Evaluation of reactions of a dilute cross-linker using solution nuclear magnetic resonance spectroscopy with isotopically labeled cross-linkers

Article abstract of DOI:10.1021/ma0346759

Cross-linked poly(methyl methacrylate) (PMMA) was synthesized to high conversion with 0.02-0.5 wt % ¹³C- or ²H-labeled ethylene glycol dimethacrylate (EGDMA). Samples were analyzed with a variety of NMR techniques to examine different cross-linker reaction products (or architectures), i.e., pendant EGDMA methacrylates, cyclized EGDMA, and cross-linked EGDMA. Solid-state ¹³C NMR spectroscopy was not applicable because poor peak resolution masked the cross-linker architecture peaks. Solid-state ²H NMR spectroscopy of deuterium-labeled EGDMA was also not useful because the deuterium line shape was convoluted by signals from natural abundance deuterium in the polymer. Solution ¹³C NMR analysis of solvent swollen polymers (gel-state ¹³C NMR spectroscopy) produced well-resolved spectra of copolymers containing less than 0.5 wt % ¹³C-labeled EGDMA. These spectra confirm that significant amounts of EGDMA were singly reacted, resulting in large numbers of pendant methacrylate units. Non-cross-linked and completely soluble model PMMA copolymers containing 0.1 wt % ¹³C-labeled pendant EGDMA methacrylates units (incorporated in a postpolymerization reaction) were used to identify the ¹³C NMR signals characteristic of EGDMA pendant units. Signals from the stereochemical triads (syndio-and heterotactic) were identified, but specific peaks for nine-membered cyclic EGDMA units were not observed. The detection limit of gel-state ¹³C NMR analysis on these MMA/EGDMA copolymers was as low as 0.02 wt % ¹³C-labeled EGDMA, which demonstrates the utility of this approach for characterizing lightly cross-linked polymers.

Full text of DOI:10.1021/ma0346759

Macromolecules 2004, 37, 3699-3706
3699
Evaluation of Reactions of a Dilute Cross-Linker Using Solution
Nuclear Magnetic Resonance Spectroscopy with Isotopically Labeled
Cross-Linkers
Lon J . Ma th ia s,* Rick D. Da vis, Scott J . Stea d m a n , a n d Willia m L. J a r r ett
University of Southern Mississippi, School of Polymers and High Performance Materials,
Hattiesburg, Mississippi 39406-0076
Rich a r d D. Red fea r n †
Ineos Acrylics, Inc., Cordova, Tennessee 38018
Ala n Bu n n ‡
ICI Technology, Middlesbrough, TS90 8J E, U.K.
Received May 21, 2003; Revised Manuscript Received March 5, 2004
ABSTRACT: Cross-linked poly(methyl methacrylate) (PMMA) was synthesized to high conversion with
2
0.02-0.5 wt % 13C- or H-labeled ethylene glycol dimethacrylate (EGDMA). Samples were analyzed with
a variety of NMR techniques to examine different cross-linker reaction products (or architectures), i.e.,
pendant EGDMA methacrylates, cyclized EGDMA, and cross-linked EGDMA. Solid-state 13C NMR
spectroscopy was not applicable because poor peak resolution masked the cross-linker architecture peaks.
Solid-state 2H NMR spectroscopy of deuterium-labeled EGDMA was also not useful because the deuterium
line shape was convoluted by signals from natural abundance deuterium in the polymer. Solution 13C
NMR analysis of solvent swollen polymers (gel-state 13C NMR spectroscopy) produced well-resolved spectra
of copolymers containing less than 0.5 wt % 13C-labeled EGDMA. These spectra confirm that significant
amounts of EGDMA were singly reacted, resulting in large numbers of pendant methacrylate units. Non-
cross-linked and completely soluble model PMMA copolymers containing 0.1 wt % 13C-labeled pendant
EGDMA methacrylates units (incorporated in a postpolymerization reaction) were used to identify the
13C NMR signals characteristic of EGDMA pendant units. Signals from the stereochemical triads (syndio-
and heterotactic) were identified, but specific peaks for nine-membered cyclic EGDMA units were not
observed. The detection limit of gel-state 13C NMR analysis on these MMA/EGDMA copolymers was as
low as 0.02 wt % 13C-labeled EGDMA, which demonstrates the utility of this approach for characterizing
lightly cross-linked polymers.
1
In tr od u ction
Landin and Macosko6 used H NMR spectroscopy to
observe pendant methacrylates in soluble samples of
poly (MMA-co-EGDMA). They constructed a series of
pendant conversion vs monomer conversion plots for low
conversion polymerizations of PMMA (<33.3%) contain-
ing 0.57-1.7 mol % EGDMA. From the positive y-
intercept on these graphs they concluded that the
extrapolated loss of pendant methacrylates at the onset
of polymerization was a result of primary EGDMA
cyclization (3-4 mol %). Primary cyclization was ex-
plained as the formation of 9- or 10-membered EGDMA
rings generated before reaction with monofunctional
monomer, similar to deliberately obtained ring struc-
tures in linear polymers via cyclopolymerization.13 The
apparent content of primary cyclization product was
unaffected by EGDMA concentration but increased with
dilution. These researchers also reported that the total
conversion of pendant methacrylates increased with an
increase in EGDMA concentration and dilution of
reactants and that the reactivity of pendant methacry-
lates was half that of monomeric methacrylates. A
kinetic model was constructed to estimate the number
of cyclics, cross-links, and pendant methacrylates at
selected monomer conversions less than 30%; no high
conversion samples were analyzed.
Cross-linked PMMA is used in a variety of commercial
materials such as membranes,1 dental fillings,2 den-
tures,2 latex coatings,3 computer-to-sensor data trans-
mission links,4 composites,5 optically clear sheeting,
limb prostheses, bathtubs, bathroom sinks, and shower
surrounds where cross-linking plays a vital role in both
processing and ultimate properties. A better under-
standing of cross-linking effects on polymerization
kinetics and cross-linker architectures would provide
insight into optimizing the industrial processes and
maximizing performance.
Several researchers since the 1970s have investigated
the polymerization kinetics of cross-linked PMMA.6-12
For example, Haung and co-workers12 used DSC to
determine monomer conversion, rate of polymerization,
reaction rate constants, and reaction orders during the
copolymerization of EGDMA and 2-hydroxylethyl meth-
acrylate (HEMA). This detailed research revealed trends
in reaction parameters that were dependent on both
EGDMA concentration and temperature. However, the
architectures of incorporated EGDMA and their com-
bined effects on the kinetics of the polymerization were
not determined.
This research provided an excellent understanding of
the fate of EGDMA in soluble PMMA copolymers at low
conversion. However, the results will not accurately
represent the EGDMA architecture in MMA/EGDMA
† Present address: Department of Chemistry, Rhodes College,
Memphis, TN.
‡ Retired.
* Corresponding author: e-mail Lon.Mathias@usm.edu.
10.1021/ma0346759 CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/17/2004

3700 Mathias et al.
Macromolecules, Vol. 37, No. 10, 2004
F igu r e 1. Possible EGDMA architectures in MMA/EGDMA
copolymers.
copolymerizations where high monomer conversion oc-
curs. This is due to inherent limitations in analysis of
cross-linked and insoluble polymers, even when poly-
merizations are stopped at conversions well below those
used in industry. Low conversion PMMA copolymers do
not experience the Trommsdorff effect (autoacceleration)
or significant changes in monomer concentration, vis-
cosity, and temperature, which will affect the individual
incorporation amounts for products from various com-
peting pathways of the propagating radicals. As a result,
much of the research in this area has been based on
soluble polymers with results extrapolated to insoluble
cross-linked systems without taking into account the
large changes in rates, molecular weights, and cross-
link densities that occur in the latter stages of reaction.
F igu r e 2. Solid-state 13C CP/MAS NMR of poly(MMA-co-1,2-
13C-EGDMA) with 0.1 wt % EGDMA. /indicates spinning
sidebands.
The overall goal of our work in this area is to explore
and develop methods for analyzing insoluble cross-
linked polymers to elucidate how changes in reaction
conditions during polymerization affect molecular com-
position and ultimate properties. This paper is aimed
at identifying dimethacrylates comonomer architec-
tures, i.e., pendant methacrylates, cyclized and cross-
linked (Figure 1). We targeted PMMA copolymerizations
with 0.02-0.5 wt % EGDMA carried to 100% conver-
sion. Possible NMR techniques for characterization of
such systems include solid-state 13C NMR, solid-state
2H NMR, and high-resolution 13C NMR spectroscopy
performed on solvent swollen samples.
This paper will first briefly discuss problems with
traditional solid-state 13C NMR and solid-state 2H NMR
methods for such analyses and then detail successful
use of gel-state 13C NMR spectroscopy for identifying
EGDMA comonomer architectures. With the last method,
the EGDMA peaks were identified using high-resolution
(solution) 13C NMR analysis of solvent swelled copoly-
mers but only after overcoming considerable difficulties
with sample preparation and related problems with
NMR shimming. The majority of this paper will discuss
the synthesis of various labeled monomers, labeled
polymers, NMR sample preparation, and the results of
gel-state 13C NMR analysis.
2
F igu r e 3. Solid-state H NMR spectra of PMMA containing
0.1 wt % (top), 1.0 wt % (middle), and 5.0 wt % (bottom) 1,1,2,2-
2H-EGDMA.
architectures in EGDMA/MMA copolymers was at-
tempted first by solid-state 13C NMR spectroscopy, then
solid-state 2H NMR spectroscopy, and finally high-
resolution gel-state 13C NMR spectroscopy. Observation
of EGDMA units at concentrations as low as 0.02-0.5
wt % and differentiation of cross-linked and pendant
EGDMA NMR peaks was only achieved using solution
13C NMR spectroscopy on solvent swollen copolymers.
Solid -St a t e 13C NMR a n d Solid -St a t e 2H NMR
An a lysis. Initially, traditional solid-state NMR char-
acterization techniques of polymers (13C and H NMR
2
spectroscopy) were used to analyze PMMA copolymers
containing 0.02-0.5 wt % 13C- or 2H-labeled EGDMA.
Solid-state 13C CP/MAS and HPD NMR spectroscopy are
extremely powerful techniques because of the flexibility
of the pulse programs and the plethora of polymers that
have been investigated. However, EGDMA ester meth-
ylene signals were convoluted by polymer backbone
carbon peaks at concentrations less than 5 wt % 13C-
labeled EGDMA (Figure 2), which is at least an order
of magnitude greater than the EGDMA concentrations
of interest.
Resu lts a n d Discu ssion
Solid-state H NMR analysis of 2H4-labeled EGDMA
in PMMA resulted in a static H line shape (Figure 3)
2
2
The three main architectures of dimethacrylates, such
as EGDMA, generated during copolymerization with
monofunctional monomers, such as MMA, are cross-
linked (incorporation of a single dimethacrylate mono-
mer into two different PMMA chains), pendant meth-
acrylates (one end of a dimethacrylate monomer incor-
porated into a PMMA chain and the other end unin-
corporated), and cyclized (incorporation of both ends of
a dimethacrylate unit into the same PMMA chain as
adjacent units) (Figure 1). Identification of these three
with a splitting of 110 kHz between 90° asymptotes.
However, at concentrations below 1 wt % the amount
of naturally abundant 2H in the PMMA chain was on
the same order as the amount of labeled EDGMA. Thus,
the signal from these deuterons (a narrow peak ∼35 kHz
in width and a 125 kHz wide-line line shape) began
overlapping with the EDGMA line shape. In addition,
this technique was time-consuming; averaging between
4 and 8 days and 60-80 000 scans per sample.

Macromolecules, Vol. 37, No. 10, 2004
Reactions of a Dilute Cross-Linker 3701
F igu r e 4. Solution 13C NMR spectrum of poly(MMA-co-AE-HEMA).
F igu r e 5. Gel-state 13C NMR spectra of 13C-labeled poly(MMA-co-EGDMA) (A, C) and 13C-labeled pendant EGDMA methacrylate
model copolymers (B, D).
An alternative method was sought that would provide
a faster and more sensitive analysis of PMMA copoly-
mers containing extremely low EGDMA content (0.02-
0.5 wt %). This technique not only had to be sensitive
enough to observe extremely low EGDMA concentra-
tions but produce well-resolved NMR signals in order
to distinguish between the different EGDMA architec-
tures (cyclized, cross-linked, and pendant methacrylate).
Thus, we examined the potential of solvent-swollen
NMR analysis.
Gel-Sta te 13C NMR An a lysis. Solution 13C NMR
spectra of solvent swollen cross-linked copolymers could
be acquired within a reasonable period of time (∼12 h).
In addition, EGDMA ester methylene signals (∼62 ppm)
were sufficiently resolved to distinguish different EGD-
MA-based structures. Assigning the splitting of the ester
peaks to specific architectures therefore became the
focus of this study. The assignments were obtained
through a series of model copolymers containing only
pendant methacrylate architecture and isotopically
labeled EGDMA.
assigned to the ester methylene units on the attached
side and pendant side, respectively, of incorporated
AE-HEMA. This model copolymer lead to the first
inclination that not only could EGDMA cross-linker be
observed by solution 13C NMR spectroscopy but also the
attached methacrylate peaks could be distinguishable
from pendant methacrylates peaks (Figure 4, peaks e
and f).
The poly(MMA-co-AE-HEMA) solution 13C NMR spec-
tra indicated the possibility of studying the nature of
the cross-link sites provided the signal of those positions
could be increased. Therefore, singly and doubly 13C-
labeled EDGMA was used to synthesize poly(MMA-co-
EGDMA), including a model system with pendant vinyl
groups. The use of labeled material permitted observa-
tion of peaks at levels of EGDMA incorporation ap-
proaching 0.02 wt %.
Figure 5 summarizes the NMR results for PMMA
polymers cross-linked with singly and doubly 13C-
labeled EDGMA. The spectrum of the model copolymer
synthesized with mono-13C-labeled EDGMA (Figure 5D)
show two doublets centered at 62.4 ppm. Using data
from poly(MMA-co-AE-HEMA) spectra as well as previ-
ously acquired spectra for substituted MMA/HEMA
systems14 as a guide, the upfield doublet is assigned to
PMMA copolymers containing pendant EGDMA meth-
acrylates was first modeled using an analogue copoly-
mer based on the acetate ester of HEMA, poly(MMA-
co-AE-HEMA) (Figure 4). Peaks labeled e and f were

3702 Mathias et al.
Macromolecules, Vol. 37, No. 10, 2004
F igu r e 6. Ester region of a model PMMA copolymer containing di-13C-labeled pendant EGDMA methacrylates, acquired at 75
and 100 MHz for carbon.
F igu r e 7. Solution 13C NMR spectra of poly(MMA-co-1,2-13C-EGDMAp) and poly(MMA-co-1,2-13C-EGDMA) before and after
bromination (0.1 wt % labeled EGDMA).
the ester methylene unit closest to the unreacted vinyl
group. The downfield doublet is associated with sites
closest to the polymer backbone. The peaks are split into
doublets due to tacticity. The splitting pattern for the
pendant doublet is consistent with a heterotactic struc-
ture (mr), while the downfield peak exhibits a syndio-
tactic pattern (rr), with a greater degree of splitting due
to its proximity to the polymer chain.15 The spectrum
of the model copolymer with doubly labeled material
(Figure 5D) exhibits a set of eight overlapping peaks
due to the addition of 13C-13C coupling (J ∼ 50 Hz).
Spectra acquired at 75 and 100 MHz for carbon confirm
this explanation (Figure 6).
important to note that the data are not completely
quantiatitve. No attempts were made to remove NOE
effects, which enhances sites with differing degrees of
mobility unequally. In addition, fully reacted EDGMA
units may be in regions of sufficient rigidity as to
broaden the signal to the point of undetectability.
However, it is believed that sufficient amount of local
mobility is present in the gel to permit these results to
be interpeted semiquantitatively.
At this point key questions concerning the assign-
ments of the ester methylenes e and f as well as the
presence of cyclic EDGMA structures need to be ad-
dressed. To confirm our previous spectral assignments,
the pendant methacrylates were brominated (Figure 7).
Referring to published tables of 13C NMR shifts,16
bromine subsitution in delocalized systems such as
CH2Br-C6H5 will move resonances for sites â, γ, and δ
from the site of halogenation approxiamately 1.2-0.7
Figure 5A,C shows the spectra for PMMA cross-linked
with singly and doubly labeled EDGMA. Both clearly
show the presence of fully reacted EDGMA (singlet at
∼62 ppm) as well a relatively large percentage of
EDGMA units with pendant methacrylates. Here it

Macromolecules, Vol. 37, No. 10, 2004
Reactions of a Dilute Cross-Linker 3703
F igu r e 8. 13C NMR gel-state spectra of EDGMA cross-linked
PMMA polymerized at 40, 60, and 80 °C.
F igu r e 9. Solution 13C NMR spectrum of the acetyl ester of
HEMA.
ppm downfield. In contrast, for aliphatic systems sites
that are two or more bonds removed from the bromine
position tend to shift upfield by approximately 1 ppm.
Using these trends as a guide, it is expected that the
pendant -CH2O- position will be significantly shifted
downfield, while the other site will be moved upfield.
The spectrum of the bominated model system in Figure
7 confirms these predictions. The pendant site shifted
∼2 ppm downfield, and the site adjacent to the main
chain moved upfield by ∼1 ppm, creating a pair of
quartet structures separated by 1 ppm. Bromination of
the MMA/EGDMA copolymer also yields a similar
quartet pattern characteristic of pendant methacrylate
groups, with the upfield peaks obscured by the cross-
linking EGDMA moeity. This again clearly shows that
a significant number of EDGMA units are not involved
in cross-linking.
Con clu sion s
Gel-state 13C NMR analysis was performed on poly-
mers containing between 0.02 and 0.5 wt % 13C-labeled
cross-linker, indicating a range of applications that
should be general to other vinyl systems. Cross-linker
concentrations higher than 0.5 wt % produced polymers
that did not swell sufficiently to resolve cross-linker
NMR signals in the system described here. Use of 13C-
labeled cross-linkers with gel-state 13C NMR analysis
of MMA/EGDMA copolymers confirmed significant
amounts of mono-reacted EGDMA. Peaks for pendant
EGDMA methacrylates and cross-linked EGDMA were
assigned and included splitting caused by tacticity in
the backbone. These assignments were based on model
copolymers synthesized specifically to mimic systems
with pendant methacrylate units. There was no evi-
dence of cyclic EGDMA units because their concentra-
tion was too low to detect or their peaks were obscured.
Finally, these results clearly demonstrate the capability
of the gel-state NMR method to see very small amounts
of appropriately labeled segments in cross-linking groups
that are normally not observable using traditional solid-
state and solution methods.
It was hoped that bromination would shift unreacted
EGDMA peaks enough to reveal the presence of any
possible cyclic structures. However, no additional peaks
were observed. In addition, all attempts to synthesize
cyclic EGDMA model compounds by varying EGDMA
concentrations, solvent, temperature, or reaction time
failed. These facts strongly suggest that cyclic units are
not present. Nevertheless, the only definite conclusions
are that pendant and cross-linked EGDMA units are
observed for EDGMA cross-linked PMMA, with cyclic
structures either absent or obscured by cross-link and
pendant peaks.
Exp er im en ta l Section
A. Syn th esis. All reagents were used as received from
Aldrich Chemical Co. (Milwaukee, WI) unless specified oth-
erwise. 13C- and 2H-labeled ethylene glycol samples were
purchased from Cambridge Isotope Laboratory (Andover, MA)
and used as received.
Tem p er a tu r e Effects in P olym er iza tion s a n d
Gel-Sta te 13C NMR An a lysis. Polymerizations of 0.1
wt % 1,2-13C-EGDMA and MMA carried out at various
temperatures produced cross-linked networks with dif-
fering degrees of swellability. Specifically, samples
prepared at 40 °C swelled much less than those obtained
at 80 °C. This indicates a higher cross-link density and
cross-linking efficiency for EGDMA at lower tempera-
tures. The MMA monomer conversion also decreased at
lower polymerization temperatures, consistent with
observations of Huang and co-workers,12 who used DSC
and FTIR to follow monomer conversion isothermally
and nonisothermally. They reported that for EGDMA
and HEMA copolymerizations lower monomer conver-
sion was observed at lower temperatures.
Figure 8 shows the 13C NMR spectra of the labeled
systems described above. Differences in these spectra
may be due to changes in NOE enhancements because
of lower mobility. However, the most probable explana-
tion is that the local mobility has been decreased to the
point where broadening mechanisms such as residual
dipolar coupling become significant. This is particularly
true for systems with low solvent uptake. This prohib-
ited semiquantitative comparison of the data.
Mon om er s. All isotopically labeled comonomers were syn-
thesized using a similar procedure. Only the synthesis of di-
13C-labeled HEMA (1,2-13C-HEMA) and di-13C-labeled EGDMA
(1,2-13C-EGDMA) comonomers will be discussed in detail. All
other labeled comonomer synthesis descriptions will contain
only information pertinent to the synthesis of those moieties.
Synthesis of the Acetate Ester of 2-Hydroxyethyl Methacrylate
(AE-HEMA). A 100 mL round-bottom flask was charged with
HEMA (2.01 g, 0.015 mol), triethylamine (2.01 g, 0.02 mol),
and 40 mL of methylene chloride. The flask was capped with
an addition funnel and cooled in an ice bath for 20 min.
Through the addition funnel, acetyl chloride (2.02 g, 0.02 mol)
was added dropwise over 45 min to the stirring solution. The
reaction proceeded in an ice bath for 2 h and then at room
temperature for an additional 2 h. The reaction mixture was
washed with three 25 mL portions of 0.1 M NaOH. The water
layers were combined and washed with three 25 mL portions
of ethyl ether. The ether layers and the organic layer were
combined and solvents were removed under high vacuum at
30 °C, leaving a clear liquid (1.98 g, 0.011 mol, 58%); 100%
purity by GC. The 13C NMR spectrum (TMS, CDCl3) is
provided in Figure 9.

3704 Mathias et al.
Macromolecules, Vol. 37, No. 10, 2004
F igu r e 11. Solution 13C NMR spectra of 1,2-13C-EGDMA.
F igu r e 10. Solution 13C NMR spectra of 1,2-13C-HEMA.
Synthesis of Di-13C-Labeled HEMA (1,2-13C-HEMA). Meth-
acryloyl chloride (1.9 g, 18 mmol) and 40 mL of methylene
chloride were combined in a 150 mL round-bottom flask fitted
with an addition funnel and magnetic stir bar. The flask was
placed in an ice bath for 25 min, and then a cooled solution of
1,2-13C-ethylene glycol (1.0 g, 16 mmol), triethylamine (1.8 g,
18 mmol), and 25 mL of methylene chloride were dropwise
added through the addition funnel over 30 min. The ice bath
was then removed, and the reaction proceeded at room
temperature for an additional hour. The reaction mixture was
combined with 200 mL of methylene chloride and washed with
50 mL of 0.1 N HCl. Methylene chloride was removed by rotary
evaporator, and the residue was dissolved in 200 mL of H2O
and washed with four 50 mL portions of heptane, thus
removing all EGDMA formed during acylation. The aqueous
layer was extracted with seven 50 mL portions of methylene
chloride, which separated HEMA from any residual ethylene
glycol. The combined methylene chloride washings were dried
over MgSO4, and the solvent was removed by rotary evapora-
tion to give a clear liquid (0.7 g, 34%); 100% purity by GC.
The 13C NMR spectrum (TMS, CDCl3) of the ester methylene
region is provided in Figure 10.
F igu r e 12. Solution 13C NMR spectra of poly(MMA-co-HEMA)
containing 3 wt % HEMA.
was capped with a septum and purged under nitrogen for 30
min. The polymerization proceeded at 65 °C overnight in a
nitrogen atmosphere. The transparent copolymer plug was
dissolved in chloroform, reprecipitated into methanol, and
dried overnight under high vacuum to give a hard white solid
(4.5 g, 81%); 10% AE-HEMA incorporated by 1H NMR. The
13C NMR spectrum (tetrachloroethane, DMSO-d6 inset) is
provided in Figure 4.
Synthesis of Poly(MMA-co-HEMA). Refer to the polymeri-
zation of poly(MMA-co-AE-HEMA). EGDMA impurity was
removed from commercial HEMA by extracting a HEMA/water
solution with heptane. The polymerization was conducted in
a test tube charged with purified HEMA (0.15 g, 1.2 mmol),
MMA (5.01 g, 0.05 mol), and 0.5 wt % AIBN. The product was
a hard white solid (4.4 g, 93%); 3 wt % HEMA by 1H NMR.
The 13C NMR spectrum (tetrachloroethane, DMSO-d6 inset)
is provided in Figure 12.
Synthesis of Di-13C-Labeled EGDMA (1,2-13C-EGDMA).
Refer to the synthesis of di-13C-labeled HEMA (1,2-13C-HEMA).
A 150 mL round-bottom flask was charged with methacryloyl
chloride (3.7 g, 35 mmol) and 40 mL of methylene chloride. A
chilled solution of 1,2-13C-ethylene glycol (1.0 g, 16 mmol),
triethylamine (3.5 g, 35 mmol), and 25 mL of methylene
chloride was dropwise added, and the reaction proceeded for
an hour at room temperature. After washing with dilute acid,
the methylene chloride layer was removed with a rotary
evaporator, leaving crude solid product which was dissolved
in 150 mL of diethyl ether and washed with 15 25 mL portions
of a 0.1 N NaOH(aq), 25 mL of H2O, and saturated KCl(aq).
The ether layer was dried over MgSO4, and diethyl ether was
removed with a rotary evaporator to give a clear liquid (2.0 g,
58%); 98% purity by GC. The 13C NMR spectrum (TMS, CDCl3)
is provided in Figure 11.
Synthesis of a Model Copolymer Containing Pendant EGD-
MA Methacrylates: Poly(MMA-co-EGDMAp). Reactant con-
centrations were based on the stoichiometric content of HEMA
in a copolymer. For example, a 25 mL round-bottom flask was
charged with poly(MMA-co-HEMA) (0.5 g, 3 wt % HEMA) and
a 3 wt % excess of triethylamine at a 25 wt % concentration
in chloroform. The round-bottom flask was capped with an
addition funnel, and the reaction mixture was stirred in an
ice bath for 20 min under nitrogen. A 10 wt % excess of
methacryloyl chloride (which had been distilled twice under
vacuum before use in chloroform) was added dropwise over
30 min, and the reaction was allowed to proceed for 6 h in an
ice bath. The copolymer was precipitated in a mixture of
methanol and water, reprecipitated into methanol from chlo-
roform, and then dried overnight under high vacuum at room
temperature. The product was a hard white solid (0.5 g, 95%);
100% conversion of pendant HEMA alcohols to methacrylate
Synthesis of Mono-13C-Labeled EGDMA (1-13C-EGDMA).
Refer to the synthesis of di-13C-labeled EGDMA (1,2-13C-
EGDMA). The product, 1-13C-EGDMA, was a clear liquid (2.1
g, 65%); 97% purity by GC.
Synthesis of 2H4-Labeled EGDMA (1,1,2,2-2H-EGDMA).
Refer to the synthesis of di-13C-labeled EGDMA (1,2-13C-
EGDMA). 1,1,2,2-2H-ethylene glycol was used instead of 13C-
labeled ethylene glycol. The product, 1,1,2,2-2H-EGDMA, was
a clear liquid (2.3 g, 65%); 98% purity by GC.
Cop olym er s. All PMMA copolymers were polymerized free
radically in bulk using thermal initiation, as described in the
procedure for the synthesis of poly(MMA-co-AE-HEMA). The
poly(MMA-co-AE-HEMA) procedure will be discussed in detail,
and all polymerization descriptions will contain only informa-
tion pertinent to the synthesis of that copolymer.
Synthesis of Poly(MMA-co-AE-HEMA). A large Pyrex test
tube was charged with AE-HEMA (0.5 g, 0.003 mol), MMA
(5.01 g, 0.05 mol), and 0.5 wt % AIBN. The reaction vessel
1
esters (3 wt % pendant methacrylate units) by H NMR.
Synthesis of Poly(MMA-co-1,2-13C-HEMA). Refer to the
polymerization of poly(MMA-co-AE-HEMA). The product was
1
a hard white solid (1.5 g, 95%); 0.1 wt % 1,2-13C-HEMA by H
NMR.

Macromolecules, Vol. 37, No. 10, 2004
Reactions of a Dilute Cross-Linker 3705
Synthesis of a Di-13C-Labeled Model Copolymer Containing
Pendant EGDMA Methacrylates: Poly(MMA-co-1,2-13C-EGD-
MAp). Refer to the synthesis of poly(MMA-co-EGDMAp) from
poly(MMA-co-HEMA). HEMA pendent alcohol groups in poly-
(MMA-co-1,2-13C-HEMA) were converted to pendant EGDMA
methacrylates by postpolymerization reaction with methacry-
loyl chloride. The product was a hard white solid (0.7 g, 90%);
100% conversion of alcohols to methacrylate ester groups by
1H NMR. The 13C NMR spectrum (TMS, CDCl3) of ester
methylene region is provided in Figure 5b.
B. NMR Sp ectr oscop y. Preparation of Solvent Swollen
Cross-Linked PMMA Copolymers for Solution 13C NMR Spec-
troscopy. High-resolution solution 13C NMR analysis was
conducted on CDCl3 swollen cross-linked PMMA copolymers
containing less than 0.5 wt % labeled EGDMA. Copolymer
plugs were broken into small chunks using a hammer and a
stainless steel compartment equipped with a stainless steel
piston. The polymer chunks were then chopped to a powder
in a coffee grinder and initially swollen with CDCl3 in a
scintillation vial while vigorously agitating with a stir bar. This
preswelling enabled more efficient mixing and dissolution
compared to a one-step preparation in NMR tubes. After
preswelling for 15 min, the swollen copolymers were trans-
ferred to a 5 mm NMR tube and continuously hand agitated
for 5-10 min. To increase solvent uptake, the samples were
suspended in a water bath held at 50 °C for several hours (8-
24 h, as needed) and intermittently mixed using a thin metal
wire. If a sample contained swollen particles suspended in a
solvent-rich continuum, excess solvent was present that
interferes with the NMR analysis (shimming problems). The
swelling process was started over from the beginning using a
new sample. Samples polymerized with 0.5 wt % EGDMA
swelled on the order of 800% while samples containing 0.02
wt % or less EGDMA behaved as totally soluble polymer. Air
bubbles trapped in gelled samples were removed by a down-
ward whipping action of the NMR tubes.
Synthesis of Poly(MMA-co-1-13C-HEMA). Refer to the po-
lymerization of poly(MMA-co-AE-HEMA). The product was a
hard white solid (1.8 g, 92%); 0.1 wt % 1-13C-HEMA by 1H
NMR.
Synthesis of a Mono-13C-Labeled Model Copolymer Contain-
ing Pendant EGDMA Methacrylates: Poly(MMA-co-1-13C-
EGDMAp). Refer to the synthesis of poly(MMA-co-EGDMAp)
from poly(MMA-co-HEMA). HEMA pendent alcohol groups in
poly(MMA-co-1-13C-HEMA) were converted to pendant EGD-
MA methacrylates by postpolymerization reaction with meth-
acryloyl chloride. The product was a hard white solid (0.9 g,
94%); 100% conversion of alcohols to ester groups by 1H NMR.
The 13C NMR spectrum (TMS, CDCl3) of ester methylene
region is provided in Figure 5d.
Synthesis of Poly(MMA-co-1,2-13C-EGDMA). Refer to the
polymerization of poly(MMA-co-AE-HEMA). The product was
a hard white solid (1.0 g, 90%); 0.1 wt % 1,2-13C-EGDMA by
1H NMR. The 13C NMR spectrum (TMS, CDCl3) of the ester
methylene region is provided in Figure 5a.
Considerable time was spent shimming on
a properly
swollen sample to obtain the desired resolution. In general,
shimming was easier on lightly cross-linked samples (below
0.2 wt % EGDMA). The quality of shim was judged at a low
number of scans by viewing the line width of the CDCl3 peaks.
Shim adjustments were made until good resolution of the
solvent triplet peaks was obtained, and then long-term acqui-
sition was carried out.
Synthesis of Poly(MMA-co-1-13C-EGDMA). Refer to the
polymerization of poly(MMA-co-AE-HEMA). The product was
1
a hard white solid (0.8 g, 83%); 0.1 wt % 1-13C-EGDMA by H
NMR. The 13C NMR spectrum (TMS, CDCl3) of the ester
methylene region is provided in Figure 5c.
Synthesis of Poly(MMA-co-1,1,2,2-2H-EGDMA). Refer to the
polymerization of poly(MMA-co-AE-HEMA). Insoluble copoly-
mers were synthesized containing 0.02, 0.1, 1.0, and 5.0 wt %
1,1,2,2-2H-EGDMA. The products were hard white solid (aver-
Solid-State 13C NMR (CP/ MAS and HPD) Spectroscopy.
Solid-state 13C NMR spectra were acquired on a 200 MHz
Bruker MSL NMR spectrometer at a spectral frequency of
50.31 MHz for carbon using a 7 mm multinuclear probe.
Routine cross-polarization/magic angle spinning (CP/MAS)
pulse program was used with a 5 µs 90° pulse, 5 s recycle time,
40 µs pulse delay, and a 3 ms contact time. A routine high-
power decoupled (HPD) pulse program was also used with a 5
µs 90° pulse, 3 s recycle time, and 40 µs pulse delay. The
average number of scans was 15 000.
2
age 90% yield). The solid-state H NMR spectrum is provided
in Figure 3.
Bromination and Titration of Model PMMA Copolymers
Containing Only EGDMA Pendant Methacrylates. The bro-
mine, catalyst, titration, and iodide solutions were indepen-
dently prepared according to previous work.17,18 The prepara-
tion of the solutions, the bromination of the pendant methacry-
lates, and the titration of I2 were all performed under red lights
to eliminate photochemical reactions.
A 10 mL round-bottom flask was charged with poly(MMA-
co-EGDMAp) (0.5 g, 3 wt % pendant EGDMA methacrylates)
at 10 wt % concentration in chloroform. The reaction vessel
was placed in an ice bath, and the bromine solution (10 g, 0.1
equiv) and catalyst solution (50 mL, 30 mol % mercuric
acetate) were added. The reaction proceeded in the dark for
10 h in an ice bath, and then a KI solution (20 g, 0.2 equiv)
was added. Aliquots were removed every 30 min, and excess
I2 was titrated with a 0.1 N Na2S2O4 solution.
Results from these measurements were not quantitative or
reproducible. Further studies were carried out involving
bromination of pendant methacrylates using a dilute Br2
solution with out concern for quantifying pendant methacry-
late content. Bromination of poly(MMA-co-1,2-13C-EGDMAp)
and poly(MMA-co-1,2-13C-EGDMA) was performed with 10 wt
% excess Br2 in relation to the total content of 13C-labeled
EGDMA. Solution 13C NMR spectra (TMS, CDCl3) of the ester
methylene region of the brominated copolymers are provided
in Figure 7.
Solid-State 2H NMR Spectroscopy. Solid-state 2H NMR
spectra were acquired on a 400 MHz Bruker MSL NMR
spectrometer operating at a spectral frequency of 64.4 MHz
for deuterium. A routine quadrapolar spin-echo pulse program
(90+x-τ-90+y-τ) was used with 90° pulse durations of 3.8 µs,
τ of 25 µs, recycle time of 3 s, and 60-80 000 average number
of transients. Raw data were shifted so that the FID was
Fourier-transformed at the top of the spin echo.
Gel-State 13C NMR Spectroscopy. Routine high-resolution
solution 13C NMR analysis of solvent swollen copolymers was
conducted on a 300 MHz Bruker AC NMR and 400 MHz
Bruker MSL NMR spectrometers using a 5 mm probe and
operating for carbon at 75.5 and 100.13 MHz spectral frequen-
cies, respectively. Instrument shimming was crucial to obtain-
ing well-resolved spectra. Long recovery times after adjusting
shim settings, plus high lock power and gain needed to obtain
a good lock signal, made shimming on the swollen polymer
networks especially tedious. The best spectral resolution was
obtained by first shimming on pure CDCl3 and then making
slight adjustments of the shim settings for the swollen polymer
samples.
Attempted Synthesis of Cyclic EGDMA (EGDMAc). Several
10 mL round-bottom flasks were charged with EGDMA and
AIBN. The reactants were diluted to 100, 50, 3, 2, 1, 0.1, and
0.01 wt % concentrations in various dried solvents, i.e.,
benzene, chloroform, methylene chloride, dioxane, and tet-
rahydrofuran. The reaction vessels were sealed with a septa
and degassed under nitrogen. The reactions proceeded for 48
h at 65 °C. All reactions gave cross-linked gels of EGDMA and
unreacted EGDMA.
Ack n ow led gm en t. The Strategic Research Fund
Grant No. 9612 from Imperial Chemical Industries
supported this research.
Refer en ces a n d Notes
(1) Spevaeck, J .; Schneider, B. Adv. Colloid Interface Sci. 1987,
81.
(2) Wolff, E. M. Aust. Dent. J . 1962, 439.

3706 Mathias et al.
Macromolecules, Vol. 37, No. 10, 2004
(3) Anseth, K. S.; Lauren, M. K.; Walker, T. A.; Anderson, K.
J .; Bowman, C. N. Macromolecules 1995, 28, 1.
(4) Takezawa, Y.; Tanno, S.; Taketani, N.; Ohara, S.; Asano, H.
J . Appl. Polym. Sci. 1991, 2811.
(5) Ma, C.; Chen, C. Polym. Eng. Sci. 1991, 1086.
(6) Landin, D. T.; Macosko, C. W. Macromolecules 1986, 846.
(7) Zhu, S.; Tian, Y.; Hamielec, A. E.; Eaton, D. R. J . Polym.
Sci., Part A: Polym. Chem. 1990, 154.
(8) Zhu, S.; Tian, Y.; Hamielec, A. E.; Eaton, D. R. Polymer 1990,
1145.
(9) Eastmond, G. C.; Richardson, J . E. Macromolecules 1991,
24, 9.
(12) Huang, C. W.; Sun, Y. M.; Huang, W. F. Polymer 1997,
1873.
(13) Butler, G. J . J . Am. Chem. Soc. 1957, 79, 3128.
(14) Bunn, A., unpublished 13C data.
(15) Bovey, F. A. Chain Struct, Macromol. 1982, 15, 7.
(16) Pretsch, E.; Clerc, T.; Seibl, J .; Simon, W. In Tables of
Spectral Data for Structure Determination of Organic Com-
pounds, 2nd ed.; Springer-Verlag: Berlin, 1989.
(17) Rowe, E. G.; Furnas, D. C.; Bliss, H. Ind. Eng. Chem. (Anal.
Ed.) 1944, 16, 371.
(18) Okay, E. G.; Nagash, H. J .; Capek, I. Polymer 1995, 12,
2413.
(10) O’Shaughnessy, B.; Yu, J . Macromolecules 1994, 27, 7.
(11) Anseth, K.; Bowman, C. N.; Peppas, N. A. J . Polym. Sci.,
Part A: Polym. Chem. 1994, 139.
MA0346759