Controlled nitroxide-mediated radical polymerization of methyl and phenyl vinyl ketone

Article abstract of DOI:10.1002/pola.25997

The article describes unprecedented nitroxide-mediated radical polymerization of methyl and phenyl vinyl ketone (MVK and PVK) using a sterically highly hindered alkoxyamine as initiator/regulator. It is shown that controlled polymerization of PVK is far more difficult to achieve than controlled MVK polymerization. Whereas for MVK high conversion resulting in polyvinyl ketone with low polydispersity index is readily obtained, the PVK polymerization provides good results only in the presence of free nitroxide and styrene as additives. Vinyl ketone polymerizations are analyzed by ESI mass spectrometry. These MS studies provide insights into the problems associated with the controlled nitroxide-mediated polymerization of PVK. 2012 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem, 2012 This article describes nitroxide-mediated polymerization (NMP) of methyl vinyl ketone (MVK) and phenyl vinylketone (PVK) using alkoxyamine 1 as initiator/regulator. NMP of MVK occurs well controlled. However, for controlled NMP of PVK, styrene and free nitroxide are necessary as additives. Mass spectrometric analysis is used to investigate the polymerization process. Copyright

Full text of DOI:10.1002/pola.25997

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Controlled Nitroxide-Mediated Radical Polymerization of Methyl and
Phenyl Vinyl Ketone
Johannes A. M. Hepperle, Heinrich Luftmann, Armido Studer
Department of Chemistry, Institute of Organic Chemistry, University of Munster, Corrensstraße 40, Munster 48149, Germany
¨
¨
Correspondence to: A. Studer (E-mail: studer@uni-muenster.de)
Received 20 December 2011; accepted 8 February 2012; published online 5 March 2012
DOI: 10.1002/pola.25997
ABSTRACT: The article describes unprecedented nitroxide-medi-
ated radical polymerization of methyl and phenyl vinyl ketone
(MVK and PVK) using a sterically highly hindered alkoxyamine
as initiator/regulator. It is shown that controlled polymerization
of PVK is far more difficult to achieve than controlled MVK po-
lymerization. Whereas for MVK high conversion resulting in
polyvinyl ketone with low polydispersity index is readily
obtained, the PVK polymerization provides good results only in
the presence of free nitroxide and styrene as additives. Vinyl
ketone polymerizations are analyzed by ESI mass spectrome-
try. These MS studies provide insights into the problems asso-
ciated with the controlled nitroxide-mediated polymerization of
C
V
PVK. 2012 Wiley Periodicals, Inc. J Polym Sci Part A: Polym
Chem 50: 2150–2160, 2012
KEYWORDS: alkoxyamines; ESI-MS spectrometry; living poly-
merization; nitroxide-mediated polymerization; polyvinyl
ketones
INTRODUCTION Polyvinyl ketones belong to the class of pho-
todegradable polymers that are often used as components for
packings.1 Moreover, they have also found applications as
functional materials.2 These polymers are generally prepared
with high polydispersities by conventional (uncontrolled) free
radical polymerization.3 During the past decade, controlled liv-
ing radical polymerization has been intensively studied and
several new methods have been developed: atom transfer rad-
ical polymerization (ATRP),4 RAFT (reversible addition–frag-
mentation chain transfer polymerization),5 I-group transfer po-
lymerization,6 Te, Sb and Bi-group transfer polymerization,7
Cobalt-mediated polymerization,8 and nitroxide-mediated poly-
merization (NMP)9 have been used for the preparation of
polymers with well-defined molecular weights and polydisper-
sities below the theoretical limit [polydispersity index (PDI) <
1.5]. However, Mittal et al. showed that polymerization of
methyl vinyl ketone (MVK) by ATRP is not possible due to the
fact that the monomer coordinates to the copper catalyst.10 In
contrast to ATRP, no metals are used in NMP.11 To the best of
our knowledge, there are only two reports on successful con-
trolled living radical polymerization of MVK either using RAFT
methodology or our recently introduced boron group transfer
process.12 Herein, we report preparation of well-defined poly-
vinyl ketones using NMP (Scheme 1). We will apply the
recently introduced alkoxyamine 1 which was shown to be an
excellent initiator/regulator for NMP of styrene, acrylates, and
acrylamides.13
RESULTS AND DISCUSSION
Polymerization Studies
Polymerizations with alkoxyamine 1 were performed in
sealed Schlenk-tubes with vinyl ketones dissolved in benzene
at 100ꢀC (MVK:benzene ¼ 1.7:1). For MVK, the alkoxyamine
concentration was systematically varied from 0.125% to 1%
and results obtained are summarized in Table 1. Polymeriza-
tion of MVK with 1 in benzene at 100ꢀC occurred efficiently
and well controlled (Entries 1–23). A conversion of 57% was
achieved in 8 h using 1 mol % initiator loading and poly-
methyl vinyl ketone (PMVK) with a narrow PDI (1.08) was
isolated. PMVK with larger molecular weight resulted by
reducing the alkoxyamine 1 loading from 1 mol % to 0.125
mol % (Entries 4, 8, 12, 16). At low initiator loading conver-
sion slightly decreased and polymerization remained well
controlled (PDI < 1.12). These experiments were repeated
by running polymerizations for 6, 4, and 2 h (Table 1,
Entries 1–3, 5–7, 9–11, 13–15). By lowering the initiator
loading conversion and PDI slightly decreased. On increasing
the polymerization time to 48 h (using 0.5 mol % initiator
1), the conversion was increased to 73% and narrow PDI
(1.08) was preserved (Entry 23).
We also tried styryl-TEMPO (Fig. 1) as initiator/regulator for
the polymerization of methyl vinyl ketone (MVK in benzene,
100ꢀC for 4 h), but no polymerization occurred and only ini-
tiator could be isolated. As for NMP of acrylates and
C
V
2012 Wiley Periodicals, Inc.
2150
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 2150–2160

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
for a NMP process which is controlled by the persistent radi-
cal effect (PRE).9b,c,e
The living character of the MVK polymerization was further
investigated by using a PMVK macroalkoxyamine as initia-
tor/regulator for subsequent styrene polymerization leading
to a PMVK-block-polystyrene diblockcopolymer. To this end,
MVK was polymerized at 100ꢀC using alkoxyamine 1 (0.5
mol %) for 3 h in benzene. The macroalkoxyamine was iso-
lated in 28% yield with a low PDI (1.18). The resulting
PMVK macroalkoxyamine (Mn ¼ 6700 g/mol) was then used
as an initiator for polymerization of styrene (neat styrene at
100ꢀC for 3 h) and PMVK-block-polystyrene diblockcopoly-
mer obtained was analyzed by size exclusion chromatogra-
phy (SEC). The PDI decreased in the copolymer (1.08) and
the molecular weight increased to 14,300 g/mol as expected
for a successful copolymerization (Fig. 3).
SCHEME 1 Nitroxide-mediated polymerization of vinyl ketones
with the sterically hindered alkoxyamine 1.
acrylamides more sophisticated nitroxides are obviously
necessary to control NMP of MVK.
We analyzed the controlled character of the 1-mediated MVK
polymerization by determining the conversion as a function
of time and by analyzing the molecular weight as a function
of monomer conversion [Fig. 2(a,b)]. Polymerizations were
run using 0.5 mol % initiator 1 (experimental data are
shown in Table 1, Entries 9–12 and 17–23). All plots showed
the typical behavior expected for a controlled living radical
process. In addition, the plot of ln([M]0/[M]) versus t2/3
depicted in Figure 2(c) showed the linear character expected
In addition, the living character was proved by using a
PMVK as macroalkoxyamine for the subsequent polymeriza-
tion of n-butylacrylate. To this end, MVK was first polymer-
ized at 100ꢀC using alkoxyamine 1 (0.5 mol %) for 7 h in
benzene. The PMVK isolated in 45% yield with a PDI of 1.10
(Mn ¼ 10,600 g/mol) was subsequently used as initiator/
regulator for the polymerization of n-butylacrylate (neat n-
butylacrylate, 125ꢀC for 4 h). The resulting PMVK-block-poly
n-butylacrylate was analyzed by SEC. We found that for the
diblockcopolymer the mean molecular weight increased to
116,700 g/mol while the PDI decreased (1.07).
TABLE 1 NMP of MVK in Benzene at 1008C—Variation of
Initiator Loading and Polymerization Time
To get an idea about the accuracy of the SEC-determined Mn
values for PMVK using a PMMA standard, we analyzed small
molecular weight PMVK also by mass spectrometry for com-
parison. For that analysis, a PMVK sample was prepared by
polymerization of MVK in benzene at 100ꢀC for 90 min using
0.5 mol % alkoxyamine 1 as initiator and was analyzed by
SEC and MALDI-MS spectrometry with DHB as matrix (Fig. 4
shows the MALDI-MS spectrum). The theoretical molecular
weight (Mn,th ¼ 2900 g/mol) was calculated and the experi-
mental molecular weight determined by MALDI-MS spec-
trometry was found to lie at 3300 g/mol (Mn,MALDI). SEC on
the same sample revealed a Mn of 4900 g/mol (Mn,SEC). Since
it is known that the MALDI-MS technique notoriously pro-
vides too low Mn values due to ‘‘cutting’’ of the polymer
chains with larger molecular weights, we estimate the cor-
rect Mn value of our PMVK to be in between the MALDI-MS
Initiator Time Conversion Mn,th
Mn,exp
Entry (mol %) (h)
(%)
(g/mol) (g/mol) PDI
1
0.125
0.125
0.125
0.125
0.25
0.25
0.25
0.25
0.5
2
4
16
21
29
39
24
31
35
40
23
31
38
49
32
44
51
57
16
29
33
46
52
58
73
9,600
12,300
16,800
23,300
6,800
8,800
9,600
11,800
3,200
4,700
5,600
6,700
2,400
3,200
3,700
4,400
2,200
4,000
4,900
6,800
7,500
8,600
10,500
10,800
18,300
23,900
31,000
9,800
14,400
16,300
18,400
5,500
7,800
8,700
9,700
3,900
4,700
5,800
6,800
4,600
6,900
8,200
9,100
10,700
13,100
16,100
1.17
1.11
1.07
1.07
1.18
1.12
1.10
1.09
1.22
1.16
1.14
1.11
1.21
1.16
1.13
1.12
1.23
1.16
1.13
1.11
1.09
1.08
1.08
2
3
6
4
8
5
2
6
4
7
6
8
8
9
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
0.5
4
0.5
6
0.5
8
1
2
1
4
1
6
1
8
0.5
1
0.5
3
0.5
5
0.5
7
0.5
12
20
48
FIGURE 1 Initiators/regulators used for the controlled polymer-
ization of MVK (the sterically highly hindered alkoxyamine 1
and styryl-TEMPO).
0.5
0.5
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 2150–2160
2151

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 3 SEC-chromatograms of a PMVK (a) (MVK in ben-
zene, 100ꢀC, 0.5 mol % 1) and a diblockcopolymer PMVK-b-PS
(b) after reinitiation (neat styrene, 100ꢀC).
try.14 To this end, polymerization was conducted in neat
monomer (100ꢀC, 1 mol % 1) and was stopped after 30 min
at low conversion targeting low-molecular weight PMVK suit-
able for mass spectrometry. After removing residual mono-
mer, the crude product was analyzed by ESI mass spectrom-
etry. Figure 5(a) shows a part of the MS-spectrum of a low-
molecular weight PMVK. The four most prominent peak
groups revealed the living nature of the 1-mediated MVK po-
lymerization. Peaks corresponding to polymers bearing a sty-
ryl moiety as the initiating unit and the nitroxide moiety as
chain terminating unit were identified (see structure in
Scheme 1, R ¼ Me). The peak series at 988.1077 correspond
to a double Naþ-adduct containing 22 MVK monomers. The
simulated spectrum in Figure 5(c) perfectly matches the
measured high-resolution mass spectrum. The peak series at
985.6309 represent the same polymer (22 MVK entitieþs)
which is doubly charged with one Naþ-ion and one NH4
-
ion. This assignment is supported by the calculated spectrum
depicted in Figure 5(e). The peak series at 971.6343 corre-
spond to the mono Naþ-adduct of the polymer bearing seven
MVK moieties. Again, the simulated spectrum in Figure 5(b)
shows a perfect match. The fourth prominent peak series at
FIGURE 2 (a) Monomer conversion versus time and PDI evolu-
tion versus time (MVK in benzene, 100ꢀC, 0.5 mol % 1; trian-
gles
¼
PDI; squares
¼
ln([M]0/[M])). (b) Molecular weight
versus monomer conversion (MVK in benzene, 100ꢀC, 0.5 mol
% 1). (c) Molecular weight versus t2/3 (MVK in benzene, 100ꢀC,
0.5 mol % 1).
und the SEC results. Hence, the SEC data reported for PMVK
throughout this article are likely overestimated values.
To further prove the living character of the MVK polymeriza-
tion, PMVK was additionally analyzed by ESI-MS spectrome-
FIGURE 4 MALDI-MS spectrum of
PMVK (MVK in benzene, 100ꢀC, 0.5 mol % 1).
a low-molecular weight
2152
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 2150–2160

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 5 ESI–MS spectrum of a low molecular weight PMVK sample: (a) part of the ESI mass spectrum of the sample (neat MVK,
100ꢀC, 1 mol % 1, 0.5 h polymerization time). (b–e) measured (above) and calculated (below) spectra of distinct polymer species in
different charge states: (b) Naþ-adduct with 7 MVK moieties. (c) 2 Naþ-adduct (doubly charged) with 22 MVK moieties. (d) 3 Naþ-
adduct (triply charged) with 35 MVK moieties. (e) Naþ/NH4þ-adduct (doubly charged) with 22 MVK moieties.
969.9162 belongs to a triply charged polymer (3 Naþ-
adduct) containing 35 MVK entities. The assignment is fur-
ther supported by the calculated spectrum presented in Fig-
ure 5(d). Hence, all the prominent peaks could be assigned
to living polymers bearing the alkoxyamine moiety at the
terminus of the polyketone and a styryl unit as initiating
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 2150–2160
2153

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 2 NMP of PVK in Benzene at 1008C in the Presence of 2
tiator 1 (experimental data are summarized in Table 2,
(0.025 mol %)—Variation of Polymerization Time
Entries 2–7).
We assumed that the low conversion might be a result of the
instability of the polymeric alkoxyamine. Hydroxylamine
elimination might lead to dead polymer chain ends (Scheme
2).9 In contrast to the polymerization of MVK, for PVK the
resulting a,b-unsaturated ketone is further stabilized via con-
jugation with the phenyl moiety. Such a competing hydroxyl-
Initiator Time Conversion Mn,th
Mn,exp
Entry (mol %) (h)
(%)
(g/mol) (g/mol) PDI
1a
2
0,5
0.5
0.5
0.5
0.5
0.5
0.5
0,5
0,5
0,5
0,5
1
66
18
21
24
27
28
35
38
35
56
17,700
4,200
5,100
6,500
7,500
8,200
9,200
10,300
9,400
13,500
11,700
6,500
1.55
1.11
1.12
1.10
1.11
1.13
1.12
1.15
1.18
1.09
3
2
7,200
amine fragmentation is
methacrylate.
a problem in NMP of methyl
4
3
7,600
5
4
9,300
6
5
9,400
An elegant solution to that problem was provided by Char-
leux et al.17 Addition of a small amount of styrene to MMA
ensures the ultimate monomer moiety in the polymer to be
a styrene unit, which in the dormant state (as alkoxyamine)
is not prone to undergo hydroxylamine fragmentation. We
therefore added 5% of styrene with respect to PVK and
indeed yield was increased to 56%. Molecular weight
increased to 14,700 g/mol and PDI decreased to 1.09 (Entry
10). Unfortunately, longer reaction times did not provide
polymers with higher molecular weights.
7
6
11,400
10,900
10,400
14,700
8
8
9
10b
24
6
a
In the absence of 2.
5% Styrene was added to the reaction mixture.
b
entity. This showed that polymerization (at least at low con-
version) occurred with a high degree of livingness and that
all chains were initiated by the alkoxyamine initiator 1.
Importantly, peaks derived from olefin terminated polymers,
which might result from hydroxylamine elimination, were
not detected in the ESI mass spectrum.
To get a better picture associated with problems of NMP of
PVK at higher conversion, we decided to analyze the poly-
merization process using ESI mass spectrometry. To this
end, we targeted low-molecular weight polymers that are
suitable to MS analysis (PVK:1 ¼ 4:1 in benzene at 100ꢀC).
Although peaks were not very prominent, we could unam-
biguously identify the Naþ-adducts of oligomers lacking the
hydroxylamine moiety (see Scheme 2). Note that we did not
see the corresponding peaks in the oligomer mixture
obtained in the polymerization of MVK. This shows that the
tendency for hydroxylamine elimination in a macroalkoxy-
amine is more pronounced for PPVK-2 as compared with
PMVK-2. Along with the disproportionation of the macrorad-
ical with the nitroxide leading to the hydroxylamine, we felt
that the PPVK-macroradical might further react via other
decomposition pathways. We therefore decided to investi-
gate the oligomers by MS–MS-analysis. Our group recently
reported on successful ESI–MS sequence analysis of oligom-
ers prepared by NMP.14c Such an analysis will give us infor-
mation regarding the fate of the PPVK-macroradicals in the
gas phase. The MS–MS spectra were recorded on an orbitrap
mass spectrometer which provides precise and high-resolu-
tion masses. In the MS–MS experiments, we used CID (colli-
sion-induced decomposition) by the specific excitation of
precursor ions to produce the corresponding daughter ions
(Fig. 8).
After successful polymerization of MVK, we decided to test
whether alkoxyamine 1 also allows for controlled NMP of
aryl vinyl ketones. We chose phenyl vinyl ketone (PVK) as a
test monomer and first experiments were carried out neat
or in benzene as a cosolvent at 100ꢀC. We found that poly-
merization of PVK was far more difficult to control than po-
lymerization of MVK. Polymerization of PVK under neat con-
ditions or in benzene occurred fast and not well controlled.
Reaction mixture became highly viscous in less than 30 min
and SEC-analysis revealed a PDI of 1.55, indicating that poly-
merization was not controlled (Table 2, Entry 1). It seemed
that kp for the polymerization of PVK is too large and the
concentration of free nitroxide is too small to control that
process. This is reminiscent to NMP of acrylates. To solve
that problem, we added a small amount of free nitroxide 2
(0.025 mol %, Fig. 6).15 Addition of free nitroxide ensures a
sufficient high concentration of the persistent radical which
should lead to a better control. This is well established for
NMP of methyl acrylate.9,16 As expected, polymerization in
the presence of 2 was slower and reaction delivered PPVK
with a narrow PDI (Entry 2). Conversion was increased to
38% on increasing reaction time to 8 h (Entries 3–7). How-
ever, further extension of reaction time did not allow for
preparation of PPVK with higher molecular weight (Entries
8–9). At around 35–40% conversion, reaction stopped. Poly-
merization seemed to be well controlled at low conversion
as shown by plotting the monomer conversion as a function
of time [Fig. 7(a)] and the molecular weight as a function of
monomer conversion [Fig. 7(b)]. Moreover, the linear relation
of ln([M]0/[M]) to t2/3 alluded to a controlled NMP [Fig.
7(c)]. These polymerizations were run using 0.5 mol % ini-
FIGURE 6 Nitroxide used for controlled polymerization of PVK.
2154
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 2150–2160

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
a. Gas-phase depolymerization by PVK fragmentation to
form the corresponding macroradicals (Scheme 3) identi-
fied as Naþ-adducts (m/z ¼ 1316, 1184, 1052, 920, 788,
656, 524; in Figure 8 the corresponding peaks are marked
with $), and
b. 1,5-H transfer to give the corresponding tertiary radicals
which can further react via two pathways (Scheme 4).
Fragmentation according to pathway 1 affords a macro-
radical, along with a trimer with m/z ¼ 419 designated
as l in Figure 8. Fragmentation via pathway 2 provides
an olefin terminated polymer designated as n with m/z ¼
1065, 933, 801, 669, 537 (see Scheme 4).
We assume that the chemistry which occurs in the gas phase
is also occurring in solution. Hence, these gas-phase experi-
ments gave important hints regarding possible competing
reactions. Besides the disproportionation of the nitroxide
with the macroradical we believe that the 1,5-H transfer in
the macroradical occurs in the NMP of PVK as a major side
reaction. We found that the larger the polymer the more in-
tensive the corresponding follow-up product peaks appeared
in the corresponding MS–MS spectra [compare Fig. 8(a–c)].
Since the rate for the 1,5-H transfer should not heavily
depend on the size of the oligomer, we believe that the 1,5-H
transfer likely already occurred during polymerization in so-
lution leading after nitroxide trapping to tertiary alkoxy-
amines. In the MS analysis during mass selection, we cannot
distinguish between two alkoxyamines of the same mass
where the nitroxide moiety is terminally or internally
located. Internal alkoxyamines will lead to the same gas-
phase MS-analysis (Scheme 5). Tertiary alkoxyamines are
more prone to undergo hydroxylamine elimination leading to
dead polymer chains.
EXPERIMENTAL
Materials
All reactions were carried out in heat-gun-dried glassware
under an argon atmosphere and were performed by using
standard Schlenk techniques. 1H NMR and 13C NMR spectra
were recorded on a Bruker DPX-300 (1H: 300 MHz, 13C: 75
MHz), a Varian Inova 500 (1H: 500 MHz, 13C: 125 MHz), or
Varian Unity plus 600 (1H: 600 MHz, 13C: 150 MHz). Chemi-
cal shifts d in ppm are referenced to the solvent residual
peak. Thin-layer chromatography was carried out on Merck
silica gel 60 F254 plates; detection by UV or dipping into a
solution of KMnO4 (1.5 g), NaHCO3 (5.0 g) in H2O (400 mL)
followed by heating. Flash chromatography (FC) was carried
FIGURE 7 (a) Monomer conversion versus time and PDI evolution
versus time (PVK in benzene, 100ꢀC, 0.5 mol % 1, 0.025 mol % free
nitroxide 2; triangles ¼ PDI; squares ¼ ln([M]0/[M]). (b) Molecular
weight versus monomer conversion (PVK in benzene, 100ꢀC, 0.5 mol
% 1, 0.025 mol % free nitroxide 2). (c) Molecular weight versus t2/3
(PVK in benzene, 100ꢀC, 0.5 mol % 1, 0.025 mol % free nitroxide 2).
All oligomers identified showed as first reaction in the gas
phase the homolytic CAO bond cleavage liberating nitroxide
2 and the corresponding macroradical. The macroradical
then further reacted via two different pathways:
SCHEME 2 Possible hydroxylamine elimination which leads to
dead polymer chain ends.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 2150–2160
2155

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 8 MS–MS spectra of a low molecular weight PVK oligomer. CID fragmentations of a PPVK oligomer (a) m/z ¼ 1071 Naþ-
addition, (b) m/z ¼ 1336 Naþ-addition, and (c) m/z 1601 Naþ-addition (stars ¼ depolymerization; squares ¼ 1,5-H transfers; circles
¼ 1,5-H transfer followed by elimination of the initiating styrene moiety).
out on Merck or Fluka silica gel 60 (40–63 lm) at about 1.5
bar. IR spectra were recorded on a Varian 3100 FT-IR
equipped with a MKII Golden Gate Single Reflection ATR
unit. ESI–MS (m/z) and HRMS (m/z) were performed using
a Bruker MicroTof (loop injection; resolution: 10,000), a LTQ
Orbitrap XL (nanospray inlet, 1.1 KV, resolution: 30,000),
and an Autoflex Speed TOF-MS (Bruker Daltonics). SEC was
carried out with degassed THF as eluent at a flow rate of 1.0
mL/min at room temperature on a system consisting of a
Smartline Pump 1000 (Knauer), a set of two PLgel 5 lm
MIXED-C columns (300 ꢁ 7.5 mm, Polymer Laboratories)
and a Knauer RI differential refractometer detector. Data
2156
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 2150–2160

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
distilled from K/Na, benzene was freshly distilled from Na
and CH2Cl2 was distilled from P2O5.
Mass Spectrometry
ESI spectra to prove livingness of the polymerization were
recorded on a LTQ Orbitrap XL (Thermo Scientific) with CID
and HCD capabilities. The resolution of the Orbitrap was R ¼
30,000 routinely, 100,000 if necessarily. Spectra are averaged
to improve the signal to noise ratio. Samples (0.2 mg) were
dissolved in 1 mL methanol or chloroform/methanol (1:1).
To improve the sodiation, 10 lL of a saturated methanolic
NaBF4 solution was added. All samples were introduced by
static nanospray with internal electrical contact. The capil-
lary voltage was adjusted to provide a stable spray (0.8–1.4
kV). MALDI spectra were recorded with an Autoflex Speed
TOF-MS (Bruker Daltonics) in linear mode. Samples were
prepared with 2,5-DHB containing NaBF4 to promote sodia-
tion. The resolution in this mode is not sufficient to separate
isotopes. Therefore, m/z is calculated with stoichiometric
values and peak intensities were used to calculate the
distribution.
1-tert-Butyl-3,3,5,5-tetraethyl-4-
(1-phenylethoxy)-piperazin-2-on (1)
Alkoxyamine 1 and the corresponding nitroxide 2 were pre-
pared according to our recently published procedure.13a
1-Phenylpropenone (2)
According to Martı´n-Matute et al.,18 a mixture of 3-chloro-1-
phenyl-1-propanone (10 g, 59 mmol) and potassium acetate
(6.41 g, 65 mmol) in ethanol (500 mL) was stirred under
reflux for 3 h. After cooling to room temperature, ethanol
was removed under reduced pressure. The residue was dis-
solved in EtOAc (200 mL) and washed with H2O (3 ꢁ 150
mL). The organic phase was dried over MgSO4, filtered,
and the solvent was removed in vacuo. Purification by FC
on silica gel (pentane:EtOAc: 20:1) afforded a slightly yellow
oil (5.4 g, 69%). The physical data are in agreement with
those reported in the literature: 1H NMR (CDCl3, 300 MHz):
d 7.78 (d, J ¼ 7.0 Hz 2 H), 7.42–7.27 (m, 3 H), 6.99 (dd,
J ¼ 17.1, 10.6 Hz, 1 H), 6.26 (d, J ¼ 17.1, H), 5.75 (d, J ¼
10.6, H).19
SCHEME 3 CAO bond homolysis and depolymerization.
were analyzed with PSS WinGPC Compact V.7.20 software
(Polymer Standards Service) based on calibration curves
built upon poly(methyl methacrylate) standards (Polymer
Laboratories Poly(methyl methacrylate) Medium MW Cali-
bration Kit M-M-10 to determine the molecular weight of
polyketones) with peak molecular weights ranging from
1660 to 1,000,000 g/mol. The following chemicals were pur-
chased and used as received: acetic acid (Acros, 99.8%), am-
monia solution (Acros, 25% in water), (1-bromoethyl)ben-
zene (Alfa Aesar, 97%), 3-bromopentane (Sigma–Aldrich,
95%) chloroform (Fisher Scientific, ꢂ99%), copper powder
(Sigma–Aldrich, <10 microns, 99%), copper(II) trifluorome-
thanesulfonate (Alfa Aesar, 98%), dimethyl sulfoxide (Acros,
Typical Procedure for the Polymerization of MVK
A heat-gun dried Schlenk tube was charged with initiator 1,
MVK, and benzene. The tube was subjected to three freeze–
thaw cycles and then sealed. The polymerization was carried
out under argon at 100ꢀC for 1 h to 24 h. The resulting mix-
ture was cooled to room temperature and dissolved in
CH2Cl2 (1 mL). Solvent and residual monomer were removed
under reduced pressure to afford the polymer. Conversion
was determined gravimetrically; molecular weight and PDI
were determined by SEC.
99.8%),
4,40-di-tert-butyl-2,20-bipyridyl
(Sigma–Aldrich,
98%), ethyl acetate (Acros, 99.6%), formaldehyde (Acros,
37% solution in water, stabilized with 10–15% methanol), 4-
heptanone (Acros, 98%), 3-pentanone (Sigma–Aldrich, 98%),
peroxyacetic acid (Acros, 35% solution in diluted acetic
acid), potassium hydroxide (Fluka, powder, ꢂ90%), 3-nitro-
propane (Acros, 96%), sodium nitrate (Acros, ꢂ98.5%), tert-
butylamine (Acros, 99%), zinc powder (Acros, 99.99%, 40
mesh), 3-chlorophenylpropanone (Fluka, ꢂ97%).
Typical Procedure for the Polymerization of PVK
A heat-gun dried Schlenk tube was charged with initiator 1,
nitroxide 2, PVK, and benzene. The tube was subjected to
three freeze–thaw cycles and then sealed. The polymeriza-
tion was carried out under argon at 100ꢀC for 1 to 8 h. After
the reaction mixture was allowed to cool to room
MVK (Alfa Aesar, tech. 90%, stabilized with 0.5% hydrochi-
none) was distilled under reduced pressure to remove the
stabilizer. THF was freshly distilled from K, Et2O was freshly
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 2150–2160
2157

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 4 1,5-H transfer and follow up gas-phase chemistry by two different pathways investigated by MS-MS analysis (in path-
way 1 both fragments can appear as sodiated ions in equal probabilities).
temperature, the residue was dissolved in dichloromethane
(1 mL). The polymer was precipitated three times on addi-
tion of a 4:1 mixture of pentane/DCM (5 mL) to afford poly-
phenyl vinyl ketone as a white solid. Molecular weight and
PDI were determined by SEC; conversion was determined
gravimetrically.
ing radical polymerization process of MVK. Peak series in
the mass spectrum showed that all polymers identified are
dormant and bear a nitroxide moiety. Moreover, successful
use of such a PMVK polymer as macroinitiator for the forma-
tion of block copolymers further supported the high degree
of livingness of the polymerization.
Polymerization of PVK was far more difficult to control. Ac-
ceptable results were achieved only on adding free nitroxide
2 (0.025 mol %) and 5 mol % of styrene. Using these addi-
tives, it was possible to obtain polyphenyl vinyl ketone in a
yield of 56% with a narrow PDI. Conversion and yield were
not increased upon increasing the reaction time. ESI–MS and
MS–MS studies revealed that hydroxylamine elimination
CONCLUSIONS
We showed for the first time that controlled NMP of MVKs is
possible using alkoxyamine 1 as an intiator/regulator. This
alkoxyamine is readily prepared in a large scale.13a Polymer-
izations occurred well controlled delivering PMVKs with nar-
row PDIs. ESI–MS spectrometry provided evidence for a liv-
2158
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 2150–2160

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 5 CAO-bond homolysis of the tertiary alkoxyamine and follow up gas-phase chemistry by two different pathways inves-
tigated by MS–MS analysis (in pathway 1 both fragments can appear as sodiated ions in equal probabilities).
REFERENCES AND NOTES
leading to dead polymers is a problem in NMP of PVK. In
contrast to MVK, the a,b-unsaturated ketones formed on hy-
droxylamine elimination are further stabilized by conjugation
into the phenyl substituent. MS-studies also revealed that
1,5-H transfer in the macroradical is a competing reaction.
That H-transfer can lead to chain termination as investigated
by MS–MS experiments.
1 (a) Golemba, F. J.; Guillet, J. E. Macromolecules 1972, 5,
212–216; (b) Guillet, J. E. Macromol. Symp. 1997, 123, 209–224.
2 (a) Green, O.; Smith, N. A.; Ellis, A. B.; Burstyn, J. N. J. Am.
Chem. Soc. 2004, 126, 5952–5953; (b) Sugita, K. Prog. Org.
Coat. 1997, 31, 87–95.
3 (a) Guillet, J. E.; Norrish, R. G. W. Nature 1954, 173, 625–627;
(b) Lyons, A. R. J. Polym. Sci.: Macromol. Rev. 1972, 6,
251–293; (c) Marvel, C. S.; Levesque, C. L. J. Am. Chem. Soc.
1938, 60, 280–284.
The authors thank the WWU Mu¨nster for financial
support.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 2150–2160
2159

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
4 (a) Matyjaszewski, K.; Xia, J. Chem. Rev. 2001, 101,
2921–2990; (b) Kamigaito, M.; Ando, T.; Sawamoto, M. Chem.
Rev. 2001, 101, 3689–3746; (c) Tsarevsky, N. V.; Matyjaszewski,
K. Chem. Rev. 2007, 107, 2270–2299; (d) Ouchi, M.; Terashima,
T.; Sawamoto, M. Acc. Chem. Res. 2008, 41, 1120–1132.
12 (a) Cheng, C.; Sun, G.; Khoshdel, E.; Wooley, K. L. J. Am.
Chem. Soc. 2007, 129, 10086–10087; (b) Uehara, K.; Wagner, C.
B.; Vogler, T.; Luftmann, H.; Studer, A. Angew. Chem. 2010,
122, 3137–3141.
13 (a) Miele, S.; Nesvadba, P.; Studer, A. Macromolecules
€
5 (a) Barner-Kowollik, C.; Davis, T. P.; Heuts, J. P. A.; Stenzel,
M. H.; Vana, P.; Whittaker, M. J. Polym. Sci. Part A: Polym.
Chem. 2003, 41, 365–375; (b) Rizzardo, E.; Chiefari, J.; Maya-
dunne, R. T. A.; Moad, G.; Thang, S. H. In Controlled/Living
Radical Polymerization; Matyjaszewski, K., Ed.; ACS Sympo-
sium Series 768; American Chemical Society: Washington, DC,
2000, pp 278–296; (c) Favier, A.; Charreyre, M.-T. Macromol.
Rapid Commun. 2006, 27, 653–692; (d) Barner-Kowollik, C.; Per-
rier, S. J. Polym. Sci. Part A: Polym. Chem. 2008, 46,
5715–5723; (e) Moad, G.; Rizzardo, E.; Thang, S. H. Acc. Chem.
Res. 2008, 41, 1133–1142.
2009, 42, 2419–2427; (b) Wienhofer, I. C.; Studer, A.; Rahman,
T.; Fukuyama, T.; Ryu, I. Org. Lett. 2009, 11, 2457–2460; (c)
Knoop, C. A.; Studer, A. J. Am. Chem. Soc. 2003, 125,
16327–16333; (d) Wetter, C.; Gierlich, J.; Knoop, C.; Muller, C.;
¨
Schulte, T.; Studer, A. Chem. Eur. J. 2004, 10, 1156–1166; (e)
Siegenthaler K. O.; Studer, A. Macromolecules 2006, 39,
1347–1352; (f) Chang, C.-C.; Studer, A. Macromolecules 2006,
39, 4062–4068.
14 (a) Nielen, M. W. F. Mass. Spectrom. Rev. 1999, 42, 309–344;
(b) Schulte, T.; Siegenthaler, K. O.; Luftmann, H.; Letzel, M.;
Studer, A. Macromolecules 2005, 38, 6833–6840; (c) Wienhofer, I.
€
6 David, G.; Boyer, C.; Tonnar, J.; Ameduri, B.; Lacroix-Des-
mazes, P.; Boutevin, B. Chem. Rev. 2006, 106, 3936–3962.
C.; Luftmann, H.; Studer, A. Macromolecules 2011, 44,
2510–2523; (d) Hart-Smith, G.; Barner-Kowollik, C. Macromol.
Chem. Phys. 2010, 211, 1507–1529; (e) McEwen C. N.; Peacock, P.
M. Anal. Chem. 2002, 74, 2743–2748; (f) Montaudo, G.; Samperi,
F.; Montaudo, M. S. Prog. Polym. Sci. 2006, 31, 277–357.
7 Yamago, S. Chem. Rev. 2009, 109, 5051–5068.
8 (a) Wayland, B. B.; Poszmik, G.; Mukerjee, S. L.; Fryd, M. J.
Am. Chem. Soc. 1994, 116, 7943–7944; (b) Kaneyoshi, H.; Maty-
jaszewski, K. Macromolecules 2006, 39, 2757–2763; (c) Poli, R.
Angew. Chem. Int. Ed. 2006, 45, 5058–5070.
15 Benoit, D.; Chaplinski, V; Braslau, R.; Hawker, C. J. J. Am.
Chem. Soc. 1999, 121, 3904–3920.
16 Grubbs, R. B. Polym. Rev. 2011, 51, 104–137.
9 (a) Hawker, C. J.; Bosman, A. W.; Harth, E. Chem. Rev. 2001,
101, 3661–3688; (b) Studer, A. Chem. Soc. Rev. 2004, 33,
267–273; (c) Studer, A.; Schulte, T. Chem. Rec. 2005, 5, 27–35;
17 (a) Charleux, B.; Nicolas, J.; Guerret, O. Macromolecules
2005, 38, 5485–5492; (b) Nicolas, J.; Dire, C.; Mueller, L.; Belle-
ney, J.; Charleux, B.; Marque, S. R. A.; Bertin, D.; Magnet, S.;
Couvreur, L. Macromolecules 2006, 39, 8274–8282; (c) Dire, C.;
Charleux, B.; Magnet, S.; Couvreur, L. Macromolecules 2007,
40, 1897–1903.
ꢀ
(d) Sciannamea, V.; Jeroˆ me, R.; Detrembleur, C. Chem. Rev.
2008, 108, 1104–1126; (e) Fischer, H. Chem. Rev. 2001, 101,
3581–3610; (f) Tebben, L.; Studer, A. Angew. Chem. 2011, 123,
5138–5174.
10 Mittal, A.; Sivaram, S.; Baskaran, D. Macromolecules 2006,
39, 5555–5558.
18 Bartoszewicz, A.; Livendahl, M.; Mart´ın-Matute, B. Chem.
Eur. J. 2008, 14, 10547–10550.
11 Ansong, O. E.; Jansen, S.; Wei, Y.; Pomrink, G.; Lu, H.;
Patel, A.; Li, S. Polym. Int. 2009, 58, 54–65.
19 Lee, P. H.; Lee, S. W.; Seomoon, D. Org. Lett. 2003, 5,
4963–4966.
2160
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 2150–2160