Visible-Light Photocatalytic Oxidation of DMSO for RAFT Polymerization?

Article abstract of DOI:10.1111/php.13468

The solvent is an important, yet often forgotten part of a reaction mechanism. Many photochemical polymerizations are carried out using dimethyl sulfoxide (DMSO) as a way to promote the solubility of both the reactants and products, but its reactivity is rarely considered when initiation mechanisms are proposed. Herein, the oxidation of DMSO by an excited-state quinone is used to form initiating radicals resulting in the polymerization of methacrylate monomers, and the polymerization can be controlled with the addition of a chain transfer agent. This process leads to the formation of polymers with narrow molecular weight distribution, and the polymerization is able to be carried out in the presence of oxygen. A visible light absorbing substituted anthraquinone is synthesized, and nanosecond transient absorption spectroscopy is used to monitor the intermediates involved in the initiation mechanism. Photoproduct analysis indicates formation of methyl radicals as a result of DMSO oxidation. Furthermore, we show that the solvent outcompetes the chain transfer agent for interacting with the excited-state anthraquinone. These observations have a broad impact on photoinduced polymerizations performed in DMSO as many photocatalysts are strong oxidants in the excited state and are capable of reacting with the solvent. Therefore, the role of the solvent needs to be more carefully considered when proposing mechanisms for photoinduced polymerizations in DMSO.

Full text of DOI:10.1111/php.13468

DR. DANIEL E. FALVEY (Orcid ID : 0000-0002-2185-502X)
Article type
: Special Issue Research Article
SPECIAL ISSUE RESEARCH ARTICLE
Visible-Light Photocatalytic Oxidation of DMSO for RAFT
Polymerization†
Matthew D. Thum, Donald Hong, Andrea N. Zeppuhar, and Daniel E. Falvey*
University of Maryland, College Park
Department of Chemistry and Biochemistry, MD (USA) 20942
*
Corresponding author e-mail: Falvey@umd.edu (Daniel E. Falvey)
†
This article is part of a Special Issue celebrating the career of Dr. Edward Clennan
This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process, which may lead to
differences between this version and the Version of Record. Please cite this article as doi:
1
0.1111/PHP.13468
This article is protected by copyright. All rights reserved

ABSTRACT
The solvent is an important, yet often forgotten part of a reaction mechanism. Many photochemical
polymerizations are carried out using dimethyl sulfoxide (DMSO) as a way to promote the solubility
of both the reactants and products, but its reactivity is rarely considered when initiation mechanisms
are proposed. Herein, the oxidation of DMSO by an excited state quinone is used to form initiating
radicals resulting in the polymerization of methacrylate monomers, and the polymerization can be
controlled with the addition of a chain transfer agent. This process leads to the formation of polymers
with narrow molecular weight distribution and the polymerization is able to be carried out in the
presence of oxygen. A visible light absorbing substituted anthraquinone is synthesized and
nanosecond transient absorption spectroscopy is used to monitor the intermediates involved in the
initiation mechanism. Photoproduct analysis indicates formation of methyl radicals as a result of
DMSO oxidation. Furthermore, we show that the solvent outcompetes the chain transfer agent for
interacting with the excited-state anthraquinone. These observations have a broad impact on
photoinduced polymerizations performed in DMSO as many photocatalysts are strong oxidants in the
excited state and are capable of reacting with the solvent. Therefore, the role of the solvent needs to
be more carefully considered when proposing mechanisms for photoinduced polymerizations in
DMSO.
This article is protected by copyright. All rights reserved

INTRODUCTION
Photomediated controlled radical polymerization is a popular method for preparing well-defined
polymers due to its mild reaction conditions and the ability to provide the overall process of the
reaction with spatial and temporal specificity. The ability of light to switch the polymerization process
“
on” and “off” is advantageous in the manufacturing of polymers with exact control over the
molecular properties of the final product. One area in which this approach has been adopted
wholeheartedly is reversible addition-fragmentation chain transfer (RAFT) polymerization. Light has
been used multiple ways to initiate RAFT polymerization.1–6 Traditionally, RAFT polymerization is
carried out under direct UV irradiation via a photoiniferter method. In this case, a dithioester, or
related species, is directly irradiated which results in C-S homolysis and allows the RAFT agent to
serve as both chain transfer agent (CTA) and photoinitiator.7–14 Since these processes require only
monomer and CTA, photoiniferter methods are valued for their simplicity. However,
photodissociation of the CTA generally requires UV wavelengths which can result in unwanted
degradation during photolysis, reducing end group fidelity and limiting high conversions.13–15
Furthermore, photoiniferter processes are usually incompatible with molecular oxygen, limiting their
potential applications.
An alternative approach, described by Hawker, Boyer, and others, uses photosensitizers (or
photoredox catalysts) to trigger the initiation step.16–20 Photocatalysts are strong excited state oxidants
or reductants which are designed to activate CTAs through either excited-state energy transfer, or
mediated one-electron transfer processes. The general photocatalytic RAFT scheme assumes that the
photocatalyst generates radicals through interaction with the CTA, either via direct reduction, or via
homolysis due to energy transfer.21–23 In the photoinduced electron transfer (PET) mechanism, the
excited state photocatalyst reacts either with the CTA directly, resulting in reduction of the CTA and
subsequent radical formation, or acts a mediator by first reacting with a sacrificial electron donor,
such as a tertiary amine, followed by reduction of the CTA. Photocatalytic RAFT is valued because it
2
0
can provide access to a broad spectrum of irradiation wavelengths. For example, the Boyer group
demonstrated the use of visible light absorbing dyes such as Eosin Y and chlorophyll A in the
This article is protected by copyright. All rights reserved

polymerization of methyl acrylate monomers and the Fors group described controlled cationic
polymerization of vinyl ethers initiated by the photoreduction of dithiolates by oxidizing pyrilium
salts.24–26 The photocatalytic approach is attractive as it generally results in less undesired degradation
of the products, however, these systems are more complex, often requiring photocatalyst and ground
state donor in addition to monomer and CTA.
Recently, Corrigan et al. developed a method in which metal phthalocyanines, activated by
2
7
NIR irradiation, were used in the controlled polymerization of acrylate monomers. Interestingly, the
initiation step did not involve photoreduction of the RAFT agent, rather, the excited photocatalyst
reacts with the solvent, N-methyl-2-pyrrolidone, to generate initiating radicals. This approach
eliminates the need for an additional ground-state donor or cosensitizer.
Dimethylsulfoxide (DMSO) is a widely used solvent for polymerization reactions in general,
and photo-RAFT polymerization in particular. While it is normally believed to function as an inert
medium, some reports show that DMSO can produce radicals under highly oxidizing conditions
2
8
(
Scheme 1), and in the presence of oxygen and a strong base. The oxidation potential for DMSO has
2
9
been estimated to be of +1.56 to +1.76 V (vs SCE). Furthermore, on the basis of pulse radiolysis
kinetics, it was inferred that one electron oxidation DMSO in the presence of water leads to formation
of sulfinic acid along with highly reactive methyl radicals.30–32 This was later adapted for the free
radical methylation of 2’-deoxyguanosine.33
<
Scheme 1>
The study described herein was undertaken to determine if a similar DMSO oxidation reaction would
be present in a photo-RAFT process, and how that would affect the initiation mechanism(s). Excited-
state quinones have been used extensively as excited state oxidants, as well the source for mediated
electron transfer from a donor to an acceptor.34–40 Most notably, the triplet state of most quinones is
formed with high efficiency and photoreduction via hydrogen atom abstraction, or electron transfer
3
7
are common. Of particular relevance is an earlier laser flash photolysis study by Görner which
established that triplet state 9,10-anthraquinone derivatives react rapidly with DMSO through one
4
1
electron transfer. For this reason, we examined a visible light absorbing anthraquinone derivative
This article is protected by copyright. All rights reserved

(AQ-1, Fig. 1) and investigated its ability to promote photocatalytic RAFT through radical generation
using the solvent, DMSO. Product analysis, laser flash photolysis (LFP), and chemical trapping
experiments demonstrate that this approach provides visible light driven controlled radical
polymerization through the interaction of the excited-state photosensitizer with the solvent which
outcompetes reactions with the RAFT agent contrary to more conventional photo-RAFT processes.
<
Figure 1>
MATERIALS AND METHODS
1
13
General Procedures: All H and C NMR were obtained on a Bruker 400 MHz spectrometer.
Chemical shifts () are reported in parts per million (ppm). Ultraviolet-visible (UV/Vis) spectra were
recorded on a Shimadzu UV-1800 spectrometer using UVProbe 2.43 software. Samples were scanned
from 200 to 600 nm using a fast scanning speed and a sampling interval of 1.0 nm. Mass spectra data was
acquired using a JEOL AccuTOF-CS using ESI-TOF. All synthesis and characterization can be found in
the Supplemental information.
Laser Flash Photolysis: Laser flash photolysis experiments were performed using a Nd:YAG laser
pump beam source with a Continuum Surelite II-10 capable of 266, 532, or 355 nm pulses between 4-6
ns. A LeCroy 350 MHz digital oscilloscope was used to observe all waveforms. Samples were prepared
so that their relative absorbance was between 0.75 and 1.5 at the excitation wavelength, 355 nm. The
probe beam was a 350 W Xenon arc lamp. The samples were loaded into a 1 cm quartz cuvette and N2
was purged for 10 minutes in the solution and an additional 5 minutes in the headspace. Samples were
flowed continuously thorough out the experiment to avoid buildup of photoproducts.
Polymerization Procedure: In a 3.5 dr vial the appropriate amount of AQ-1 in DMSO and 13.1 mg
(46.9 mmol) of CPADB was added and the solution was sonicated in the absence of light until everything
dissolved. Methyl methacrylate (MMA), 1 mL, was added and the mixture was thoroughly mixed. If
polymerizing in the presence of air, the cap was sealed and the vial was photolyzed in the Rayonet
photoreactor (419 nm broadband irradiation) for the appropriate amount of time. If polymerizing under an
inert atmosphere, the vial was sealed with a septum and purged with N for 10 minutes in the solution and
2
This article is protected by copyright. All rights reserved

an additional 5 minutes in the headspace. Unless otherwise noted, the conversion was measured by taking
1
a small ( 10 µL) amount of the mixture and adding it to 0.5 mL of CD CN for H NMR. The amount of
3
monomer loss was measured using the signal of DMSO, 2.5 ppm, as the internal standard and comparing
it to the loss of signal at 3.7 ppm corresponding to MMA.
RESULTS AND DISCUSSION
3
4
The photocatalyst (AQ-1) which was synthesized according to previously published methods, has an
absorption tail that extends out to 440 nm allowing for the polymerizations to be carried out with
visible light (Figure S2). Polymerization reactions were performed with 4-cyanopentanoic acid
dithiobenzoate (CPADB) as a chain transfer agent due to its photolytic stability under visible light
irradiation, and AQ-1 in DMSO was used as the photoinitiator.
Polymerization
Unless otherwise noted, polymerizations were carried out at 419 nm under broadband irradiation in a
3
5 W, 16 bulb Rayonet photoreactor with an operating temperature of 35°C. Photolysis of AQ-1 in a
mixture of DMSO/MMA resulted in polymer formation (Table 1, Entry 1). This demonstrates that
photolysis of AQ-1 in DMSO results in the formation of initiating radicals in the absence of a co-
sensitizer. The polymerization process was able to be controlled with the addition of CPADB
resulting in low polydispersity and good agreement between predicted and measured molecular
weights (Table 1, Entries 4-12.). The presence of the dithiobenzoate end group was confirmed by 1H
NMR of the products, along with detection of its characteristic UV/vis absorption at 300 nm (Figures
S9, S14). Although purging the solution with nitrogen prior to photolysis resulted in slightly higher
conversions than those carried out in air, the effect was minimal (Table S1).
<
Table 1>
To assess the feasibility of the photocatalyst in low concentrations, polymerizations were
carried out using sub-mM amounts of AQ-1 (Table 1, Entries 5-10. Fig. 2). Encouragingly, all
experiments showed an increase in conversion and molecular weight relative to background
This article is protected by copyright. All rights reserved

polymerization while maintaining control over the polymerization process. Examples where the
solution was left unexposed to light resulted in no conversion, and although control experiments
confirm the polymerization to be photocatalytically initiated, some background photolysis of the CTA
is observed (Table 1, Entries 2-3). One possible explanation for the observed background
polymerization is due to contributions from a photoiniferter mechanism involving the UV edge of the
broadband emission of 419 nm bulbs (Figure S2).
<
Figure 2>
Increasing photolysis time shows a general increase in conversion and molecular weight of the
formed polymers, but also leads to larger disagreements between observed and theoretical molecular
weights (Table 1, Entries 9-10). This suggests that some degradation of the end group occurs with
prolonged photolysis which is consistent with both the background polymerization experiments and
has been observed previously using UV irradiation.13, 14
The polymerization process was applied to multiple methacrylate monomers (Table S2). Well-
defined polymers with good agreement between predicted and measured molecular weights were
achieved for both MAA and GMA, but DMAEMA showed significant degradation of the CTA and,
consequently, poor control over the molecular weight. As shown in Fig. 1, DMAEMA has an alkyl
amino substituent at one of the terminal positions. Anthraquinones are well known to act as excited
state oxidants of alkyl amines and substituted amines are commonly used as sacrificial electron
3
7
donors in photoinduced electron transfer process. The reaction of the excited state AQ-1 with
DMAEMA likely results in unwanted radical formation and redox process which, in turn, may be
responsible for the poor polymer properties.
Figure 3 illustrates ‘On’ / ‘Off’ experiments that were performed to demonstrate the temporal
control over the reaction process. In the absence of light, no conversion is observed, while during
photolysis, a noticeable increase in the conversion of monomer is seen. Additionally, a polymerization
was carried out in which a sample was photolyzed for a short period of time followed by an extended
incubation period in the dark (Table S2, Entry 2). After the incubation period, low amounts of
conversion were measured which indicates that radical species formed are quickly quenched and do
This article is protected by copyright. All rights reserved

not propagate extensively in the absence of light. This further demonstrates the necessity of light for
continuous formation of initiating radicals and rules out “dark” polymerization that have been
reported previously.42
<
Figure 3>
Photopolymerization reactions were also carried out in different solvents to assess the viability
of the proposed mechanism and examine the versatility of this approach. Polymerization was
observed irrespective of the solvent used, but with DMSO having the highest conversion and Mn
while still maintaining a controlled process (Table 2, Entry 1). Other solvents studied all showed
much lower conversion suggesting that AQ-1 is unable to efficiently generate initiating radicals and
the polymerization is dominated by an iniferter mechanism. Interestingly, methanol, which is known
to be a weak H-atom donor and has been shown to serve as a H-atom donor in the photochemical
production of semiquinones, showed very poor control over the polymerization process and very low
conversion which indicates that any semiquinone formed during photolysis is not beneficial to the
polymerization process (Table 2, Entry 3).
<
Table 2>
Mechanistic Analysis
The feasibility of excited-state electron transfer can be estimated using the Rehm-Weller equation
3
and, using this relationship, electron transfer from DMSO to AQ-1 * is expected to be exergonic by
-1
ca. 12 kcal•mol (Supplemental Information, Eq 1-3). To support this, pulsed photolysis (354.7 nm,
0 mJ, 5-7 ns) of AQ-1 in DMSO was performed (Fig. 4). Under an atmosphere of either N or air, an
1
2
intermediate is formed with absorption bands at 490 and 620 nm and a lifetime of ca. 0.5 µs. This
3
early species is assigned to AQ-1 * on the basis of its resemblance to triplet-state spectra of
structurally similar anthraquinone derivatives, reactivity towards molecular oxygen (Figure S3), and
similarity to the spectra from pulsed photolysis of AQ-1 in a non-reactive solvent, acetonitrile (Figure
3
7
S4). Following the decay of the triplet, a longer-lived species is formed with a peak at 550 nm after
ca.1 µs which is quenched by air (Figure S5). This longer-lived signal is assigned to the anion radical
This article is protected by copyright. All rights reserved


•
37
(
AQ-1 ) formed after the triplet oxidation of DMSO. This signal is consistent with the anion radical
formed from pulsed photolysis of AQ-1 in acetonitrile with 1,4-diazabicyclo[2.2.2]octane (DABCO),
a ground state electron donor (Fig. 4C > 700 ns). The results from pulsed photolysis demonstrate that

•
photolysis of AQ-1 in DMSO results in the formation AQ-1 from one-electron oxidation of DMSO
as predicted by the Rehm-Weller equation.
<
Figure 4>
One commonly invoked initiation mechanism in photocatalytic RAFT reactions is electron
transfer from the reduced photocatalyst to the CTA. The effect of CPADB on AQ-1 • was explored.
The latter was generated by pulsed photolysis of AQ-1 in methanol using triethylamine (TEA) as an
electron donor (Figure S6). Kinetic traces showing the decay of AQ-1 in the presence of CPADB
are provided in Fig. 5. No noticeable change in the decay rate for the AQ-1 is observed, however,
the magnitude of the initial signal decreases with increasing concentrations of CPADB (Fig. 5).
•
•
3
Stern–Völmer analysis (Fig. 6) demonstrates that CPADB quenches AQ-1 * at the diffusion limit by

•
averaging the signal corresponding to AQ-1 over 1 µs (blue area, Fig. 5) and plotting the change in
signal with respect to the concentration of quencher. A linear relationship was observed which is
3
indicative of energy transfer from AQ-1 * to CPADB which agrees with previous work in a non-
3
9 -1 21
reactive solvent demonstrating AQ-1 * energy transfer to CPADB with a rate of 3.4x10 s .
<
<
Figure 5>
Figure 6>
The pulsed photolysis experiments help to explain the mechanism of initiation. Namely, AQ-
13
* oxidizes DMSO to form AQ-1 and the corresponding DMSO cation radical. Although CPADB
 •
3
quenches AQ-1 * as well, significant contribution to the polymerization process of a PET-Energy
Transfer-RAFT pathway can be ruled out given the high concentration of DMSO, concentration of air
present in the polymerization procedure with respect to the CTA, and previous work demonstrating
that triplet energy transfer to the first excited triplet of CPADB does not result in significant C-S bond
2
1
 •
dissociation. Additionally, the lifetime of AQ-1 is independent of the amount of CPADB present
This article is protected by copyright. All rights reserved

which shows that the reduced quinone is not responsible for the activation of the chain transfer agent
leading to polymerization. Thus, we can exclude contribution due to a PET-RAFT process in which

•
AQ-1 acts to reduce the CTA.
<
Scheme 2>
Chemical trapping experiments support the formation of methyl radicals. As shown in Scheme
, the methyl radical is expected to react with CPADB forming the methyl-adduct, and an alkyl
2
radical. For these experiments AQ-1 was photolyzed in DMSO in the presence of CPADB, in the
absence of monomer and the photoproduct was analyzed by MS. After photolysis, a mass
corresponding to the methyl ester was found which matched an independently synthesized standard
(Figures S17, 18).
In addition to generating methyl radicals, the proposed mechanism also suggests formation of
sulfinic acid. This is supported by photolysis of AQ-1 in DMSO followed by detection of acid using
malachite green carbinol base as a pH indicator (Figures S7, 8). With increasing photolysis time, the
yield of acid increased accordingly. Control experiments conducted in the absence of AQ-1 or in the
absence of irradiation resulted in no acid detection, providing further mechanistic support.
Interestingly, sulfinic acid may also be a source of reactive alkyl radicals resulting from oxidation, but
this possibility was not investigated in the current study.43
<
Scheme 3>
The proposed mechanism of contributions to the initiation of polymerization from photo
oxidation of DMSO is shown in Scheme 3. The photosensitizer, AQ-1, absorbs a photon and, after
intersystem crossing, forms its triplet state. The latter oxidizes DMSO leading to the formation of the
anthraquinone anion radical, and a DMSO cation radical. The cation radical is trapped by latent water
to yield sufinic acid and a methyl radical. The methyl radical can combine with monomer to initiate
polymerization and enter into the RAFT process. In the presence of air, AQ-1 • can react with
molecular oxygen to regenerate AQ-1.
CONCLUSION
This article is protected by copyright. All rights reserved

The solvent plays a crucial role in any reaction mechanism. This study demonstrates that
photochemical oxidation of DMSO can be used to generate radicals and prepare methacrylate
polymers with high conversions and a narrow PDI. Through the use of a suitably substituted
anthraquinone derivative, these reactions can be carried out with visible light and in the presence of
air. Pulsed photolysis results and chemical trapping experiments support contributions to the initiation
proposed in Scheme 3 wherein the DMSO cation radical reacts with water to produce sulfinic acid
along with reactive methyl radicals. Given that DMSO is a widely used solvent for photocatalytic
RAFT polymerizations, and many of the photocatalysts are highly effect excited-state oxidants, the
role of the solvent must be more carefully considered when discussing the mechanisms involved in
photo-RAFT processes.
ACKNOWLEDGEMENTS: This work has been supported in part by the National Science
Foundation (CHE-1112018).
SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the
end of the article:
Figure S1. Synthesis of methyl benzodithioate.
Figure S2. Normalized absorption spectra of AQ-1 and CPADB in MeCN. Normalized emission
spectrum of the 419 nm bulbs from the Rayonet photoreactor.
Figure S3. Left: Waveforms from pulsed photolysis of AQ-1 in DMSO 640 nm. The lifetime of the
3
signal corresponding to AQ-1 * is affected little by the presence of air suggesting efficient quenching
3
of AQ-1 * by DMSO. Right: Waveforms from pulsed photolysis of AQ-1 in DMSO 550 nm. The
-•
lifetime of the signal corresponding to AQ-1 , on the other hand, is significantly decreased in air due
-•
to quenching of AQ-1 by molecular oxygen. On the other hand, at early times (< 0.5 s), where the
3
signal for AQ-1 * dominates, little effect is seen.
This article is protected by copyright. All rights reserved

Figure S4. Left: Transient absorption spectrum of AQ-1 in MeCN under N . The peaks corresponding
2
3
3
to AQ-1 * are seen at 490 and 630 nm. Right: Waveforms showing quenching of AQ-1 * in air
orange) at 490 nm.
(
Figure S5. Waveforms from transient absorption spectrum of AQ-1 in MeCN with 2 mM DABCO
under N at 550 nm. Waveforms show the decay of the anion radical of AQ-1 in the presence of air
2
(orange).
Figure S6. Left: Transient absorption spectrum of AQ-1 in MeOH with 25 mM TEA under N . The
2
sample was continuously flowed to avoid the buildup of photoproducts. A very long-lived peak at 490
-•
nm is assigned to AQ-1 . Also notable is the semiquinone which can be seen at 410 nm. Right:
-•
Waveforms illustrating the quenching of AQ-1 by molecular oxygen.
Figure S7. Monitoring formation of acid over time from photolysis of AQ-1 in DMSO using
malachite green carbinol base as a pH indicator. Blue: 370 nm irradiation of AQ-1 in DMSO; Orange:
3
70 nm irradiation of DMSO in the absence of AQ-1; Gray: AQ-1 in DMSO without irradiation.
Figure S8. Monitoring formation of acid over time from photolysis of AQ-1 in DMSO using
malachite green carbinol base as a pH indicator. Acid formation is detected by the growth of the band
at 633 nm.
Table S1. RAFT Polymerization of MMA with AQ-1 in DMSO under Nitrogen
Table S2. Polymerization scope and limitations in Air
Figure S9. UV/Vis spectra of Table 1, entry 11.
Figure 10. UV/Vis spectra of Table 1, entry 4.
1
Figure S11. H NMR of Table 1, entry 11.
1
Figure S12. H NMR of Table 1, entry 4.
1
Figure S13. H NMR of Table S2, entry 3.
1
Figure S14. H NMR of Table S2, entry 4.
1
Figure S15. H NMR of methyl benzodithioate.
1
3
Figure S16. C NMR of methyl benzodithioate.
Figure S17. Mass Spec (DART-) of methyl benzodithioate in CD CN.
3
Figure S18. Mass Spec (DART-) showing formation of methyl benzodithioate from photolysis of
AQ-1 and CPADB in DMSO. The peaks corresponding to the photoproduct are shown starting at M/Z
This article is protected by copyright. All rights reserved

1
68.8652.
REFERENCES
(1)
Hill, M. R., Carmean, R. N., Sumerlin, B. S. (2015) Expanding the Scope of RAFT
Polymerization: Recent Advances and New Horizons. Macromolecules 48, 5459–5469.
(2)
Chen, M., Zhong, M., Johnson, J. A. (2016) Light-Controlled Radical Polymerization:
Mechanisms, Methods, and Applications. Chem. Rev., 116, 10167–10211.
(3)
Pan, X., Tasdelen, M. A., Laun, J., Junkers, T., Yagci, Y., Matyjaszewski, K. (2016)
Photomediated Controlled Radical Polymerization. Prog. Polym. Sci. 62, 73–125.
(4)
McKenzie, T. G., Fu, Q., Uchiyama, M., Satoh, K., Xu, J., Boyer, C., Kamigaito, M., Qiao, G.
G. (2016) Beyond Traditional RAFT: Alternative Activation of Thiocarbonylthio Compounds for
Controlled Polymerization. Adv. Sci. 3, 1500394.
This article is protected by copyright. All rights reserved

(
5)
Reversible Deactivations through Electron and Energy Transfer. Macromol. Rapid Commun. 38,
700143.
Shanmugam, S., Xu, J., Boyer, C. (2017) Photocontrolled Living Polymerization Systems with
1
(6)
Phommalysack-Lovan, J., Chu, Y., Boyer, C., Xu, J. (2018) PET-RAFT Polymerisation:
Towards Green and Precision Polymer Manufacturing. Chem. Commun. 54, 6591– 6606.
(7)
McKenzie, T. G., Fu, Q., Wong, E. H. H., Dunstan, D. E., Qiao, G. G. (2015) Visible Light
Mediated Controlled Radical Polymerization in the Absence of Exogenous Radical Sources or
Catalysts. Macromolecules 48, 3864–3872.
(8)
Otsu, T. (2000) Iniferter Concept and Living Radical Polymerization. J. Polym. Sci. Part A
Polym. Chem. 38, 2121–2136.
(9)
Khan, M. Y., Cho, M. S., Kwark, Y. J. (2014) Dual Roles of a Xanthate as a Radical Source
and Chain Transfer Agent in the Photoinitiated RAFT Polymerization of Vinyl Acetate.
Macromolecules 47, 1929–1934.
(10) Shanmugam, S., Cuthbert, J., Kowalewski, T., Boyer, C., Matyjaszewski, K. (2018) Catalyst-
Free Selective Photoactivation of RAFT Polymerization: A Facile Route for Preparation of Comblike
and Bottlebrush Polymers. Macromolecules 51, 7776–7784.
(11) Quinn, J. F., Barner, L., Barner-Kowollik, C., Rizzardo, E., Davis, T. P. (2002) Reversible
Addition - Fragmentation Chain Transfer Polymerization Initiated with Ultraviolet Radiation.
Macromolecules 35, 7620–7627.
(12) You, Y., Hong, C., Bai, R., Pan, C., Wang, J. (2002) Photo-Initiated Living Free Radical
Polymerization in the Presence of Dibenzyl Trithiocarbonate. Macromol. Chem. Phys. 203, 477–483.
(13) Lu, L., Zhang, H., Yang, N., Cai, Y. (2006) Toward Rapid and Well-Controlled Ambient
Temperature RAFT Polymerization under UV−Vis Radiation:ꢀ Effect of Radiation Wave Range.
Macromolecules 39, 3770-3776.
This article is protected by copyright. All rights reserved

(14) Quinn, J. F., Barner, L., Barner-Kowollik, C., Rizzardo, E., Davis, T. P. (2002) Reversible
Addition - Fragmentation Chain Transfer Polymerization Initiated with Ultraviolet Radiation.
Macromolecules 35, 7620–7627.
(15) Carmean, R. N., Becker, T. E., Sims, M. B., Sumerlin, B. S. (2017) Ultra-High Molecular
Weights via Aqueous Reversible-Deactivation Radical Polymerization. Chem. 2, 93– 101.
(16) Xu, J., Jung, K., Boyer, C. (2014) Oxygen Tolerance Study of Photoinduced Electron
Transfer–Reversible Addition–Fragmentation Chain Transfer (PET-RAFT) Polymerization Mediated
by Ru(Bpy) Cl . Macromolecules 47, 4217–4229.
3
2
(17) Xu, J., Jung, K., Atme, A., Shanmugam, S., Boyer, C. (2014) A Robust and Versatile
Photoinduced Living Polymerization of Conjugated and Unconjugated Monomers and Its Oxygen
Tolerance. J. Am. Chem. Soc. 136, 5508–5519.
(18) Tu, K., Xu, T., Zhang, L., Cheng, Z., Zhu, X. (2017) Visible Light-Induced PET-RAFT
Polymerization of Methacrylates with Novel Organic Photocatalysts. RSC Adv. 7, 24040–24045.
(
19) Niu, J., Page, Z. A., Dolinski, N. D., Anastasaki, A., Hsueh, A. T., Soh, H. T., Hawker, C. J.
2017) Rapid Visible Light-Mediated Controlled Aqueous Polymerization with In Situ Monitoring.
(
ACS Macro Lett. 6, 1109–1113.
(20) Shanmugam, S., Xu, J., Boyer, C. (2015) Exploiting Metalloporphyrins for Selective Living
Radical Polymerization Tunable over Visible Wavelengths. J. Am. Chem. Soc. 137, 9174–9185.
(21) Thum, M. D., Wolf, S., Falvey, D. E. (2020) State-Dependent Photochemical and
Photophysical Behavior of Dithiolate Ester and Trithiocarbonate Reversible Addition–Fragmentation
Chain Transfer Polymerization Agents. J. Phys. Chem. A 124, 4211– 4222.
(22) Allegrezza, M. L., Konkolewicz, D. (2021) PET-RAFT Polymerization: Mechanistic
Perspectives for Future Materials ACS Macro Lett. 10, 433−446.
This article is protected by copyright. All rights reserved

(23) Corrigan, N., Xu, J., Boyer, C., Allonas, X. (2019) Exploration of the PET-RAFT Initiation
Mechanism for Two Commonly Used Photocatalysts. ChemPhotoChem. 3, 1193– 1199.
(24) Kottisch, V., Michaudel, Q., Fors, B. P. (2017) Photocontrolled Interconversion of Cationic
and Radical Polymerizations. J. Am. Chem. Soc. 139, 10665–10668.
(25) Xu, J., Shanmugam, S., Duong, H. T., Boyer, C. (2015) Organo-Photocatalysts for
Photoinduced Electron Transfer-Reversible Addition–Fragmentation Chain Transfer (PET-RAFT)
Polymerization. Polym. Chem. 6, 5615-5624.
(26) Michaudel, Q., Chauviré, T., Kottisch, V., Supej, M. J., Stawiasz, K. J., Shen, L., Zipfel, W.
R., Abruña, H. D., Freed, J. H., Fors, B. P. (2017) Mechanistic Insight into the Photocontrolled
Cationic Polymerization of Vinyl Ethers. J. Am. Chem. Soc. 139, 15530–15538.
(27) Corrigan, N., Xu, J., Boyer, C. A (2016) Photoinitiation System for Conventional and
Controlled Radical Polymerization at Visible and NIR Wavelengths. Macromolecules 49, 3274–3285.
(28) Evoniuk, C. J., Gomes, G., Hill, S. P., Fujita, S., Hanson, K., Alabugin, I. V. (2017) Coupling
3
N–H Deprotonation, C–H Activation, and Oxidation: Metal-Free C(sp )–H Aminations with
Unprotected Anilines J. Am. Chem. Soc. 139, 16210–16221.
(
29) Kishore, K., Asmus, K. D. (1989) Radical Cations from One-Electron Oxidation of Aliphatic
Sulphoxides in Aqueous Solution. A Radiation Chemical Study. J. Chem. Soc. Perkin Trans. 2. 12,
079-2084.
2
(30) Gollnick, K., Stracke, H. U. (2008) Direct and Sensitized Photolysis of Dimethyl Sulphoxide
in Solution. Pure Appl. Chem. 33, 217–246.
(31) Garg, S., Rose, A. L., Waite, T. D. (2007) Production of Reactive Oxygen Species on
Photolysis of Dilute Aqueous Quinone Solutions. Photochem. Photobiol. 83, 904–913.
(32) Veltwisch, D., Janata, E., Asmus, K. D. (1980) Primary Processes in the Reaction of Hydroxyl
Radicals with Sulfoxides. J. Chem. Soc., Perkin Trans. 2. 1, 146–153.
This article is protected by copyright. All rights reserved

(33) Crean, C., Geacintov, N. E., Shafirovich, V. (2009) Methylation of 2′-Deoxyguanosine by a
Free Radical Mechanism. J. Phys. Chem. B. 113, 12773–12781.
(34) Thum, M. D., Falvey, D. E. (2018) Photoreleasable Protecting Groups Triggered by
Sequential Two-Photon Absorption of Visible Light: Release of Carboxylic Acids from a Linked
Anthraquinone- N -Alkylpicolinium Ester Molecule. J. Phys. Chem. A 122, 3204–3210.
(35) Ly, D., Kan, Y., Armitage, B., Schuster, G. B. (1996) Cleavage of DNA by Irradiation of
Substituted Anthraquinones: Intercalation Promotes Electron Transfer and Efficent Reaction at GG
Steps. J. Am. Chem. Soc. 118, 8747–8748.
(36) Breslin, D. T., Schuster, G. B. (1996) Anthraquinone Photonucleases: Mechanisms for GG-
Selective and Nonselective Cleavage of Double-Stranded DNA. J. Am. Chem. Soc. 118, 2311–2319.
(37) Görner, H. (2003) Photoreduction of 9,10‐Anthraquinone Derivatives: Transient Spectroscopy
and Effects of Alcohols and Amines on Reactivity in Solution. Photochem. Photobiol. 77, 171–179.
(38) Bardagi, J. I., Ghosh, I., Schmalzbauer, M., Ghosh, T., König, B. (2018) Anthraquinones as
Photoredox Catalysts for the Reductive Activation of Aryl Halides. European J. Org. Chem. 1, 34–40.
(39) Kuss-Petermann, M., Wenger, O. S. (2016) Electron Transfer Rate Maxima at Large Donor–
Acceptor Distances. J. Am. Chem. Soc. 138, 1349–1358.
(40) Kuss-Petermann, M., Orazietti, M., Neuburger, M., Hamm, P., Wenger, O. S. (2017)
Intramolecular Light-Driven Accumulation of Reduction Equivalents by Proton-Coupled Electron
Transfer. J. Am. Chem. Soc. 139, 5225–5232.
(41) Görner, H. (2005) Photoreactions of P-Quinones with Dimethyl Sulfide and Dimethyl
Sulfoxide in Aqueous Acetonitrile. Photochem. Photobiol. 82, 71-77.
(42) Shanmugam, S., Xu, J., Boyer, C. (2017) Photoinduced Oxygen Reduction for Dark
Polymerization. Macromolecules 50, 1832–1846.
This article is protected by copyright. All rights reserved

(
43) Gomes, G.; Wimmer, A.; Smith, J.; König, B.; Alabugin, I.V. (2019) CO or SO : Should It
2 2
Stay, or Should It Go? J. Org. Chem. 84, 6232-6243.
FIGURE CAPTIONS
Scheme 1. Photo-oxidation of DMSO leads to the formation of reactive methyl radicals.
Figure 1. Compounds used in this study. The synthesis of AQ-1 has been reported previously.34
MAA, DMAEMA, GMA, stand for methacrylic acid, N,N-diaminoethyl methacralate, glycidyl
methacralate, respectively.
Figure 2. GPC traces showing narrow molecular weight distributions and an increasing in the Mn
with increasing concentrations of photoinitiator used.
1
Figure 3. Reaction process was monitored by H NMR during polymerization. During photolysis
(white regions) noticeable amounts of monomer were consumed while, in the dark (grey regions), no
significant change was noted. Conversions were estimated to ± 2%. The seeming decrease monomer
conversion in dark periods shown in Fig. 3 is likely due to errors inherent in the conversion
measurement.
Figure 4. Transient absorption spectrum of AQ-1 in DMSO in air (A) and under N (B). Transient
2
absorption spectrum of AQ-1 in MeCN with 2 mM DABCO.
Figure 5. Waveforms at 490 nm of AQ-1 generated by pulsed photolysis of AQ-1 in MeOH with
•
TEA under N . The addition of increasing concentrations of CPADB result a smaller magnitude for
2

•
signal for AQ-1 due to competitive quenching between CPADB and TEA.
Figure 6. Stern Völmer analysis. The change in O.D. averaged over 1 µs is plotted with respect to
the concentration of CPADB. The τ is estimated from the waveforms in Fig. 6 to calculate the k from
q
the slope of a linear fit to the data.
This article is protected by copyright. All rights reserved

Scheme 2. Formation of methyl benzodithioate and sulfinic acid from photolysis of AQ-1 in DMSO.
Scheme 3. Mechanism for the photo-oxidation of DMSO by AQ-1 followed by radical formation
and subsequent RAFT polymerization.
This article is protected by copyright. All rights reserved

Table 1: RAFT Polymerization with AQ-1 in a DMSO/Monomer mixture in the presence of Air
b
c
a
d
Mn, GPC
Mn, theo
Entry
M/CTA
[AQ-1] , equiv.
Photolysis Time, hrs
PDI
Conv.
1
2
3
4
5
6
7
8
9
0
0.1
0
4
42.1k
7.6k
9.8k
-
-
1.61
1.03
1.05
-
21%
26%
59%
0%
200
200
200
200
200
200
200
200
200
200
200
8
5.6k
12.2k
-
0
16
0.1
DARK
0.01
0.02
0.03
0.04
0.05
0.05
0.1
8
8
10.8k
8.7k
8.7k
9.1k
9.9k
13.1k
9.1k
10.1k
7.8k
9.5k
10.5k
10.9k
12.3k
19.2k
9.5k
15.7k
1.16
1.06
1.02
1.03
1.14
1.23
1.06
1.03
37%
45%
50%
52%
59%
93%
45%
76%
8
8
8
1
1
1
0
1
2
16
8
0.1
16
a
c
b
Relative to the amount of CTA-1. Measured by GPC using THF as the solvent and calibration using polystyrene standards.
Calculated according to the following equation: Mn, theo = [M] /[CTA] • MW • conv. + MWCTA. Where [M] , [CTA] ,
0
0
M
0
0
MW , conv., MWCTA, correspond to initial concentration of monomer, M, the initial concentration of RAFT agent, the
M
molecular weight of the monomer, the fraction of monomer converted into polymer and the molecular weight of the RAFT
d
1
agent respectively. Determined using H NMR.
This article is protected by copyright. All rights reserved

Table 2: Solvent dependence on the RAFT polymerization of MMA with AQ-1
Photolysis Time,
hrs
b
a
b
c
Entry CTA/I [AQ-1] , equiv. Solvent
Mn
PDI
Conv.
1
2
3
4
5
200
200
200
200
200
0.05
0.05
0.05
0.05
0.05
DMSO
Toluene
MeOH
MeCN
THF
16
16
16
16
13.1k
11.0k
8.4k
1.23
1.30
2.09
1.12
1.06
93%
34%
21%
40%
32%
11.9k
9.0k
16
a
b
Photolysis at 419 nm. Relative to the amount of CTA-1. Measured by GPC using THF as the
c
1
solvent and calibration using polystyrene standards. Determined using H NMR.
This article is protected by copyright. All rights reserved

php_13468_f1.jpg
Thisꢀarticleꢀisꢀprotectedꢀbyꢀcopyright.ꢀAllꢀrightsꢀreserved

php_13468_f2.jpg
Thisꢀarticleꢀisꢀprotectedꢀbyꢀcopyright.ꢀAllꢀrightsꢀreserved

php_13468_f3.jpg
Thisꢀarticleꢀisꢀprotectedꢀbyꢀcopyright.ꢀAllꢀrightsꢀreserved

php_13468_f4.jpg
Thisꢀarticleꢀisꢀprotectedꢀbyꢀcopyright.ꢀAllꢀrightsꢀreserved

php_13468_f5.jpg
Thisꢀarticleꢀisꢀprotectedꢀbyꢀcopyright.ꢀAllꢀrightsꢀreserved

php_13468_f6.jpg
Thisꢀarticleꢀisꢀprotectedꢀbyꢀcopyright.ꢀAllꢀrightsꢀreserved