Kinetic modeling of radical thiol-ene chemistry for macromolecular design: Importance of side reactions and diffusional limitations

Article abstract of DOI:10.1021/ma302619k

The radical thiol-ene coupling of thiol-functionalized polystyrene (PS-SH) with dodecyl vinyl ether (DVE) and the polystyrene-b-poly(vinyl acetate) (PS-b-PVAc) polymer-polymer conjugation using 2,2-dimethoxy-2-phenylacetophenone (DMPA) as photoinitiator are modeled to assess the importance of diffusional limitations and side reactions. Intrinsic chemical rate coefficients are determined based on a kinetic study of the coupling of benzyl thiol (BT) and DVE. The addition and transfer reactions are chemically controlled, whereas diffusional limitations on termination slightly increase the coupling efficiency. Termination by recombination of carbon-centered radicals and addition of DMPA derived radicals to DVE are shown to be mainly responsible for the reduced coupling efficiency in case polymeric species are involved. The obtained results confirm the idea to disregard radical thiol-ene chemistry as a true member of the family of click chemistry techniques for polymer-polymer conjugation and show that the initial conditions have a significant impact on the coupling efficiency.

Full text of DOI:10.1021/ma302619k

Article
pubs.acs.org/Macromolecules
Kinetic Modeling of Radical Thiol−Ene Chemistry for Macromolecular
Design: Importance of Side Reactions and Diffusional Limitations
Pieter Derboven,† Dagmar R. D’hooge,† Milan M. Stamenovic,‡ Pieter Espeel,‡ Guy B. Marin,†
Filip E. Du Prez,*,‡ and Marie-Francoise Reyniers*,†
̧
†Laboratory for Chemical Technology, Ghent University, Ghent B-9000, Belgium
‡Polymer Chemistry Research Group, Ghent University, Ghent B-9000, Belgium
S
* Supporting Information
ABSTRACT: The radical thiol−ene coupling of thiol-functionalized
polystyrene (PS-SH) with dodecyl vinyl ether (DVE) and the polystyrene-
b-poly(vinyl acetate) (PS-b-PVAc) polymer−polymer conjugation using 2,2-
dimethoxy-2-phenylacetophenone (DMPA) as photoinitiator are modeled to
assess the importance of diffusional limitations and side reactions. Intrinsic
chemical rate coefficients are determined based on a kinetic study of the
coupling of benzyl thiol (BT) and DVE. The addition and transfer reactions
are chemically controlled, whereas diffusional limitations on termination
slightly increase the coupling efficiency. Termination by recombination of
carbon-centered radicals and addition of DMPA derived radicals to DVE are
shown to be mainly responsible for the reduced coupling efficiency in case
polymeric species are involved. The obtained results confirm the idea to
disregard radical thiol−ene chemistry as a true member of the family of “click”
chemistry techniques for polymer−polymer conjugation and show that the initial conditions have a significant impact on the
coupling efficiency.
UV-initiated thiol−ene “click” chemistry enables an unique
spatial and temporal control with applications in the fields of
materials science and synthetic chemistry, including surface and
polymer modification, biofunctionalization, and uniform net-
work formation.11,13
INTRODUCTION
■
Recent research1−3 has shown that the “click” chemistry
concept allows the efficient and straightforward synthesis of
advanced polymer architectures. Using this concept, complex
macromolecular architectures can be formed by coupling
macromolecules, ideally with high yield, excellent modularity,
and orthogonality under a variety of mild conditions. In
addition, for a true “click” reaction inoffensive byproducts are
easily removable via nonchromatographic methods.4
The principle of the UV-initiated thiol−ene process is shown
in Scheme 1. Initially, carbon-centered radicals are generated by
cleavage of a photoinitiator (I1I2), typically 2,2-dimethoxy-2-
phenylacetophenone (DMPA),17 and subsequent transfer to
the thiol (R1SH) leads to the formation of thiyl radicals (R1S•),
which in turn can add to an ene molecule (CH2CHR2). The
desired thio−ether coupling product is formed, once the
resulting carbon-centered radical abstracts a hydrogen from
another thiol molecule. In this last step, a new thiyl radical is
formed, allowing further reaction if the ene-containing species
is still present. In case of DMPA, applied wavelengths typically
are close to the visible light (λ = 365 nm), implying the
possibility to generate thiyl radicals using sunlight as radiation
source.18,19 Interestingly, thiol−ene chemistry can even be
conducted in the absence of a photoinitiator, as specific thiols
lead to self-initiation at a well-chosen wavelength (e.g., alkyl 3-
mercaptopropionate at λ = 254 nm).20,21
Over the past decade, the Huisgen Cu(I)-catalyzed 1,3-
dipolar cycloaddition of alkynes and azides5 has evolved as the
norm for “click”-termed reactions, in particular when tailored
design features are required, as is the case for the synthesis of
functionalized polymers and polymer−polymer conjugates.6−8
Despite its very high reaction rate, the Cu(I)-mediated “click”
reaction suffers from toxicity issues related to the employed
catalyst. Therefore, significant efforts have been devoted to the
exploitation of metal-6 and additive-free9 “click” reactions,
extending the realm of the “click” philosophy to, among others,
(hetero-)Diels−Alder reactions,8,9 radical and nucleophilic
thiol−ene10−13 chemistry, and ultrarapid synthetic approaches
such as norbornene−tetrazine chemistry.14
Within the large repertoire of these new “click” chemistry
approaches, the radical addition of thiyl radicals (R1S•) to
nonactivated double bonds (CH2CHR2), i.e. thiol−ene
reactions,15 offers in principle the combination of high
efficiency, good robustness, and oxygen16 and water tolerance,
starting from commercially available compounds. In particular,
Ideally, the final thio−ether coupling product concentration
should be equal to the initial ene or thiol concentration
Received: December 21, 2012
Revised: February 4, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ma302619k | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
a
Scheme 1. Principle of UV-Initiated Thiol−Ene “Click” Chemistry
a
R1 and R2: low or high molar mass compound; side reactions are indicated in gray.
depending on the ene or thiol being the limiting reagent, and a
coupling efficiency of 100% should therefore be obtained.
However as indicated in Scheme 1, several side reactions can
interfere, leading to a reduced coupling efficiency. For example,
consumption of the ene compound might occur via
homopolymerization if a (homo)polymerizable ene, such as
an acrylate,22 is used or via addition of photoinitiator derived
radicals to the ene. Also, termination reactions can interrupt the
thiol−ene coupling process, among which thiyl−thiyl radical
coupling, head-to-head coupling of the carbon-centered
radicals, and cross-termination between carbon-centered and
thiyl radicals are the most prominent termination reactions.23
The mechanism of the radical thiol−ene chemistry presented
in Scheme 1 suggests that the overall kinetics might be first
order in both the initial thiol and ene concentration. However,
differential scanning calorimetry (DSC) and Fourier transform
infrared (FTIR) monitoring of radical thiol−ene reactions23,24
indicate that different vinyl compounds exhibit different overall
reaction orders and that the overall rates mainly depend on the
as required by the adapted macromolecular “click” definition.30
In addition, size exclusion chromatography (SEC) analysis
suggested the undesired formation of termination products.
Also, the star polymer assembly of a poly(butyl acrylate)
macromonomer with a trimethylolpropane tris(2-mercaptoace-
tate) trifunctional thiol core was only successful to a small
extent, as indicated by mass spectrometry and SEC analysis. In
a first attempt to explain these observations, the authors
proposed that unwanted termination by recombination
reactions and addition of photoinitiator derived radicals to
the ene as well as a diffusion-controlled addition reaction could
contribute to the reduction of the coupling efficiency. Because
of the lack of an extensive set of experimental data, these
hypotheses could unfortunately not be attested.
Kinetic simulations by Preuss and Barner-Kowollik31
revealed that diffusional effects might indeed play a role in
polymer−polymer conjugation. In case the coupling reactivity
for low molar mass species is similar as for termination (109 L
mol−1 s−1), the obtained molar mass distribution (MMD)
displays a marked conversion dependence with the effect being
most pronounced for initially broad MMDs. Based on literature
reports, the DMPA photoinitiated radical thiol−ene polymer−
polymer conjugation can thus be expected to be influenced
both by side reactions and diffusional limitations on the
reaction rates. However, it remains unclear to which extent each
of these phenomena influence the coupling efficiency.
In this work, a combined experimental and modeling study is
presented, allowing to assess the main factors controlling the
coupling efficiency in radical thiol−ene reactions photoinitiated
by DMPA. As model compounds, benzyl thiol (BT), dodecyl
vinyl ether (DVE), thiol-functionalized polystyrene (PS-SH)
and allyl ether-functionalized poly(vinyl acetate) are consid-
ered, the macromolecular compounds being the same as in the
earlier study of Koo et al.19 It is shown that both for coupling of
small species and coupling reactions involving polymeric
species the coupling efficiency crucially depends on the chosen
reaction conditions. Lower initial concentrations of the ene and
thiol compound and higher initial photoinitiator concentrations
decrease the ratio of the addition/transfer rate to the
photoinitiator decomposition rate, thus favoring the formation
of side products. Furthermore, the lower coupling efficiencies
for polymer−polymer conjugation cannot be attributed to
diffusional limitations on addition and transfer reactions, since
their intrinsic reactivity is too low as compared to termination
reactions.
ratio of the addition to the transfer rate coefficient kadd,chem
/
ktr,chem. For the thiol−allyl ether system, the overall reaction
order was even observed to be independent of the initial ene
concentration and could be very well described by a kadd,chem to
ktr,chem ratio of 10.23,24 Contrary to these experimental findings,
a recent quantum-chemical study reported by Northrop and
Coffey25 suggests a ratio of kadd,chem to ktr,chem lower than 1 for
the thiol−allyl ether system and that for thiol−propene, thiol−
vinyl ether, thiol−vinylsilazane, and thiol−allyl ether the overall
reaction rate might be slowed down considerably due to the
occurrence of the β C−S scission reaction, i.e., the reverse of
the addition reaction. However, Reddy et al.24,26 were able to
model conversion−time profiles for a range of thiol−ene
systems without the need to consider the occurrence of the
reverse β C−S scission.
Recently, Koo et al.19 employed thiol−ene chemistry in
combination with reversible addition−fragmentation chain
transfer (RAFT) polymerization27−29 to assess its potential as
a tool for polymer functionalization as well as star polymer and
block copolymer synthesis. However, when targeting con-
jugation of thiol-functionalized polystyrene (PS-SH) and allyl
ether-functionalized poly(vinyl acetate) (PVAc), nuclear
magnetic resonance (NMR) spectroscopy and elemental
analysis revealed only a limited success of the radical-initiated
thiol−ene reaction (coupling efficiency: ∼25%), particularly in
case equimolar starting prepolymer concentrations were used,
B
dx.doi.org/10.1021/ma302619k | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Furthermore, it enables to study the effect of side reactions with
photoinitiator-derived radicals and to determine the influence of
diffusional limitations on the coupling efficiency (fcoupling), which is
defined as the molar ratio of the concentration of the thio−ether
coupling product (cthio−ether) to the initial concentration of the limiting
reagent (cthiol,0 or cene,0):
EXPERIMENTAL SECTION
■
Materials. Dodecyl vinyl ether (DVE; 98% pure) and dibenzyl
disulfide (BDS; 98% pure) were purchased from Sigma-Aldrich.
Benzyl thiol (BT; 99% pure) and the solvent 1,4-dioxane (>99.5%
pure, inhibitor free) were purchased from Acros Organics. As a radical
source, 2,2-dimethoxy-2-phenylacetophenone (DMPA; Acros Organ-
ics) was used. All chemicals were used as received.
Methods. Analysis. A VLX365 radiometer (bandwidth: 355−375
nm) was used to measure the intensity profile of the UV lamps (setup
with nine Philips PL-S 9 W UV-A lamps, bandwidth: 340−400 nm and
maximum at 365 nm), while the temperature profiles of the sample
mixtures during reaction were measured with a digital VWR
thermometer (EU 620-0919). For the coupling of DVE and BT,
online Fourier transform infrared (FTIR; carbon−carbon double bond
stretch absorbance peak at 1620 cm−1; Mettler Toledo ReactIR 4000),
gas chromatography (GC; Hewlett-Packard (Agilent) 5890; flame
cthio−ether
cene,0
cthio−ether
cthiol,0
fcoupling
=
or fcoupling =
(1)
In addition, for macromolecular species the coupling reaction should
be ideally performed using equimolar initial concentrations of the ene
and thiol groups.30 It should be kept in mind that the definition given
in eq 1 only holds when the limiting reagent reaches full conversion
during the time span of the considered reaction. An alternative
definition is presented by eq 2, in which the obtained coupling product
concentration is related to the amount of limiting reagent that has
been consumed. However, throughout this article eq 1 will be used,
1
ionization detector), and 300 MHz H nuclear magnetic resonance
(1H NMR; Bruker Avance 300) analyses were performed. A Hewlett-
Packard GCD-Plus GC/MS (G1800B; MS: mass spectrometry) was
used for qualitative characterization of the samples.
1
since this quantity is directly accessible by H NMR analysis for the
studied system.
cthio−ether
cene,0 − cene
cthio−ether
cthiol,0 − cthiol
Coupling Reaction between BT and DVE (Small−Small). An
overview of the experiments performed is given in Table 1 (entries 1−
f′coupling
=
or f′coupling =
(2)
A detailed description of the kinetic model and its assumptions are
given in section S4 of the Supporting Information. The corresponding
intrinsic chemical rate coefficients are taken from literature (see
section S4 in Supporting Information) or determined based on
experimental data as discussed below.
Table 1. Overview of Reaction Conditions Used for the
Kinetic Study of the Coupling of Benzyl Thiol (BT) and
Dodecyl Vinyl Ether (DVE) in 1,4-Dioxane, Photoinitiated
by 2,2-Dimethoxy-2-phenylacetophenone (DMPA) (T = 21
°C; Irradiation Time = 10 min)
a
b
RESULTS AND DISCUSSION
[thiol]0/
[ene]0
[ene]0 (
[DMPA]0
vsol
(%)
■
entry
components
mol L−1
)
(mol L−1
)
In this section, it is first explained how approximate values for
the intrinsic kinetic parameters are obtained for the BT/DVE
small−small model system. Next, the influence of the initial
concentrations on the coupling efficiency and the side product
spectrum for this small−small model system is illustrated.
Finally, it is demonstrated that coupling reactions involving
polymeric species are not influenced by diffusional limitations
on addition nor transfer. The reduced coupling efficiency for
polymer−small and polymer−polymer conjugations is shown
to be mainly due to an increased importance of termination
reactions with carbon-centered radicals and of addition
reactions involving photoinitiator-derived radicals. The latter
is a direct consequence of the lower applied initial ene and thiol
concentrations due to the limited solubility of the polymers.
Coupling of BT and DVE (Small−Small). As mentioned
above, the coupling of BT and DVE has been monitored using
online FTIR and GC. Figure 1 (entry 1 in Table 1) shows that
online FTIR (squares) and GC (triangles) measurements lead
to the same DVE conversion profile. For the online FTIR
measurements, the carbon−carbon double bond stretch
absorbance peak at 1620 cm−1 is tracked, while for GC analysis
1
2
3
4
BT/DVE/DMPA
BT/DVE/DMPA
BT/DVE/DMPA
1
0.3
0.4
0.2
0.002
0.002
0.002
0.004
89
0.5
2
87
90
c
BDS/DMPA
∼100
a
In entries 2−4, 0.01 equiv of DMPA with respect to limiting reacting
agent (i.e., thiol, ene, or benzyl disulfide (BDS)) is used, whereas in
entry 1 this value is 0.0067. vsol: volume percentage of solvent in the
b
c
sample. Concentration of BDS/DMPA: 0.004 mol L−1.
3). For a typical experiment (e.g., entry 1 in Table 1; infrared (IR)
detection) first, DMPA (1.0 × 10−5 mol, 2.55 mg) is dissolved in 4.5
mL of 1,4-dioxane in a 25 mL two-necked, glass flask. Then, DVE (1.4
× 10−3 mol, 0.37 mL) is added, and after attaching the flask to an IR-
probe, the obtained mixture is purged with argon for 30 min and
stirred at 800 rpm. After 30 min, BT (1.4 × 10−3 mol, 0.17 mL) is
added. In case GC is used as an analysis method, an additional 15 μL
of the internal standard n-dodecane (DD) is added. The sample is
purged for another 20 min and afterwards irradiated for 10 min with
365 nm UV-light.
To study disulfide dissociation/formation in the absence of ene or
thiol (entry 4 in Table 1), a similar approach as for entries 1−3 in
Table 1 is followed as explained in section S.3 of the Supporting
Information.
In the present kinetic study, reliable analysis under maximum light
intensity during the entire reaction was not possible due to practical
reasons and the change of the intensity of the UV lamps had therefore
to be taken into account in the kinetic model (see section S1 of the
Supporting Information). Furthermore, no significant temperature
increase of the mixture during the coupling reaction was measured,
which allows to assume isothermal conditions for the kinetic modeling
study and determination of the intrinsic chemical rate coefficients.
Kinetic Model. Similar to the work of Cramer et al.23,24 and Back et
al.,15 kinetic modeling is used as a tool to study the coupling reaction
between BT and DVE. However, in this work a more extended
reaction scheme, including the reverse β C−S scission as proposed by
Northrop and Coffey,25 is considered. The model thus allows to assess
the importance of the reverse reaction for the addition step.
Figure 1. Conversion profiles of dodecyl vinyl ether (DVE)
determined by online Fourier transform infrared (FTIR; squares),
gas chromatography (GC; triangles), and 300 MHz proton nuclear
magnetic resonance (1H NMR; circles); entry 1 in Table 1.
C
dx.doi.org/10.1021/ma302619k | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 2. Overlay of the proton nuclear magnetic resonance (1H NMR) 300 MHz reference spectrum: dodecyl vinyl ether (DVE)/2,2-dimethoxy-2-
phenylacetophenone (DMPA)/1,4-dioxane (ref), the spectrum just before the reaction (0″), and the spectrum after 45 s of reaction (45″);
conditions: entry 1, Table 1.
the ratio of the DVE peak to the internal standard peak (DD) is
compared to the initial ratio.
The 300 MHz H NMR overlay of the reference spectrum
1). It can be observed that a good description of the
experimental DVE conversion data is obtained for all three
conditions. Moreover, an excellent description of the photo-
initiator conversion under typical coupling conditions (entry 1
in Table 1) results, as shown in Figure 4.
1
(before BT was added: DVE/DMPA/1,4-dioxane; ref), the
spectrum just before reaction (0″), and the spectrum after 45 s
of reaction (45″) are shown in Figure 2. The peaks
corresponding to 1,4-dioxane, BT, and DVE are labeled. The
coupling efficiency is calculated by relating the area of peak a,
which is the resonance signal of the two hydrogens next to the
thio−ether bond (see Figure 2), to the area of peak d, which
corresponds to the methyl hydrogens at the tail of the dodecyl
group in the DVE molecule. A coupling efficiency of 100% is
obtained at sufficiently high reaction time (around 4 min), so
that its corresponding time profile can be used as extra
measurement for the conversion−time profile of DVE or BT.
As can be seen in Figure 1, this profile (circles) is in good
agreement with the FTIR (squares) and GC (triangles) results.
Figures 3a−c show the experimentally measured and
simulated DVE conversion profiles (full lines) for an initial
molar ratio of BT to DVE of 2, 1, and 0.5 (entries 1−3 in Table
For the intrinsic rate coefficient for the addition of BT to
DVE (kadd,chem) and the intrinsic rate coefficient for the
subsequent transfer reaction yielding the thio−ether coupling
product (ktr,chem), values of 1.2 × 106 and 1.1 × 106 L mol−1 s−1
are respectively obtained, which are similar to those reported by
Reddy et al.22,26 It should be stressed that the measured DVE
conversion data allow a reliable assessment of these intrinsic
chemical rate coefficients, since diffusional limitations can be
neglected for the coupling reaction between the low molar mass
compounds BT and DVE and isothermicity holds (see
Experimental Section). Although Northrop and Coffey25
highlighted the importance of the β C−S scission reaction for
vinyl ether coupling reactions, simulations revealed that a
proper description of the experimental conversion data
recorded in this work is not possible using their reported rate
D
dx.doi.org/10.1021/ma302619k | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 3. Dodecyl vinyl ether (DVE) conversion (full line) and the coupling efficiency (eq 1) as a function of time for (a/d) entry 3 in Table 1,
[BT]0/[DVE]0 = 2; (b/e) entry 1 in Table 1, [BT]0/[DVE]0 = 1; (c/f) entry 2 in Table 1, [BT]0/[DVE]0 = 0.5; points: experimental data; line:
simulated using the kinetic parameters in Table S1; BT = benzylthiol.
profile is shown in Figure 5 (points). As explained in section S3
of the Supporting Information, additionally, GC/MS measure-
Figure 4. Conversion of the photoinitiator as a function of time;
conditions: entry 1 in Table 1; points: experimental data; line:
simulated using the parameters in Table S1.
Figure 5. Concentration of dibenzyl disulfide as a function of time
(entry 4 in Table 1); points: experimental data; line: simulated using
kinetic parameters in Table S1.
coefficients. However, since these authors focused on methyl
vinyl ether/methyl mercaptan instead of DVE/BT,25 the
importance of the β C−S scission reaction is re-evaluated
from the equilibrium coefficient for the addition/ β C−S
ments are performed to identify the different products at
varying irradiation times.
scission calculated using Benson’s group additivity method.32
A
very high value for the equilibrium coefficient of 1011 L mol−1
results, indicating that the addition of the thiyl radical to the
double bond of DVE can be considered as irreversible and,
hence, that the β C−S scission can indeed be safely neglected
for the BT/DVE system (see section S5 of the Supporting
Information).
The values of the other intrinsic kinetic parameters are
discussed in section S4 of the Supporting Information. Except
for termination by recombination, literature data are
used.22,33,34 The intrinsic chemical rate coefficients for
termination by recombination (ktc,chem) are determined from
an independent experiment (entry 4 in Table 1), in which a
known amount of dibenzyl disulfide (BDS) and DMPA are
dissolved in 1,4-dioxane in the absence of DVE and BT,
followed by irradiation at the same wavelength used in an actual
thiol−ene coupling experiment (UV, λ = 365 nm). The
concentration of BDS during the irradiation period is
monitored using GC, and the resulting concentration−time
The experiment is designed based on the observations of
Sayamol et al.,35 who reported the UV-induced dissociation of
disulfides. Since the results of Reddy et al.22,26 imply that, in a
first approximation, the intrinsic chemical recombination rate
coefficients in the DVE/BT system can be assumed equal, the
experimental data in Figure 5 allow the determination of the
intrinsic rate coefficient ktc,chem taking into account that mainly
recombination reactions between the thiyl and the photo-
initiator radicals occur (see section S3 of the Supporting
Information). Using a value of 4 × 109 L mol−1 s−1 for ktc,chem, a
good agreement between experiment and simulation is
obtained, as shown in Figure 5. Furthermore, this value is in
agreement with the early work of Fischer et al.36 and the more
recent evaluation of the intrinsic termination rate coefficient of
small radicals by Smith et al.37 The photodissociation
parameters considered in the kinetic model for this specific
experiment are discussed in sections S1 and S3 of the
Supporting Information.
E
dx.doi.org/10.1021/ma302619k | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
The simulated coupling efficiencies (eq 1) for entries 1−3 in
Table 1 are shown in Figure 3d−f. On the basis of these results,
it can be concluded that the coupling reaction of BT and DVE
shows “click” characteristics, since values of ca. 100% are
obtained under equimolar conditions (entry 1 in Table 1), in
agreement with the 300 MHz 1H NMR data (Figures 1 and 2).
However, as illustrated in Figure 6, simulations reveal that even
concentration range typically applied in synthetic chemistry is
presented. Additionally in Figure 6, the coupling efficiencies
corresponding to typically applied initial photoinitiator to ene
molar ratios, i.e. 0.1 and 0.2, are indicated. It follows that such
conditions allow a high coupling efficiency (>0.8).
It is important to realize that a proper understanding of the
effect of the initial concentrations on the coupling efficiency is
crucial with regard to radical thiol−ene reactions involving
polymeric species. In such reactions, the limited solubility of the
polymer species directly determines the initial end-group
concentrations for the ene/thiol species that can be used in
the coupling reaction. Typically, the initial concentration of the
polymeric species is some 10 times lower than the initial
concentrations used for small−small coupling reactions. Hence,
from Figure 6, an increased importance of the occurrence of
side reactions is expected when going from typical small−small
conditions (≥0.1 mol L−1) to conditions in which polymeric
species are involved (0.01−0.05 mol L−1).
The observed trend in Figure 6 can be understood by taking
into account that the photodecomposition of DMPA is
unimolecular, while the addition and transfer rates are
bimolecular. Hence, a typical 10-fold decrease of the initial
concentrations of ene, thiol, and photoinitiator when going
from small−small radical thiol−ene reactions to coupling
reactions involving polymeric species will have a more
pronounced effect on the addition and transfer rates. This
difference in decrease between the radical generation rate by
photoinitiation and the rate at which the radicals react to yield
the desired coupling product indeed favors the formation of
side products. The latter is illustrated in Figure 8 for two initial
conditions which are typical for small−small (Figure 8a,b) and
polymer−polymer (Figure 8c,d) thiol−ene coupling reactions.
Although the initial DMPA to ene molar ratio remains
unchanged, i.e., 0.2, lowering of the initial ene/thiol
concentration from 0.3 mol L−1 (Figure 8a,b) to 0.03 mol
L−1 (Figure 8c,d) implies that the relative contribution of side
products increases (Figure 8b,d). In particular, it can be seen
that the formation of recombination products and products
derived from addition of photoinitiator radicals to the ene
increases with increasing overall dilution.
Moreover, this decrease in the initial concentrations results in
an important increase of the photoinitiator conversion from 20
to 60% (Figure 8e; full and dotted line, right axis) at full
consumption of the ene, leading to a lower selectivity of the
thiyl radical toward addition on the ene double bond (Figure
8f; full and dotted line) and a concomitant reduction of the
coupling efficiency (Figure 8e; full and dotted line, left axis).
It can thus be concluded that a small−small radical thiol−ene
coupling reaction only shows “click” characteristics when
sufficiently low initial photoinitiator and sufficiently high initial
ene/thiol concentrations are used.
Figure 6. Final coupling efficiency as a function of the logarithm of the
initial photoinitiator (2,2-dimethoxy-2-phenylacetophenone, DMPA)
and ene (dodecyl vinyl ether, DVE) concentration; the absolute initial
concentrations of the ene and the photoinitiator (mol L−1) have been
added as well on the surface for clarity; coupling efficiency: eq 1;
simulated using the parameters in Table S1; T = 21 °C, reaction
performed in 1,4-dioxane; final = full ene conversion.
for the small−small BT/DVE/DMPA system, the coupling
efficiency crucially depends on the applied initial concen-
trations. This figure shows that at full ene conversion both an
increase of the initial photoinitiator concentration and/or a
decrease of the initial ene concentration significantly reduce the
coupling efficiency for an equimolar radical thiol−ene reaction.
Clearly, the highest coupling efficiencies are obtained for initial
ene/thiol concentrations greater than 0.1 mol L−1 and
photoinitiator concentrations lower than 0.05 mol L−1. A
further decrease of the initial ene and/or thiol concentration or
an increase of the initial photoinitiator concentration induce an
increased formation of side products (see further) and hence
reduce the coupling efficiency, as also shown in Figure 7 in
which a two-dimensional projection of Figure 6 for the initial
Polymeric Thiol−Ene Reactions: Coupling of PS-SH
and DVE (Polymer Functionalization). When BT is
replaced by its macromolecular analogue, i.e., thiol-function-
alized polystyrene (PS-SH), Koo et al.19 reported a coupling
efficiency of ∼90% for the coupling with DVE photoinitiated by
DMPA ([DMPA]0/[PS-SH]0 = 0.2). Such value was measured
both in case a 5-fold excess and an equimolar amount of DVE
were used. These authors suggested that diffusional limitations
and side reactions, such as termination and addition of
photoinitiator-derived radicals, might be responsible for the
reduced coupling efficiency. However, these hypotheses could
not be validated from the experimental data. To evaluate the
Figure 7. Final coupling efficiency as a function of the initial initiator
to ene molar ratio for various initial ene concentrations for the dodecyl
vinyl ether (DVE)/benzyl thiol (BT)/2,2-dimethoxy-2-phenylaceto-
phenone (DMPA) model system in 1,4-dioxane; in all simulations
equimolar amounts of ene and thiol are used; bulk: isothermal
conditions assumed, T = 21 °C; final = full ene conversion.
F
dx.doi.org/10.1021/ma302619k | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
For the other reactions involving macromolecular species
(addition and transfer), no power law model is available to
assess the importance of diffusional limitations, and the
encounter pair model39−42 is used instead:
1
1
1
kdiff,1
=
+
kapp,1
kchem,1
(4)
In eq 4, kdiff,l and kchem,l are the diffusional and intrinsic chemical
rate coefficient of the considered reaction 1. For the calculation
of kdiff,l the reader is referred to section S6 of the Supporting
Information. On the other hand, kchem,l can be assessed based
on the corresponding small−small system, i.e., BT/DVE, taking
into account a typical chain length dependency when going
from nonmacromolecular to macromolecular species.
Several literature reports43−46 indicate a decrease of the
intrinsic addition rate coefficient by 1 order of magnitude when
going from a small species to a polymer chain consisting of
more than ca. 10 monomer units, which is mainly attributed to
differences in the pre-exponential Arrhenius factor. A similar
chain length dependency is expected for transfer reactions, as
the pre-exponential Arrhenius factors of both addition and
transfer reactions are determined by similar physical parame-
ters.45,47 Therefore, in this kinetic model a 10-fold lower
reactivity of macroradicals toward addition and transfer
reactions has been assumed, taking into account that the
average chain length of the polymer molecules is 40.19 Hence,
for k
and k
, intrinsic values of 1.2 × 105 and 1.1 ×
macro
macro
add,chem
tr,chem
105 L mol−1 s−1 are considered, respectively.
However, both intrinsic chemical rate coefficients are too low
to be influenced by diffusional limitations. They are 4 orders of
magnitude lower (∼105 L mol−1 s−1) than for termination
(∼109 L mol−1 s−1) refuting, for the studied system, the
hypothesis of Koo et al.,19 which states that diffusional
limitations on the addition reaction could explain the reduced
coupling efficiency when macromolecular species are involved.
Only in case addition and transfer reactions have an intrinsic
chemical rate coefficient of ca. 108−109 L mol−1 s−1, the lower
mobility of longer macromolecules can lead to a diffusion
controlled coupling.31
Figure 8. Concentration of ene (full line; dodecyl vinyl ether, DVE),
photoinitiator (dashed line; 2,2-dimethoxy-2-phenylacetophenone,
DMPA), and contribution of main side products as a function of
time; molar ratio of photoinitiator to ene of 0.2; (a/b) [ene]0
=
[thiol]0 = 0.3 mol L−1 and (c/d) [ene]0 = [thiol]0 = 0.03 mol L−1; (b/
d) contribution = concentration of the corresponding side product
normalized with respect to the initial DVE concentration; (e) coupling
efficiency and DMPA conversion as function of ene conversion (full
line: conditions (a/b), dotted line: conditions (c/d)); (f) selectivity of
the thiyl radical toward addition on the ene double bond (full line:
conditions (a/b), dotted line: conditions (c/d)); reaction performed
in 1,4-dioxane.
In Table 2, the simulated final coupling efficiency, as
obtained by a kinetic model considering the chain length
macro
tr,chem
dependency of kmacro
add,chem and k
(CLD) on the one hand and
neglecting this dependency (CLI) on the other are compared
with the experimental data of Koo et al.19 for the coupling of
PS-SH with DVE. It follows that the former assumption results
in a slight underestimation of the experimental coupling
effect of diffusional control on the coupling efficiency, the
kinetic model was extended to account for diffusional
limitations on termination, addition, and transfer reactions.
For termination reactions, a power law approach was used:37,38
ii
11
i−α
Table 2. Final Coupling Efficiency (Eq 1) for Thiol-
k
= k
ii
tc,app
tc,app
(3)
Functionalized Polystyrene (PS-SH)/Dodecyl Vinyl Ether
a
(DVE) (Polymer Functionalization)
in which k
is the apparent rate coefficient for termination
tc,app
by recombination of macromolecules of chain length i in dilute
solution and α is a fitting parameter available in the literature. α
is taken equal to 0.5 as reported for PS radicals in the short
chain length regime.38 It should however be stressed that for
high chain lengths (i > 50) its value decreases to 0.15, and
experimental
simulation
[DVE]0/
[PS-SH]0
fcoupling (CLD):
fcoupling (CLI):
macro
entry
fcoupling
k
kchem
chem
1
2
1
5
∼0.90
∼0.90
0.82
0.87
0.97
0.98
hence, its beneficial effect on the coupling efficiency will be less
11
tc,app
pronounced. In this work, for k
the value obtained for the
a
Mn/PDI of PS-SH: 4000 g mol−1/1.20; intrinsic kinetic parameters:
coupling reaction between BT and DVE (small−small) is used.
For further details the reader is referred to section S6 of the
Supporting Information. It has to be emphasized that
diffusional limitations on termination lead to increased values
of the coupling efficiency.
Table S1 in Supporting Information; diffusion parameters: see section
S6 of the Supporting Information; reaction time: 1 h; [PS-SH]0 = 0.05
mol L−1; [DMPA]0 = 0.01 mol L−1 (0.2 equiv with respect to PS-SH);
DMPA = 2,2-dimethoxy-2-phenylacetophenone; final = full ene
conversion.
G
dx.doi.org/10.1021/ma302619k | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 9. Selectivity of the thiol R1-SH (Scheme 1) toward transfer reactions with photoinitiator-derived radicals and carbon-centered radicals
leading to the coupling product, for both chain length dependent (CLD) and independent (CLI) addition and transfer rate coefficients; R1:
polystyrene chain, R2: OC12H25; results shown until full conversion of the ene; conditions: entry 1 in Table 2.
efficiency, whereas the latter clearly leads to an overestimation,
favoring the consideration of a chain length dependent addition
and transfer reaction in agreement with literature reports.43−46
In other words, Table 2 demonstrates that a reduction of the
intrinsic rate coefficients for addition and transfer affects the
coupling efficiency significantly. Bearing in mind the above
discussion, this can be easily understood since a lowering of the
rate coefficients for addition and transfer directly reduces the
addition and transfer rates while the initiation rate remains
unaffected.
addition and transfer reactions can be neglected. Moreover,
these side reactions are a direct consequence of the lower
applied initial concentrations and the reduction of the addition
and transfer rate coefficients due to their chain length
dependency, both leading to a significant decrease of the
addition and transfer rates, whereas the rate at which radicals
are generated by photodecomposition of the initiator is affected
less. Additionally, it can be anticipated that similar factors
explain the failure of the star formation synthesis starting from a
poly(butyl acrylate) macromonomer and a trifunctional thiol, as
observed by Koo et al.19
Polymeric Thiol−Ene Reactions: Polymer−Polymer
Conjugation (PS-b-PVAc). Koo et al.19 reported a reduced
coupling efficiency of ca. 25% for the conjugation of equimolar
amounts (0.03 mol L−1) of an allyl ether-functionalized
poly(vinyl acetate) (PVAc, Mn = 1800 g mol−1, PDI = 1.5)
segment and a PS-SH (Mn = 4000 g mol−1, PDI = 1.2) segment
using an initial molar ratio of photoinitiator to ene equal to 0.2.
However, as it is impossible to quantify experimentally the
different polymer−polymer and polymer-initiator radical
adducts, simulations are necessary to explain the strong
reduction of the coupling efficiency. From the previous
simulation results involving PS-SH, it can be expected that
the low polymer−polymer coupling yield can be attributed to a
high importance of termination by recombination reactions and
side reactions with photoinitiator radicals, confirming the
hypothesis of Koo et al.19
This leads to a favoring of side reactions, as illustrated in
Figure 9, which shows a higher selectivity of the thiol for
transfer reaction with a photoinitiator-derived radical than for
an intermediate carbon-centered radical leading to the desired
coupling product, in case chain length dependent addition and
transfer rate coefficients are assumed. Note that the selectivity
is defined as the fraction of the consumed thiol molecules that
have reacted via the corresponding reaction pathway.
Furthermore, as already anticipated by Koo et al.,19 Figure 10
demonstrates that coupling of intermediate carbon-centered
In the kinetic model the chain length dependency of the
addition and transfer rate coefficients is taken into account, and
the ratio between both is set equal to 10 based on the results of
Cramer et al.23,24 A 10 times lower reactivity toward transfer
allows to reproduce the experimental trends, which is however
not the case when the rate coefficients reported by Northrop
and Coffey25 for methyl allyl ether/methyl mercaptan are used.
In particular, the rate coefficients calculated by Northrop and
Coffey25 result in an overall reaction rate being dependent both
on the ene and the thiol concentration, while for an allyl ether
system the overall reaction rate is experimentally observed to be
independent of the initial ene concentration.23,24 Therefore, to
evaluate the importance of the β C−S scission reaction, the
equilibrium coefficient for the addition/β C−S scission reaction
has again been calculated using Benson’s group additivity
method.32 An equilibrium coefficient of 102 L mol−1 (see
section S5 in Supporting Information) results, implying a
higher importance of the β C−S scission reaction than for the
vinyl ether system. Hence, combined with the known value for
the addition rate coefficient, the rate coefficient for the β C−S
scission is obtained. Simulation results confirm that this
Figure 10. Concentration of the main side products as a function of
time; R1: polystyrene chain, R2: OC12H25; results shown until full
conversion of the ene; conditions: entry 1 in Table 2 (CLD) with final
simulated coupling efficiency of 0.82.
radicals is the main termination pathway, whereas addition of
initiator radicals to the ene compound has a significant
contribution to the loss of coupling efficiency as well.
Hence, it can be concluded that the reduced coupling
efficiency for the PS-SH/DVE system, compared to the BT/
DVE system, is mainly caused by the occurrence of side
reactions with photoinitiator fragments and termination by
recombination reactions, whereas diffusional limitations on
H
dx.doi.org/10.1021/ma302619k | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 11. Concentration of the ene (full line; coinciding with the thiol profile), the photoinitiator (dotted line), and the total concentration of the
main side products (dashed line) as a function of time for equimolar coupling of thiol-functionalized polystyrene (PS-SH) and allyl ether-
functionalized poly(vinyl acetate) (PVAc); [PS-SH]0 = [PVAc]0 = 0.03 mol L−1 (a) [photoinitiator]0 = 0.006 mol L−1 (0.2 equiv), (b)
[photoinitiator]0 = 0.015 mol L−1 (0.5 equiv) and (c) [photoinitiator]0 = 0.03 mol L−1 (1 equiv); (d−f) corresponding coupling efficiency as a
function of time.
approach did not affect the zero-order dependency of the
overall reaction rate on the initial ene concentration, while the
values of the other rate coefficients are taken the same as
determined for the vinyl ether system.22,26
To reach full conversion of the thiol and ene compound, and
aspiring in that way a higher coupling efficiency, it could be
expected that a higher initial photoinitiator concentration might
be beneficial. However, Figure 11d−f shows that increasing the
initial photoinitiator concentration does not lead to the desired
increase of the coupling efficiency in spite of the increased
conversion of the thiol and ene compound (Figure 11a−c; full
line). In line with the observations for the small molecule
system (BT/DVE/DMPA; Figure 6), simulations revealed that
a higher initial photoinitiator concentration favors side product
formation and thus reduces the coupling efficiency even more.
This can also be derived from Figure 11a−c, in which the total
concentration of the main side products is also plotted as a
function of reaction time (dashed line).
In addition, as shown in section S7 of Supporting
Information, the carbon−carbon coupling of intermediate
carbon-centered radicals is the main side reaction, in agreement
with the simulations results on the coupling of PS-SH/BT and
DVE. It is also shown in section S6 that the addition of
photoinitiator-derived radicals to the ene and its subsequent
transfer and recombination product(s) have a significant
contribution.
Finally, in Figure 12 the effect of stepwise addition of small
amounts of photoinitiator (DMPA) on the coupling efficiency
is evaluated. It can be seen that somewhat higher coupling
efficiencies can be obtained (∼50%), since this approach allows
to reach full ene conversion while keeping the photoinitiator
concentration low. However, simulations in which the initial
photoinitiator concentration, the amount of photoinitiator
added, and the frequency of the addition were varied revealed a
maximal coupling efficiency of only 60%. This implies that even
with stepwise addition of the photoinitiator high coupling
efficiencies can never be achieved.
Hence, it can be concluded that photoinitiated radical thiol−
ene chemistry is not an efficient way to synthesize block
copolymers when making use of an allyl ether end-group
functionality. A more reactive end group, such as norbornene,
can increase the coupling efficiency, as was recently addressed
It has to be stressed that also for this polymer−polymer
system diffusional limitations are only important on termi-
nation reactions, which lead to increased values of the coupling
efficiency. Since the transfer and addition reactions are even
more chemically controlled, the reduced mobility when shifting
from the polymer−small to the polymer−polymer system
cannot explain the limited coupling efficiency. On the contrary,
the low initial applied concentrations of the thiol and ene
polymeric species due to their limited solubility, and the
reduced reactivity toward transfer compared to the vinyl ether
system, are the actual reasons for the limited coupling
efficiency. Both factors result in a decrease of the addition
and transfer rates and, as highlighted above, lead to an
increased selectivity for side reactions. This is further illustrated
in Figure 11d, which shows the simulated coupling efficiency as
a function of time for the conditions mentioned above. A final
coupling efficiency of 0.3 is obtained, which is close to the
experimental one of 0.25. Although the same initial conditions
have been applied as for the simulations in Figure 8e (dotted
line) for the small−small system (BT/DVE/DMPA), the
significant reduction of the coupling efficiency from 0.9 (Figure
8e, dotted line) to 0.3 (Figure 11d) is solely attributed to the
above indicated decrease in reactivity toward addition and
transfer.
Furthermore, Figure 11a clearly indicates that the low
coupling efficiency cannot be completely attributed to side
reactions, since no full conversion of the ene and thiol (full
line) is reached during the time span of complete
decomposition of the photoinitiator (dotted line) for the
considered condition. It should be stressed that both the
experimentally determined coupling efficiency via 1H NMR and
the simulated efficiencies in Figure 11d−f correspond to the
definition in eq 1, implying that an incomplete conversion of
the ene or thiol would also lead to a reduction of the coupling
efficiency.
I
dx.doi.org/10.1021/ma302619k | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
products as a function of time for polymer−polymer
conjugation. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: MarieFrancoise.Reyniers@UGent.be (M.-F.R.); Filip.
Duprez@UGent.be (F.E.D.).
■
Notes
The authors declare no competing financial interest.
Figure 12. Equimolar coupling of thiol-functionalized polystyrene
(PS-SH) and allyl ether-functionalized poly(vinyl acetate) PVAc; [PS-
SH]0 = [PVAc]0 = 0.03 mol L−1. (a) Concentration of the initiator
(dashed line) and ene compound (full line) versus time. (b) Coupling
efficiency versus time; [initiator]0 = 0.003 mol L−1; stepwise addition
of initiator every 400 s.
ACKNOWLEDGMENTS
■
The Institute for the Promotion of Innovation through Science
and Technology in Flanders (IWT Vlaanderen), the Fund for
Scientific Research Flanders (FWO), the Long Term Structural
Methusalem Funding by the Flemish Government, and the
Interuniversity Attraction Poles Programme initiated by the
Belgian State, the Prime Minister’s office (IAP/IUAP/PAI P7/
05: “Functional Supramolecular Systems”) and the European
Science Foundation−Precision Polymer Materials (P2M)
program are acknowledged for financial support.
by Stamenovic et al.,48 although side reactions will still be
inevitable.
CONCLUSIONS
■
Kinetic modeling is used to understand the reduced coupling
efficiency for the functionalization reaction of PS-SH with DVE
and the PS-b-PVAc polymer−polymer conjugation, photo-
initiated by DMPA. The simulations show that the reduced
coupling efficiencies cannot be attributed to a diffusion-
controlled addition or transfer reaction but are caused by an
increased importance of in particular termination by recombi-
nation reactions and addition of the DMPA derived radicals to
the ene compound, as previously proposed by Koo et al.19
This study also revealed that even for small−small systems
the coupling efficiency crucially depends on the choice of the
reaction conditions. Too high initial concentrations of
photoinitiator and too low initial ene/thiol concentrations
enhance the occurrence of side reactions. The limited coupling
efficiency for polymer−polymer thiol−ene conjugation can
hence be traced back to the limited solubility of polymeric
species and the resulting low end-group concentrations.
In principle, a higher coupling efficiency can be achieved by
stepwise addition of DMPA, as full conversion of the thiol and
ene compound can be reached while maintaining a low DMPA
concentration. However, the obtained coupling product yields
are still far away from the desired 100%, confirming that radical
thiol−ene chemistry has to be excluded as a true member of the
family of “click” chemistry techniques for polymer−polymer
conjugation.
NOMENCLATURE
■
Cu(I) copper atom, oxidation state +I
I1I2 photoinitiator which dissociates upon UV-radiation
yielding the radical fragments I1 and I2
RSH general thiol molecule
RS general thiyl radical
R1,2 low or high molar mass compound
[thiol]0 initial thiol concentration (mol L−1)
[ene]0 initial concentration of ene compound (mol L−1)
[I1I2] concentration of photoinitiator (mol L−1)
f photoinitiator efficiency
h the Planck constant (J s)
NA the Avogadro constant (mol−1)
fcoupling coupling efficiency (−)
I irradiation intensity (mW cm−2 s−1)
c speed of light (m s−1)
Greek Symbols
λ wavelength (nm)
ν frequency (s−1)
ε molar absorptivity (L mol−1 s−1)
α fitting parameter (−)
Abbreviations
UV ultraviolet light
DVE dodecyl vinyl ether
DMPA 2,2-dimethoxy-2-phenylacetophenone
BT benzyl thiol
PS polystyrene
PS-SH thiol-functionalized polystyrene
PVAc poly(vinyl acetate)
ASSOCIATED CONTENT
* Supporting Information
■
S
Determination of the molar absorptivity of DMPA in 1,4-
dioxane at 365 nm; recorded intensity versus time profile of the
UV-lamps used; experimental procedure, GC(/MS) analysis
results, and determination of kinetic parameters for the UV-
induced dissociation of benzyl disulfide in the presence of
DMPA in 1,4-dioxane; model assumptions and a complete list
of all the reactions considered in the kinetic model together
with the values of the corresponding rate coefficients;
calculation of the equilibrium coefficients for the addition/β
C−S-scission via Benson’s group additivity method, including a
list of the literature-based group additive values necessary for
the calculation; evaluation of diffusional limitations for both the
addition/transfer and the termination reactions in case
macromolecular species are involved; plot of the main side
DD n-dodecane
CRP controlled radical polymerization
RAFT reversible addition−fragmentation chain transfer
NMR nuclear magnetic resonance
SEC size exclusion chromatography
FTIR Fourier transform infrared
GC gas chromatography
MS mass spectroscopy
EPR electron paramagnetic resonance
MMD molar mass distribution
THF tetrahydrofuran
J
dx.doi.org/10.1021/ma302619k | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(24) Cramer, N. B.; Reddy, S. K.; O’Brien, A. K.; Bowman, C. N.
Macromolecules 2003, 36, 7964−7969.
BDS benzyl disulfide
rpm rotations per minute
vol % volume percentage (−)
LSODA Livermore Solver for Ordinary Differential Equa-
tions
PDI polydispersity index
CLD chain length dependent
CLI chain length independent
(25) Northrop, B. H.; Coffey, R. N. J. Am. Chem. Soc. 2012, 134,
13804−13817.
(26) Reddy, S. K.; Cramer, N. B.; Bowman, C. N. Macromolecules
2006, 39, 3673−3680.
(27) Moad, G.; Rizzardo, E.; Thang, S. H. Polymer 2008, 49, 1079−
1131.
(28) Chiefari, J.; Chong, Y. K.; Ercole, F.; Krstina, J.; Jeffery, J.; Le, T.
P. T.; Mayadunne, R. T. A.; Meijs, G. F.; Moad, C. L.; Moad, G.;
Rizzardo, E.; Thang, S. H. Macromolecules 1998, 31, 5559−5562.
(29) Roth, P. J.; Boyer, C.; Lowe, A. B.; Davis, T. P. Macromol. Rapid
Commun. 2011, 32, 1123−1143.
(30) Barner-Kowollik, C.; Du Prez, F. E.; Espeel, P.; Hawker, C. J.;
Junkers, T.; Schlaad, H.; Van Camp, W. Angew. Chem., Int. Ed. 2011,
50, 60−62.
(31) Preuss, C. M.; Barner-Kowollik, C. Macromol. Theory Simul.
2011, 20, 700−708.
(32) Poling, B. E.; Prausnitz, J. M.; O’Connell, J. P. The Properties of
Gases and Liquids; McGraw-Hill: New York, 2001.
(33) Colley, C. S.; Grills, D. C.; Besley, N. A.; Jockusch, S.;
Matousek, P.; Parker, A. W.; Towrie, M.; Turro, N. J.; Gill, P. M. W.;
George, M. W. J. Am. Chem. Soc. 2002, 124, 14952−14958.
(34) Hristova, D.; Gatlik, I.; Rist, G.; Dietliker, K.; Wolf, J. P.;
Birbaum, J. L.; Savitsky, A.; Mobius, K.; Gescheidt, G. Macromolecules
2005, 38, 7714−7720.
Subscripts
sol solvent
I regarding initiator (derived) species
chem intrinsic chemical (contribution)
add addition
tr transfer
tc termination by recombination
Superscripts
i, j chain length
macro macromolecular
REFERENCES
■
(1) Binder, W. H.; Sachsenhofer, R. Macromol. Rapid Commun. 2007,
28, 15−54.
(2) Binder, W. H.; Sachsenhofer, R. Macromol. Rapid Commun. 2008,
29, 952−981.
(35) Sayamol, K.; Knight, A. R. Can. J. Chem. 1968, 46, 999−1003.
(36) Fischer, H.; Paul, H. Acc. Chem. Res. 1987, 20, 200−206.
(37) Smith, G. B.; Russell, G. T.; Heuts, J. P. A. Macromol. Theory
Simul. 2003, 12, 299−314.
(3) Moses, J. E.; Moorhouse, A. D. Chem. Soc. Rev. 2007, 36, 1249−
1262.
(4) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed.
2001, 40, 2004−2021.
(38) Barner-Kowollik, C.; Russell, G. T. Prog. Polym. Sci. 2009, 34,
1211−1259.
(5) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B.
Angew. Chem., Int. Ed. 2002, 41, 2596−2599.
(6) Becer, C. R.; Hoogenboom, R.; Schubert, U. S. Angew. Chem., Int.
Ed. 2009, 48, 4900−4908.
(39) Gilbert, R. G. Pure Appl. Chem. 1992, 64, 1563−1567.
(40) D’Hooge, D. R.; Reyniers, M.-F.; Marin, G. B. Macromol. React.
Eng. 2009, 3, 185−209.
(41) D’Hooge, D. R.; Reyniers, M.-F.; Stadler, F. J.; Dervaux, B.;
Bailly, C.; Du Prez, F. E.; Marin, G. B. Macromolecules 2010, 43,
8766−8781.
(7) Hansell, C. F. E., P.; Stamenovic, M. M.; Barker, I. A.; Dove, A.
P.; Du Prez, F. E.; O’Reilly, R. K. J. Am. Chem. Soc. 2011, 133, 13828−
13831.
(42) De Roo, T.; Wieme, J.; Heynderickx, G. J.; Marin, G. B. Polymer
2005, 46, 8340−8354.
(8) Tasdelen, M. A. Polym. Chem. 2011, 2, 2133−2145.
(9) Hizal, G.; Tunca, U.; Sanyal, A. J. Polym. Sci., Part A: Polym.
Chem. 2011, 49, 4103−4120.
(43) Fischer, H.; Radom, L. Angew. Chem., Int. Ed. 2001, 40, 1340−
1371.
(10) Hoyle, C. E.; Bowman, C. N. Angew. Chem., Int. Ed. 2010, 49,
1540−1573.
(44) Heuts, J. P. A.; Russell, G. T. Eur. Polym. J. 2006, 42, 3−20.
(45) Heuts, J. P. A.; Russell, G. T.; Smith, G. B. Aust. J. Chem. 2007,
60, 754−764.
(11) Hoyle, C. E.; Lee, T. Y.; Roper, T. J. Polym. Sci., Part A: Polym.
Chem. 2004, 42, 5301−5338.
(46) Gridnev, A. A.; Ittel, S. D. Macromolecules 1996, 29, 5864−5874.
(47) Heuts, J. P. A. In Handbook of Radical Polymerization;
Matyjaszewski, K., Davis, T. P, Eds.; Wiley and Sons: New York,
2002; pp 1−76.
(12) Kade, M. J.; Burke, D. J.; Hawker, C. J. J. Polym. Sci., Part A:
Polym. Chem. 2010, 48, 743−750.
(13) Hoyle, C. E.; Lowe, A. B.; Bowman, C. N. Chem. Soc. Rev. 2010,
39, 1355−1387.
(48) Stamenovic, M. M.; Espeel, P.; Van Camp, W.; Du Prez, F. E.
Macromolecules 2011, 44, 5619−5630.
(14) Inglis, A. J.; Barner-Kowollik, C. Macromol. Rapid Commun.
2010, 31, 1247−1266.
(15) Back, R. M., C.; Sivertz, C. Can. J. Chem. 1954, 32, 1078−1091.
(16) Kharasch, M. N., W.; Mantell, G. J. Org. Chem. 1951, 16, 524−
532.
(17) Uygun, M.; Tasdelen, M. A.; Yagci, Y. Macromol. Chem. Phys.
2010, 211, 103−110.
(18) ten Brummelhuis, N.; Diehl, C.; Schlaad, H. Macromolecules
2008, 41, 9946−9947.
(19) Koo, S. P. S.; Stamenovic, M. M.; Prasath, R. A.; Inglis, A. J.; Du
Prez, F. E.; Barner-Kowollik, C.; Van Camp, W.; Junkers, T. J. Polym.
Sci., Part A: Polym. Chem. 2010, 48, 1699−1713.
(20) Cramer, N. B.; Reddy, S. K.; Cole, M.; Hoyle, C.; Bowman, C.
N. J. Polym. Sci., Part A: Polym. Chem. 2004, 42, 5817−5826.
(21) Cramer, N. B.; Scott, J. P.; Bowman, C. N. Macromolecules 2002,
35, 5361−5365.
(22) Reddy, S. K.; Cramer, N. B.; Bowman, C. N. Macromolecules
2006, 39, 3681−3687.
(23) Cramer, N. B.; Davies, T.; O’Brien, A. K.; Bowman, C. N.
Macromolecules 2003, 36, 4631−4636.
K
dx.doi.org/10.1021/ma302619k | Macromolecules XXXX, XXX, XXX−XXX