Acylgermanes: Photoinitiators and sources for Ge-centered radicals. insights into their reactivity

Article abstract of DOI:10.1021/ja404433u

Acylgermanes have been shown to act as efficient photoinitiators. In this investigation we show how dibenzoyldiethylgermane 1 reacts upon photoexcitation. Our real-time investigation utilizes femto- and nanosecond transient absorption, time-resolved EPR (50 ns), photo-chemically induced dynamic nuclear polarization, DFT calculations, and GC-MS analysis. The benzoyldiethylgermyl radical G? is formed via the triplet state of parent 1. On the nanosecond time scale this radical can recombine or undergo hydrogen-transfer reactions. Radical G? reacts with butyl acrylate at a rate of 1.2 ± 0.1 × 10⁸ and 3.2 ± 0.2 × 10⁸ M^-1 s^-1, in toluene and acetonitrile, respectively. This is ～1 order of magnitude faster than related phosphorus-based radicals. The initial germyl and benzoyl radicals undergo follow-up reactions leading to oligomers comprising Ge-O bonds. LC-NMR analysis of photocured mixtures containing 1 and the sterically hindered acrylate 3,3-dimethyl-2-methylenebutanoate reveals that the products formed in the course of a polymerization are consistent with the intermediates established at short time scales.

Full text of DOI:10.1021/ja404433u

Article
pubs.acs.org/JACS
Acylgermanes: Photoinitiators and Sources for Ge-Centered Radicals.
Insights into their Reactivity
Dmytro Neshchadin,† Arnulf Rosspeintner,‡ Markus Griesser,† Bernhard Lang,‡
Sandra Mosquera-Vazquez,‡ Eric Vauthey,‡ Vitaly Gorelik,†,# Robert Liska,∥,∇ Christian Hametner,∥
Beate Ganster,∥ Robert Saf,§ Norbert Moszner,⊥ and Georg Gescheidt*,†
§
†Institute of Physical and Theoretical Chemistry and Institute of Chemistry and Technology of Materials, Graz University of
Technology, Stremayrgasse 9, 8010 Graz, Austria
‡Dep
́ ́ ̀ ̀
artement de Chimie Physique, Universite de Geneve, 30, quai Ernest-Ansermet, 1211 Geneve 4, Switzerland
∥Institute of Applied Synthetic Chemistry, Vienna University of Technology, Getreidemarkt 9/163/MC, 1060 Vienna, Austria
⊥Ivoclar Vivadent AG, Bendererstrasse 2, 9494 Schaan, Liechtenstein
∇Christian Doppler Laboratory for Photopolymers in Digital and Restorative Dentistry, Getreidemarkt 9, 1060 Vienna, Austria
S
* Supporting Information
ABSTRACT: Acylgermanes have been shown to act as efficient photoinitiators. In this investigation we show how
dibenzoyldiethylgermane 1 reacts upon photoexcitation. Our real-time investigation utilizes femto- and nanosecond transient
absorption, time-resolved EPR (50 ns), photo-chemically induced dynamic nuclear polarization, DFT calculations, and GC-MS
analysis. The benzoyldiethylgermyl radical G• is formed via the triplet state of parent 1. On the nanosecond time scale this
radical can recombine or undergo hydrogen-transfer reactions. Radical G• reacts with butyl acrylate at a rate of 1.2 0.1 × 108
and 3.2
0.2 × 108 M−1 s−1, in toluene and acetonitrile, respectively. This is ∼1 order of magnitude faster than related
phosphorus-based radicals. The initial germyl and benzoyl radicals undergo follow-up reactions leading to oligomers comprising
Ge−O bonds. LC-NMR analysis of photocured mixtures containing 1 and the sterically hindered acrylate 3,3-dimethyl-2-
methylenebutanoate reveals that the products formed in the course of a polymerization are consistent with the intermediates
established at short time scales.
be found at ∼350 nm. Up to now the highest shift for type I
photoinitiators was observed for bis-acylphosphine oxides with
shifts <400 nm.
1. INTRODUCTION
Germanium-centered radicals are key intermediates in a broad
range of chemical reactions.1−7 The reactivity of the related tin-
and silicon-based radicals has been well described, but the
reactivity of the related Ge-centered radicals is much less
explored.8−13
In the field of photoinitiated free-radical polymerization14,15
Ge-centered radicals have started to play an important role.
Especially for applications requiring longer excitation wave-
lengths (e.g., dental fillings), optimizations of photoinitiators
are highly desirable. The recently developed dibenzoyldiethyl-
germane 116,17 shows the absorption maximum of its n−π*
transition, which is required for α-cleavage reactions leading to
radicals, at red-shifted 418 nm. The overlap of the d-orbitals of
Ge with the π* orbital of the carbonyl group is responsible for
that shift. In classical hydroxyalkylphenones this transition can
The efficiency of 1 has been demonstrated,16−19 in terms of
photo-DSC measurements and product analysis. However, the
primary species leading to the polymerization process have not
been established by experiment. As shown in Scheme 1, the
Received: May 3, 2013
Published: October 15, 2013
© 2013 American Chemical Society
17314
dx.doi.org/10.1021/ja404433u | J. Am. Chem. Soc. 2013, 135, 17314−17321

Journal of the American Chemical Society
Article
EPR spectra. During a TR-EPR experiment the desired magnetic field
range was scanned by recording the accumulated (50−100
accumulations) EPR time responses to the incident laser pulses at a
given static magnetic field. The experimental setup was controlled
using fsc2, software developed and kindly provided by J. Toerring
(Berlin). Argon-saturated solutions were irreversibly pumped through
a quartz tube (inner diameter 2 mm, flow ∼2−3 mL/min) in the
rectangular cavity of the EPR spectrometer.
Scheme 1. Reaction Scheme of the Relevant Photophysical
Processes of 1 upon Laser Excitation
a
1H NMR and CIDNP experiments were carried out on a 200 MHz
Bruker AVANCE DPX spectrometer equipped with a custom-made
CIDNP probe head. A Spectra Physics INDI Nd:YAG laser (355 nm,
∼60 mJ/pulse, pulse length ∼8−10 ns) operating at 20 Hz and a
Hamamatsu Hg−Xe lamp (150 W, pulse length 300 ms) served as
light sources. The typical CIDNP timing sequence consists of the
following parts: composite pulse presaturation, laser/lamp flash, 90°
radiofrequency detection pulse (2.2 μs), and FID. Sample decom-
position during the CIDNP experiment did not exceed 1%, as
controlled by NMR.
Geometry optimizations, hyperfine coupling constant (hfc)
calculations and frequency analysis were done using the Gaussian
0327 software package at the B3LYP28,29 level of theory with the
TZVP30 basis set. Single point energy calculations were carried out
using the M052X/6-311+G(2d,2p) protocol.31
a
The grey boxes refer to the processes observable with the respective
experimental techniques. The colors of the species correspond to
those in the transient absorption spectra (Figure 1).
primary reaction expected is the α-cleavage of a Ge−C(O)
bond. Whereas the reactivity of the benzoyl radical is rather
well established,20−23 the knowledge on the Ge-centered
radicals requires some expansion.10
The aim of our work is to establish the mechanism of the
primary cleavage of 1 and to investigate the properties of the
thus formed Ge-centered radical. This germyl radical will be
described in terms of its structure and its intrinsic conversions;
furthermore its reactivity toward acrylate double bonds will be
explored quantitatively. This is accomplished by time-resolved
optical spectroscopy (fs−μs), time-resolved electron para-
magnetic resonance (TR-EPR) (ns), photo-chemically induced
dynamic nuclear polarization (CIDNP), EI-MS, and DFT
calculations. In addition to the primary reactions of the germyl
radical, we show conversions of 1 at long-term irradiation (up
to 600 s) in the presence of butyl acrylate and the sterically
hindered acrylate 3,3-dimethyl-2-methylenebutanoate (to avoid
formation of insoluble polymers). We predominately focus on
the chemistry of the Ge-centered radicals, showing some
general features of their characteristic transformations.
GC-TOF MS analyses of irradiated solutions of 1 in CD3CN (c =
4.8 × 10−6 mol/L, irradiation time up to 2 min, under Ar) were
performed on an Agilent 7890A (Agilent, USA) gas chromatograph
coupled to an orthogonal TOF mass spectrometer (GCT Premier,
Waters). The GC conditions were as follows: column, DB-5MS (J &
W Scientific, CA, USA) 30 m length × 0.25 mm ID × 0.25 μm phase
thickness; injection temperature, 250 °C; transfer line temperature,
250 °C; column temperature program, 40 °C for 4 min, ramp at 20
°C/min to 280 °C, with a final 4 min hold at 280 °C; carrier gas,
helium 1.00 mL/min; injection mode, split 1:100; and injection
volume, 1 μL. Mass spectra were acquired in electron ionization mode
(70 eV, source temperature 250 °C), over the mass range m/z 50−800
Da, with ∼7000 fwhm resolution and an acquisition rate of 4 spectra/s.
Data were processed using MassLynx (version 4.1) software.
Photolysis experiments in the presence of the monomer 3,3-
dimethyl-2-methylenebutanoate (t-BAM)32 were performed with a
Bluephase C8 long-term irradiation (LED) lamp (430−530 nm). The
corresponding LC-NMR spectra were recorded on a 400 MHz Bruker
Avance DRX400 FT-NMR spectrometer with a 60 μL triple-resonance
flow probe head. LC was carried out on an Agilent 1100 HPLC with
quaternary pump, UV detector, and solid-phase extraction unit SEP
Prospekt 2.
2. EXPERIMENTAL SECTION
The synthesis of 1 is described in refs 16 and 17. Toluene, toluene-d8
(Fluka), and acetonitrile-d3 were used as received. Samples were
saturated with Argon prior to use to remove the oxygen from the
solution.
For time-resolved optical spectroscopy, toluene (Fisher Scientific,
>99.8%), acetonitrile (Fisher Scientific, 99.98%), and butylacrylate
(Sigma-Aldrich) were used as received.
The apparatus used for femtosecond transient absorption (fs-TA)
spectra with excitation at 400 nm was already described.24 The
instrument response function had a full width at half-maximum
(fwhm) of ∼200−500 fs as obtained from measurements of the optical
Kerr effect in toluene and acetonitrile. Samples were excited with
∼1015 photons/(cm2 pulse). Samples were bubbled with nitrogen
during the experiment to constantly refresh the excitation volume, thus
avoiding sample decomposition. Changes in the sample concentration
due to degradation and/or solvent evaporation were negligible as
judged from absorption spectra before and after the experiments.
Nanosecond transient absorption (ns-TA) experiments were
performed with excitation at 355 nm using a setup described in ref
25. Samples were saturated with nitrogen and constantly bubbled with
nitrogen to refresh the sample and avoid sample decomposition.
Rate constants were extracted from the experimental spectra by
global target analysis.26
3. RESULTS AND DISCUSSION
3.1. Photophysics. Time-resolved optical spectroscopy was
used to obtain information how and on which time scale a Ge-
centered radical is formed. We performed femto- and
nanosecond broadband transient absorption experiments.
Directly after photoexcitation we observe a transient absorption
spectrum with a maximum around 470 nm, which we can
attribute to the singlet excited state of 1. This species
disappears within tens of picoseconds (cf. Table 1) to give
rise to the triplet state of 1, which exhibits absorption maxima
at 460 nm and above 700 nm (green line, upper part of Figure
1).
On the nanosecond time scale, it can be observed that the
triplet state of 1 decays within ∼15 ns and a new band at 500
Table 1. Rate Constants of Inter System Crossing (ISC) and
α-Cleavage of 1
TR-EPR experiments were performed using light pulses (355 nm,
∼10 mJ/pulse, pulse length ∼10 ns) from a Continuum Surelight II
Nd:YAG laser, operating at 20 Hz. A Bruker ESP 300E X-band
spectrometer with unmodulated static magnetic field and a LeCroy
9400 dual 125 MHz digital oscilloscope were used to acquire the TR-
solvent
kISC, s−1
(3 0.3)·1010
(2 0.1)·1010
kcleav, s−1
(6 2)·107
(7 2)·107
toluene
acetonitrile
17315
dx.doi.org/10.1021/ja404433u | J. Am. Chem. Soc. 2013, 135, 17314−17321

Journal of the American Chemical Society
Article
1
calculations show that the remaining H hfcs (γ- and aromatic
protons) in G• are at least 1 order of magnitude smaller and
are not resolved in the experimental TR-EPR spectrum (see
Supporting Information).40
The TR-EPR experiments clearly establish radicals B• and
G• as the primary species formed on a 50 ns time scale upon
irradiation of 1 after intersystem crossing from the excited
singlet to the excited triplet state and subsequent α-cleavage
(Scheme 1).
1
3.3. H TR-CIDNP. When an initiator is photolyzed and
undergoes a homolytic bond cleavage, the corresponding
radical pair is created in a correlated spin state. The follow-
up reactions of this initially formed radicals depend on the
relative orientations of their spins, e.g., two radicals with an
overall triplet state cannot recombine, causing a non-Boltzmann
population of magnetic energy levels (up to a μs time scale)
and leading to ‘polarized’ NMR signals. Therefore, reaction
pathways proceeding via radical pairs can preferentially be
followed using this phenomenon (CIDNP).41,42 We have
applied the time-resolved 1H CIDNP technique43−45 to
investigate primary transformations of the radicals G• and B•.
Figure 1. Femto- and nanosecond transient absorption spectra of 1 in
toluene. The upper part shows the time regime between 1 and 200 ps,
whereas the lower part represents the nanosecond experiment 10−60
ns (see Experimental Section). Note, that the dashed green spectrum
3
in the lower part corresponds to a spectrum comprising 1 and G•.
nm emerges (Figure 1, lower part). This absorption can
straightforwardly be ascribed to the Ge-centered radical G•
since the simultaneously formed benzoyl radical B• is known
not to be detectable in the 400−700 nm range.21
1
In Figure 3, the H NMR spectrum of parent 1 and the
corresponding TR-CIDNP resonances with delays of 1 and 20
μs (in toluene-d8) are presented. The spectra taken 1 and 20 μs
after the laser flash show compatible polarization patterns but
indicate that additional follow-up products emerge after 20 μs.
The resonances can be assigned in the following way: In the
1 μs TR-CIDNP spectrum (Figure 3B) methylene protons of
parent 1 appear in strong enhanced absorption. This can be
attributed to the (geminate or cage) reformation of 1 from G•
and B•.46 Using Kaptein’s rule47−49 for the net polarization and
Accordingly, the reaction sequence sketched in Scheme 1 can
be adopted. Whereas the rate constant for inter system crossing
has an expectable value of ∼3 × 1010 s−1, the rate constant for
α-bond cleavage is relatively small (∼7 × 107 s−1). This is in
remarkable contrast to the photophysics of the P containing
analog, the intersystem crossing rate of which is reported to be
8 × 109 s−1 and its cleavage rate being much faster with 1010
s−1.21 The corresponding rate constants are summarized in
Table 1.
3.2. TR-EPR. With a time resolution of ∼50 ns, and its rather
high sensitivity based on radical-pair effects,33−36 TR-EPR is
ideally suited to get insight into the structure of the primary
Ge-centered radical G•. Laser-flash photolysis of 1 leads to the
TR-EPR spectrum shown in Figure 2. It consists of two
1
taking into account the g factor and the H hfcs of G• and B•
(enhanced absorption, Δg > 0, hfc(CH2) > 0, derived from
EPR and DFT calculations, respectively, see above) it can be
deduced that the primary radical pair is created in the triplet
excited state, in full agreement with the emissive polarization
pattern of the TR-EPR signals (Figure 2) and in line with
related photoinitiators.50
The initial radicals also diffuse apart and yield ‘escape’
products. A well-established reaction is the dimerization of two
identical radicals leading to 2 (formed from two G•, related to
the reactivity of phosphinoyl radicals)51 and diphenylethane-
1,2-dione (benzil, 6, from two B•).
The signals assigned to 2 are the emissive resonances at δ =
1.31 and 1.74 ppm, for CH3 and CH2 respectively, in agreement
with Kaptein’s rule for the net CIDNP polarization. 1H CIDNP
resonances of 6 can hardly be distinguished because the small
hfcs of phenyl protons of the precursor (benzoyl radical) lead
to very weak polarization intensities and the aromatic region
contains many overlapping peaks.
Even after the short delay of 1 μs, new resonances appear
with distinct polarizations. In particular, a singlet in enhanced
absorption at δ = 9.59 ppm and a negatively polarized multiplet
at δ = 7.54 ppm emerge. These resonances are typical for the
carbonyl and ortho-phenyl protons of benzaldehyde 3 and have
been frequently observed for several photoinitiating systems
containing the benzoyl moiety.52 Formation of 3 is usually
attributed to a disproportionation reaction (β-hydrogen
transfer)52 in the primary radical pairs consisting of a benzoyl
radical and radicals, which are able to donate hydrogen atoms.
In the present case, the hydrogen has to be transferred from the
ethyl group of G• as shown in Scheme 2.
Figure 2. TR-EPR spectrum of 1 in toluene, taken 250 ns after the
laser flash, showing two different EPR signals: G• (blue) and B•
(gray).
different overlapping signals appearing in emission, indicating
that radicals are polarized via the triplet mechanism,34,37 in line
with the results obtained by optical spectroscopy. The rather
narrow and unresolved signal (g = 2.000, marked in gray in
Figure 2) is attributed to the benzoyl radical B•.20,38 The
dominating pattern (g = 2.006, blue in Figure 2) is a quintet
(intensity ratio 1:4:6:4:1), with an isotropic 1H hfc of 0.44 mT.
It can be assigned readily to the four equivalent methylene
protons (“β-hydrogens”) of the two ethyl groups in the Ge-
centered radical39 G• (calcd H hfc, 0.61 mT). Moreover,
1
17316
dx.doi.org/10.1021/ja404433u | J. Am. Chem. Soc. 2013, 135, 17314−17321

Journal of the American Chemical Society
Article
1
Figure 3. H NMR (A), 1 μs (B) and 20 μs (C) TR-CIDNP spectra of 1 in toluene-d8.
calculations, which predict the bond dissociation energy of the
Ge−C bond in G• being only 183 kJ mol−1. Since Ge(II)-
containing species are known to be highly reactive without
additional stabilization (e.g., by donor−acceptor substitu-
tion),53−55 5 is rapidly converted to more persistent products
(see Section 3.4) and cannot be established after longer
reaction periods.
Scheme 2. Possible Hydrogen Transfer Reaction within the
Primary Radical Pair (G• and B•) Formed upon
Photoinduced α-Cleavage of 1
In the CIDNP spectrum taken 1 μs after the laser flash, two
weaker emissive signals are observed at δ = 4.67 and 4.81 ppm.
These signals presumably do not stem directly from the initial
radicals G• and B•. These resonances might be due to
secondary products generated via rapid follow-up reactions of
an intermediate germene 4.
According to Kaptein’s rules, aldehyde 3 is observed on the
time scale of our experiment as a cage product (enhanced
absorption, triplet precursor, Δg > 0, hfc(CH2) > 0).
The resonances of vinyl hydrogens at δ = ∼5.7 ppm point at
E- and Z-isomers of a (transient) germene 4 (Scheme 2).
The strongly emissive quartet at δ = 2.44 ppm (J = 7.1 Hz)
can be provisionally assigned to the methylene protons of
diethylgermylene 5 (Figure 3B) as the result of a second
cleavage at the stage of G• leading to 5 and a second benzoyl
radical B• (Scheme 3). This assignment is provisional since we
Germenes are reported to undergo fast and almost
quantitative [2 + 2] and [2 + 4] cycloadditions56−59 in the
presence of saturated and unsaturated ketones, respectively.
The thus formed heterocycles are likely to undergo further
transformations. Accordingly, the signals at δ = 4.67 and 4.81
ppm, the emissively polarized signals at ca. δ = 6.15 ppm and
two multiplets (δ = 4.47 and 4.88 ppm) in the 1 and 20 μs TR-
CIDNP spectra (Figure 3B,C) could originate from such
products. Scheme 4 summarizes the short-lived intermediates
(up to 20 μs), which have been established (and, partly,
1
neither found H NMR spectroscopic information for Ge(II)
nor for similar Si(II) and Sn(II) species. The decomposition of
G• is feasible according to M052X/6-311+G(2d,2p) DFT
1
anticipated) from TR-EPR and H TR-CIDNP experiments.
Scheme 3. Cleavage of Ge-Centered Radical G•
3.4. GC-MS Analysis of “Relaxed” Reaction Mixtures.
The above experiments have shown reaction products obtained
upon laser-flash photolysis (355 nm, 10 ns pulse) up to a
microsecond time scale, creating the basis for conversions,
following at longer irradiation times. In technical applications,
however, irradiation is performed over a period in the order of
seconds. To obtain insight into products formed at extended
time scales and under thermodynamic control, GC-TOF MS
analysis of reaction solutions treated with different periods of
irradiation was performed.
Generally, follow-up products compatible with the formation
of benzoyl radical B• could be well established, e.g.,
benzaldehyde 3 and benzil 6 were found as the main products.
17317
dx.doi.org/10.1021/ja404433u | J. Am. Chem. Soc. 2013, 135, 17314−17321

Journal of the American Chemical Society
Article
Scheme 4. Reaction Pattern of the Primary Radicals B• and G•
Additionally, small amounts of phenol and benzoic acid were
detected. In the case of products formed via Ge-centered radical
G•, where H CIDNP and H NMR only provided early
products, mass spectra were observed, that can be assigned to
products revealing Ge−O bonds, especially cyclic products
(GeEt2O)n (7n).
3.5. Addition of Radical G• to Double Bonds. It has
been shown that the acylgermane photoinitiator 1 is rather
efficient from a technical point of view.16−18 Obviously, both
radicals G• and B• add to (acrylate) double bonds. The kinetic
data for B• were determined by time-resolved IR spectrosco-
py.21
1
1
We have measured the rate constant of the addition of G• to
butyl acrylate using ns-TA spectroscopy, by following the decay
of the absorption at ∼500 nm (cf. Figure 1). The plots showing
the dependence of the observed (pseudo first order) rate
constant vs the concentration of butyl acrylate are shown in
Figure 4.
As an example, the most important product containing
germanium was detected at a retention time tR = 12.7 min and
is assigned to 73. The base peak in the corresponding mass
spectrum was observed at m/z = 410.9473 Da, which
corresponds well to the calculated mass of the fragment ion
[73-Et]+ (C10H25Ge3O3; m/zcalc = 410.9469 Da). The molecular
+
ion 73 was observed too, but only with very low relative
intensity and consequently low mass accuracy. The homolo-
gous products 7n with n = 4 and 5 were observed at tR = 14.6
and 16.4 min, respectively. Furthermore, other volatile products
were detected at low concentration, where the fragmentation
pattern as well as the accurate mass data indicated species like
(GeEt2n−1O)nBz (Bz = benzoyl-), e.g., (GeEt2n−1O)4Bz at a
retention time of tR = 17.9 min.
Figure 4. (a) Decay curves obtained from ns-TA experiments at
different butyl acrylate concentrations (0, 0.014, 0.028, 0.048, 0.077 M;
observation wavelength, 507 nm; solvent, acetonitrile); (b)
pseudofirst-order decay rate constants, kobs, obtained from (a) as a
function of butyl acrylate concentration in toluene and acetonitrile for
The formerly suggested oligomeric Ge−Ge bonded
the determination of kadd
.
species,19 anticipated from broadened H NMR lines, could
1
not be discerned. This is not astonishing in view of the
relatively weak Ge−Ge bond (∼270 kJ mol−1) compared to,
e.g., a Ge−O bond (662 kJ mol−1), which is certainly more
likely to be formed under these reaction conditions favoring
thermodynamic control. The results discussed above imply that
oligomeric and probably also polymeric (cyclic) molecules with
O−Ge−O connectivity are formed.
The corresponding addition rate constants, kadd, are 1.2 × 108
and 3.2 × 108 M−1 s−1, in toluene and acetonitrile, respectively,
are compatible with those determined for the triphenylgermyl
radical PhG•50 (Table 2).
The kadd values for the Ge-centered radicals are substantially
higher than those of the radical B• (3 orders of magnitude).
17318
dx.doi.org/10.1021/ja404433u | J. Am. Chem. Soc. 2013, 135, 17314−17321

Journal of the American Chemical Society
Article
analogy to the B•-derived molecules B1 and B2/3, products
G1 and G2 represent the addition of t-BAM to G•. Whereas
the saturated product B1 dominates in the case of B• as the
precursor, unsaturated G2 shows a higher LC peak intensity.
Owing to the characteristics of the LED utilized, further
photochemical decomposition of G1 and G2 is not observed.
In irradiation experiments at 365 nm, these products occurred
only in traces because immediate photochemical decomposition
occurred at shorter-wavelength irradiation. Another remarkable
product is GMB. It can be either based on a B-(t-BAM) radical
and addition of G• or a G-(t-BAM) radical reacting with B•.
Table 2. Experimentally Established Rate Constants for the
Addition of the Radicals G• to Butyl Acrylate and PhG• to
Methyl Acrylate
radical
solvent
kadd, M−1 s−1
ref
G•
G•
PhG•
B•
toluene
(1.2 0.1) × 108
(3.2 0.2) × 108
this work
acetonitrile
acetonitrile
acetonitrile
this work
50
a
2.2 × 108
21
2.7 × 105 ( 10%)
a
No error limits given.
Thus, the G• is the prominently active radical generated from
photoinitiator 1. It is moreover, noteworthy, that the
substituents do not markedly alter the rate constants for the
addition of the Ge center to double bonds.
3.6. Photolysis in the Presence of 3,3-dimethyl-2-
methylenebutanoate (t-BAM). For the application of
photoinitiator 1 in polymerizations, usually LEDs and longer
irradiation times are applied. To investigate if the species
established by the time-resolved experiments at very short time
scales described above also are decisive during a polymer-
ization, photolysis experiments (LED lamp, 430−530 nm) of 1
in the presence of t-BAM (6 equiv) were performed in
acetonitrile. The products were analyzed by LC-NMR.
The LC presents nine prominent components (Figure 7,
individual NMR spectra see Supporting Information). These
can be clearly assigned to photoproducts of the primary radicals
G• and B•. Benzaldehyde 3 has been identified as a product
being formed at very early stages of the decomposition of 1,
based on B•. It has, however, to be borne in mind that 3 can
also be formed by a partly reversible polymerization process.60
When the benzoyl radical reacts with t-BAM, the saturated
derivative B1 and the two unsaturated isomers B2 and B3 are
formed connected with hydrogen transfer. The occurrence of
B4, where an additional benzoyl moiety is attached to B1
indicates that benzoyl radicals are even present during the
initial steps of the polymerization and are able to add to the
growing polymer chain. Interestingly, 2, the dimer of the
benzoyldiethylgermyl radical G•, also established by the
CIDNP technique still can be detected by LC-NMR. In
4. CONCLUSION
The germanium based photoinitiator 1 displays a substantial
efficiency for initiating radical polymerization. Its red-shifted
n−π* absorption provides excellent usability. Parent 1 cleaves
upon irradiation to give the radical pair B• (benzoyl) and G•
(germyl). These two radicals are the primary species initiating
radical polymerizations.
On short time scales, G• radicals can form germylenes (H-
transfer) or dimerize, however, the substantial preference of Ge
toward oxygen leads to the formation of products displaying
(Ge−O)n moieties.
It is remarkable that long-term irradiation (LED) and short-
time scale experiments (laser) reveal rather similar products.
This can be traced back to the low energy of the LED utilized
for these experiments but is also consistent with a partially
reversible addition−elimination mechanism of the parent
radicals toward the double bonds followed by hydrogen
transfer.60,61
Our investigations suggest that the substituents around a Ge
radical center should not have a very high impact on the
reactivity of the radical toward (acrylate) double bonds. This is
in line with published data50 but certainly needs substantiation.
Another important factor will be the investigation of the
reactivity of Ge-centered radicals with oxygen.
Figure 7. Optimized LC chromatogram for the NMR detection of the photolysis solution of 2 in acetonitrile (detection wavelength 225 nm).
17319
dx.doi.org/10.1021/ja404433u | J. Am. Chem. Soc. 2013, 135, 17314−17321

Journal of the American Chemical Society
Article
(22) Dietliker, K.; Broillet, S.; Heltrung, B.; Rzadek, P.; Rist, G.; Wirz,
J.; Neshchadin, D.; Gescheidt, G. Helv. Chim. Acta 2006, 89, 2211.
(23) Voll, D.; Junkers, T.; Barner-Kowollik, C. Macromolecules 2011,
44, 2542.
(24) Banerji, N.; Duvanel, G.; Perez-Velasco, A.; Maity, S.; Sakai, N.;
Matile, S.; Vauthey, E. J. Phys. Chem. A 2009, 113, 8202.
(25) Lang, B.; Mosquera-Vazquez, S.; Lovy, D.; Sherin, P.; Markovic,
V.; Vauthey, E. Rev. Sci. Instrum. 2013, 84, 73107.
ASSOCIATED CONTENT
■
S
* Supporting Information
Geometry and calculated hfcs of radicals B• and G•, LC-NMR
spectra. This information is available free of charge via the
Internet at http://pubs.acs.org/.
AUTHOR INFORMATION
■
(26) Fita, P.; Luzina, E.; Dziembowska, T.; Radzewicz, C.;
Grabowska, A. J. Chem. Phys. A 2006, 125, 184508.
(27) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J.
B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R.
E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J.
J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.;
Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision
E.01; Gaussian, Inc.: Wallingford, CT, 2004.
Corresponding Author
g.gescheidt-demner@tugraz.at
Notes
The authors declare no competing financial interest.
#Deceased.
ACKNOWLEDGMENTS
■
Financial support by the Austrian Science Fund (P19769), TU
Graz, and Ivoclar (Liechtenstein) is gratefully acknowledged.
G.G. thanks Dr. Roland Fischer (TU Graz, Institute of
Inorganic Chemistry) for valuable discussions.
REFERENCES
■
(1) Lloyd, R. V.; Rogers, M. T. J. Am. Chem. Soc. 1973, 95, 2459.
(2) Drost, C.; Griebel, J.; Kirmse, R.; Lonnecke, P.; Reinhold, J.
Angew. Chem., Int. Ed. 2009, 48, 1962.
(3) Herrmann, W. A.; Denk, M.; Behm, J.; Scherer, W.; Klingan, F.
R.; Bock, H.; Solouki, B.; Wagner, M. Angew. Chem. 1992, 104, 1489.
(4) Sheberla, D.; Tumanskii, B.; Tomasik, A. C.; Mitra, A.; Hill, N. J.;
West, R.; Apeloig, Y. Chem. Sci. 2010, 1, 234.
(28) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(29) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623.
(30) Schafer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
5829.
(5) Power, P. P. Chem. Rev. 2003, 103, 789.
(6) Lee, V. Y.; Sekiguchi, A. In Reviews of Reactive Intermediate
Chemistry; Platz, M. S., Moss, R. A., Jones, J., M., Eds.; John Wiley &
Sons, Inc.: Hoboken, NJ, 2007, pp 47−120.
(7) Lee, V. Y.; Sekiguchi, A. Organometallic Compounds of Low-
Coordinate Si, Ge, Sn and Pb: From Phantom Species to Stable
Compounds; Wiley & Sons Inc.: Chichester, 2010.
(8) Albert, A.; Seconi, G.; Pedulli, G. F.; Degl’innocenti, A. J.
Organomet. Chem. 1983, 253, 291.
(9) Bernardoni, S.; Lucarini, M.; Pedulli, G. F.; Valgimigli, L.;
Gevorgyan, V.; Chatgilialoglu, C. J. Org. Chem. 1997, 62, 8009.
(10) Ingold, K. U.; Lusztyk, J.; Scaiano, J. C. J. Am. Chem. Soc. 1984,
106, 343.
(11) Krusic, P. J.; Kochi, J. K. J. Am. Chem. Soc. 1971, 93, 846.
(12) Taraban, M. B.; Volkova, O. S.; Kruppa, A. I.; Leshina, T. V. In
Patai’s Chemistry of Functional Groups, Organic Germanium, Tin and
Lead Compounds; Rappoport, Z., Liebman, J. F., Marek, I., Eds.; Wiley:
Hoboken, NJ, 2003, p 579.
(31) Zhao, Y.; Schultz, N. E.; Truhlar, D. G. J. Chem. Theory Comput.
2006, 2, 364.
(32) Seidl, B.; Liska, R. Macromol. Chem. Phys. 2007, 208, 44.
(33) Chemically induced magnetic polarisation; Muus, L. T.; Atkins, P.
W.; McLauchlan, K. A.; Pedersen, J. B., Eds.; D. Reidel: Dordrecht,
1977.
(34) Nagakura, S.; Hayashi, H.; Azumi, T. Dynamic Spin Chemistry;
Kodansha and John Wiley&Sons co-publication: Tokyo and New
York, 1998.
(35) Rodgers, C. T. Pure Appl. Chem. 2009, 81, 19.
(36) Yurkovskaya, A.; Morozova, O.; Gescheidt, G. In Encyclopedia of
Radicals in Chemistry, Biology and Materials; Chattilialoglu, C., Studer,
A., Eds.; Wiley: Hoboken, NJ, 2012; Vol. 1, p 175.
(37) Adrian, F. J. Res. Chem. Intermed. 1991, 16, 99.
(38) Williams, R. M.; Khudyakov, I. V.; Purvis, M. B.; Overton, B. J.;
Turro, N. J. J. Phys. Chem. B 2000, 104, 10437.
(39) Sakurai, H.; Mochida, K.; Kira, M. J. Am. Chem. Soc. 1975, 97,
929.
(40) Gerson, F.; Huber, W. Electron Spin Resonance Spectroscopy of
Organic Radicals; Wiley-VCH: Weinheim, 2003.
(41) Bargon, J.; Fischer, H.; Johnsen, U. Z. Naturforsch., A: Phys. Sci.
(13) Wille, U. Chem. Rev. 2013, 113, 813.
(14) Yagci, Y.; Jockusch, S.; Turro, N. J. Macromolecules 2010, 43,
6245.
(15) Dietliker, K.; Jung, T.; Benkhoff, J.; Kura, H.; Matsumoto, A.;
Oka, H.; Hristova, D.; Gescheidt, G.; Rist, G. Macromol. Symp. 2004,
217, 77.
1967, 22, 1551.
(42) Ward, H. R.; Lawler, R. G. J. Am. Chem. Soc. 1967, 89, 5518.
(43) Closs, G. L.; Miller, R. J. J. Am. Chem. Soc. 1979, 101, 1639.
(44) Miller, R. J.; Closs, G. L. Rev. Sci. Instrum. 1981, 52, 1876.
(45) Schaublin, S.; Wokaun, A.; Ernst, R. R. Chem. Phys. 1976, 14,
285.
(46) Taraban, M. B.; Maryasova, V. I.; Leshina, T. V.; Rybin, L. I.;
Gendin, D. V.; Vyazankin, N. S. J. Organomet. Chem. 1987, 326, 347.
(47) Kaptein, R. Chem. Commun. 1971, 732.
(16) Ganster, B.; Fischer, U. K.; Moszner, N.; Liska, R. Macro-
molecules 2008, 41, 2394.
(17) Moszner, N.; Fischer, U. K.; Ganster, B.; Liska, R.; Rheinberger,
V. Dent. Mater. 2008, 24, 901.
(18) Moszner, N.; Zeuner, F.; Lamparth, I.; Fischer, U. K. Macromol.
Mater. Eng. 2009, 294, 877.
(19) Durmaz, Y. Y.; Kukut, M.; Moszner, N.; Yagci, Y. Macro-
molecules 2009, 42, 2899.
(48) Closs, G. L.; Trifunac, A. D. J. Am. Chem. Soc. 1969, 91, 4554.
(49) Kaptein, R.; Oosterhoff, L. J. Chem. Phys. Lett. 1969, 4, 214.
(50) Lalevee, J.; Allonas, X.; Fouassier, J. P. Chem. Phys. Lett. 2009,
469, 298.
(51) Kolczak, U.; Rist, G.; Dietliker, K.; Wirz, J. J. Am. Chem. Soc.
1996, 118, 6477.
(20) Hristova, D.; Gatlik, I.; Rist, G.; Dietliker, K.; Wolf, J.-P.;
Birbaum, J.-L.; Savitsky, A.; Moebius, K.; Gescheidt, G. Macromolecules
2005, 38, 7714.
(21) Colley, C. S.; Grills, D. C.; Besley, N. A.; Jockusch, S.;
Matousek, P.; Parker, A. W.; Towrie, M.; Turro, N. J.; Gill, P. M. W.;
George, M. W. J. Am. Chem. Soc. 2002, 124, 14952.
17320
dx.doi.org/10.1021/ja404433u | J. Am. Chem. Soc. 2013, 135, 17314−17321

Journal of the American Chemical Society
Article
(52) Borer, A.; Kirchmayr, R.; Rist, G. Helv. Chim. Acta 1978, 61,
305.
(53) Becerra, R.; Boganov, S. E.; Egorov, M. P.; Lee, V. Y.; Nefedov,
O. M.; Walsh, R. Chem. Phys. Lett. 1996, 250, 111.
(54) Smith, G. R.; Guillory, W. A. J. Chem. Phys. 1972, 56, 1423.
(55) Thimer, K. C.; Al-Rafia, S. M. I.; Ferguson, M. J.; McDonald, R.;
Rivard, E. Chem. Commun. 2009, 7119.
(56) Barrau, J.; Escudie, J.; Satge, J. Chem. Rev. 1990, 90, 283.
(57) El Kettani, S. E. C.; Lazraq, M.; Ranaivonjatovo, H.; Escudie, J.;
Gornitzka, H.; Ouhsaine, F. Organometallics 2007, 26, 3729.
(58) El Kettani, S. E. C.; Lazraq, M.; Ranaivonjatovo, H.; Escudie, J.;
Couret, C.; Gornitzka, H.; Atmani, A. Organometallics 2005, 24, 5364.
(59) Lazraq, M.; Escudie, J.; Couret, C.; Satge, J.; Soufiaoui, M.
Organometallics 1992, 11, 555.
(60) Griesser, M.; Neshchadin, D.; Dietliker, K.; Moszner, N.; Liska,
R.; Gescheidt, G. Angew. Chem., Int. Ed. 2009, 48, 9359.
(61) Voll, D.; Neshchadin, D.; Hiltebrandt, K.; Gescheidt, G.; Barner-
Kowollik, C. Macromolecules 2012, 45, 5850.
17321
dx.doi.org/10.1021/ja404433u | J. Am. Chem. Soc. 2013, 135, 17314−17321