Biomaterials 23 (2002) 18191829

Eect of chemical structure on degree of conversion in light-cured

dimethacrylate-based dental resins
I. Sideridou*, V. Tserki, G. Papanastasiou
Department of Chemistry, Aristotle University of Thessaloniki, GR-54006, Thessaloniki, Hellas, Greece
Received 16 March 2001; accepted 5 September 2001

Abstract
In this work the room-temperature photopolymerization of Bis-GMA, Bis-EMA, urethane dimethacrylate (UDMA) and
triethylene glycol dimethacrylate (TEGDMA) induced by camphoroquinone/N;N-dimethylaminoethyl methacrylate, as photoinitiator system, was followed by FT-IR. The results obtained were then tted by a non-linear least square method to a rational
function, which permitted the accurate calculation of the limiting degree of conversion. The latter was found to increase in the order
Bis-GMAoBis-EMAoUDMAoTEGDMA. This trend is discussed in connection with the chemical structure of dimethacrylates.
The photopolymerization of mixtures of Bis-GMA/TEGDMA, Bis-GMA/UDMA and Bis-GMA/Bis-EMA showed a good linear
relationship of degree of conversion with the mole fraction of Bis-GMA and in the case of the rst pair also with the Tg of the initial
monomer mixture. r 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Dental materials; Composites; Copolymers; Dimethacrylates; Degree of conversion; Glass transition temperature; Bis-GMA; Bis-EMA;
UDMA; TEGDMA

1. Introduction
The polymerization of dimethacrylates produces
densely crosslinked networks, resins, which nd wide
applications in dentistry such as dental composites, pit
and ssure sealants, dentine bonding agents and cements
[13], dental adhesives [4] dentures and elastomeric
impression materials [5].
Dental composites are used for the restoration of
teeth and consist of two principal components, an
organic matrix and inorganic ller. The organic matrix
is formed by free radical polymerization of dimethacrylates, which are non-toxic and capable of rapid
polymerization in the presence of oxygen and water,
because the restorations are polymerized in situ in a
tooth cavity. This matrix when used unlled for the
restoration of teeth shows a poor wear resistance. This
can be improved by the inclusion of particulate llers,
which are harder than the polymeric matrix. An
ambitious goal would be to match the remarkable
properties of dental enamel, which contains more than
95 vol% of hydroxyapatite crystallites tightly packed
*Corresponding author. Tel.: +30-3199-7825; fax: +30-3199-7769.
E-mail address: siderid@chem.auth.gr (I. Sideridou).

into an intricate microstructure. In comparison the

current composite restorative materials have crude
microstructure with no more than 65 vol% of inorganic
ller [6]. A wide range of llers of varying shapes and
sizes, ranging from colloidal dimensions to tens of
microns is being used in varying combinations. Vinyl
silane coupling agents are also used to promote matrixller adhesion. The current composite materials have
good color and translucency, but much lower wear
resistance than the silver amalgams, which they are
designed to replace. The lifetime for anterior polymeric
restorative materials is about 8 years, but for posterior
materials is often not longer than 24 years. In
comparison with traditional dental amalgams, which
have the time of use of about 1020 years, this seems to
be a very short period [7]. Despite this deciency there
are diverse cogent reasons, such as esthetics and
avoidance of mercury pollution of the environment,
which spur on their further development.
The most common dimethacrylate monomer in
current commercial dental composites is the so-called
Bis-GMA (Scheme 1), which is the reaction product of
bisphenol A and glycidyl ester methacrylate (GMA).
Advantages of using Bis-GMA over the rst used smallsized dental monomers, such as methyl methacrylate,

0142-9612/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 1 4 2 - 9 6 1 2 ( 0 1 ) 0 0 3 0 8 - 8

1820

I. Sideridou et al. / Biomaterials 23 (2002) 18191829

CH3

CH3

CH2 = C

C=CH2

C=O

CH3

OCH2 CHCH 2 O

OH

C=O
OCH2 CHCH2 O
OH

CH3

Bis-GMA
2,2-bis-[4-(2-hydroxy-3-methacryloyloxyprop-1-oxy)phenyl]propane]
or
Bisphenol A glycol dimethacrylate

CH3

CH3
C=CH2

CH2=C
C=O

C=O

OCH2CH2OCH2CH2OCH2CH2O

TEGDMA
Triethyleneglycol dimethacrylate
CH3

CH3

CH2=C
C=O

C=CH2
C=O

CH3 CH3
OCH2CH2OCNHCH2CHCH2CCH2CH2NHCOCH2CH2O
CH3
O
O
UDMA

1,6-bis-(methacryloyloxy-2-ethoxycarbonylamino)-2,4,4-trimethylhexane
or
Urethane dimethacrylate

CH3

CH3
C=CH2
C=O

CH2=C
C=O

CH3

O(CH2CH2O)n

(OCH2CH2) nO

CH3

Bis-EMA
Bisphenol A ethoxylated dimethacrylate
(4 ethylene oxide groups/bisphenol A group)

Scheme 1. Chemical structures of the dimethacrylate monomers used.

include less shrinkage, higher modulus and reduced

toxicity due to its lower volatility and diusivity into
tissues. These desirable properties of Bis-GMA are
partially negated by a relatively high viscosity, which
does not permit the use of high amount of ller. The
increased ller content tends to improve mechanical
properties and to reduce curing shrinkage and the

thermal expansion coecient. Since the viscosity of the

resultant past limits the amount of ller, which can be
incorporated, it is common practice to also use a less
viscous monomer as diluent comonomer, normally
triethylene glycol dimethacrylate (TEGDMA). However, TEGDMA has been shown to adversely aect the
properties of the matrix resin by increasing the water
sorption and curing shrinkage. Investigations are being
carried out in identifying new dimethacrylates, which
will have moderately low viscosities to eliminate or
minimize the use of the diluent monomer [8,9]. But the
only signicant changes, which have found their way
into some commercial composites, were the alternative
employment of low viscosity structural analogous of
Bis-GMA [10] and the introduction of urethane
dimethacrylates [11].
The polymerization of dimethacrylates is chemical or
visible light-initiated with the latter being more preferable because of allowing a ner control of the entire
polymerization process. Thus initiation can be started
and stopped almost at will. The room-temperature
polymerization of dimethacrylates usually leads to
glassy resins in which only a part of the available
double bonds are reacted. Before the completion of
conversion the vitrication process decelerates the
reaction to a hardly perceptible rate. Only very exible
monomers in which the reactive methacrylate groups are
relatively far apart can be completely reacted at ambient
temperature. The degree of conversion of resins is a
major factor inuencing their bulk physical properties.
In general, the higher the conversion of double bonds,
the greater the mechanical strength. The unreacted
double bonds may either be present in free monomer or
as pendant groups on the network. The unreacted
monomer may leach from the polymerized material and
irritate the soft tissue. For example TEGDMA is
suspected to be propitious to bacterial growth around
the restoration [12]. Furthermore, monomer trapped in
the restoration may reduce the clinical serviceability of
composite through oxidation and hydrolytic degradation, which may be manifested in forms such as
discoloration of the llings and accelerated wear [13].
The nal degree of conversion of a resin depends on
the chemical structure of the dimethacrylate monomer
and the polymerization conditions i.e., atmosphere,
temperature, light intensity and photoinitiator concentration [14].
The aim of this work was to investigate the inuence
of the chemical structure of dimethacrylates most
commonly used in the preparation of dental composites
on the degree of conversion. These are Bis-GMA,
TEGDMA, bisphenol A ethoxylated dimethacrylate
(Bis-EMA) and urethane dimethacrylate (UDMA)
(Scheme 1). The light-induced homopolymerization of
these monomers and copolymerization of two or more
monomers of varying compositions were carried out

1821

I. Sideridou et al. / Biomaterials 23 (2002) 18191829

Table 1
Abbreviations and compositions of samples used in this study
Abbreviation

Composition (mol%) of sample

Bis-GMA

G
T
U
E
GT (1)
GT (2)
GU (1)
GU (2)
GE (1)
GE (2)
GTUE (1)
GTUE (2)
GUE
a
b

TEGDMA

UDMA

Bis-EMA

100
100
100
100
35.8a
56.6b
35.8
56.6
35.8
56.6
56.6
56.6
56.6

64.2a
43.4b
64.2
43.4

23.4
13.4

10
15
21.7

64.2
43.4
10
15
21.7

This molar composition corresponds to 50/50 (wt%).

Corresponds to 70/30 (wt%).

under exactly the same conditions in order to obtain

comparable results. A comparison of these results will
provide valuable information on the relationship between the chemical structure of dimethacrylates and
degree of conversion, which can help in the better
understanding of the behavior of composites contained
in these dimethacrylates and also in the development of
improved or new dental resins in a future work.

2. Experimental procedures
2.1. Materials
The dimethacrylates used were Bis-GMA (Polysciences Europe GmbH), Bis-EMA (Aldrich Chem.
Co.), UDMA (Ivoclar AG) and TEGDMA (Aldrich
Chem. Co.). They were used as received without further
purication. Nine mixtures of these monomers were
prepared, the composition of which is shown in Table 1.
In the commercial dental composites based on mixtures
of Bis-GMA and TEGDMA, they are used in a ratio
varied between 50 : 50 and 70 : 30 by weight, in order to
obtain viscosities of 12 Pa, suitable for the incorporation of the appropriate amounts of inorganic llers [15].
These weight ratios correspond to molar ratios
35.8 : 64.2 and 56.6 : 43.4 and these molar ratios were
used in the preparation of all mixtures studied, in order
to keep the concentration of double bonds constant. To
make the samples light cured, 2 mol% of camphoroquinone (CQ) (Polysciences) used as photosensitizer, and
2 mol% of N,N-dimethylaminoethyl methacrylate
(DMAEMA) (Riedel-de Haen) used as reducing agent,
was added to each sample. This photoinitiator system is
the most common one used in the current photoactivated polymer-based dental materials. All samples

except TEGDMA were viscous liquid, so the CQ and

DMAEMA were rst dissolved in dichloromethane,
then a certain amount of this solution was added to the
sample and the solvent was subsequently evaporated
under vacuum.
2.2. Degree of conversion
The degree of photopolymerization of a very thin lm
formed from dimethacrylate monomer or a mixture of
monomers has been determined by using an FT-IR
spectrophotometer (PerkinElmer 1600). A small drop
of each sample contained in the photoinitiator system
was placed between two translucent polyethylene strips,
which were pressed between two NaCl crystals to
produce a very thin layer. Polyethylene lm was used
mainly to avoid adhesion of the formed resin on NaCl
crystals and to also to prevent oxygen inhibition of
polymerization. The sample was irradiated for successive short periods of time, which gave cumulative
exposure times of 10, 20, 40, 60, 80, 120, 180 and
240 s. The FT-IR spectrum was recorded at zero time t0
and after each period of exposure to visible-light, over a
certain range of frequency for each monomer. This was
16601550 cm
1 for Bis-GMA and Bis-EMA, 1860
1600 cm
1 for UDMA and 18001570 cm
1 for TEGDMA. The spectra were recorded immediately after the
end of photoirradiation, in transmission with 16 scans at
a resolution of 1 cm
1.
Photopolymerization was initiated with a XL 3000
dental photocuring source (3M, USA). This source
consisted of a 75 W tungsten halogen lamp, a series of
optical lters and lenses and a fused ber optic light
guide with a 7 mm exit window. This unit emitted
radiation predominantly in the 420500 nm range,
where also CQ absorbed (lmax 470 nm, e 3:8

1822

I. Sideridou et al. / Biomaterials 23 (2002) 18191829

Table 2
Coecients of the rational functions (Equation 1) used for the determination of the limiting degree of conversion of the neat dimethacrylates and
their mixtures
Sample

a1  10
14

a2

a3

b1

b2

b3  102

Sy;x a

R2

Bis-GMA
TEGDMA
UDMA
Bis-EMA
GT (1)
GT (2)
GU (1)
GU (2)
GE (1)
GE (2)
GTUE (1)
GTUE (2)
GUE

9.51

5.19
1.29

51.1

16.9

12.3

37.5

36.4

10.3

2.74

12.6
2.85

14.6

22.67
0.4038
1.311
13.08
44.26
43.42
75.78
70.23
2.349
1.822
46.13
8.135
42.98

0.96027
0.77943
2.4564
2.5250
1.6691
1.5398
0.45165
0.93958
3.6127
1.6321
2.6871
4.4730
3.4067

2.10
3.48

0.259
19.0
3.60
2.47
8.28
8.01
2.65
0.671
2.58

0.592
3.17

0.9507
0.1458
0.1954
0.4636
0.9399
0.9852
1.566
1.613
0.7571
0.3892
1.421
0.8062
1.740

2.4633
1.0299
3.5318
4.7465
2.7397
2.8068
7.9046
1.8579
7.9542
3.8539
5.6415
9.9568
7.6584

0.2019
4.0581
0.1259
0.6985
0.3654
0.2611
0.2575
0.9306
0.2200
0.2322
0.1038
0.2894
0.1093

0.9999
0.9907
1.0000
0.9995
0.9999
0.9999
0.9999
0.9986
0.9999
0.9999
1.0000
0.9998
1.0000

a
b

Standard error of estimate.

Coecient of determination.

104 cm2 mol
1) [16]. The unit was used without the light
guide, at a distance of 6 mm from the sample. The curing
light and heat (infrared) intensity at the position of the
sample measured by a Hilux curing light meter, was 200
and 0 mW cm
2 correspondingly.
The amount of double vinyl bonds remaining in the
sample exposed to irradiation is shown by the intensity
of the peak at 1637 cm
1 referring to the C C
stretching of the vinyl group and 816 cm
1 which refers
to the C C twisting. Both have been used in the study
of polymerization of acrylates and methacrylates [17].
We have chosen the 1637 cm
1 absorption, because it is
stronger than the absorption at 816 cm
1 and then will
provide less experimental deviation. The degree of
conversion was directly related to the decrease of
1637 cm
1 absorption on the FT-IR spectra as follows:
Degree of conversion

A0
At
100;
A0

where A0 is the absorption of the peak at 1637 cm
1

when time is equal to zero and At is the absorption at
time t:
In the case of resins of Bis-GMA and Bis-EMA the
peak at 1608 cm
1 assigned to aromatic C C bond was
used as an internal standard [18]. In the case of UDMA
and TEGDMA the sharp well-dened peak at
1720 cm
1 assigned to the carbonyl stretching vibration
could not be used as an internal standard, because the
position and the intensity of this peak change during
polymerization [19]. In the uncured state the carbonyl
group is conjugated with the C C bond and on curing
this conjugation is lost. This results in a shift of the
carbonyl peak to a higher frequency in the cured state,
as the bond becomes stronger, owing to the fact that the
electrons are no longer delocalized. A signicant loss of
intensity of this peak was also observed.
When mixtures of monomers were used, one of the
components was always the Bis-GMA, so the peak at

1608 cm
1 of the aromatic C C bond was used as an

internal standard.
All experiments were carried out in triplicate and the
results were averaged. The mean values were then tted
by a non-linear least square method to the following
rational function (i.e. the ratio of two polynomials):
Degree of conversion

a1 a2 x a3 x2
:
b1 b2 x b3 x2

The values of aI and bI ; are listed in Table 2 along with

the corresponding values of the standard error of
estimate (Sy;x ) and the coecient of determination
(R2 ). The reported values of Sy;x and R2 show that
Eq. (1) closely represents the experimental data.
Eq. (1) then permitted the calculation with accuracy
of the limiting degree of conversion, from the ratio a3 =b3
for t-N:
2.3. Glass transition temperature (Tg )
The Tg values of monomers were determined by using
a dierential scanning calorimeter (DSC-Pyris 1, PerkinElmer) at a scanning rate of 51C min
1. The Tg
values were determined from the mid-point in the
thermogram, as measured from the extensions of the
pre- and post-transition baselines.
The Tg could not be readily discerned by DSC in the
case of cured dimethacrylates, due to the breadth of the
transition region [20]. So the Tg of polymer networks
was determined using a thermal mechanical analyzer
(TMS-2, PerkinElmer) at a heating rate of 51C min
1
and a penetration probe loading with 150 g. Specimen
discs 5 mm in diameter and 1 mm in height were
fabricated in aluminum mold between two glass slides
covered with polyethylene lm. They were irradiated for
200 s on each side with the XL 3000 dental photocuring
unit, without the light guide, at a distance B1.5 cm from
the sample, where the light intensity was 60 mW cm
2.

1823

I. Sideridou et al. / Biomaterials 23 (2002) 18191829

Table 3
Basic physical properties of dimethacrylate monomers studied
Monomer

MW

Conc. of double bonds (mol/kg)

Viscosity (Pa)

Refractive index (n20

D)

Bis-GMA
TEGDMA
UDMA
Bis-EMA

510.6
286.3
470.0
540

3.90
6.99
4.25
3.70

1200a,b
0.011b
23.1a/7.054c
0.9

1.5497
1.460
1.485d
1.5320

As cited in Ref. [21].

Cited in Ref. [15].
c
Cited in Ref. [22].
d
Cited in Ref. [9].
b

All specimens showed a very broad transition region

well illustrating the problem involving assignment Tg in
highly crosslinked polymers. The Tg values were
obtained by the use of the derivative curve. It is
noteworthy that generally the TMS Tg values agree
with those determined by DSC with the same heating
rate, because both are static methods (no movement of
the sample). This fact was also veried in this work with
the use of a poly(ethylene terephthalate) (PET) sample,
which showed by DSC a Tg 771C (H.R. 51C min
1)
and by TMS with 150 g load penetration (H.R.
51C min
1) a Tg 761C

3. Results and discussion

Table 3 shows some basic physical properties of the
four dimethacrylates used in this study. Bis-GMA and
Bis-EMA have about the same size in contrast to
UDMA and mainly of TEGDMA, which have smaller
size and therefore higher concentration of double bonds;
so the latter at equal degrees of conversion will exhibit
higher density of crosslinking and will form tighter
networks.
Bis-EMA is structurally analogous to Bis-GMA with
a sti central phenyl ring core, without, however, the
two pendant hydroxyl groups, which are responsible for
the high water sorption [23] and mainly for the
extremely high viscosity of Bis-GMA due to the strong
hydrogen bonding [20,24,25]. UDMA shows a higher
viscosity than TEGDMA and Bis-EMA, due to the
hydrogen bonding between the aNHa and >C O
groups, which, however, is much lower than the
viscosity of Bis-GMA, because imino groups form
weaker hydrogen bonds compared to hydroxyl groups
[26]. The viscosity is a measure of the resistance of
molecules to ow and a high viscosity value is indicative
of the presence of intermolecular interactions. These
interactions can cause a decreased mobility of monomer
molecules during polymerization and also a decreased
exibility of the corresponding polymeric network.
Table 3 also shows the refractive indices of dimethacrylates, because an important consideration in the

formulation of esthetic dental composites with high

conversions and depths of cure is how well the refractive
indices of polymer matrices match those of reinforcing
llers. Typical radiopaque llers, such as those containing Ba, Sr and Zr have refractive indices of about 1.55.
It is noteworthy that the polymers often have a dierent
refractive index, slightly higher than their monomer
precursors [26,27]. From this point of view Bis-GMA
and Bis-EMA seem to be more suitable for use in dental
composites.
The monomers and the mixtures of monomers
(Table 1) were activated for visible light photopolymerization by the addition of CQ 2 mol% (varies between
0.621.16 wt% depending on sample) and DMAEMA
also 2 mol% (varies between 0.581.10 wt%). In several
commercially available visible-light-cured dental
composites, Taira et al detected CQ and DMAEMA
over a large concentration range of CQ (0.171.03 wt%)
and DMAEMA (0.091.39 wt%) [28]. At CQ concentrations of above 2 mol%, the initiator eciency was
found to be independent of the concentration of
DMAEMA [29].
Photopolymerization of samples was followed by FTIR spectroscopy. The degree of conversion observed at a
certain polymerization time for the neat monomers is
shown in Fig. 1. These data are the mean values of three
experiments for each polymerization time. The solid
lines resulted from the tting of the experimental data
by a non-linear least square method to the rational
Function (1) reported in the experimental part. The
advantage resulting from the use of rational functions
rather than linear polynomials for tting experimental
data was discussed in the literature [30,31]. In accordance with this discussion, it was found in this
investigation that Eq. (1) provided a better t to the
experimental data than the conventional polynomials
with the same number of adjustable coecients. Eq. (1)
permitted the accurate calculation of the limiting degree
of conversion of dimethacrylate monomers (Table 4).
This equation could also be used for the calculation of
degree of conversion at any time of photopolymerization and vice-versa, without the performance of polymerization.

1824

I. Sideridou et al. / Biomaterials 23 (2002) 18191829

Table 4
Dependence of limiting degree of conversion on Tg for neat dimethacrylate monomers
Monomer

Limiting degree of conversion

Tg (1C) of monomer

Tg (1C) of polymer

Bis-GMA
TEGDMA
UDMA
Bis-EMA

39.0
75.7
69.6
52.2

7.7

83.4

35.3

46.1

67
65
68
68

80

70

Degree of conversion (%)

60

50

40

30
Bis.GMA
20

TEGDMA
UDMA
Bis.EMA

10

0
0

50

100

150

200

250

Polymerization time (sec)

Fig. 1. Degree of conversion of C C double bonds (%) as a function

of irradiation time, for the neat dimethacrylates. The solid lines are
calculated ts of the experimental data to the Eq. (1).

The curves in Fig. 1 show considerable dierences,

despite the similar initiation rates of polymerization due
to the constant initiator concentration and light
intensity used in all polymerizations.
UDMA and Bis-GMA showed much higher initial
polymerization reactivity than TEGDMA and BisEMA. After 10 s of polymerization 49.5% and 22.9%
of the double bonds were reacted in the case of UDMA
and Bis-GMA and only 13.7% and 13.5% in TEGDMA
and Bis-EMA correspondingly. However later TEGDMA and Bis-EMA continued to polymerize with much
higher rate than UDMA and Bis-GMA and nally the
TEGDMA showed the highest limiting degree of
conversion (Table 4). This quite dierent behavior of
monomers in reactivity during the polymerization
process is more clearly seen in Fig. 2, where the rates
of polymerization as a function of time are presented.

The photopolymerization of UDMA and Bis-GMA

under the experimental conditions used starts with a
maximum rate which is rapidly decreased, while the
photopolymerization of TEGDMA and Bis-EMA,
shows a maximum rate 1.9 and 1.6 s
1 after 9.6 and
8.4 s when 13.1% and 11% of the double bonds were
reacted correspondingly. The dierences in the polymerization behavior of the four dimethacrylates studied
must be due to the dierent chemical structure of the
spacer group connecting the methacrylate groups
(Scheme 1). For a better understanding of this eect, it
is worth mentioning the main characteristics of kinetics
of the free-radical bulk polymerization of dimethacrylates.
This polymerization generally exhibits certain complex features such as autoacceleration, autodeceleration
and a maximum limiting conversion, which is signicantly less than unity, in spite of the presence of
unreacted double bonds and trapped radicals. This
complex behavior is due to the fact that the mobility of
the reacting medium decreases as the polymerization
proceeds. At very low conversions, where the reacting
medium is in liquid state, propagation and termination
step of polymerization are chemically controlled and the
polymerization proceeds with a constant rate. However,
as an insoluble innitely large network forms (i.e.,
gelation) the movements of the macroradicals are
restricted and termination step, which involves the
reaction of two macroradicals, becomes diusionlimited. A decrease in the termination rate leads to a
corresponding increase in the polymerization rate which
is known as autoacceleration or gel eect. This eect is
more pronounced in the case of viscous monomers.
During the phase of gelation the reaction system
transforms from a liquid to a rubber and consists of
two species: the sol component consisting primarily of
residual monomer and the gel (insoluble but swellable in
good solvents fraction) consisting of branched and
mainly of crosslinked polymer chains. As the polymerization progresses, the system becomes even more
crosslinked and restricted, so the propagation step,
which involves the reaction of the smaller monomer
molecules with macroradicals, also becomes diusionlimited. A balance between diusion-controlled propagation (which decreases the rate of polymerization) and
diusion-controlled termination (which increases the

I. Sideridou et al. / Biomaterials 23 (2002) 18191829

1825

system changes as a function of conversion

[32,33,37,38].
The Tg of the polymerizing system can be estimated
from the Tg s of its separate components using the
known Fox equation applied to plasticizers [20]:

3.0
Bis-GMA

2.5

1=Tgnet wsol =Tgmonomer wgel =Tggel ;

Polymerization rate R p (s-1 )

UDMA

2.0
vitrification

1.5

TEGDMA

1.0
Bis-EMA

gelation

0.5

0.0
0

10

20

30

40

50

60

Polymerization time (sec)

Fig. 2. Rate of polymerization Rp (expressed in s
1) as a function of

time for the neat monomers.

rate of polymerization) results to a maximum in the rate

max
of polymerization (Rmax
is observed when the
p ). The Rp
glass transition temperature (Tg ) of the reacting system
becomes coincident with the polymerization temperature. Shortly after the polymerization reaches the Rmax
p ;
solidication (i.e. vitrication) starts with transformation of the network from the rubbery to the glassy state
[5,20,3236].
During isothermal vitrication the polymerization
rate is controlled by diusion, i.e. by free volume
activation and the overall rate constant can be described
by a WLF-type equation of the form [20,33,34]:
lnk c1 T
Tg =T
Tg c2 ;

where c1 and c2 are constants, T the curing temperature

and Tg the glass transition temperature of the reacting
system. When Tg approaches the value of T c2 the rate
goes to zero, leaving residual unreacted monomer,
pendant double bonds and trapped radicals frozen in
the glass. So, the nal degree of conversion of the double
bonds in an isothermal polymerization depends exclusively on the polymerization temperature and the
relationship of degree of conversion with the Tg of the
reacting system, which expresses its mobility. Studies of
the bulk polymerization of various dimethacrylates
showed that and the shape of the rate curve also
depends on how rapidly the mobility the of reacting

where wsol and wgel are the weight fractions of sol

(monomer, acting as plasticizer) and gel; Tggel is the Tg
of the gel fraction, which depends on the exibility of
the monomer units consisted and the crosslinking
density. It is obvious that the lower the Tg of a
dimethacrylate monomer and the lower the concentration of the vinyl bonds per unit mass, the lower the Tg of
the monomer units of the corresponding formed
polymer.
In Table 4 the Tg of dimethacrylate monomers are
shown. This is a measure of chain exibility of
monomer, which depends on the nature and the size of
the groups of the chain. Large and polar groups, which
are responsible for intra- and inter-molecular interactions, decrease the exibility of the chain and increase
the Tg value.
Bis-GMA exhibits the highest Tg (
7.71C) due to the
presence of the rigid aromatic nuclei and mainly due to
the strong hydrogen bonding, if we take into account the
much lower Tg value (
46.11C) of its structural analog
Bis-EMA. The hydroxyl groups, positioned diametrically across the rigid bisphenol core structure of the
molecule, cannot form intramolecular hydrogen bonds
but only intermolecular, resulting in a quasi-network
hydrogen bonded structure [21]. This structure results in
an extremely viscous and hindered reaction environment, which must be responsible for the immediate
autoacceleration region observed in the polymerization
of Bis-GMA (Fig. 2). The newly formed radicals attach
to a rigid network; it is dicult to terminate them by
combination or disproportionation, leading to a pronounced maximum in the polymerization rate. Nevertheless the polymerization was carried out at room
temperature (about 231C) and the polymerization
window up to Rmax
was very low, from Tgmonomer
p

7:71C up to Tgnet Tcure 231C: The restricted
mobility of the network then also rapidly suppressed
the propagation reaction, by hindering the diusion of
the free monomer and mainly of the pendant bonds. The
polymerization stopped when the formed polymer
network showed a Tg 671C (Table 4) obtained with a
very low degree of conversion (39%). At this stage
unreacted pendant double bonds, macroradicals and
free monomer became trapped among network units
and completely immobilized.
UDMA contains an aliphatic spacer group between
the methacrylates, but exhibits a relatively high Tg ; due
to the presence of the urethane groups aNHCOOa
which contain rigid quasi-conjugated double bonds [39]

1826

I. Sideridou et al. / Biomaterials 23 (2002) 18191829

and also form hydrogen bonds. Although the viscosity

and the Tg of UDMA is much lower than that of BisGMA, the shape of the rate versus time curve is
analogous to that of Bis-GMA (Fig. 2). The immediate
auto-acceleration region occurred in the polymerization
of UDMA most probably due to the labile hydrogen
atoms of aNHa groups, which greatly favor chaintransfer reactions [40]:
*
BMd aNHa-BMH aNa;
where BMd is a macroradical. The aNHa group may
be part of a monomer or a polymer molecule, so the
*
newly formed aNa
radical can correspondingly cause
the initiation of polymerization or crosslinking. Thus in
the polymerization of UDMA the initiation rate is
higher than that for all the other monomers and the
formed network is more dense and rigid than is
predicted by its structure, considering as crosslinking
sites only those of the two vinyl groups.
Bis-EMA and TEGDMA are non-viscous liquids with
a very low Tg which must be responsible for a
suppressed diusion eect on the termination reaction
during gelation and on propagation during vitrication,
resulting in the broadness of the shape of the rate versus
time curves of these monomers (Fig. 2). TEGDMA
showed the highest limiting degree of conversion, in
spite of the highest concentration of the double bonds
per unit mass (Table 3) because of the very low value of
Tg
83:41C: It is also worth mentioning that UDMA
with a higher Tg value (
35.31C) than Bis-EMA
(Tg
46:11C) gave polymer network with much higher
degree of conversion. This high value (69.6%) may be
due to chain transfer reactions caused by the aNHa
groups with the polymer, which increase the mobility of
radical sites on the network and thereby oer an
alternative path of continuation of the polymerization
until the sample is deeply in the glassy state.
In Table 4 the Tg values of the prepared polymer
networks are also shown. They are all about the same

(65681C), which is 42451C above the curing temperature. This result is in accordance with the general
observation that typically: Tg Tcure 202401C [35].
In Table 5 the results obtained from the study of
photo-polymerization of mixtures of the four dimethacrylates are presented, the composition of which are
shown in Table 1. The values of limiting degree of
conversion of mixtures, calculated on the basis of Eq. (1)
as in the case of neat monomers, showed a good linear
relationship between the conversion and the mole
fraction of Bis-GMA in the mixture (Fig. 3).
As far as the rate behavior during the polymerization
of mixtures is concerned, the mixtures of Bis-GMA/
UDMA and Bis-GMA/Bis-EMA showed an intermediate behavior between those of the corresponding neat
monomers. Especially the mixtures of Bis-GMA/BisEMA with 35.8 and 56.6 mol% content of Bis-GMA,
correspondingly showed a maximum rate 2.8 and 3.3 s
1
after 2.5 and 1 s, when 5.6 and 2.9% of the double bonds
had been reacted. These results show an approximate
additive eect of the mixture components on the
polymerization rate. On the contrary, the mixtures of
Bis-GMA/TEGDMA showed a maximum rate from the
onset of the polymerization and a generally higher
initial polymerization reactivity than that of neat BisGMA. This result show a synergistic eect of components of this mixture on the polymerization rate, which
may be most probably be due to the better plasticizing
eect of TEGDMA on Bis-GMA than UDMA and BisEMA.
In Table 5 the Tg values of the monomer
mixtures experimentally determined by a DSC and
the theoretically values calculated from the Fox
equation are shown. It is observed that in all mixtures
the experimental value is lower than the calculated
value and this dierence is higher in the case of BisGMA/TEGDMA mixtures. Bis-GMA as has already
been mentioned has a structure of a quasi-network
formed by strong hydrogen bonding. In the mixtures

Table 5
Limiting degree of conversion of mixtures of Bis-GMA with various dimethacrylates
Monomer mixture

GT (1)
GT (2)
GU (1)
GU (2)
GE (1)
GE (2)
GTUE (1)
GTUE (2)
GUE
a
b

Limiting degree of conversion

60.9
54.9
57.1
50.6
45.4
42.3
47.6 (51.8)b
44.9 (50.4)b
44.5 (48.5)b

Tg (1C) of monomer mixture

a

Tg (1C) of polymer network

Exp.

Fox

Exp.

Foxa

61.1

47.0

29.4

24.6

36.4

30.4

51.8

36.1

25.6

19.8

34.1

26.4

75
76
73
73
67
68

66.0
66.4
67.6
67.4
67.6
67.4

Fox equation is applied for Tg in K.

These are theoretical values of degree of conversion calculated on the basis of the law of mixtures.

1827

I. Sideridou et al. / Biomaterials 23 (2002) 18191829

80

80
2

Bis-GMA/TEGDMA R = 0.997
Bis-GMA/UDMA R2= 0.995
2

Bis-GMA/Bis-EMA R2= 0.950

R = 0.982
70

Degree of conversion (%)

Degree of conversion (%)

70

60

50

40

60

50

40

0.0

0.2

0.4

0.6

0.8

1.0

Mole fraction of Bis-GMA

-80

-60

-40

-20

Monomer Tg ( C)

Fig. 3. Limiting degree of conversion of various mixtures of BisGMA, versus mole fraction of Bis-GMA.

Fig. 4. Variation of limiting degree of conversion with Tg of the initial

monomer mixture Bis-GMA/TEGDMA.

many of these bonds are destroyed, the exibility

of the Bis-GMA molecules increases and so the Tg
value of Bis-GMA in the mixtures is lower than
that in neat Bis-GMA. This eect, which is not predicted
by the Fox equation, is probably responsible for the
lower experimental values than the calculated values.
Also, the higher dierence in these values in the
Bis-GMA/TEGDMA mixtures must be attributed
to the much smaller size of TEGDMA (Table 3)
resulting in a better compatibility of Bis-GMA and
TEGDMA.
In Fig. 4 the dependence of limiting degree of
conversion as a function of the Tg of monomer mixture
Bis-GMA/TEGDMA is shown. A good linear relationship is observed (R2 0:982) which shows that in this
mixture the network mobility depends mainly on the Tg
of the initial monomer mixture. In the case of monomer
mixture of Bis-GMA/UDMA and of Bis-GMA/BisEMA, the above coecient of linear correlation was
correspondingly R2 0:935 and R2 0:834:
In Table 5 the Tg values, experimental and theoretical,
of the prepared copolymer networks are also presented.
For the calculation of these Tg from the Fox equation,
the Tg values of the homopolymer networks of the

corresponding monomers presented in Table 4 were

used. It is observed that while in mixtures of Bis-GMA/
Bis-EMA the experimental and theoretical values are
about the same, in the mixtures of Bis-GMA/UDMA
and mainly in Bis-GMA/TEGDMA the experimental
values are higher. These higher values could not
attributed to a higher degree of conversion of these
copolymers than the expected, since this was found to
follow the additive law of mixtures (Fig. 3). However,
this result could be explained if it is accepted that the
polymerization rate goes to zero when the temperature
Tb and not the Tg of the polymerizing system coincides
with the curing temperature, as has already been
suggested by Kloosterboer and Lijten [36]. Tg (atransition) is characteristic of the onset of segmental
motion of the main chain of polymer, while Tb (btransition) is usually ascribed to localized group
motions, in polymethacrylates with the rotation of
aCOOR group and in polydimethacrylates with polymer segments or domains containing pendant unreacted
end groups [41]. The higher Tg values of copolymers BisGMA/UDMA and Bis-GMA/TEGDMA is probably
due to the restriction of the exibility of the UDMA and
TEGDMA monomer units by the rigid and high polar

1828

I. Sideridou et al. / Biomaterials 23 (2002) 18191829

Bis-GMA monomer units. So, the Tg values of the

UDMA and TEGDMA monomer units seem to be
higher in the copolymers than those in the corresponding homopolymers.
In this work the mixtures of Bis-GMA with more than
one comonomer were also studied. The limiting degree
of conversion was slightly less than that predicted by the
additive law of mixtures (Table 5).

4. Conclusions
The room-temperature photopolymerization of the
most widely used dimethacrylates in dentistry was
studied by FT-IR. The results obtained were then tted
by a non-linear least square method to a rational
function, which permitted the accurate calculation of the
limiting degree of conversion. The latter was found to
increase in the order:
Bis-GMAoBis-EMAoUDMAoTEGDMA:
This trend is connected with the mobility of the
polymerizing system, which depends on Tg of the
formed network and mainly on the Tg of the unreacted
monomer. Especially in the case of UDMA, the degree
of conversion is higher than that expected, most
probably due to chain transfer reactions caused by the
aNHa groups, which increase the mobility of radical
sites on the network. These chain-transfer reactions may
also be responsible for the high polymerization reactivity of UDMA.
The nal degree of conversion is not aected by the
polymerization reactivity of monomers. The maximum
rate of polymerization of UDMA and Bis-GMA was
higher than that of TEGDMA and Bis-EMA.
The limiting degree of conversion of mixtures of BisGMA with the other dimethacrylates showed a good
linear relationship with the mole fraction of Bis-GMA
and in the case of Bis-GMA/TEGDMA mixture, also
with the Tg of the initial monomer mixture.
TEGDMA was found to have a better plasticizing eect on Bis-GMA than UDMA and Bis-EMA,
which is responsible for the synergistic eect on the
polymerization rate observed in the mixture Bis-GMA/
TEGDMA.

References
[1] Craig RG, editor. Restorative dental materials, 10th ed. St. Louis,
Missouri: C.V. Mosby Company, 1997.
[2] Linden LA. Photo curing of polymeric dental materials, plastic
composite resins. In: Fouassier JP, Rabek JE, editors. Radiation
curing in polymer science and technology, vol. IV. Essex: Elsevier
Ltd, 1993. p. 387466.

[3] Linden LA. Dental polymers. In: Salamone CJ, editor. Polymeric
materials encyclopedia, vol. 3D-E. New York: CRC Press, 1996.
p. 1839.
[4] Silikas N, Watts DC. Rheology of urethane dimethacrylate and
dilute formulations. Dent Mater 1999;15:25761.
[5] Cook WD. Kinetics and properties of a photopolymerized
dimethacrylate oligomer. J Appl Polym Sci 1991;42:220922.
[6] Kalachandra S, Taylor DF, McGrath JE, Sankarapandian M,
Shobha HK. StructureFproperty relationships in dental composites based on polydimethacrylates. Polym Prepr 1997;38:945.
[7] Bogdal D, Pielichowski J, Boron A. Application of diol
dimethacrylates in dental composites and their inuence on
polymerization shrinkage. J Appl Polym Sci 1997;66:23337.
[8] Sankarapandian M, Shobba HK, Kalachandra S, Mc Grath JE,
Taylor DF. Characterization of some aromatic dimethacrylates
for dental composite applications. J Mater Sci: Mater Med
1997;8:4658.
[9] Labella R, Braden M, Clarke RL, Davy KWM. THFMA in
dental monomer systems. Biomaterials 1996;17:4316.
[10] Wang G, Culbertson BM, Xie D, Seghi RR. Eect of uorinated
triethylene glycol dimethacrylate on the properties of unlled,
light cured dental resins. J Macromol Sci Pure Appl Chem
1999;36:23752.
[11] Asmussen E, Peutzfeldt A. Inuence of UEDMA, Bis-GMA and
TEGDMA on selected mechanical properties of experimental
resin composites. Dent Mater 1998;14:516.
[12] Hansel C, Leyhausen G, Mai U, Geurtsen W. Eect of various
components extracts on the growth of Streptococcus sobrinus.
J Dent Res 1998;77(1):607.
[13] Tanaka K, Taira M, Shintani H, Wakasa K, Yamaki M. Residual
monomers (TEGDMA and Bis-GMA) of a set visible-light-cured
dental composite resin when immersed in water. J Oral Rehab
1991;18:35362.
[14] Selli E, Bellobono IR. Photopolymerization of multifunctional
monomers: kinetic aspects. In: Fouassier JP, Rabek JE, editors.
Radiation curing in polymer science and technology, vol III.
Essex: Elsevier Ltd, 1993. p. 132.
[15] Kalachandra S, Taylor DF, DePorter CD, McGrath JE.
Polymeric materials for composite matrices in biological environments. Polymer 1993;34:77882.
[16] Cook WD. Photopolymerization kinetics of dimethacrylates using
the camphoroquinone/amine initiator system. Polymer
1992;33:6009.
[17] Decker C. Kinetic analysis and performance of UV-curable
coatings. In: Pappas SP, editor. Radiation curing, science and
technology. New York: Plenum Press, 1992. p. 13579.
[18] Heatley F, Pratsitsilp Y, McHugh N, Watts DC, Derlin H.
Determination of extent reaction in dimethacrylate-based dental
composites using solid-state 13C m.a.s.n.m.r spectroscopy and
comparison with FT-IR spectroscopy. Polymer 1995;36:185967.
[19] Chambers S, Guthrie J, Otterburn MS, Woods J. Factors
aecting residual unsaturation and curing rates in photocured
crosslinked compositions. Polym commun 1986;27:20911.
[20] Cook WD. Thermal aspects of the kinetics of dimethacrylate
photopolymerization. Polymer 1992;33:215261.
[21] Davy KWM, Kalachandra S, Pandain MS, Braden M. Relationship between composite matrix molecular structure and properties. Biomaterials 1998;19:200714.
[22] Silikas N, Watts DC. Rheology of urethane dimethacrylate and
dilute formulations. Dent Mater 1999;15:25761.
[23] Kalachandra S, Kusy RP. Comparison of water sorption by
methacrylate and dimethacrylate monomers and their corresponding polymers. Polymer 1991;32:242834.
[24] Sankarapandian M, Shobha HK, Kalachandra S, Taylor DF,
Shultz AR, McGrath JE. Synthesis of new dental composite
matrix dimethacrylates. Polym Prepr 1997;38:923.

I. Sideridou et al. / Biomaterials 23 (2002) 18191829

[25] Kalachandra S, Taylor DF, McGrath JE, Sankarapandian S,
Shobha HK. Structureproperty relationships in dental composites based on polydimethacrylates. Polym Prepr 1997;38:945.
[26] Khatri CA, Antonucci JM, Stransbury, Schultheisz CR. Synthesis, characterization and evaluation of urethane derivatives of
Bis-GMA. Polym Prepr 2000;41:1724.
[27] Wang G, Culbertson BM, Xie D, Seghi RR. Physical property
evaluations of peruorotriethylene glycol dimethacrylate as a
potential reactive diluent in dental composite resins. J Macromol
Sci, Pure Appl Chem 1999;36:22536.
[28] Taira M, Urabe H, Hirose T, et al. Analysis of photoinitiators in
visible-light cured dental composite resins. J Dent Res
1988;67:248.
[29] King MB, Queen NM. Use of rational functions for representing
data. J Chem Eng Data 1979;24:178.
[30] Kolling OW. Nonlinear regression model for the representation of
dielectric constants for binary aprotic solvents. Anal Chem
1985;57:1721.
[31] Yoshida K, Greener EH. Eect of photoinitiator on degree of
conversion of unlled light-cured resin. J Dent 1994;22:2969.
[32] Cook WD. Photopolymerization kinetics of oligo(ethylene Oxide)
and oligo(methylene) oxide dimethacrylates. J Polym Sci, Polym
Chem 1993;31:105367.
[33] Lovell LG, Stransbury JW, Syrpes DC, Bowman CN. Eects of
composition and reactivity on the reaction kinetics of dimethacrylate/dimethacrylate
copolymerizations.
Macromolecules
1999;32:391321.

1829

[34] Zhu S, Tian Y, Hamielec AE, Eaton DR. Termination of trapped

radicals at elevated temperatures during copolymerization of
MMA/EGDMA. Polymer 1990;31:172634.
[35] Prime RB, Turi EA, editors. Thermosets In: Thermal Characterization of polymeric materials, 2nd ed. vol. 2. New York:
Academic Press, 1997 [Chapter 6].
[36] Kloosterboer JG, Lijten GFC. Photopolymers exhibiting a large
dierence between glass transition and curing temperature.
Polymer 1990;31:95101.
[37] Scranton AB, Bowman CN, Klier J, Peppas NA. Polymerization
reaction dynamics of ethylene glycol methacrylates and dimethacrylates by calorimetry. Polymer 1992;33:16839.
[38] Anseth KS, Kline LM, Walker TA, Anderson KJ, Bowman CN.
Reaction kinetics and volume relaxation during polymerizations
of multiethylene glycol dimethacrylates. Macromolecules
1995;28:24919.
[39] Tager A. Polymer chain exibility. In: Physical chemistry of
polymers, 2nd ed. Moscow: Mir Publishers, 1978. p. 105.
[40] Decker C. New developments in UV-curable acrylic monomers. In: Fouassier JP, Rabek JE, editors. Radiation curing in
polymer science and technology, vol III. Essex: Elsevier Ltd, 1993
[Chapter 2].
[41] Jeppesen MT, Rawls HR. Dynamic mechanical analysis of
amorphous-phase organization in acrylic polymers. J Adhes
1994;47:1919.