Trialkyl phosphites and diaryliodonium salts as co-initiators in a system for radical-promoted visible-light-induced cationic polymerization(1)

Article abstract of DOI:10.1021/jo302830r

Trialkyl phosphites ((RO)₃P) can act as co-initiators for the diaryliodonium-induced cationic polymerization of cyclohexene oxide (CHO) or THF. A radical initiation step is also required, consistent with the essential role of a radical chain reaction of the phosphite with the iodonium salt to form polymerization-starting aryltrialkoxyphosphonium salts (ArP⁺(OR) ₃). We used the visible photolysis of phenylazoisobutyronitrile (PAIBN) as the radical initiation step. The presence of multiple fluorine substituents on the phosphite, as in tris(2,2,2-trifluoroethyl) phosphite (TFP), allows polymerization to proceed with a minimal amount of chain transfer from nucleophilic attack by the phosphite. In a typical experiment, a CHO solution of bis(4-tert-butylphenyl)iodonium hexafluorophosphate (0.05 M), TFP (0.1 M), and PAIBN (0.02 M) was illuminated with a 65-W compact fluorescent bulb for 1 h, resulting in a 78% conversion to poly(cyclohexene oxide) with an average molecular weight (M_W) of 25000. We also used competition experiments to determine approximate rate constants for reactions of phenyl radicals with CHO (k = 2 × 10⁶ M^-1 s^-1) and with TFP (k = 2 × 10⁸ M^-1 s^-1).

Full text of DOI:10.1021/jo302830r

Article
pubs.acs.org/joc
Trialkyl Phosphites and Diaryliodonium Salts as Co-initiators in a
System for Radical-Promoted Visible-Light-Induced Cationic
Polymerization1
Thomas W. Nalli,* Lee G. Stanek, Rebekka H. Molenaar, Krysia L. Weidell, Justin P. Meyer,
Brandon R. Johnson, Timothy T. Steckler, Jay Wm. Wackerly, Missy Jo Studler, Kevin L. Vickerman,
and Stephen A. Klankowski
Chemistry Department, Winona State University, Winona, Minnesota 55987-5838, United States
S
* Supporting Information
ABSTRACT: Trialkyl phosphites ((RO)3P) can act as co-initiators for the
diaryliodonium-induced cationic polymerization of cyclohexene oxide (CHO) or
THF. A radical initiation step is also required, consistent with the essential role
of a radical chain reaction of the phosphite with the iodonium salt to form
polymerization-starting aryltrialkoxyphosphonium salts (ArP+(OR)3). We used
the visible photolysis of phenylazoisobutyronitrile (PAIBN) as the radical
initiation step. The presence of multiple fluorine substituents on the phosphite,
as in tris(2,2,2-trifluoroethyl) phosphite (TFP), allows polymerization to proceed with a minimal amount of chain transfer from
nucleophilic attack by the phosphite. In a typical experiment, a CHO solution of bis(4-tert-butylphenyl)iodonium
hexafluorophosphate (0.05 M), TFP (0.1 M), and PAIBN (0.02 M) was illuminated with a 65-W compact fluorescent bulb
for 1 h, resulting in a 78% conversion to poly(cyclohexene oxide) with an average molecular weight (MW) of 25000. We also used
competition experiments to determine approximate rate constants for reactions of phenyl radicals with CHO (k = 2 × 106 M−1
s−1) and with TFP (k = 2 × 108 M−1 s−1).
give an arylphosphonium ion (ArP+(OMe)3) that can
methylate the monomer (eq 5, R = Me), leading to
polymerization (eq 6). A key step in the radical chain is the
single-electron transfer (SET) from the arylphosphoranyl
radical (ArP•(OMe)3) to the iodonium salt (eq 1).
INTRODUCTION
■
Diaryliodonium salts (Ar2I+Y−) with non-nucleophilic anions
−
−
(e.g., PF6 , BF4 ) have long been known to function as
photoinitiators of cationic polymerizations.2 The mechanism
involves photolysis of the iodonium salt to form an aryl radical
and an iodoarene radical cation leading to the formation of
Brønsted acids, which start the polymerization.2−6 With certain
monomers, such as THF and 1,3-dioxolane, radical-chain
chemistry of the aryl radicals with the monomer can lead to
monomer-derived cations that are also capable of starting
polymerization.7−9
•
+
Ar2I+ + ArPZ3 → Ar I• + ArPZ3 (propagation)
(1)
(2)
2
Ar2I• → ArI + Ar• (propagation)
•
Ar• + PZ3 → ArPZ3 (propagation)
(3)
(4)
However, the use of iodonium salts in applications such as
photocurable coatings and composites, printing inks, and
stereolithography is limited by the weak absorption of these
salts in the visible region.3−6 Therefore, there has been
considerable interest in developing iodonium-based systems
for visible-light-induced cationic polymerization (VLICP).10−23
A number of approaches have been employed including the
modification of iodonium structure to extend the λmax and the
use of photosensitizers that can do photoinduced electron
transfer to the iodonium salt. Another approach involves using
light to initiate a reductive radical chain reaction of the
iodonium salt with an additive (co-initiator) to form reactive
cations.20−23
+
Ar2I+ + PZ3 → ArI + ArPZ3 (net)
(Z = OR or Ph)
Initiation of the radical chain can be accomplished by visible
photolysis of phenylazoisobutyronitrile (PAIBN, λmax = 395
We reported previously that trimethyl phosphite (TMP)
functions as a co-initiator for the iodonium-induced polymer-
ization of cyclohexene oxide (CHO).23 The mechanism
involves a radical-chain reaction24,25 (eqs 1−4, Z = OMe) to
Received: December 31, 2012
Published: April 2, 2013
© 2013 American Chemical Society
3561
dx.doi.org/10.1021/jo302830r | J. Org. Chem. 2013, 78, 3561−3569

The Journal of Organic Chemistry
Article
nm)26 to form phenyl radicals (eq 7). Thus, a diaryliodonium
salt, TMP, and PAIBN comprise one system for visible-light-
induced cationic polymerization (VLICP).23 However, pro-
longed irradiation times were required, and relatively small
polymers were obtained. This was attributed to the nucleophilic
phosphite both consuming itself by an Arbuzov reaction (eqs 8
and 9) and bringing about polymerization chain transfer by
competing with the monomer for electrophilic sites on the
growing polymer (eq 10).
initial experiments that compared TMP, TFP, and dimethyl
phenylphosphinite (DMPP) (Table 1). Bis(4-tert-butylphenyl)-
iodonium hexafluorophosphate 1a (0.01 M) and PAIBN (0.02
M) were also included in the reaction solutions. Illumination
was provided by a 500-W halogen lamp. TMP and TFP both
gave a modest 36% yield of polyCHO ([phosphite] = 0.04−
0.05 M, 5 h, entries 1a and 3a). In contrast, DMPP gave only
3% conversion (entry 2a). Control experiments that left out the
phosphite, the iodonium salt, or light did not polymerize. With
TFP, higher phosphite concentrations gave higher polymer
yields (65−78%, entry 5), but with TMP the opposite effect
was observed (17−24%, entry 1). Polymers with significantly
higher Mw (20000−30000) were obtained in the reactions that
used TFP compared to those that used TMP (11000−18000,
entry 1). The use of a UV-filtering material slowed the TFP-
induced polymerization, but a significant yield of polymer was
still obtained (21%, entry 6b). Thus, visible light is sufficient for
bringing about polymerization in the TFP/iodonium/PAIBN
system.
•
•
PhNNC(CH3)2CN → Ph + N2 + C(CH3)2CN
(7)
(8)
(9)
+
+
ArP(OR)3 + P(OR)3 → ArP(O)(OR)2 + RP(OR)3
+
+
RP(OR)3 + P(OR)3 → RP(O)(OR)2 + RP(OR)3
All three phosphites (TFP, TMP, DMPP) react rapidly with
phenyl radicals (eq 3)27 to form arylphosphoranyl radicals
(ArP•Z3) that do not possess fast unimolecular decomposition
pathways.28 Thus, chain-propagating single-electron-transfer to
the iodonium salt (eq 1) is not precluded, and the observed
large differences in co-initiation ability are not due to different
propensities to undergo the required radical-chain chemistry
(eqs 1−3). In contrast, a clear difference exists in these
phosphites’ relative nucleophilicities. Literature reports on
phosphine reactivity29 and indirect measurements of the
electron availability on phosphorus,30 as well as DFT
calculations done as part of this study (see later discussion)
all point to a nucleophilic reactivity series of DMPP > TMP >
TFP. Of course, in valence-bond terms, TFP’s non-nucleophilic
nature is simply due to the electron withdrawing effect of the
fluorines. Thus, TFP is a superior polymerization co-initiator
because its counterproductive reactions as a nucleophile (eqs
8−10) are slow. This interpretation also explains why higher
[TFP] resulted in faster photopolymerizations to form
For this work, we decided to explore the use of other
alkoxyphosphines (phosphites) as co-initiators in otherwise
identical systems for VLICP. We reasoned that halogen-
substituted alkoxy groups would reduce the nucleophilicity of
the phosphite, thus allowing polymerization to proceed without
undue competition from the nonproductive Arbuzov and chain-
transfer reactions (eqs 8−10). Accordingly, we tested tris(2,2,2-
trifluoroethyl) phosphite (P(OTfe)3) (TFP) and found it to be
a superior co-initiator for iodonium-induced VLICP. In this
paper, we report the results of experiments comparing polymer
yields and molecular weights afforded by various phosphite co-
initiators including TMP and TFP. Also reported are the results
of mechanistic experiments aimed at providing greater
understanding of the TFP/iodonium polymerization system.
RESULTS AND DISCUSSION
■
Isolated yields and absolute molecular weights of poly-
(cyclohexene oxide) (polyCHO) were determined in some
a
Table 1. Photopolymerization of CHO Using 1a, PAIBN, and TMP, DMPP, or TFP
b
c
d
d
d
entry
1
PZ3
expt
[PZ3] (M)
time (h)
yield (%)
Mw
Mw/Mn
Xn
TMP
a
b
c
a
b
a
b
a
b
c
d
a
b
c
d
a
0.04
0.08
0.17
0.04
0.04
0.05
0.05
0.04
0.04
0.04
0.04
0.01
0.10
0.15
0.19
0.04
0.04
5 + 0
36
24
17
3
18000
16900
11100
na
1.54
1.61
1.45
119
107
78
5 + 0
5 + 0
e
e
2
3
4
DMPP
TFP
5 + 0
5 + 20
5 + 0
8
na
36
38
9
21700
22700
20800
22600
28600
26800
29800
32200
32000
na
1.56
1.52
1.49
1.56
1.51
1.52
1.49
1.44
1.57
142
152
142
148
193
180
205
228
208
5 + 20
2 + 20
3 + 20
4 + 20
5 + 20
5 + 20
5 + 20
5 + 20
5 + 20
5 + 20
5 + 20
TFP
20
51
43
38
65
69
78
40
21
5
6
TFP
TFP
e
e
e
na
f
b
na
a
Irradiation with 500-W halogen bulbs at ca. 18 °C. [1a] = 0.01 M. [PAIBN] = 0.02 M. The experiments grouped in each entry used common stock
solutions and were irradiated simultaneously. Entry 1 results were reported previously.23 Time of irradiaton + postirradiation dark time. Isolated
percent yield of polyCHO. Absolute molecular weights determined by GPC-MALLS. Not analyzed. Irradiation with UV filter film in place.
b
c
d
e
f
3562
dx.doi.org/10.1021/jo302830r | J. Org. Chem. 2013, 78, 3561−3569

The Journal of Organic Chemistry
Article
polymers with undiminished molecular weights (entry 5),
whereas higher [TMP] gave lower yields and smaller polymers
(entry 1). On one hand, high [TMP] leads to more of the
detrimental side reactions (eqs 8−10), but TFP is not
nucleophilic enough to undergo these reactions even when its
concentration is high. On the other hand, high [TFP] promotes
more efficient phenyl radical scavenging (eq 3) with less
termination due to reaction of the phenyl radicals with the
monomer.
experimentation on optimizing reaction conditions and using
product studies to establish the mechanism of the TFP/
iodonium/PAIBN photopolymerization system. These experi-
ments and the nonpolymer products that were formed are
summarized by eq 11.
Upon further experimentation we found it possible to use
NMR to simultaneously observe both the iodonium/phosphite
radical chemistry and the cationic polymerization. Conversion
to polymer was determined by integrating the α-proton
resonances of CHO (3.0 ppm) and polyCHO (3.4 ppm),
and reactions of the phosphite and iodonium salt were
1
observed by both 31P and H NMR. Unless otherwise noted,
all NMR experiments used a 65-W compact fluorescent lamp
(CFL) as the visible light source.
TFP was compared to five other commercially available
phosphites in a series of experiments that were run under
identical conditions (Table 2). Two of these, tris(isopropyl)
Table 2. Photopolymerization of CHO Using 1a, PAIBN,
a
and Various Co-initiators
Mw/
b
c
In attempting to optimize the TFP system, we found that
relatively high concentrations of both the phosphite (0.20 M)
and the iodonium salt 1a (0.05 M) were advantageous (Table
3, entry 8i). However, higher concentrations of the azo initiator
entry
co-initiator
% polym
% ArI
Mw
Mn
Xn
d
7a
7b
7c
7d
7e
7f
P(OCH2CF3)3
P(O-i-Pr)3
45
43
29
9
59
65
nd
19000
15600
15500
13500
na
1.49
1.40
1.48
1.34
130
113
107
103
20
30
10
10
10
60
P(OC2H4Cl)3
P(O-n-C13H27)3
P(OSiMe3)3
BnOP(OEt)2
PhNMe2
66
61
24
59
47
e
Table 3. TFP-Co-initiated CHO Polymerization: Effect of
9
a
Reactant Concentrations
16
42
9700
1.34
1.43
74
99
7g
13800
b
entry [1a] [TFP] [PAIBN] % polym
Mw
Mw/Mn
Xn
a
[1a] = 0.01 M, [PAIBN] = 0.7 mM, [co-initiator] = 0.10 M.
Illumination with a 65-W CFL at ca. 34 °C for 2.0 h. Absolute
molecular weights determined by GPC-MALLS. Percent conversion
to polyCHO by H NMR initially after the illumination period and
after standing for an additional 5 days dark period. Percent conversion
of 1a to 4-tert-butyliodobenzene. Not detected. Limit of detection
8a
8b
8c
8d
8e
8f
0.01
0.02
0.05
0.01
0.02
0.05
0.01
0.02
0.05
0.01
0.02
0.05
0.10
0.10
0.10
0.10
0.10
0.10
0.20
0.20
0.20
0.20
0.20
0.20
0.02
0.02
0.02
0.10
0.10
0.10
0.02
0.02
0.02
0.10
0.10
0.10
65
84
78
65
67
69
80
84
96
72
81
80
25400
na
1.60
162
c
c
c
c
b
25000
25400
na
1.62
1.62
158
160
1
c
d
e
22900
24700
na
1.50
1.55
156
162
estimated at 5%. Not analyzed.
8g
8h
phosphite (43%, entry 7b) and tris(2-chloroethyl) phosphite
(29%, entry 7c), gave comparable polymer yields to that
afforded by TFP (45%, entry 7a). However, the others gave
sharply reduced yields (entries 7d−f), and all five gave
polymers of diminished Mw indicating more chain transfer
due to greater phosphite nucleophilicity. It is also interesting to
note that all of the tested phosphites reacted with iodonium salt
1a to produce significant NMR yields of 4-tert-butyliodoben-
zene (10−30% ArI, entry 7b−f) except for TFP (<5% ArI,
entry 7a).
d
8i
8j
8k
8l
24300
25400
na
1.54
1.47
161
176
20300
1.65
126
a
Concentrations are in mol/L. Solutions were illuminated with a 65-W
CFL at ca. 34 °C for 1.0 h. Absolute molecular weights were
b
1
determined by GPC-MALLS. Percent conversion to polymer by H
c
d
NMR. Not analyzed. 2a (∼6%) and TFPO (∼10%) were detected
by 31P NMR.
N,N-Dimethylaniline (DMA) was reported to function as a
co-initiator for iodonium-initiated VLICP of CHO when
various organic dyes were used as photoinitiators.20 The
mechanism was proposed to involve a radical-chain leading to
the formation of an iminium ion (CH2N+(CH3)Ph), which
starts the polymerization. Therefore, we included DMA in our
comparison of alternative co-initiators (Table 2, entry 7g). The
fact that DMA gave a 42% polymer yield when using PAIBN as
a radical initiator substantiates the previously proposed radical
chain mechanism.20
generally gave lower polymer yields (entries 8d,e,f,j,k,l).
Moreover, under the optimum conditions of entry 8i,
conversion to polymer with Mw = 24300 was excellent at
96% after only 1 h photolysis with the 65-W CFL. We also note
here that the completely polymerized solutions afforded by
these experiments are transparent and nearly colorless unlike
that afforded by VLICP systems that use a dye as the light-
absorbing component.20
Small product peaks at 22.4 and 21.8 ppm were observed in
the 31P NMR of the optimized run (entry 8i) as well as in
several of the experiments reported in subsequent tables. The
Having established that TFP is the most effective co-initiator
among commercially available phosphites, we focused further
3563
dx.doi.org/10.1021/jo302830r | J. Org. Chem. 2013, 78, 3561−3569

The Journal of Organic Chemistry
Article
peak at 22.4 ppm was identified as bis(2,2,2-trifluoroethyl) tert-
butylphenylphosphonate (ArP(O)(OTfe)2) 2a. This assign-
ment was verified by adding a synthetic sample of 2a to the
NMR tube and reacquiring the spectrum. The resonance at
21.8 ppm was due to the unsubstituted phenyl phosphonate,
PhP(O)(OTfe)2 2c (lit.31 shift = 21.3 ppm). The formation of
the phosphonates 2a and 2c is consistent with the idea that
intermediate phosphonium salts (ArP+(OTfe)3 and
PhP+(OTfe)3) start polymerization by transferring a trifluor-
oethyl group to the monomer (eq 5, R = Tfe). Thus, unlike
most other photoinduced cationic polymerization systems
involving iodonium salts, this system does not rely on the
generation of Brønsted acids.
This observation provides further evidence for the alkylation
step (eq 5) as the key to starting the cationic polymerization. In
contrast, the 31P NMR of the purified polymers did not show
resonances attributable to phosphorus-containing end groups,
consistent with our proposal that TFP does not react with
active centers on the growing polymer. Using p-bromofluor-
obenzene as an internal standard, integration of the 19F
spectrum of polyCHO from entry 8a showed a 1:8 ratio of
TfeO end groups to polymer chains. Most likely, chain transfer
to monomer due to traces of water33 causes the number of
polymer molecules to be greater than the number of initiating
end groups. The somewhat elevated molecular weight
distributions observed (Mw/Mn = 1.3−1.6) are higher than
the ideal for a living polymerization most likely due to this same
factor.34 Using the 1:8 end group to polymer ratio in
conjunction with the degree of polymerization in entry 8a
(Xn = 162) leads to a value for the chain transfer constant (C =
ktr/kp) of C = 20. Corroboration of this conclusion comes from
the 20:1 mol ratio of polyCHO (from Xn = 161 and polymer
yield =65%) to 2a (6% yield based on 0.05 M iodonium salt)
observed in entry 8i.
A multiplet observed at −75 ppm in the 19F NMR spectra of
highly purified polyCHO from entries 5a, 8a, and 8i revealed
the presence of 2,2,2-trifluoroethoxy end groups (Figure 1).32
The TFP/iodonium/PAIBN combination can also bring
about the cationic polymerization of THF (entry 10a, Table 4),
although as expected,35 THF polymerization was slower than
CHO polymerization under these conditions (entry 9a).
Control experiments that omitted the light (entry 10e) or
PAIBN (entries 9b and 10b) did not polymerize. Omission of
the TFP (entries 9c and 10c) gave much lower conversions to
polymer, thus demonstrating TFP’s essential role as a co-
initiator.
Air oxidation of TFP to tris(2,2,2-trifluoroethyl) phosphate
(TFPO) was a side reaction in the polymerizing solutions.
Integration of the 31P NMR spectra revealed yields of TFPO of
approximately 20% based on the starting TFP concentration
(entries 9a and 10a, Table 4). This product (ca. 10%) was also
detected in entry 8i (Table 3). This oxidation reaction must be
a photoinitiated radical process because in the experiments
which omitted light or the azo initiator, significantly less
oxidized product was observed (entries 9b, 9e, and 10b).
Because this reaction did not appear to adversely affect the
ability of the iodonium/TFP/azo combination to bring about
photopolymerization we did not concern ourselves further with
it and our experiments used nondegassed solutions unless
otherwise noted.
Figure 1. 19F NMR of polyCHO after successive reprecipitations from
CHCl3/MeOH: (a) crude; (b) after one precipitation; (c) after two
precipitations; (d) after three precipitations.
a
Table 4. Photopolymerization of CHO and THF Using 1a, PAIBN, and TFP
b
% yields
c
d
e
e
e
entry
expt
[TFP]
[PAIBN]
polymer
TFPO
2a
ArI
5
ArH
9, CHO
a
0.10
0.10
none
none
0.10
0.10
0.10
none
none
0.02
none
0.02
none
0.02
0.02
none
0.02
none
47
nd
2
52
nd
5
20
7
2
na
na
na
na
na
12
na
21
na
b
c
na
na
na
nd
5
nd
nd
nd
nd
18
nd
20
nd
nd
nd
4
d
nd
nd
6
nd
nd
22
f
e
10, THF
a
30
9
g
b
c
nd
0.2
nd
nv
nv
nv
nd
nd
nd
g
g
nd
nd
d
a
b
[1a] = 0.05 M. 65-W CFL with UV filter at ca. 30 °C for 1.0 h. Reactant concentrations are given in mol/L. NMR yields. “na” = not analyzed. “nd”
c
d
= not detected. Conversion of monomer to poly(CHO) or poly(THF). The second column is after an additional 18−20 h in the dark. Yield of
e
tris(2,2,2-trifluoroethyl) phosphate based on TFP from 31P NMR integrals. Yields of ArP(O)(OTfe)2 2a, 4-tert-butyliodobenzene (ArI), and tert-
f
g
butylbenzene (ArH) based on 1a. Kept in the dark for 24 h. No increase in viscosity was observed so not analyzed.
3564
dx.doi.org/10.1021/jo302830r | J. Org. Chem. 2013, 78, 3561−3569

The Journal of Organic Chemistry
Article
a
Table 5. Product Studies and Thermal Initiation
c
% yields
b
d
e
f
f
f
g
entry
11a
solvent
none
[1a] (M)
init
time (min)
poly(CHO)
TFPO
2a
2c
ArI
2a/2c
0.01
PAIBN
60
250
60
32
53
0
25
15
6
4
nd
14
nd
nd
nd
nd
14
44
13
33
nd
10
na
na
20
na
na
nd
nd
33
85
28
77
26
na
na
0.69
10
nd
nd
nd
nd
27
62
20
59
16
7
11b
11c
12a
none
none
0.01
none
0.02
none
250
60
20
0
16
6
none
250
60
0
21
15
32
8
h
CD3CN
CD3CN
PAIBN
PAIBN
1
2.0
1.4
1.5
1.8
270
60
3
i
h
h
12b
0.02
1
270
270
120
120
4
11
16
90
8
12c
CD3CN
none
0.02
0.02
0.02
none
1
j
j
j
13a
BPO, Δ
none, Δ
34
2
0.72
j
13b
none
na
a
b
[TFP] = 0.10 M. Illumination with a 65-W CFL at ca. 34 °C unless otherwise noted. Radical initiator included at 0.02 M. “BPO” = benzoyl
c
d
e
f
peroxide. NMR yields. Percent conversion of CHO to polymer. Yield of TFP oxide based on TFP. Yields of ArP(O)(OTfe)2 2a, PhP(O)(OTfe)2
g
2c, and 4-tert-butyliodobenzene based on the iodonium salt. Ratio of arylphosphonate 2a to phenyl phosphonate 2c as determined by 31P NMR.
h
i
j
CD3CN cosolvent, [CHO] = 3.7 M. Reactions run in NMR tubes. Solution was degassed by N2 purge. Thermal initiation; reactions run at 92 °C
in dark.
In the THF experiments (entry 10, Table 4), we were able to
a phosphonium salt, two polymerization−starting cations are
produced per initiation event.) When acetonitrile-d3 was used
as a cosolvent with CHO (entry 12, [CHO] = 3.7 M) the
radical chain reaction to form ArI and 2a was faster than in neat
CHO ([CHO] = 9.9 M). Furthermore, the 2a/2c ratio
increased by the same factor of about 2.5 that the CHO
concentration had been decreased by (2a/2c = 1.5−2.0, entries
12a,b). Thus, reactions of propagating radicals with the
monomer (CHO) represent a significant set of termination
reactions. Most likely, H-abstraction from CHO by aryl radicals
is followed by radical coupling of monomer-derived radicals.38
Yet, even when these termination reactions are removed by
omitting CHO and using nonreactive CD3CN as the solvent39
the TFP/iodonium reaction is surprisingly slow (Table 6). For
example, the reaction of iodonium salt 1a with TFP (entry 14)
is ca. 10 times slower than the reaction of 1a with Ph3P (entry
15). Therefore, the kinetic chain length of the TFP/iodonium
reaction must be much shorter than that of the very efficient
Ph3P/iodonium radical chain (eqs 1−4, Z  Ph).24 However,
TFP reacts rapidly when Ph3P is present (20−50% yield of 2a,
entry 16−22).40 Moreover, the yield ratio of aryl phosphonate
to aryl phosphonium salt (2a/ArP+Ph3) is proportional to the
starting ratio of [TFP] to [Ph3P] (Figure 2). Thus, TFP and
Ph3P compete nearly equally for reaction with aryl radicals
(while Ph3P continues to propagate a chain reaction for their
formation).
determine the yields of all aryl-containing products by 1H NMR
because the tert-butyl protons’ resonances were not obscured
by the spectrum of the monomer (as with CHO). The results
provide further evidence for the key role of the arylphospho-
nium salt (ArP+(OTfe)3) in starting polymerization. The
diaryliodonium salt was consumed at the same rate in the
absence of TFP (18% ArI, entry 10a) as it was when TFP was
present (20% ArI, entry 10c), but polymerization only occurred
to a significant extent in the TFP-containing solution.36 Clearly,
the THF polymerization is facilitated greatly when aryl radicals
from the iodonium salt are diverted to reaction with the
phosphite (5% yield 2a, entry 10a). Thus, alkylation of the
monomer by the arylphosphonium salt to form phosphonate 2a
(eq 5) is implicated as the main polymerization-starting step.
The role of the azo compound PAIBN is made clear by the
experiments reported in Table 5. With solutions containing
PAIBN (0.02 M, entry 11a), polymerization was relatively fast
(32%, 60 min), and phenylphosphonate 2c was formed (14%,
250 min). In the absence of PAIBN (entry 11b), polymer-
ization was much slower (0%, 60 min) and no 2c was
detected.37 Hence, phenyl radicals from PAIBN photolysis (eq
7) have the same fate as chain-propagating aryl radicals,
meaning that PAIBN initiation of the radical chain comes about
through the reaction of phenyl radicals with TFP to form
phosphoranyl radicals (eq 3), which do SET with the iodonium
salt (eq 1) and ultimately end up as phosphonate 2c.
Polymerization can also be brought about by thermal radical
initiation using benzoyl peroxide (entry 13a), and the observed
10% yield of 2c reveals the mode of initiation as being the same
as with PAIBN photolysis. Hence, it is not a photolytic reaction
of the iodonium salt that brings about polymerization, but
rather it is a free-radical chain reaction (eqs 1−3) that can be
initiated by any source of phenyl radicals.
The slope of Figure 2 corresponds to the relative rate
constants of TFP and Ph3P in their respective reactions with
phenyl radicals (slope = kTFP/kPh3P = 0.65). The triphenyl-
phosphine rate constant has been estimated previously, kPh3P
=
3 × 108 M−1 s−1.24 Thus, we can estimate a rate constant of kTFP
= 2 × 108 M−1 s−1 for the reaction of TFP with phenyl radicals
(eq 3, Z  OTfe). More importantly, we conclude that the
inefficiency of the TFP/iodonium chain is not due to this
reaction being slow.
The yields of aryl phosphonate 2a and phenyl phosphonate
2c are nearly equal under these conditions (2a/2c = 0.7, entries
11a and 13a), indicating that the radical chain reaction of
iodonium 1a with TFP is not a true chain reaction (average
kinetic chain length = approximately 1.0) under these
conditions. (Still, because the initiation sequence also produces
Similar experiments in which CHO was allowed to compete
with triphenylphosphine for 4-methylphenyl radicals generated
from the radical chain reaction of Ph3P with bis(4-
methylphenyl)iodonium hexafluorophosphate 1b are reported
in Table 7. The results are graphed in Figure 3, the slope of
3565
dx.doi.org/10.1021/jo302830r | J. Org. Chem. 2013, 78, 3561−3569

The Journal of Organic Chemistry
Article
Table 6. Competition Experiments: Reactivity of TFP vs
a
Ph3P toward Aryl Radicals
b
reactant molarities
[1a] [TFP] [Ph3P] time (min) ArI
% yield
entry
2a
ArP+Ph3
c
14
0.02
0.02
0.02
0.01
0.01
0.001
0.01
0.02
0.01
0.15
none
0.15
0.15
0.2
none
0.16
0.3
3
3
2
na
na
35
86
57
64
33
42
48
55
24
32
31
39
29
53
42
59
10
3
12
44
88
72
89
70
89
67
86
66
87
59
81
50
81
63
85
8
c
15
na
na
18
19
39
45
23
31
39
49
30
41
18
32
21
30
10
3
16
17
18
19
20
21
22
10
15
45
3
0.15
0.25
0.15
0.15
0.15
0.15
Figure 3. Yield ratio vs [reactant] ratio for CHO vs Ph3P competition
experiments.
10
3
0.3
10
3
0.3
reason for the observed positive correlation of polymer yield
with starting TFP concentration (Table 3).
10
3
0.15
0.15
We also carried out DFT calculations as part of this study
(Table 8). As expected, the phosphite HOMO energies show
that TFP (EHOMO = −7.4 eV) is inherently much less
nucleophilic than the nonfluorinated phosphites TMP and
DMPP (EHOMO = −6.2 and EHOMO = −5.6 eV). More
importantly, the calculations allow an estimate of the
corresponding phenylphosphonium salts’ reduction potentials,
which in turn allow us to gauge ΔG for the crucial single
electron transfer step (eq 1). The validity of the obtained
reduction potentials is bolstered by the excellent agreement of
the value obtained for tetraphenylphosphonium with a
literature value for its half-wave reduction potential (av calcd
10
3
10
a
CD3CN solutions were irradiated with a 500-W halogen lamp at ca.
b
c
15 °C. NMR yields based on 1a. PAIBN was also included at 0.02
M.
E
red = −2.0 V vs SCE, lit.41 E1/2 = −1.87 V vs SCE). Using the
computed phosphonium reduction potentials and the (in our
opinion) best literature value available for E1/2 for diphenylio-
donium salts (E1/2 = −0.7 V vs SCE)42 leads to estimates of
ΔGSET for the various phosphines that are instructive. The
result calculated for TFP (ΔGSET = −1.5 4.5 kcal/mol, entry
30) points to the most probable reason for the inefficiency of
the TFP/iodonium chain reaction, this being a nearly
thermoneutral (nonexergonic) and, therefore, slow SET step
(eq 1). In contrast, trimethyl phosphite (ΔGSET = −27
kcal/mol, entry 28) and triphenylphosphine (ΔGSET = −31
3
3
Figure 2. Yield ratio vs [reactant] ratio for TFP vs Ph3P competition
experiments.
kcal/mol, entry 27), both of which react rapidly in radical-chain
fashion with iodonium salts, form phosphoranyl radicals that
can transfer an electron to the iodonium salt in a very exergonic
process. Of course, the phosphoranyl radical from TFP
(PhP•(OTfe)3) is a relatively poor reductant due to the same
fluorine inductive effect that makes TFP a weak nucleophile.
Thus, in the further refinement of this system for VLICP, it
might be feasible to design a relatively non-nucleophilic
phosphite co-initiator that can also form a phosphoranyl
radical capable of an exergonic electron transfer to the
iodonium salt.
Table 7. Competition Experiments: Reactivity of CHO vs
Ph3P with Aryl Radicals
a
b
b
b
entry
[CHO]
[Ph3P]
TolI
TolH
TolP+Ph3
23
24
25
26
5.0
4.0
2.8
2.5
0.10
0.10
0.10
0.11
76
100
99
16
17
12
9
64
83
89
93
99
a
CD3CN solutions of bis(4-methylphenyl)iodonium hexafluorophos-
phate 1b (0.01 M) were irradiated with a 65-W CFL for 60 min at ca.
b
30 °C. Reactant concentrations are in mol/L. NMR yields based on
the iodonium salt.
CONCLUSION
■
The combination of a trialkyl phosphite, PAIBN, and an
iodonium salt can be an effective system for visible-light
induced cationic polymerization of cyclic ethers such as CHO
and THF. When TFP is used as the phosphite, the resulting
polymers have high molecular weights and low polydispersities
and are nearly colorless, unlike other systems that make use of
dyes as photosensitizers. Other trialkyl phosphites also serve as
co-initiators but result in lower Mw polymers. TFP’s non-
which (slope = kCHO/kPh3P = 0.0059) allows the rate constant
for reaction of phenyl radicals with cyclohexene oxide to be
estimated as kCHO = 2 × 106 M−1 s−1. Using these data along
with our value for kTFP, one calculates that only about 50% of
the aryl radicals were captured by the phosphite in polymer-
ization experiments that used [TFP] = 0.1 M. Thus, the
efficiency of phenyl radical scavenging is probably the main
3566
dx.doi.org/10.1021/jo302830r | J. Org. Chem. 2013, 78, 3561−3569

The Journal of Organic Chemistry
Article
a
Table 8. Energies Determined by DFT Calculations on Phosphines (PZ3), Phosphoniums (PhP+Z3), and Phosphoranyl
Radicals (PhP•Z3)
PhP+Z3 − PhP•Z3
b
c
PZ3 HOMO (eV)
(kcal/mol)
Ered vs SCE (V)
ΔGSET (kcal/mol)
entry
PZ3
6-31G
DZVP
6-31G
DZVP
6-31G DZVP
6-31G
DZVP
d
d
27
28
29
30
31
32
33
PPh3
−5.30
−6.05
−5.63
−7.21
−6.75
−5.98
−5.49
−5.54
−6.34
−5.89
−7.62
−7.01
−6.24
−5.74
57.8
62.0
69.2
85.8
73.2
54.2
46.2
63.8
67.2
76.3
94.2
77.9
59.3
52.5
−2.17
−1.99
−1.68
−0.96
−1.51
−2.33
−2.68
−1.91
−1.76
−1.36
−0.58
−1.29
−2.10
−2.40
−34
−30
−22
−6
−19
−38
−69
−28
−24
−15
+3
P(OMe)3
PhP(OMe)2
e
P(OTfe)3
P(OC2H4Cl)3
BnOP(OEt)2
P(OSiMe3)3
−14
−32
−62
Geometries were optimized using MNDO/d.50 DFT/B3LYP calculations used either the 6-31G(d,p)51 or the DZVP52 basis set. Ered vs SCE
a
b
calculated from E(PhP+Z3) − E(PhP•Z3) by multiplying by 23.06 kcal mol−1 V−1 and subtracting the absolute voltage of the standard calomel
electrode (4.68 V).53 ΔG for single-electron transfer to diaryliodonium salt calculated from the DFT-derived phosphonium Ered and the literature
c
E1/2 for diphenyliodonium, E1/2 = −0.7 V vs SCE.42 Literature E1/2 for Ph4P+ Cl− is −1.87 V vs SCE.41 Tfe = 2,2,2-trifluoroethyl.
d
e
m/z 378 (M+), 363, 335; HRMS (ESI) m/z calcd for [C14H17F6O3P +
nucleophilic nature allows polymer chains to grow without
excessive chain transfer. The mechanism is as shown in eqs
Na]+ 401.0717, obsd 401.0725.
Photochemical Conditions. The light source was either a 500-W,
1−7. However, the reaction of TFP with iodonium salt (eq 4)
118-mm tungsten−halogen bulb (also known as a “work lamp”) or a
is slow because the radical chain reaction (eqs 1−3) is impaired
65-W (200-W equivalent) compact fluorescent lamp (CFL). Some
by a nearly thermoneutral SET propagation step (eq 1).
Alkylation of the monomer by the arylphosphonium salt
experiments used a UV-filtering sheet from Edmund Scientific placed
between the light source and the reaction tubes. This material had a
ArP+(OTfe)3 (eq 5) is responsible for starting the cationic
transmission <4% at wavelengths <400 nm. The experiments grouped
polymerization as evidenced by both the formation of aryl
in each entry used common stock solutions and were irradiated
phosphonate 2a and the 19F NMR detection of CF3CH2O end
groups on the polymer.
simultaneously on a merry-go-round apparatus.
Photopolymerization Experiments. Reaction solutions (2.0
mL) were prepared under N2 in dry 1.0-cm borosilicate test tubes
EXPERIMENTAL SECTION
capped with rubber septa. In early experiments (Table 1), the merry-
go-round apparatus was placed in a plastic container containing a
water bath held at 18 2 °C, and illumination was from three work
lamps arranged around the outside of the plastic container as described
previously.23 After irradiation, the solutions were poured into 10 mL of
5% NH4OH in methanol. The precipitated poly(cyclohexene oxide)
was vacuum filtered, air-dried, and weighed to give the conversions
reported in Table 1.
The experiments reported in Tables 2−5 used the 65-W CFL
placed in the middle of the merry-go-round apparatus. This light
source stays relatively cool so no cooling other than air circulation
provided by the merry-go-round motor was provided. A thermometer
placed on the merry-go-round with the tubes invariably registered 30−
35 °C at the end of the irradiation period. After irradiation, the
reaction solutions were immediately diluted in CDCl3 and the NMR
spectra were obtained.
■
General Methods. NMR spectra were obtained on a 300 MHz
instrument with chemical shifts referenced to solvent deuterium
signals. 19F NMR spectra were processed using backward linear
interpolation (BLIP) to remove the broad resonance due to
poly(tetrafluoroethene) in the probe. Cyclohexene oxide was filtered
through alumina and vacuum distilled from CaH2. THF was distilled
from Na/benzophenone. The iodonium salts2 and PAIBN26 were
prepared using literature methods. Iodonium salt 1b required four
recrystallizations from H2O to reduce the amount of o,p impurity to
<2%: mp 175.7−177.2 °C; lit.2 mp 169−173 °C. All other compounds
were commercial products used as received.
GPC used THF as eluant with absolute molecular weight
determination by a multiangle light scattering (MALLS) detector.
We found dn/dc = 0.114 mL/g for polyCHO. Polydispersities (Mw/
Mn) ranged from 1.3 to 1.6 according to no apparent pattern. Prior to
GPC analysis, polyCHO was purified by dissolution in CHCl3
followed by precipitation with MeOH. Semiempirical MO and DFT
calculations were carried out using Hyperchem software and a personal
computer with full details given in the Supporting Information.
Integration of the 1H and/or 31P NMR was employed to determine
relative product yields. Key 1H NMR resonances (ppm) used for yield
determination were as follows: CHO, 3.05; polyCHO, 3.4; THF, 3.65,
1.75; polyTHF, 3.3, 1.5; TFP, 4.3; 2a/2c/TFPO (overlap), 4.6; 2a,
1.36; 1a, 8.0, 1.25; ArI, 7.1−7.2, 7.5−7.6, 1.23; ArH, 1.26, 1b, 2.37;
Competition Experiments. Reactions were run in NMR tubes.
The NMR tube was either placed in a Pyrex cooling water jacket (ca.
15 °C) at a distance of 2.5 cm from the work lamp bulb (Table 6) or
rotated around the 65-W CFL on the merry-go-round apparatus
(Table 7).
ASSOCIATED CONTENT
■
−
TolI, 2.25; TolH, 2.30; TolP+Ph3 PF6 , 2.48. Key 31P NMR resonances
S
* Supporting Information
(ppm) used for yield determination were: TFP, +139; TFPO, −2; 2a,
+22; 2c, +21. When possible, yield results from 31P NMR were
checked using 1H NMR, and the results generally agreed within 2%.
2a. Bis(4-tert-butylphenyl)iodonium chloride (429 mg, 1.0 mmol)
was reacted with TFP (83 μL, 1.5 mmol) under UV irradiation in
DMSO-d6. Ether/H2O extraction and flash chromatography (alumina/
hexanes/EtOAc) gave 162 mg (43%) 2a as a viscous liquid: 1H NMR
(CDCl3) δ 1.35 (s, 9H), 4.35 (m, 2H), 4.45 (m, 2H), 7.54 (dd, J = 8.5,
4.4 Hz, 2H), 7.76 (dd, J = 14, 8.3 Hz, 2H). 13C NMR δ 31.1, 35.4 (d, J
= 1.4 Hz), 62.3 (dq, J = 5.1, 38 Hz), 121.4 (d, J = 198 Hz), 122.7 (dq,
J = 9.4, 277 Hz), 126.1 (d, J = 17 Hz), 131.9 (d, J = 12 Hz), 157.8 (d, J
= 3.6 Hz); 31P NMR δ 22.6; 19F NMR δ −75.01 (t, J = 8.5 Hz); MS
NMR spectra and GC−MS results for compound 2a.
Computational method and detailed results. This material is
available free of charge via the Internet at http://pubs.acs.org/.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: tnalli@winona.edu.
Notes
The authors declare no competing financial interest.
3567
dx.doi.org/10.1021/jo302830r | J. Org. Chem. 2013, 78, 3561−3569

The Journal of Organic Chemistry
Article
Chem. Soc. 1977, 99, 7589−7600. (b) Fu, J-J. L.; Bentrude, W. G.;
Griffin, C. E. J. Am. Chem. Soc. 1972, 94, 7717−7722.
(28) α-Scission of Ph• or RO• and β-scission of R• from
phosphoranyl radicals are slow processes. See ref 24 and: (a) Roberts,
B. P.; Scaiano, J. C. J. Chem. Soc., Perkin Trans. 2 1981, 905−911.
(b) Davies, A. G.; Parrott, M. J.; Roberts, B. P. J. Chem. Soc., Perkin
Trans. 2 1976, 1066−1071.
ACKNOWLEDGMENTS
■
We are very grateful for the suggestions and encouragement of
the late Professor Jack A. Kampmeier of the University of
Rochester. Thanks to Dr. Robert Kopitzke and Dr. Saeed Ziaee
for their assistance with the light-scattering measurements and
to Dr. Frank Saeva for helpful suggestions about this
manuscript. We acknowledge the support of the National
Science Foundation, CCLI Grant No. 0126470. We also thank
the Minnesota Chromatography Forum for an Undergraduate
Research Grant to L.G.S. and Winona State University for
Undergraduate Research Grants to K.L.V., R.H.M., T.T.S.,
J.W.W., M.J.S., and S.A.K.
(29) Chapman, S.; Kane-Maguire, L. A. P. J. Chem. Soc., Dalton Trans.
1993, 2021−2026.
(30) Tolman, C. A. J. Am. Chem. Soc. 1970, 92, 2953−2956.
(31) Lowther, N.; Crook, P.; Hall, C. D. Phosphorus, Sulfur Silicon
Relat. Elem. 1995, 102, 195−204.
(32) Literature values for OTfe groups are in the range −73 to −78
ppm. (a) Everett, T. S. In Chemistry of Organic Fluorine Compounds II;
American Chemical Society: Washington, DC, 1995; pp 1037−1086.
(b) Kubota, T.; Miyashita, S.; Kitazuma, T.; Ishikawa, N. J. Org. Chem.
1980, 45, 5052−5057.
(33) Odian, G. Principles of Polymerization, 3rd ed.; Wiley: New York,
1991; pp 541−547.
(34) Matyjaszewski, K. J. Phys. Org. Chem. 1995, 8, 197−207.
(35) Hrkach, J.; Matyjaszewski, K. J. Polym. Sci., Polym. Chem. Ed.
1995, 33, 285−298.
(36) The products and slight polymerization observed in entry 10c
come about through a radical-chain reaction of the iodonium salt with
THF to form α-THF cations capable of starting cationic polymer-
ization.7
(37) The 20% conversion to polymer observed upon 250 min of
irradiation when the azo compound was not included (entry 11b) was
due to unintended UV photolysis of the iodonium salt. When a UV
filter was used between the light source and the tubes, control
experiments like this one did not polymerize (see entries 9b and 10b,
Table 4).
(38) A reviewer suggested that monomer-derived radicals may lead to
polymerization through their reaction with the phosphine, with
subsequent electron transfer leading to a reactive phosphonium ion,
which could start polymerization either by alkylating the monomer or
by expelling TFP to form a monomer-derived oxonium ion. We
obtained no specific evidence for the formation of CHO radical dimers
and we cannot rule out the possibility that some CHO radicals
propagated by reacting with TFP, but we think that the observed
effects of lowering the monomer concentration (entry 12) and
increasing the TFP concentration (entries 5 and 8) argue strongly for
radical termination as the primary mode of reactivity for the CHO-
derived radical.
(39) The rate constant for the reaction of phenyl radicals with
acetonitrile has been reported in the literature, kMeCN ≅ 1.0 × 105 M−1
s−1. Figuring in the expected primary deuterium isotope effect, the
corresponding rate constant for acetonitrile-d3 must be on the order of
only 2 × 104 M−1 s−1. Scaiano, J. C.; Stewart, L. C. J. Am. Chem. Soc.
1983, 105, 3609−3614.
(40) No PAIBN photoinitiator was required for these reactions
(entry 16−22) because triphenylphosphine and iodonium salt form a
visible-sensitive charge-transfer complex.24
REFERENCES
■
(1) (a) Nalli, T. W.; Stanek, L. G.; Steidler, R. H.; Weidell, K. L.;
Steckler, T. T.; Wackerly, J. W.; Studler, M. J. Abstracts of Papers, 235th
National Meeting of the American Chemical Society, New Orleans,
LA, March 2008; American Chemical Society: Washington, DC, 2008;
ORGN 125. (b) Nalli, T. W.; Stanek, L. G.; Steidler, R. H.; Weidell, K.
L.; Steckler, T. T.; Wackerly, J. W.; Studler, M. J. Abstracts of Papers,
39th National Organic Symposium 2005, University of Utah, June
2005; C47. (c) Nalli, T. W.; Stanek, L. G. Abstracts of Papers, 219th
National Meeting of the American Chemical Society, San Francisco,
CA, March 2000; American Chemical Society: Washington, DC, 2000;
ORGN 405.
(2) Crivello, J. V.; Lam, J. W. Macromolecules 1977, 10, 1307−1315.
(3) Crivello, J. V.; Sangermano, M. J. Polym. Sci., Polym. Chem. 2001,
39, 343−356.
(4) Crivello, J. V. Polym. Sci., Polym. Chem. 1999, 37, 4241−4254.
(5) Sangermano, M. J.; Crivello, J. V. In Photoinitiated Polymerization;
American Chemical Society:, Washington, DC, 2003; pp 242−252.
(6) Fouassier, J. P. In Photoinitiation, Photopolymerization and
Photocuring: Fundamentals and Applications; Hanser: Munich, 1995.
(7) Kampmeier, J. A.; Nalli, T. W. J. Org. Chem. 1994, 59, 1381−
1388.
(8) Ledwith, A. Makromol. Chem., Suppl. 1979, 3, 348−358.
(9) Ledwith, A. Polymer 1978, 19, 1217−1219.
(10) Crivello, J. V. Adv. Polym. Sci. 1984, 62, 1−48.
(11) He, J.-H.; Mendoza, V. S. J. Polym. Sci., Polym. Chem. 1996, 34,
2809−2816.
(12) Scranton, A. B.; Moorjani, S. K.; Sirovatka, K. J.; Nelson, E. W.
Polym. Mater. Sci. Eng. 1996, 75, 192−193.
(13) Yonet, N.; Bicak, N.; Yagci, Y. Macromolecules 2006, 39, 2736−
2738.
(14) Durmaz, Y. Y.; Moszner, N.; Yagci, Y. Macromolecules 2008, 41,
6714−6718.
(15) Yagci, Y.; Hepuzer, Y. Macromolecules 1999, 32, 6367−6370.
(16) Hua, Y.; Crivello, J. V. Macromolecules 2001, 34, 2488−2494.
(17) Aydogan, B.; Durmaz, Y. Y.; Kahveci, M. U.; Uygun, M.;
Tasdelen, M. A.; Yagci, Y. Macromol. Symp. 2011, 308, 25−34.
(18) Tasdelen, M. A.; Kumbaraci, V.; Jockusch, S.; Turro, N. J.;
Talinli, N.; Yagci, Y. Macromolecules 2008, 41, 295−297.
(19) Chen, S.; Cook, W. D.; Chen, F. Macromolecules 2009, 42,
5965−5975.
(41) White, C. K.; Rieke, R. D. J. Org. Chem. 1983, 43, 4638−4641.
(42) There has been disagreement about the proper value of the half-
wave reduction potential to use for Ph2I+, with many work-
ers3,10,13,14,16,18,20 continuing to use the value first reported by
Beringer in 1958, E1/2 = −0.2 V.43 However, five much more recent
papers have reported measurements of E1/2 in the −0.6 to −1.0 V
range.44−48 In addition, Previtali and co-workers49 have argued for the
−0.7 V value (used herein) based on the triplet quenching rates of a
series of photosensitizers by diphenyliodonium chloride. We think the
recent measurements along with the Previtali observations make it
clear that the true value is much closer to −0.7 V than it is to −0.2 V.
(43) Bachofner, H. E.; Beringer, F. M.; Meites, L. J. Am. Chem. Soc.
1958, 80, 4274−4278.
(20) Bi, Y.; Neckers, D. C. Macromolecules 1994, 27, 3683−3693.
(21) Crivello, J. V. J. Polym. Sci., Polym. Chem. 2009, 47, 866−875.
́
(22) Lalevee, J.; El-Roz, N.; Allonas, X.; Fouassier, J. P. J. Polym. Sci.,
Polym. Chem. 2008, 46, 2008−2014.
(23) Muneer, R.; Nalli, T. W. Macromolecules 1998, 31, 7976−7979.
(24) Kampmeier, J. A.; Nalli, T. W. J. Org. Chem. 1993, 58, 943−949.
(25) Ptitsyna, O. A.; Pudeeva, M. E.; Reutov, O. A. Dokl. Akad. Nauk
SSSR 1965, 165, 582−585.
(26) Ford, M. C.; Rust, R. A. J. Chem. Soc. 1958, 1297−1298.
(27) Rate constants reported in the literature for phenyl radical
additions to P(III) compounds are invariably in the 108 M−1 s−1 range.
(a) Kryger, R. G.; Lorand, J. P.; Stevens, N. R.; Herron, N. R. J. Am.
(44) Fung, S.; Moratti, S. C.; Graham, S. C.; Friend, R. H. Synth. Met.
1999, 102, 1167−1168.
3568
dx.doi.org/10.1021/jo302830r | J. Org. Chem. 2013, 78, 3561−3569

The Journal of Organic Chemistry
Article
(45) Kunz, A.; Muller, U.; Fouassier, J.-P.; Morlet-Savary, F. J.
̈
Photochem. Photobiol. A: Chemistry 1997, 110, 115−122.
(46) Bruce, A. C.; Klein, J. J.; Sahyun, M. R. V. J. Photochem.
Photobiol. A: Chemistry 2000, 131, 27−31.
(47) Vase, K. H.; Holm, A. H.; Pedersen, S. U.; Daasbjerg, K.
Langmuir 2005, 21, 8085−8089.
(48) Weissmann, M.; Baranton, S.; Coutanceau, C. Langmuir 2010,
26, 15002−15009.
́
(49) Gomez, M. L.; Montejano, H. A.; Previtali, C. M. J. Photochem.
Photobiol. A: Chemistry 2008, 197, 18−24.
(50) Thiel, W.; Voityuk, A. A. J. Phys. Chem. 1996, 100, 616−626.
(51) Petersson, G. A.; Al-Laham, M. A. J. Chem. Phys. 1991, 94,
6081−6090.
(52) Godbout, N.; Salahub, D. R.; Andzelm, J.; Wimmer, E. Can. J.
Chem. 1992, 70, 560−571.
(53) Trasatti, S. Pure Appl. Chem. 1986, 58, 955−966.
3569
dx.doi.org/10.1021/jo302830r | J. Org. Chem. 2013, 78, 3561−3569