data were collected
with a frequency of 1 Hz before and 5 Hz during irradiation.
The double bond conversion (DBC) was determined by
recording a set of single spectra (time interval 5 0.47 s) and
then integrating the respective double bond bands at
ꢁ6160 cm21. The ratio of the double bond signal at the start
and the end of the measurement were used to calculate
the ultimate DBC. The final DBC is defined as the average
DBC of the last 30 data points. DBC at the point of gelation
(DBC g) is defined as the conversion when the storage and
loss modulus intersect. The time to 95% of total conversion
t95DBC is estimated graphically from the DBC diagram as the
time till the moving average (5 measuring points) reaches
95% of total conversion. (c.f. Supporting Information).
ple chamber was purged with a N2 flow (ꢁ 20 mL min21
)
for 4 min and, subsequently, the samples were irradiated
with filtered UV-light (400–500 nm) from an Exfo OmniCure
S2000 broadband Hg-lamp with a light intensity of 1 W
cm22 at the exit of the light guide. The samples were
exposed to light for 2 3 5 min, and the heat flow was
recorded as a function of time. Time to attain 95% conver-
sion (t95%) is defined as the time required for generated
heat to equal 95% of DHp.25
NMR Spectroscopic Analysis
NMR spectra (200 MHz for 1H and 50 MHz for 13C, respec-
tively) were recorded with a Bruker AC 200 spectrometer.
To determine the conversion of CTAs and BMA, the monomer
mixtures and polymerized samples from photo-DSC measure-
ments were dissolved in CDCl3 (deuteration >99.5%). The
solvent signal was used as internal reference. The conversion
was calculated from the integral difference between the dou-
ble bond peaks before and after curing. The signal of the
tosyl methyl group from the CTAs appears as a singlet at
about 2.6 ppm. This peak was used as reference, as it is not
changed by polymerization. By normalizing the double bond
signals of the CTA and of the monomer to this peak, it is
possible to quantify the decrease of these signals after
polymerization.
GPC
Following NMR, the samples were diluted with 0.5 mL THF
and examined by GPC. GPC was performed with a Waters
GPC using three columns connected in series (Styragel HR
0.5, Styragel HR 3 and a Styragel HR 4) and a Waters 2410
RI detector. The columns were maintained at 40 8C and a
flow rate of 1.0 mL min21 was used. Polystyrene standards
were used for molecular weight calibration and THF was
used as solvent. The polymers were not precipitated prior to
GPC characterization.
Dynamic Mechanical Thermal Analysis
DMTA was performed with an Anton Paar MCR 301 with a
CTD 450 oven and an SRF 12 measuring system. Rectangular
(ꢁ 5 3 2 3 40 mm3) samples for DMTA were formed in a
silicone mold placed into a Lumamat 100 light chamber (Ivo-
clar Vivadent), which provides light of 400–580 nm. Expo-
sure was performed for 10 min on the top and 10 min on
the back side. Samples were polished with sandpaper before
placement into the DMTA. The samples were tested in tor-
sion mode with a frequency of 1 Hz and strain of 0.1% and
the temperature was increased from 2100 to 200 8C with a
heating rate of 2 8C min21. The glass transition temperature
was defined as the temperature at the maximum dissipation
factor (tan d).
Laser Flash Photolysis
Nanosecond transient absorption experiments were per-
formed on a LKS80 Spectrometer (Applied Photophysics,
UK). Excitation was performed using the third harmonic
(355 nm, 10–20 mJ/pulse, ca. 8 ns) of a Spitlight Compact
100 (InnoLas, Germany) Nd:YAG laser. Concentration of
CGI403 in acetonitrile was adjusted to achieve an absorb-
ance of ca. 0.3 at 355 nm. The transient absorption spectra
were recorded in a quartz cuvette (1 3 1 cm2). Samples
were purged and constantly bubbled with argon to refresh
the sample and avoid sample decomposition. The rate con-
Dynstat Impact Resistance
Rectangular (ꢁ 10 3 2 3 15 mm3) samples for Dynstat
tests were formed by curing in a silicone mold (details see
DMTA). Samples were polished with sandpaper before
1420
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1417–1427

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
SCHEME 2 Synthetic routes to b-allyl sulfone CTAs.
placement into a Karl Frank Type 573 Dynstat machine. The
samples were destroyed with a 5 kpcm hammer and the val-
ues were read from the scale. The results in kpcm were con-
verted into kJ and divided by the area of fracture in m2.
Testing of CTAs in a Monofunctional Model Formulation
To better understand the reactivity of the synthesized CTAs,
their copolymerization behavior and the influence on the
molecular weight of formed polymers, photo-DSC, NMR, and
GPC measurements were performed with a monofunctional
methacrylate monomer (formulation a). Benzyl methacrylate
(BMA) was used as monomer since the benzyl moiety does
not interfere with the 1H NMR signals of the CTAs. In addi-
tion BMA is a reasonable solvent for additives and is high
boiling, which improves gravimetric analysis. Samples poly-
merized via photo-DSC could be used directly for NMR and
GPC, such that the results of these three analytical methods
should correlate directly for each sample.
RESULTS AND DISCUSSION
Synthesis of b-Allyl Sulfone CTAs
Generally, b-allyl sulfones can be easily prepared by radical
reaction of tosyl iodide with methacrylic acid derivatives.26
The alkyl iodide intermediate addition product is then
directly transformed (without workup) into the unsaturated
sulfone by eliminating HI with TEA. This procedure was
used for the synthesis of nitrile ASN in a yield of 21%
[Scheme 2(a)] and can also be used for the synthesis of ester
ASEE. Nevertheless, ASEE can be obtained more easily in
high yield (79%) by condensation of ethyl 2-(bromomethyl)-
acrylate with sodium p-toluenesulfinate [Scheme 2(b)].27
Amide ASA could be synthesized by this method as well, but
it was impossible to obtain the pure substance, as countless
side reactions occurred during the process. Therefore, amide
ASA was synthesized in two steps: (i) formation of 2-(tosyl-
methyl)-acrylic acid from 2-(bromomethyl)acrylic acid and
sodium p-toluenesulfinate and (ii) reaction of the acyl chlo-
ride of the intermediate with N,N-methylpropylamine (13%
overall yield) [Scheme 2(c)]. As the previously described
general procedure did not lead to the desired product, two
different synthetic routes to styrene-based BAS were tested.
The first route was based on the addition of tosyl hydrazide
to a-methylstyrene and provided the product in 41% yield.28
By comparison, the copper catalyzed addition of p-tosyl chlo-
ride to methyl styrene followed by elimination with TEA pro-
vided BAS in 53% yield [Scheme 2(d)].29
Assessing the Photoreactivity of CTAs by Photo-DSC
Photo-DSC provides information on the kinetics and thermo-
dynamics of the polymerization of BMA in the presence of
different CTAs. In some cases, the solubility of the CTAs in
the monomer was poor and limited the experiment to mix-
tures containing only 5 db% CTA (ASN, BAS). By compari-
son, ester ASEE and amide ASA were more soluble and
could also be tested at higher concentrations. The time to
reach the maximum rate of polymerization (tmax) is in the
same range for all tested mixtures (Table 1). BMA is a mono-
functional monomer which does not gel but does solidify
when polymerized neat. Auto-acceleration is much less pro-
nounced with BMA than with typical difunctional monomers
and thus tmax is reached much later.23 While CTAs moderate
auto-acceleration of difunctional monomers and thus delay
tmax, the effect is different with BMA. In this case, tmax is not
delayed but actually comes slightly earlier. Also the time to
reach 95% conversion (t95%) is reduced by addition of the
tested CTAs. This indicates that the entire polymerization
process (and not just the initial stages) is faster with CTAs.
Explanation for both kinetic terms could be given by the fact
that chain transfer should prevent runaway propagation and
thus lower the viscosity of the reacting mixture. This should
in turn allow monomers to diffuse more easily. Only mixture
None of the synthesized substances showed UV–vis absorp-
tion above 350 nm, so interference during polymerization
using 400–500 nm light can be excluded.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1417–1427
1421

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 1 Analysis of BMA Polymerized Neat and with Different Concentrations of CTA (statistics c.f. Supporting Information)
Photo-DSC
NMR
DBC – BMA (%)
GPC
Mn (Da)
Mixture
CTA
[CTA] (db%)
tmax (s)
t95% (s)
DBC (%)
DBC – CTA (%)
PDI
a
b
c
d
e
f
–
0
20.6
19.0
17.7
18.5
18.6
16.7
15.6
122
107
104
113
135
94.8
102
47
43
52
43
72
33
26
45
45
56
53
73
36
33
–
6,580
1,530
890
2.9
2.3
1.5
2.3
1.8
1.7
1.6
ASEE
ASEE
ASA
ASA
ASN
BAS
5
56
41
9.5
14
52
36
20
5
5,500
3,920
1,900
1,880
20
5
g
5
e containing 20 db% of amide ASA is slightly retarded com-
pared to the pure monomer. This may be caused by steric
hindrance of the CTA. From literature it is known that acryl-
amides are quite reactive, even more so than acrylic esters.
Methacrylamides are by comparison much less reactive (a
difference in reactivity even more dramatic than that
between acrylic and methacrylic esters). Substitution on the
N atom in methacrylamides further decreases the reactivity
in homopolymerization dramatically. Disubstitution, even
with methyl groups completely suppresses homopolymeriza-
tion due to steric hindrance.30
to be substantiated in the subsequent NMR experiments. The
5 db% mixtures of styrene-based BAS and nitrile ASN (mix-
tures g and f) show significantly reduced monomer DBC val-
ues. This can be rationalized by the good resonance
stabilization of the generated radicals.
1
Conversion Determination by H NMR Spectroscopy
To quantify the conversion of monomer and chain transfer
reagent, 1H NMR spectra of photo-DSC samples were
recorded before and after irradiation. The difference in the
integral of the vinyl protons directly correlates to the DBC of
monomer BMA and of the CTA. As BMA and all of the CTAs
have two vinyl protons, both signals were integrated and the
arithmetic average was used for calculation of the DBC. In
the case that vinyl proton signals of the CTA overlap with
those of the monomer, only non-overlapping signals were
integrated.
Defining the transfer reaction as energy neutral,21 which was
recently proven for esters analogous to ASEE, heat measured
in photo-DSC arises principally from the polymerization of
BMA. The heat of polymerization was calculated from both
NMR and DSC. To do so, pure samples of BMA were poly-
merized in the DSC device and the DBC of BMA in the sam-
ples was afterwards determined by NMR. Correlating the
known conversion with the heat of polymerization measured
in DSC gives the experimental DH0 value for BMA (DH0 5 52
kJ mol21), which is close to 56 kJ mol21 from literature.31
Taking into account that the mixtures contain 80–95 mol %
BMA, the measured heat of polymerization has to be cor-
rected by the weight fraction of BMA (mass of BMA mBMA
divided by the total mass of the formulation mtot). With eq
1, the DBC of BMA could be calculated from DSC.21
The center of Table 1 shows the conversion of neat BMA
(mixture a) and of the CTA and BMA in all other formula-
tions. The pure monomer BMA reaches a DBC of 45% as
determined by NMR, which is very close to the value deter-
mined by DSC. Ester ASEE (mixtures b and c) and the mono-
mer are consumed at comparable rates, so the molar ratio to
BMA is nearly constant during the entire polymerization.
This indicates a transfer constant CT close to 1. On the other
hand, DBC of amide ASA is very low at both tested concen-
trations (mixtures d and e). While DBC of the CTA is low,
the conversion of BMA for the mixture with 20 db% ASA is
much higher than all of the other samples. The low conver-
sion of ASA supports the theory of a thinning effect pro-
posed before. As the amide activated CTA is very unreactive,
only small amounts take part in the polymerization while
diluting the concentration of free radicals. Nitrile ASN (mix-
ture f) is consumed at a much higher rate than the mono-
mer, and therefore a CT > 1 can be assumed. This seems
reasonable, as this compound is highly activated by the
nitrile moiety. In this set of experiments the styrene-based
transfer agent BAS (mixture g) shows equal conversion as
the monomer BMA, similar to ASEE, while reaching a much
lower overall conversion. This indicates that the nitrile moi-
ety of ASN is much more activating than the phenyl moiety
of BAS and supports the hypothesis that propagation is
slowed down by resonance stabilization of the aromatic ring.
DH
DBCDSC5
(1)
mBMA
mtot
DH0
Polymerization of BMA without CTA provided a DBC of ꢁ
45%. This rather low conversion is expected as BMA is not
among the most reactive methacrylate monomers. As mix-
tures b and c show, ester ASEE has only marginal effects on
the DBC of the monomer. On the other hand, mixture e con-
taining 20 db% of amide ASA reaches a much higher DBC.
Assuming that the transfer reaction is energy neutral, the
less reactive and liquid amide could cause a thinning effect.
Under these conditions, propagating radicals are less likely
to recombine at low conversion and chains continue to grow
for a longer time (t95% is also much higher). This theory is
supported by the fact that addition of 5 db% amide (ASA)
shows no significant effect on the DBC in mixture d and has
1422
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1417–1427

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
When comparing the DBC results for BMA determined from
DSC and from NMR, differences are in some cases observed.
Although it is known that the transfer reaction can be
assumed energy neutral for ester ASEE, the energy balance
of the other CTAs may vary. In the case of mixture g with
BAS, DBC of BMA as measured by DSC is significantly lower
than the NMR value. This indicates some endothermic reac-
tions. In no case, however, is DBC from photo-DSC greater
and thus exothermic side reactions including propagation of
CTA seem unlikely.
The Effects of CTAs on Molecular Weight, Established by
GPC
Following NMR, GPC was performed on the BMA mixtures to
determine the effect of the different CTAs on molecular
weight. All of the CTAs lower the molecular weight and poly-
dispersity of BMA (last two columns of Table 1). The effect
is most pronounced in mixtures with ester ASEE, which was
not surprising based on results from DSC and NMR. The
number-average molecular weight Mn is drastically reduced
from 6580 to 890 Da by addition of 20 db% of ASEE (mix-
ture c). Moreover, the PDI decreases from 2.9 to 1.5, which
indicates very good regulation and a homogenous incorpora-
tion of CTA. While the effect is more dramatic with 20 db%
CTA, addition of 5 db% of ASEE (mixture b) also shows a
strong regulating effect. By comparison, amide ASA (mix-
tures d and e) has a similar effect on PDI at both concentra-
tions but molecular weight stays high. Considering the very
low DBC of the CTA as measured by NMR, it looks like ASA
does not transfer many chains during polymerization of BMA
but rather reduces early stage chain termination by recombi-
nation. This still leads to a sharper molecular weight distri-
bution, but is of little interest considering the low reactivity
of the CTA. Nitrile ASN is very reactive and therefore, even
in solutions with only 5 db% (mixture f), an effect is clearly
visible. The molecular weight of BMA is strongly reduced to
1900 Da, while the PDI is lowered significantly as well. Also
the reactive phenyl compound BAS shows a strong regulat-
ing effect. Only 5 db% reduces the molecular weight to 1880
Da while lowering the PDI to 1.6. While these results are
indeed promising, the conversion of monomer in these mix-
tures only reached about 30%.
FIGURE 2 Observed addition rates versus concentration of
monomers and CTAs. [Color figure can be viewed in the online
issue, which is available at wileyonlinelibrary.com.]
determined kadd (pseudo first order) are collected in Table
2. It can be seen that for the better performing CTAs, the
addition rate constant is comparable to that of methacry-
lates (ꢁ4 3 107 L mol21 s21). Generally, all CTAs, except
amide ASA, exhibit addition rate constants that are in a
comparable range. Amide ASA shows a much lower kadd
rate constant, which correlates well with the low CTA con-
version values in photopolymerizations, determined by NMR
experiments. The low addition rate is most likely caused by
both sterical hindrance from the disubstituted amide group
and subsequently weak radical stabilization because of the
distorted geometry.32 Styrene-based BAS shows the highest
k
add, which is due mainly to the good resonance stabiliza-
tion of the formed intermediate radical by the phenyl moi-
ety. The high addition rate compared to methacrylates will
lead to preferential addition to the CTA over the monomer.
While addition is fast, the fragmentation reaction is likely to
be the slowest among the tested CTAs, which explains the low
conversion measured by NMR. It should be noted, that in the
LFP experiment the attacking radical is a phosphinoyl moiety,
which has a slightly different reactivity compared to the Ger-
manium initiator radicals used in the other experiments, as
well as polymer chain radicals. Nevertheless, the trends in
reactivity determined from LFP likely hold true for the other
experiments as well and are in general reflected in the conver-
sion data.
Determination of Addition Rates by LFP
To better understand the effect of different vinyl activating
groups on the reactivity of b-allyl sulfones, laser flash photol-
ysis (LFP) experiments were performed. As a model reaction,
the addition rate constant (kadd) of an initiator radical to a
CTA (or monomer) was determined. It should be noted that
a different initiator (CGI403, bisacylphosphine oxide initia-
tor) was used for LFP experiments. The reasons for this
choice have been stated previously.21 The reaction mixture
was excited at 355 nm and the decay of the phosphinoyl rad-
ical absorption was followed at 450 nm.
It was not possible to determine significant differences for
b-scission of the intermediate radicals as described previ-
ously.21 Although the rate of b-scission is expected to be
lowered when the intermediate CTA radical is highly reso-
nance stabilized as in the case of nitrile ASN and phenyl
modified BAS. Still, the b-scission reaction is preferred and
Figure 2 shows the observed radical addition rates at vary-
ing concentration for the different CTAs and the two refer-
ence monomers benzyl and methyl methacrylate (MMA). The
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1417–1427
1423

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 2 Addition Rate Constants for Monomers and CTAs
Compound
kadd (107 L mol21 s21
)
BMA
MMA
ASEE
ASA
ASN
BAS
4.2 6 0.1
4.0 6 0.1
3.2 6 0.1
0.77 6 0.05
2.9 6 0.1
4.9 6 0.1
addition of the intermediate radical to monomer was never
observed.
Testing of Difunctional Monomers
A lot of industrial photocurable resins are based on dimetha-
crylate monomers. Cure kinetics and mechanical properties
are important for virtually all applications. Therefore, after
studying the b-allyl sulfones in mixtures with monofunctional
BMA, they were tested with a dimethacrylate resin as well. As
the resulting networks are insoluble, analysis by NMR and
GPC is impractical. Instead, further information about the
behavior of CTAs in difunctional monomer mixtures could be
gathered by RT-NIR photorheology and DMTA. For photo-
rheology measurements, the formulations consist of an equi-
molar mixture of commercially used dimethacrylates UDMA
and D3MA (mixture A), 0.3 wt % Ivocerin as photoinitiator
and 5–20 db% of the b-allyl sulfone.
FIGURE 3 Higher conversion at the gel point results in reduced
shrinkage stress (liquid 5 dotted, solid 5 dashed).
reach a higher DBCg than the reference. Impressively, the
DBCg is increased to nearly 30% with the addition of 5 db%
of ester ASEE (mixture B). Adding 20 db% of ASEE increases
the DBCg up to 41%. Mixtures D and E containing amide
ASA show a slightly improved DBCg, but nearly no delay of
tg is witnessed. These results are in good agreement with
the monofunctional methacrylate experiments and once
again attest to the low reactivity of amide ASA. By compari-
son, adding only a small amount of nitrile ASN (mixture F)
causes an enormous delay of tg while showing a lower DBCg
than the reference. Both metrics are indicative of poor poly-
merization performance. As initiator concentration and light
intensity are low in this experiment, the retardation caused
by the strongly resonance stabilized CTA is more pro-
nounced. Styrene-based BAS shows better performance, as
the tg of mixture G is strongly delayed while the DBCg is
raised slightly.
In Situ Investigation of Chemical and Rheological
Behavior During Photocuring
The coupling of RT-NIR spectroscopy with photorheology is a
very powerful tool for characterizing the curing reaction of a
photopolymerizable formulation. Important metrics gathered
from the technique include the time until gelation (tg), conver-
sion at the gel point (DBCg), t95DBC, final conversion (DBC),
and shrinkage stress (normal force measurements, FN).
In radical polymerization of dimethacrylates, crosslinking
occurs already at fairly low conversion. After reaching the
gel point, the polymer chains lack long-range mobility. As
monomer conversion further increases, the material is
exposed to high shrinkage stress as the gel cannot accommo-
date to the reduction of volume (see Fig. 3). By comparison,
when gelation occurs at higher conversion, the system is
flexible for longer and can rearrange to minimize stress. Fol-
lowing gelation, there will be less remaining monomer to
contribute to the ultimate shrinkage stress. RT-NIR photo-
rheology can be used to follow the process of gelation, by
measuring tg as the time where the storage modulus G0 and
loss modulus G00 plots intersect (G00/G0 5 1). At this time
point, the DBCg can also be determined.
TABLE 3 Results from RT-NIR-Photorheology with Commercial
Monomers UDMA and D3MA in Equimolar Ratio, Showing the
Time to Gelation (tg), the Double Bond Conversion at the Time
of Gelation (DBCg), the Overall Conversion (DBC), and the
Shrinkage Stress as Maximal Normal Force (FN) (statistics c.f.
Supporting Information)
[CTA] tg
DBCg
(%)
t95DBC
DBC
(%)
FN (N)
Network CTA
(db%) (s)
(s)
A
B
C
D
E
F
–
0
5
4.2
9.9
22
29
87
73
74
75
73
74
48
57
229.6
218.6
214.7
214.0
211.2
213.1
217.6
ASEE
108
184
108
119
244
227
ASEE 20
36.5 41
Gel Point
ASA
ASA
ASN
BAS
5
5.2
6.1
27
29
As can be seen in Table 3, gelation of the reference mixture
A occurs after only 4.2 s, at which point a DBCg of 22% is
measured. In all AFCT moderated systems tg generally occurs
later than in the reference. More importantly, most samples
20
5
45.7 17
38.6 30
G
5
1424
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1417–1427

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 4 Results of DMTA Measurements with Commercial
Time to 95% of Total Conversion
Monomers UDMA and D3MA in Equimolar Ratio and Various
CTAs
The time to reach 95% of total conversion (t95DBC) is a good
metric for retardation. It is clearly visible that all transfer
agents cause some retardation. At 0.5 W cm22 light intensity
and with only 0.3 wt % of initiator, the retardation caused
by the highly resonance stabilized nitrile ASN and styrene-
based BAS is significant (factor ꢀ 3).
[CTA]
(db%)
G0208C
FWHM
Tg
G0r
Network
CTA
(MPa)
(8C)
(8C)
(MPa)
A
B
C
D
E
F
-
0
1,060
1,020
1,070
953
144
47
144
121
48
813
280
45
ASEE
ASEE
ASA
ASA
ASN
BAS
5
Overall Double Bond Conversion
20
5
28
RT-NIR allows the tracking of DBC, by monitoring the
decrease of the area of the double bond peak at
ꢁ6160 cm21 during the curing reaction. The reference sam-
ple without CTA reaches a final DBC around 73%. By com-
parison, the mixtures containing ester ASEE and amide ASA
reach this value as well. On the other hand, the resonance
stabilized CTAs based on nitrile (ASN) and styrene (BAS)
cause a significant decrease in final conversion. From the
basic investigations on BMA, this was well expected. In both
cases, the formed intermediate radical is so well stabilized,
that fragmentation happens significantly slower.
126
58
133
66
637
191
470
390
20
5
888
982
93
126
124
G
5
1,090
82
Ester ASEE produced the most homogenous polymer net-
works with an extraordinary low and sharp glass transition.
(Fig. 4) This is due to its radical addition rate being compa-
rable to the methacrylate monomer, and the CTA is therefore
built into the network statistically. While the other CTAs also
lower and sharpen the glass transition, the effect is less pro-
nounced. It is also important to note that the modulus at
room temperature of the ASEE the modified polymers (mix-
tures B and C) is in the same range as the reference. By
comparison, the amide ASA has a much weaker influence on
the glass transition while consequently lowering modulus
(mixtures D and E). As prior tests with ASA all showed poor
reactivity and conversion, the reagent is likely doing little
more than softening the network.
Shrinkage Stress
Monitoring of the force normal to the cure surface (FN) dur-
ing photorheometry gives an indication of shrinkage stress
as it develops during polymerization. The measured FN for
the dimethacrylate reference mixture A was nearly 30 N. FN
for all of the other mixtures is greatly reduced. In prior
investigations,22 a good correlation between the DBCg and
the evolution of FN has been found. Indeed mixtures with
higher DBCg values show lower FN values (c.f. Table 3).
The formulations with 5 db% of ASN (F) and BAS (G) exhibit
similar DMTA curves. The storage modulus and dissipation
factor curves of both mixtures run congruent with each other,
although the polymer formed from mixture F shows a much
lower storage modulus at room temperature (Table 4). Both
CTAs are more effective than ASA at 5 db% in sharpening and
lowering the glass transition. As these reagents were found in
previous measurements to be highly reactive, it is likely that
most of the CTAs reacted at the beginning of the polymerization.
Thus, very short chains were produced in early stages while the
majority of monomer propagation was left unregulated. This is
supported by the fact that there are at least two glass transitions
overlaying each other in the tan d diagram (c.f. Supporting Infor-
mation) Furthermore, one has to keep in mind that DBC is sig-
nificantly lower than in previously mentioned samples.
Interestingly, the amount of CTA needed to reduce the
shrinkage stress seems to be rather low, as all mixtures with
5 db% CTA show good FN values (B, D, F, and G). The mix-
ture with 20 db% of unreactive amide ASA (mixture E) has
the highest reduction in shrinkage stress. This result can be
caused by a combination of chain transfer reaction and thin-
ning of the monomer formulation causing a lower network
density at the gel point. Formulations with 5 wt % of ASN
and BAS have greatly reduced shrinkage stress, which is to
be expected with these slower polymerizations reaching
lower final monomer conversion. ASEE does a very good job
of lowering shrinkage stress, and in mixture C with 20 db%
the reduction is 50%.
DMTA
To investigate the glass transition of the dimethacrylate net-
works, DMTA measurements were performed. In these
experiments, the storage modulus (G0) and loss factor (tan d)
plots of the b-allyl sulfone modified dimethacrylate networks
are tracked relative to the unmodified polymer. Regulation of
polymerization with CTAs affects the glass transition temper-
ature of the resultant network. In all cases, the glass transi-
tion was both lowered and sharpened, which is indicative of
a more regular and homogenous network. The sharpness of
the transition is quantified as the full width at half maximum
(FWHM) of the tan d peak (see Table 4).
Unfortunately, it was not possible to measure samples with
concentrations of ASN or BAS higher than 5 db% as they
tended to crystallize during preparation.
Dynstat Impact Resistance
To determine the effect of chain transfer on impact resist-
ance, Dynstat tests were performed with all of the cured
polymer mixtures. The Dynstat tests clearly show that the
impact resistance is improved in all cases with addition of
CTA (Table 5). Ester ASEE causes the largest effect, and in
polymer network C with 20 db% of this CTA impact
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1417–1427
1425

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 4 Storage modulus (G0) and loss factor (tan d) plots of reference network A, and ASEE containing formulations B and C.
[Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
resistance is increased by a factor of 2.5. Amide ASA is also
capable of improving the toughness but the effect is smaller.
Phenyl substituted BAS allows a rather significant improve-
ment in impact resistance already at 5 db%. Nitrile ASN, by
comparison, provides a polymer with impact resistance little
different from the control.
polymerization behavior and final material properties of
mono- and difunctional methacrylates.
The behavior of b-allyl sulfones is dramatically changed by
variation of the vinyl activating moiety. Activating groups
such as phenyl and nitrile, which offer good resonance stabi-
lization for the intermediate CTA radical, significantly
increase the rate of radical attack on the CTA. However, good
resonance stabilization also slows down the b-scission frag-
mentation and therefore the polymerization process. All
tested CTAs were able to reduce shrinkage stress and led to
materials with lowered glass transitions and improved impact
strength. The ester ASEE generally gave the best results,
explained principally by its structural similarity to the meth-
acrylate monomers. As a result, the CTA was incorporated
rather statistically leading to well-regulated networks. The
addition of ASEE to the dimethacrylate resin provided poly-
mers with lower and sharper glass transition temperatures
with impact strength 2.5 times that of the control.
From DMTA it was determined that the polymer containing
5 db% ASN had a slightly higher modulus than the refer-
ence, which indicates a hard but also brittle material. The
polymer with 5 db% BAS, on the other hand, is less brittle
but also less hard. Where high values of both modulus and
impact strength are desired, the polymers with ASEE (B and
C) are the most satisfactory.
CONCLUSIONS
In this article, we have presented b-allyl sulfones with differ-
ent activating groups and assessed their influence on the
TABLE 5 Results of the Dynstat Impact Resistance Tests with
Commercial Monomers UDMA and D3MA in Equimolar Ratio
and Various CTAs
ACKNOWLEDGMENT
Financial support from Ivoclar Vivadent AG and the Christian
Doppler Research Association is gratefully acknowledged. The
authors thank Georg Gescheidt and TU Graz/NAWI Graz for the
use of the LFP equipment.
Impact Resistance
Network
CTA
[CTA] (db%)
(kJ mm22
)
A
B
C
D
E
F
–
0
4 6 1
10 6 1
11 6 1
6 6 1
7 6 2
5 6 1
9 6 3
ASEE
ASEE
ASA
ASA
ASN
BAS
5
20
5
REFERENCES AND NOTES
ꢀ
1 B. Husar, S. C. Ligon, H. Wutzel, H. Hoffmann, R. Liska, Prog.
20
5
Org. Coat. 2014, 77, 1789–1798.
2 N. Moszner, U. Salz, Prog. Polym. Sci. 2001, 26, 535–576.
3 Fleischer, J. E. Mod. Paint Coat. 2001, 91, 21–22, 25.
G
5
1426
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1417–1427

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
4 S. Baudis, F. Nehl, S. C. Ligon, A. Nigisch, H. Bergmeister, D.
Bernhard, J. Stampfl, R. Liska, Biomed. Mater. (Bristol, U. K.).
2011, 6, 055003 (12pp).
19 W. K. Busfield, C. I. Zayas-Holdsworth, S. H. Thang, J.
Polym. Sci. Part A: Polym. Chem. 2001, 39, 2911–2919.
20 H. Lu, J. A. Carioscia, J. W. Stansbury, C. N. Bowman, Dent.
Mater. 2005, 21, 1129–1136.
5 J. Stampfl, S. Baudis, C. Heller, R. Liska, A. Neumeister, R.
Kling, A. Ostendorf, M. Spitzbart, J Micromech. Microeng.
2008, 18, 125014
21 C. Gorsche, M. Griesser, G. Gescheidt, N. Moszner, R. Liska,
Macromolecules 2014, 47, 7327–7336.
6 H. Lu, J. W. Stansbury, C. N. Bowman, Dent. Mater. 2004, 20,
979–986.
22 C. Gorsche, T. Koch, N. Moszner, R. Liska, Polym. Chem.
2015, 6, 2038–2047.
7 N. Moszner, Macromol. Symp. 2004, 217, 63–75.
23 S. C. Ligon, K. Seidler, C. Gorsche, M. Griesser, N. Moszner,
R. Liska, J. Polym. Sci. Part A: Polym. Chem. 2015, early view,
doi: 10.1002/pola.27788.
8 F. Sanda, T. Endo, J. Polym. Sci. Part A: Polym. Chem. 2000,
39, 265–276.
9 W. D. Emmons, A. W. Gross, (Rohm & Haas Company) Pat-
ent US 5399642, 1993.
24 B. Ganster, U. K. Fischer, N. Moszner, R. Liska, Macromole-
cules 2008, 41, 2394–2400.
10 C. E. Hoyle, C. N. Bowman, Angew. Chem. Int. Ed. 2010, 49,
1540–1573.
25 C. Dworak, S. Kopeinig, H. Hoffmann, R. Liska, J. Polym.
Sci. Part A: Polym. Chem. 2009, 47, 392–403.
11 C. E. Hoyle, T. Y. Lee, T. Roper, J Polym. Sci. Part A: Polym.
Chem. 2004, 42, 5301–5338.
26 I. W. Harvey, E. D. Phillips, G. H. Whitham, Tetrahedron
1997, 53, 6493–6508.
12 A. Mautner, X. Qin, H. Wutzel, S. C. Ligon, B. Kapeller, D.
Moser, G. Russmueller, J. Stampfl, R. Liska, J. Polym. Sci. Part
A: Polym. Chem. 2013, 51, 203–212.
27 D. Colombani, C. Navarro, M. Degueil-Castaing, B. Maillard,
Synth. Commun. 1991, 21, 1481–1487.
28 X. Q. Li, X. S. Xu, C. Zhou, Chem. Commun. 2012, 48,
13 P. Esfandiari, S. C. Ligon, J. J. Lagref, R. Frantz, Z.
Cherkaoui, R. Liska, J. Polym. Sci. Part A: Polym. Chem. 2013,
51, 4261–4266.
12240–12242.
29 A. A. Pudikova, N. P. Gerasimova, Y. A. Moskvichev, E. M.
Alov, A. S. Danilova, O. S. Kozlova, Russ. J. Org. Chem. 2010,
46, 352–354.
14 D. Colombani, P. Chaumont, Prog. Polym. Sci. 1996, 21, 439–503.
15 G. Moad, E. Rizzardo, S. H. Thang, Acc. Chem. Res. 2008,
41, 1133–1142.
30 T. Otsu, M. Inoue, B. Yamada, T. Mori, J. Polym. Sci. Part C:
Polym. Lett. 1975, 13, 505–510.
16 G. Moad, E. Rizzardo, S. H. Thang, Polymer 2008, 49, 1079–1131.
31 J. Brandrup, E. H. Immergut, E. A. Grulke, Polymer Hand-
book, 4th ed.; Wiley: New York, 2004.
17 D. Zhuo, Y. Ruan, X. Zhao, R. Ran, J. Appl. Polym. Sci.
2011, 121, 660–665.
32 X. Xie, T. E. Hogen-Esch, Macromolecules 1996, 29, 1746–
18 S. Yamago, Y. Nakamura, Polymer 2013, 54, 981–994.
1752.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1417–1427
1427