Synthesis and structure-activity relationship of several aromatic ketone-based two-photon initiators

Article abstract of DOI:10.1002/pola.24806

Several novel aromatic ketone-based two-photon initiators containing triple bonds and dialkylamino groups were synthesized and the structure-activity relationships were evaluated. Branched alkyl chains were used at the terminal donor groups to improve the solubility in the multifunctional monomers. Because of the long conjugation length and good coplanarity, the evaluated initiators showed large two-photon cross section values, while their fluorescence lifetimes and quantum yields strongly depend on the solvent polarity. All novel initiators exhibited high activity in terms of two-photon-induced microfabrication. This is especially true for fluorenone-based derivatives, which displayed much broader processing windows than well-known highly active initiators from the literature and commercially available initiators. While the new photoinitiators gave high reactivity in two-photon-induced photopolymerization at concentration as low as 0.1% wt, these compounds are surprisingly stable under one photon condition and nearly no photo initiation activity was found in classical photo DSC experiment. Copyright

Full text of DOI:10.1002/pola.24806

Synthesis and Structure-Activity Relationship of Several Aromatic
Ketone-Based Two-Photon Initiators
ZHIQUAN LI,1 MARTON SIKLOS,1 NIKLAS PUCHER,1 KLAUS CICHA,2 ALIASGHAR AJAMI,3 WOLFGANG HUSINSKY,3
2
ARNULF ROSSPEINTNER,4 ERIC VAUTHEY,4 GEORG GESCHEIDT,5 JURGEN STAMPFL, ROBERT LISKA1
¨
1Institute of Applied Synthetic Chemistry, Division of Macromolecular Chemistry, Vienna University of Technology,
Getreidemarkt 9/163/MC, 1060 Vienna, Austria
2Institute of Materials Science and Technology, Vienna University of Technology, Favoritenstrasse 9-11, 1040 Vienna, Austria
3Institute of Applied Physics, Vienna University of Technology, Wiedner Hauptstrasse 8, 1060 Vienna, Austria
4Department of Physical Chemistry, University of Geneva, 30, Quai Ernest Ansermet CH-1211 Geneva 4, Switzerland
5Graz University of Technology, Institute of Physical and Theoretical Chemistry, Stremayrgasse 9, 8010 Graz, Austria
Received 8 April 2011; accepted 25 May 2011
DOI: 10.1002/pola.24806
Published online 14 June 2011 in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: Several novel aromatic ketone-based two-photon ini-
tiators containing triple bonds and dialkylamino groups were syn-
thesized and the structure-activity relationships were evaluated.
Branched alkyl chains were used at the terminal donor groups to
improve the solubility in the multifunctional monomers. Because
of the long conjugation length and good coplanarity, the eval-
uated initiators showed large two-photon cross section values,
while their fluorescence lifetimes and quantum yields strongly
depend on the solvent polarity. All novel initiators exhibited high
activity in terms of two-photon-induced microfabrication. This is
especially true for fluorenone-based derivatives, which displayed
much broader processing windows than well-known highly active
initiators from the literature and commercially available initiators.
While the new photoinitiators gave high reactivity in two-photon-
induced photopolymerization at concentration as low as 0.1% wt,
these compounds are surprisingly stable under one photon con-
dition and nearly no photo initiation activity was found in classical
C
V
photo DSC experiment. 2011 Wiley Periodicals, Inc. J Polym Sci
Part A: Polym Chem 49: 3688–3699, 2011
KEYWORDS: laser-induced polymers; NLO; photochemistry;
photopolymerization; radical polymerization; two-photon
absorption photoinitiator
INTRODUCTION Two-photon-induced
photopolymerization
which often result in damage to the polymeric structures are
usually required. Therefore, great efforts were made to
improve the two-photon sensitivity. In the last decades, plenty
of chromophores with large rTPA were synthesized.10 How-
ever, most of the reported chromophores exhibit high fluores-
cence yields as desirable for fluorescence imaging applica-
tions.11,12 On the other hand, to be effective as a TPA initiator,
low fluorescence quantum yields are preferred as this leads
to a higher population of the triplet state (triplet quantum
yield), which is usually the active state of the initiators pro-
ducing radicals or ions for initiating the polymerization.13
Thus, a large rTPA value combined with a high photoreactivity
just like for normal one-photon initiators seems to be the goal
to achieve.
(TPIP) has been intensively studied recently as the interest
in the fabrication of smaller features for nanotechnology, in-
formation technology, and other industries requiring high
precision has significantly increased.1–5 With this unique
process, it is possible to produce complex 3D structures
with spatial resolution (120 nm) below the diffraction limit
of the utilized wavelength.6
TPIP is a solid free-form fabrication technique where a resin,
which contains mainly
a multifunctional acrylate-based
monomer and a photoinitiator (PI), is cured inside the focal
point of a femtosecond pulsed laser. To obtain an efficient
and clean polymerization and therefore high-quality struc-
tures, a highly active two-photon absorption (TPA) PI plays a
key role.
To design an ideal molecular structure with a high rTPA
value, several molecular key features have been identified.14
To utilize intramolecular charge-transfer as the ‘‘driving
force’’ for TPA,15,16 electron-donor and electron-acceptor
groups are required. In addition, coplanar p conjugated
Originally commercially available radical PIs were usually
used in the initial stages of TPIP research.7,8 As these PIs have
rather low cross sections (rTPA) and therefore poor absorp-
tion behavior,9 high excitation power and long exposure time,
Additional Supporting Information may be found in the online version of this article.
Correspondence to: R. Liska (E-mail: robert.liska@tuwien.ac.at)
C
V
Journal of Polymer Science Part A: Polymer Chemistry, Vol. 49, 3688–3699 (2011) 2011 Wiley Periodicals, Inc.
3688
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE 1 New TPA PIs and reference compounds.
bridges which lead to states with extended charge separa-
tion,17,18 are critical in enhancing the efficiency of intramo-
lecular charge transfer. Usually double bonds, due to their
simple synthetic accessibility, are used in the backbone of
the PIs to extend the conjugation length although the so
obtained rTPA is normally lower than that of the triple-bond-
containing counterparts.19,20 However, double bonds tend to
attenuate the desired photochemical processes by cis-trans
isomerization.
gated in this article. To study structure-activity relationships,
we used different aromatic ketones (benzophenone, fluore-
none and anthraquinone) as acceptors in the central part of
the p bridge and dialkylamino groups as electron-donor
moieties.
Compared to B3K, the benzophenone-based initiator B3BP
exhibits a longer conjugation length of the whole p system
and therefore a higher rTPA can be expected. Structure-activ-
ity relationship tests in the recent literature showed that
benzophenone-bearing dyes exhibit much larger rTPA than
their analogues without a benzophenone unit.24
Recently, we have reported on the successful syntheses and
evaluation of a series of radical PIs optimized for TPIP pro-
cesses.21 1,5-Bis(4-(N,N-dibutylamino)phenyl) penta-1,4-diyn-
3-one (B3K) containing a triple-bond p-bridge and a car-
bonyl group as electron withdrawing functional group serves
as a highly sensitive and efficient TPA PI in comparison with
typical one-photon UV PIs and other very potent two-photon
initiators well known from literature,22 such as E,E-1,4-
Additionally, aromatic ketones with stereorigid carbon skele-
ton (B3FL, 3,6-B3FL, and B3AN) were chosen to ensure
good coplanarity, thus facilitating the intermolecular charge
transfer process, which is critical in enhancing rTPA. Anthra-
quinone was used because of its excellent electron accepting
ability and because its 2-, 6-positions can be easily modified
by introducing p conjugated systems with electron donor
groups resulting in a D-p-A-p-D (A ¼ electron acceptor and
D ¼ electron donor) type molecule. Recently, a double-bond
containing anthraquinone compound was synthesized and
exhibited a large two-photon cross section value of 1635 GM
(1 GM ¼ 10ꢂ50 cm4 s photonꢂ1) at 800 nm, as well as low
fluorescence quantum yield.13
bis[4’-(N,N-di-n-butylamino)
styryl]-2,5-dimethoxybenzene
(R1) and E,E-1,4-bis-[4’-(N,N-di-n-butylamino)styryl]benzene
(R2). Nicely shaped structures and broad processing
windows can be obtained with PI concentrations as low as
ꢀ 0.05 wt %. It is worth to mention that compared to R1,
which shows strong fluorescence, the triple bond containing
ketone-based PIs either do not show any or an extremely
weak (Uem < 5 ꢁ 10ꢂ3) fluorescence emission. This is
rationalized by an efficient intersystem crossing to the
excited triplet state.23
B3FL was selected because 2,7-substituted fluorene cores
are known to exhibit high thermal and photochemical stabil-
ity.25–27 Furthermore, the electron-withdrawing carbonyl
group at C-9 stabilizes the LUMO of fluorene, making it
remarkably electrophilic.28 Substitution at positions 3 and 6
Based on the molecular structure of B3K containing triple
bonds and cross-conjugated D-p-A-p-D lead structures, sev-
eral aromatic ketone-based TPA PIs (Fig. 1) were investi-
SEVERAL AROMATIC KETONE-BASED TWO-PHOTON INITIATORS, LI ET AL.
3689

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
SCHEME 1 Reaction scheme for
the synthesis of the new TPA PIs.
theoretically offers good electronic communication through
the entire fluorenone skeleton.28 To investigate the effect of
different substituted positions on structure-reactivity rela-
tionships, 2,7- and 3,6-substituted fluorenone compounds
were synthesized and are being discussed in this article.
pling reactions of the appropriate iodo dialkylamino ben-
zenes with either (trimethylsilyl)acetylene or 2-methyl-3-
butyn-2-ol followed by a subsequent cleavage of the protec-
tion group in two steps29,30 or a Wittig-type one pot synthe-
sis starting from the corresponding aldehydes.31 Although
we have previously proven that both methods are suitable to
obtain the terminal aryl acetylenes in high yields,21 the for-
mer method was chosen because various alkyl chains could
be easily realized for alternative PIs in further studies, while
the counterpart aldehyde compounds are not always com-
mercially available.
Moreover, the solubility in the monomers is an important
issue for the efficiency of TPA initiators. This issue has been
addressed by the introduction of branched-alkyl chains at
the end of the amino moieties, as in BB3FL and BB3AN.
For comparison, the highly efficient TPA PI R1, well-known
from the literature,22 and Irgacure 369 (a typical commer-
cially available one-photon PI, that is, also frequently applied
in TPIP) were tested. For the characterization of the initia-
tors, UV-vis absorption and emission measurements were
carried out. Two photon cross-section measurements via
z-scan, as well as TPIP structuring tests were performed to
evaluate the TPA properties of the initiators. Ideal building
parameters for each initiator were determined by changing
the laser intensity, the feed rate, and the shape size using
the same molar PI concentrations.
The 4-iodobenzeneamine derivatives (1a-b) required for the
Sonogashira reaction had to be synthesized via alkylation of
4-iodobenzeneamine with different alkyl halides. The butyl
derivative 1a was obtained with 1-iodobutane in high yields
(78%). For economical reasons, 3-(bromomethyl)heptane was
use instead of its iodo analogues to prepare branched precur-
sor in the presence of potassium iodide and a phase transfer
catalyst (tetrabutylammonium bromide). As a second substitu-
tion with the branched alkyl chain exhibited quite low yield
(less than 15% from GC-MS) due to the steric hindrance, one
butyl group was used instead to give 1b in 60% yield.
RESULTS AND DISCUSSION
The reaction of these 4-iodobenzeneamine derivatives with
2-methyl-3-butyn-2-ol under Sonogashira conditions gave the
protected alkyne products 2a-b in yields between 83 and
88%. Finally, by deprotection under basic conditions the ter-
minal alkynes 3a-b were obtained in good yields (80–87%).
Synthesis
For the synthesis of the new initiators (Scheme 1), there are
two key intermediates 3a and 3b based on the p-amino aryl
alkyne group containing different alkyl chains at the N-atom.
These intermediates can be prepared by Sonogashira cou-
3690
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE 2 (a) Absorption spectra of 3,6-B3FL, B3FL, B3AN, and B3BP in dichloromethane. The half wavelength of the TPA cross-
section measurements (400 nm) is shown as vertical gray line. (b) Reduced fluorescence spectra (I(m)/m3) of 3,6-B3FL, B3FL, B3AN,
and B3BP in cyclohexane (blue), butyl ether (green), ethyl ether (yellow), butyl acetate (orange, for B3BP propyl acetate) and
dichloromethane (red). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
The dihalide aromatic ketones 4a-d for the coupling reaction
with terminal alkynes 3a-b could be prepared from the cor-
responding aromatic compounds via direct iodination or a
subsequent oxidation in one pot reactions with the reaction
yield ranging from 60 to 75%.32–34 3,6-Dibromo-fluorenone
was synthesized from the light-induced bromination of phen-
anthrene-9,10-dione followed by elimination under basic
condition.35,36 Multiple recrystallizations in DMSO must be
performed to remove trace amounts of 3,6-dibromophenan-
threne-9,10-dione, which is detrimental toward efficient
Sonogashira couplings.28
sochromic shift with increasing solvent polarity (Table 2). A
thorough solvatochromic analysis of these absorption bands,
for example, of Lippert-Mataga (LM) type, is, however
obscured by the fact that the absorption band is rather
broad, slightly changes shape with the solvent (not shown)
and overlaps with the adjacent transitions. Any attempts of
performing a LM analysis using simply the maxima of the
absorption band resulted in suboptimal (i.e., nonlinear) LM
plots. This might indicate, that the nature of the transition is
changing with the solvent, from a mainly local excitation in
apolar solvents to a partial charge transfer absorption in po-
lar solvents (or at least the amount of charge transfer char-
acter is changing).
On the basis of these intermediates, the final PIs were
obtained by Sonogashira coupling reactions of the terminal
alkyne derivatives with different aromatic ketone halides
with yields of 36–57%.
Inspection of Figure 2(b) and Table 2 shows that the situa-
tion is only slightly better concerning the interpretation of
the emission spectra. The 4 TPIPs investigated (B3AN, B3BP,
B3FL, and 3,6-B3FL) show a relative intense and structured
fluorescence in apolar cyclohexane. By increasing the solvent
Spectroscopy
Figure 2(a) depicts the absorption spectra of B3AN, B3BP,
B3FL, and 3,6-B3FL in dichloromethane (also Table 1 for
some prominent values of the extinction coefficient). All
TPIPs evaluated do not exhibit any linear absorption beyond
600 nm, which could thus interfere at the wavelength used
in the z-scan experiments and the two-photon polymeriza-
tion microfabrication. The absorption spectra of these novel
PIs exhibit a low energy absorption band, which appears at
significantly lower energies than those of either the parent
central accepting molecule (i. e. fluorenone, benzophenone,
anthraquinone) or the 4-(dimethylamino)ethynylbenzene do-
nor moiety.37 This ‘‘new’’ band is rather intense, with its
maximum molar decadic extinction coefficient, e, being in the
TABLE 1 Molar Extinction Coefficients of 3,6-B3FL, B3FL,
B3AN, and B3BP in Dichloromethane for the Two Lowest
Energy Absorption Maxima and the Half Wavelength of the
TPA Cross-Section Measurements (400 nm)
Substance
k (nm)
10ꢂ3e (Mꢂ1 cmꢂ1
)
B3AN
B3BP
485/400/387
405/400/300
495/400/396
460/400/315
28/38/50
60/60/50
15/100/105
40/20/85
B3FL
range from 104 to 5ꢃꢁ 104 Mꢂ1
s
ꢂ1, and shows a slight hyp-
3,6-B3FL
SEVERAL AROMATIC KETONE-BASED TWO-PHOTON INITIATORS, LI ET AL.
3691

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
TABLE 2 Basic Photophysical Properties of 3,6-B3FL, B3FL,
B3AN, and B3BP in Six Solvents at 25 8C
On the basis of the present data, which comprise only
steady-state and subnanosecond time-resolved emission
experiments, the following tentative conclusions may be
drawn. A charge transfer state is populated either via direct
excitation, or indirectly, via excitation of a locally excited
state.38 The current data does not allow speculations of pos-
sible populations of the triplet state of the PIs. However fem-
tosecond time resolved transient absorption and emission
spectroscopy are currently being performed to help shed
light onto the photophysics and possible mechanism(s) of
photoinitiation for these TPIPs.
Solvent
e
n
mf (eV)
ma (eV)
/f 10ꢂ3
s (ns)
B3AN
2.18
1.87
1.71
1.62
–
CX
BE
EE
BA
DC
2.01
3.08
4.20
5.01
8.93
1.4235
1.3968
1.3495
1.3918
1.4210
2.67
2.64
2.66
2.61
2.53
520
130
17
4.3
2.3
0.6
–
0.7
–
–
B3BP
2.95
2.57
2.43
2.11
1.96
B3FL
2.10
1.99
1.98
1.84
1.80
1.65
3,6-B3FL
2.52
2.17
2.04
1.84
1.81
1.69
TPA Cross Section
CX
BE
EE
PA
DC
2.01
3.08
4.20
6.00
8.93
1.4235
1.3968
1.3495
1.3828
1.4210
3.16
3.15
3.16
3.13
3.05
70
340
430
230
160
0.1
1.7
1.8
1.7
1.8
To investigate the TPA properties of the new PIs, an open
aperture z-scan analysis was performed to determine the
TPA cross section at 800 nm. THF was used as solvent for
the TPA characterization of all compounds. A nonrecycling
flow cell geometry was used for the measurements because
of the strong photodegradation of some PIs during the tests
(especially the ketone-based PIs). This phenomenon was also
described by Schafer et al.9 and was furthermore described
as an indicator for an effective two-photon initiator. In the
Supporting Information, the photo-degradation of BB3FL is
shown as a representative example at two different z-scans:
without flow and with 4 mL/h flow rate. The results indi-
cated that a flow rate of 4 mL/h is sufficient for all PIs to
wash the irradiated sample out of the focal volume in a rea-
sonable time.
CX
BE
EE
BA
PA
DC
2.01
3.08
4.20
5.01
6.00
8.93
1.4235
1.3968
1.3495
1.3918
1.3828
1.4210
2.61
2.57
2.57
2.54
2.52
2.49
440
150
110
15
4.6
2.3
2.0
0.5
0.4
–
11
0.7
CX
BE
EE
BA
PA
DC
2.01
3.08
4.20
5.01
6.00
8.93
1.4235
1.3968
1.3495
1.3918
1.3828
1.4210
2.93
2.87
2.83
2.77
2.76
2.68
210
650
680
90
0.8
3.6
4.1
0.9
0.9
0.3
The experimental data were fitted using the adopted equa-
tions of Sheik-Bahae et al.39 To obtain the TPA cross section
(r) [Fig. 3(a)]. To exclude excited-state absorption and to
verify that a pure r is determined, the measurements were
repeated at different peak intensities and the calculated pa-
rameter q0 scales linearly with intensity [Fig. 3(b)].39 All cal-
culated r values are given in Table 3.
44
12
The properties comprise the energies of the maximum of the reduced
fluorescence spectrum (I(m)/m3), mf, the lowest energy maximum of the
reduced absorption spectrum (e(m)/m), ma, the fluorescence quantum yield
and the fluorescence lifetime. Additionally the dielectric constant, e, and
refractive index, n, of the solvent are given.
The cross section value of the reference compound R1 is in
good agreement with our previous result21 and the litera-
ture,22 verifying the reliability of the experimental setup.
With longer conjugation length, benzophenone-based initia-
tor B3BP showed a larger cross section value than B3K of
238 GM, which is almost equal to that of R1. By changing
the acceptor group in B3BP from a benzophenone group to
an anthraquinone moiety as in B3AN, the value decreased to
235 GM. The reduction may be due to the relatively lower
absorption in one-photon spectra at about 400 nm, which
indicates that the TPA peaks in TPA spectra are shifted in
wavelength away from 800 nm. As expected, the substitution
of the butyl group at the amine by branched alkyl chains
alters the two photon cross section only marginally.
CX, cyclohexane; BE, butyl ether; EE, ethyl ether; BA, butyl acetate; PA,
propyl acetate; DC, dichloromethane.
polarity (from cyclohexane to dichloromethane), the emis-
sion spectra undergo more (for 3,6-B3FL and B3BP) or less
(B3FL) strong bathochromic shifts in the range of 0.5–1 eV.
Additionally the emission spectra lose their vibronic struc-
ture upon increasing the solvent polarity, with their band-
shape resembling more and more that of a pure charge
transfer transition. Since these trends are more or less prom-
inent for certain TPIPs, it is worth emphasizing their differ-
ences: For both B3AN and B3FL, lifetime and emission quan-
tum yield decrease monotonously with increasing solvent
polarity. For 3,6-B3FL the emission quantum yield and life-
time pass through a maximum for the ethers but hold a
smaller value in cyclohexane. For B3BP, whose emission and
absorption spectrum are also at higher energies than for the
other PIs, the quantum yield and lifetime show no pro-
nounced dependence with solvent polarity, except for the
fact that these values in cyclohexane are rather low.
Interestingly, the 3,6-substituted fluorenone-based PI, which
theoretically should facilitate an intramolecular electron trans-
fer process, exhibits a lower cross section value compared to
the 2,7-substituted analogue B3FL. An explanation can be
given by the red-shift-induced low TPA absorption at the given
wavelength. Among the aromatic ketone derivatives, B3FL had
the highest TPA cross section value with 440 GM. One reason is
the excellent coplanarity of the stereo rigid fluorenone carbon
3692
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE 3 (a) Experimental z-scan data (dots) and fitting curves (full lines) for B 3FL at different pulse energies. (b) q0 plotted ver-
sus intensity for the investigated PIs and an example for standard measurement and fitting data. [Color figure can be viewed in
the online issue, which is available at wileyonlinelibrary.com.]
frame and the suitably strong absorption around 400 nm,
which ensured the considerable TPA at 800 nm.
procedure. Above that, class 1 defines excellent structures
with fine hatch-lines and class 2 good structures with thick
hatch-lines (compared to class 1) or slightly contorted struc-
tures (Fig. 5). Generally, broader ‘‘perfect’’ processing win-
dows (class 1 and 2) and lower laser intensities are desired
in terms of high throughput in mass production because this
allows a splitting of the initial laser beam for parallel proc-
essing with multiple laser heads at high feed rates, while
thermally induced decomposition of the material can be
avoided. Samples rated as class 3 have shapes that can be
identified but with small mistakes (e.g., holes, exploded
regions caused by overexposure). Parts structured with laser
intensities rated as class 4 no longer showed acceptable
results. The shapes are no longer identifiable with com-
pletely missing walls and/or vast holes.
Determination of the TPA cross section gives only informa-
tion about the absorption behavior (similar to the extinction
coefficient, e, of one-photon absorption in Lambert-Beer’s
law). Thus, TPIP structuring tests were performed to further
characterize the new PIs.
TPIP Structuring Test
To estimate the activity of the PIs, defined test structures
(lateral dimension: 50 ꢁ 50 lm, 10 lm hatch-distance, 0.7
lm layer-distance, 20 layers) were written into the monomer
formulation by means of TPIP. The laser intensity was
screened in a range of 1–32 mW (measured after passing
the 100ꢁ microscope objective). A 1:1 mixture of Trimethy-
lolpropane triacrylate (TTA) and ethoxylated (20/3)-trime-
thylolpropane triacrylate (ETA) as an acrylate-based test
resin with the same molar PI concentration of 6.3 ꢁ 10ꢂ6
mol PI/g resin was used, since good results had been previ-
ously achieved under these conditions.40
In our measurements, the well-known initiator from the liter-
ature R1 can be used to build nicely shaped structures at low
laser intensities. On the other hand, the commercially avail-
able PI Irgacure 369 showed no good results at 800 nm at
the given PI concentration (6.3 ꢁ 10ꢂ6 mol PI/g resin), which
is in good agreement with the low TPA cross section.9 How-
ever, by increasing the concentration of Irgacure 369 up to
1 wt %, nice shaped structures could be obtained at 600 nm
excitation.21 The same concentration effect was also observ-
able when using the presented initiators.
The different color of the bars and their corresponding num-
bers in Figure 4 indicate the quality of the structures. At
write speeds beyond the ideal processing window, the initia-
tion threshold is not reached and incompletely connected
structures are obtained after the standard post processing
The benzophone-based initiator B3BP gives nice structures
at similarly low laser intensities as R1, but its ideal process-
ing window is significantly smaller. By changing the acceptor
group in B3BP from a benzophenone group to an anthraqui-
none moiety as in B3AN, the solubility was dramatically
decreased. In the structuring tests, we were not able to
obtain any good results with B3AN due to its poor solubility
(data not shown). The introduction of branched-alkyl chains
at the end of the amino moieties of the anthraquinone-based
initiator BB3AN could, to some extent, resolve the solubility
problem. However, high laser intensities are required to
obtain good structures in a small area. This might be
explained by the relatively small TPA cross section and the
significantly lower solubility in the resin.
TABLE 3 TPA Cross Section Values of PIs in THF by z-Scan
Measurement at 800 nm
PI
rTPA (GM)
R1
328
238
336
440
235
308
385
250
B3K21
B3BP
B3FL
B3AN
3,6-B3FL
BB3FL
BB3AN
SEVERAL AROMATIC KETONE-BASED TWO-PHOTON INITIATORS, LI ET AL.
3693

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
FIGURE 4 PI TPIP screening tests
(PI concentration: 6.3 ꢁ 10ꢂ6 mol
PI/g resin).
The fluorenone-based initiators (3,6-B3FL, B3FL, and
BB3FL) showed very good results in these tests. The ideal
processing windows are broader than that of the references,
but the required laser intensities are slightly higher. Despite
the lower cross section compared to that of B3BP, the ideal
window of 3,6-B3FL is much broader. This might be
explained by the stereo rigidity of the carbon frame of fluo-
renone, which increases the coplanarity of the entire mole-
cule and thus facilitates the electron transfer process (the
mechanism of photopolymerization is currently under inves-
tigation). The triple bonds attached at the 2- and 7-positions
of B3FL, exhibit much broader ideal processing windows
than those of 3,6-B3FL, which is in good agreement with the
z-scan values. By changing the alkyl side chains from butyl
to branched alkyl chains of BB3FL, the ideal processing win-
dows (class 1 and 2) became smaller. The difference may be
explained by the lower migration ability of the PI with the
increase of its molecule size. The lower laser intensity
required for BB3FL might be attributed to the increased sol-
ubility. It is worth mentioning that the introduction of
branched-alkyl chains significantly improved the solubility of
the fluorenone-based PIs. Laser intensities as low as 2.47
mW could be used to obtain nice structures with fairly high
concentrations (3.2 ꢁ 10ꢂ5 mol PI/g resin) of BB3FL. B3FL
turned out to be the best performing initiator in our tests
having the broadest ideal structuring process window at low
concentration (6.3 ꢁ 10ꢂ6 mol PI/g resin) and good solubil-
ity in the resin.
The tests were repeated also with lower molar PI concentra-
tions to determine the lowest possible PI concentration
under these conditions. For the lower concentration (1.6 ꢁ
10ꢂ6 mol PI/g resin, 0.1% wt of B3FL) the results for all ini-
tiators are very similar to those at higher concentration,
with the only difference having a slight shift of the ideal
processing window towards higher laser intensities.
Additionally, more complex 3D structures (Fig. 6) were
inscribed into the material volume using B3FL (6.3 ꢁ 10ꢂ6
mol PI/g resin) as initiator. These shapes demonstrate per-
fectly the advantages of TPIP compared to other rapid proto-
typing techniques. High spatial resolution, which is otherwise
inaccessible, and complex 3D structures with massive over-
hangs such as the St. Stephen’s Cathedral of Vienna [Fig.
6(a)] can be easily obtained.
EXPERIMENTAL
Materials
All reagents were purchased from Sigma-Aldrich, Fluka, and
ABCR and were used without further purification. The sol-
vents were dried and purified by standard laboratory
FIGURE 5 Classification of the structures by the typical quality
of their shapes.
3694
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE 6 3D structures (resin ETA/TTA 1:1, B3FL as initiator): (a) St. Stephen’s Cathedral; (b) Tarantula Spider; (c) detail of the
London Tower Bridge; (d) detailed view of the woodpile structure.
methods or were dried over Al2O3 cartridges. Trimethylol-
propane triacrylate (TTA, Genomer 1330) and ethoxylated
(20/3)-trimethylolpropane triacrylate (ETA, Sartomer 415)
were received as a gift from Rahn and Sartomer, respectively.
Petroleum ether (PE) refers to the fraction boiling in the
range 40-60 ꢄC. Column chromatography was performed
with conventional techniques on VWR silica gel 60 (0.040–
0.063 mm particle size). Aluminum-backed silica gel plates
were used for TLC analyses.
until the aqueous layer was colorless. The combined organic
layers were washed with water (3 ꢁ 200 mL) and brine
(2 ꢁ 100 mL), dried over sodium sulfate, filtered and con-
centrated. Purification by column chromatography (hexane)
yielded 23.26 g (60%) of the product 1b as colorless oil.
1H NMR (200 MHz, CDCl3) d 7.39 (d, J ¼ 8.9 Hz, 2H), 6.40
(d, J ¼ 9.0 Hz, 2H), 3.23 (t, J ¼ 7.4 Hz, 2H), 3.08 (d, J ¼ 7.3
Hz, 2H), 1.66-1.60 (m, 1H), 1.58 – 1.11 (m, 12H), 1.04 – 0.74
(m, 9H). 13C NMR (50 MHz, CDCl3) d 147.94, 137.56, 114.49,
75.48, 55.29, 51.60, 37.42, 30.72, 28.79, 28.36, 23.98, 23.21,
20.33, 14.13, 14.01, 10.82. Anal. Calcd for C18H30IN: C,
55.81; H, 7.81; N, 3.62. Found: C, 55.96; H, 7.33; N, 3.51.
Synthesis
N,N-dibutyl-4-iodobenzeneamine (1a),21 Bis(4-iodophenyl)-
methanone (4a),34 2,7-diiodo-9H-fluoren-9-one (4b),33 3,6-
dibromophenanthrene-9,10-dione,36 3,6-dibromo-9H-fluoren-
9-one (4c),35 2,6-diiodo-9,10-anthraquinone (4d)32 were
prepared according to the literature. A detailed synthesis
procedure is also given in the Supporting Information.
4-(4-(N,N-Dibutylamino)phenyl)-2-methylbut-3-yn-2-ol
(2a)
To a solution of 3.43 g (10.36 mmol) of N,N-dibutyl-4-iodo-
benzeneamine (1a) in 100 mL of degassed dry triethylamine
under argon atmosphere, 1.21 g (14.37 mmol) of 2-methyl-
but-3-yn-2-ol, 0.038 g (0.20 mmol) of copper(I)iodide and
0.146 g (0.21 mmol) of bis(triphenylphosphine)palladium(II)
dichloride (PdCl2(btpp)) were added and the reaction mix-
ture was stirred at 40 ꢄC. The progress of the reaction was
monitored by TLC and GC-MS analysis. After the entire con-
version of 1a, the red colored solution was filtered off, the
solid residue was washed with ethyl acetate (3 ꢁ 100 mL)
and the volatile compounds were removed in vacuo. The
crude product was dissolved in 100 mL of ethyl acetate and
extracted with water (3 ꢁ 100 mL) and brine (1 ꢁ 100 mL).
N-Butyl-N-(2-ethylhexyl)-4-iodoaniline (1b)
To a solution of 21.91 g (100 mmol) of 4-iodoaniline in
300 mL degassed dry DMF under argon atmosphere, 50 mL
(300 mmol) of 3-(bromomethyl)heptane, 41.46 g Na2CO3
powder (300 mmol), 8.31 g (50 mmol) of potassium iodide
and 2 g (6 mmol) tetrabutylammonium bromide were added
ꢄ
and the reaction mixture was stirred at 110 C until 4-iodoa-
niline was completely consumed (GC-MS analysis). Then
34 mL (300 mmol) of iodobutane were added, and heating
was continued for 3 days. Afterwards, the solution was
poured onto 500 mL water and extracted by ethyl acetate
SEVERAL AROMATIC KETONE-BASED TWO-PHOTON INITIATORS, LI ET AL.
3695

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
The combined organic layers were dried over sodium sulfate,
filtered and concentrated. Purification by column chromatog-
raphy (PE: EE ¼ 20:3) yielded 2.63 g (88%) of 2a as light
yellow, highly viscous liquid.
Bis(4-((4-(dibutylamino)phenyl)ethynyl)phenyl)metha-
none (B3BP) (5a)
0.43 g (1.01 mmol) of 4a and 14 mg of bis(triphenylphos-
phine) palladium (II) dichloride [PdCl2(btpp)] (0.02 mmol)
were dissolved in a three-neck flask in 30 mL of degassed
dry THF under argon atmosphere. Then 0.57 g (2.49 mmol)
of 3a, 30 mL of degassed dry triethylamine and 3.8 mg
(0.02 mmol) of copper(I)iodide were added successively. The
reaction mixture was stirred at 80 ꢄC. The progress of the
reaction was monitored by TLC. The red solution was poured
onto 200 mL water and extracted by ethyl acetate until the
aqueous layer was colorless. The combined organic layers
were washed with water (3 ꢁ 50 mL) and brine (2 ꢁ
50 mL), dried over sodium sulfate, filtrated and
concentrated. Purification by column chromatography (PE:
CHCl3 ¼ 1:2) yielded 0.29 g (45%) of product as yellow
powder. Mp: 67-68 ꢄC. 1H NMR (200 MHz, CDCl3) d 7.75
(d, J ¼ 8.2 Hz, 4H), 7.56 (d, J ¼ 8.2 Hz, 4H), 7.38 (d, J ¼ 8.7
Hz, 4H), 6.58 (d, J ¼ 8.9 Hz, 4H), 3.28 (t, J ¼ 7.3 Hz, 8H),
1.63-1.49 (m, 8H), 1.45 – 1.18 (m,8H), 0.93 (t, J ¼ 7.2 Hz,
12H). 13C NMR (50 MHz, CDCl3) d 195.24, 148.32, 135.77,
133.16, 130.89, 130.00, 128.85, 111.19, 107.96, 94.76, 86.92,
50.68, 29.36, 20.32, 14.04. Anal. Calcd for C45H52N2O:
C, 84.86; H, 8.23; N, 4.40; Found: C, 84.60; H, 8.08; N, 4.28.
1H NMR (CDCl3): (ppm) ¼ 7.24 (d, J ¼ 8.80 Hz, 2H), 6.53 (d,
J ¼ 9.00 Hz, 2H), 3.26 (t, J ¼ 7.72 Hz, 4H), 2.11 (s, 1H), 1.60
(s, 6H) 1.64-1.48 (m, 4H), 1.43-1.25 (m, 4H), 0.95 (t J ¼ 7.24
Hz, 6H). 13C NMR (CDCl3): 148.02, 132.95, 111.24, 108.27,
91.31, 83.31, 65.87, 50.76, 31.84, 29.45, 20.42, 14.10.
4-(4-(Butyl-(2-ethylhexyl)amino)phenyl)-2-methylbut-3-yn-
2-ol (2b)
2b was obtained from 1b according to the procedure
described for 1a as a highly viscous orange liquid with a
yield of 85%.
1H NMR (200 MHz, CDCl3) d 7.22 (d, J ¼ 8.9 Hz, 2H), 6.52
(d, J ¼ 8.9 Hz, 2H), 3.26 (t, J ¼ 7.4 Hz 2H), 3.13 (d, J ¼ 7.3
Hz, 2H), 2.06 (s, 1H), 1.70-1.58 (m, 1H), 1.58 (s, 6H), 1.43 –
1.11 (m, 12H), 0.95-0.83 (m, 9H). 13C NMR (50 MHz, CDCl3)
d ¼ 148.21, 132.72, 111.57, 108.13, 91.22, 83.14, 65.74,
55.12, 51.46, 37.51, 31.71, 30.68, 28.75, 28.49, 23.95, 23.16,
20.29, 14.08, 13.96, 10.79. Anal. Calcd for C23H37NO: C,
80.41; H, 10.86; N, 4.08; Found: C, 80.23; H, 10.98; N, 4.21.
4-(Ethynyl)-N,N-dibutylbenzeneamine (3a)
2,7-Bis((4-(dibutylamino)phenyl)ethynyl)-9H-fluoren-9-one
(B3FL) (5b)
0.58 g (10.42 mmol) of dry KOH powder were added to a so-
lution of 2.22 g (7.71 mmol) of alcohol 2a in 100 mL of dry
and degassed toluene under argon atmosphere. The suspen-
sion was heated to reflux and the acetone formed was con-
tinuously removed (the progress of the reaction was moni-
tored by TLC and GC-MS). The mixture was then cooled to
room temperature. The solid residue was filtered off and the
solution was washed with water (3 ꢁ 100 mL) and brine
(1 ꢁ 100 mL). The combined organic layers were dried over
sodium sulfate, filtered off and the solvent was evaporated.
The crude product was purified by column chromatography
(PE: EE ¼ 40:1) giving 1.54 g (87%) of 3a as pale yellow,
highly viscous oil.
{5b was prepared from 3a and 4b according to the proce-
dure described for 5a. Purification by column chromatogra-
phy (PE: EE ¼ 40:1) yielded the product 5b as red crystals
with a yield of 57%. Mp: 158-159 ꢄC. 1H NMR (200 MHz,
CDCl3) d 7.73 (d, J ¼ 1.4 Hz, 2H), 7.57 (dd, J ¼ 7.8, 1.4 Hz,
2H), 7.42 (d, J ¼ 7.8 Hz, 2H), 7.34 (d, J ¼ 8.8 Hz, 4H), 6.56
(d, J ¼ 8.9 Hz, 4H), 3.31 – 3.23 (t, J ¼ 7.3 Hz, 8H), 1.71 –
1.46 (m, 8H), 1.33-1.25 (m, 8H), 0.95 (t, J ¼ 7.2 Hz, 12H).
13C NMR (50 MHz, CDCl3) d 192.84, 148.17, 142.45, 137.13,
134.41, 133.00, 126.95, 125.32, 120.29, 111.18, 108.08,
93.19, 86.71, 50.68, 29.36, 20.31, 13.99. Anal. Calcd for
C45H50N2O: C, 85.13; H, 7.94; N, 4.41;. Found: C, 85.24; H,
1H NMR (CHCl3): d (ppm) ¼ 7.33 (d, J ¼ 9.00 Hz, 2H), 6.54
(d, J ¼ 9.00 Hz, 2H), 3.27 (t, J ¼ 7.72 Hz, 4H), 2.96 (s, 1H),
1.64-1.49 (m, 4H), 1.44-1.26 (m, 4H), 0.96 (t, J ¼ 7.23 Hz,
6H); 13C NMR (CDCl3): 148.33, 133.47, 111.15, 107.52,
85.22, 74.52, 50.78, 29.45, 20.43, 14.11.
7.81; N, 4.46.
3,6-Bis((4-(dibutylamino)phenyl)ethynyl)-9H-fluoren-9-one
(3,6-B3FL) (5c)
5c was prepared from 3a and 4c according to the procedure
described for 5a. Purification by column chromatography
(PE: CHCl3 ¼ 1:3) yielded the product 5c as red crystals
with a yield of 36%. Mp: 126-127 ꢄC. 1H NMR (200 MHz,
CDCl3) d 7.56 (s, 2H), 7.52 (s, 2H), 7.34 (s, 2H), 7.32 (d, J ¼
8.8 Hz, 4H), 6.52 (d, J ¼ 8.8 Hz, 4H), 3.12 (t, J ¼ 7.4 Hz, 8H),
1.51 (m, 8H), 1.29 (m, 8H), 0.90 (t, J ¼ 7.2 Hz, 6H). 13C NMR
(50 MHz, CDCl3) d 193.17, 148.41, 143.83, 133.25, 132.89,
131.81, 130.89, 124.14, 122.80, 111.19, 107.79, 95.57, 87.58,
50.69, 29,35, 20.31, 13.99. Anal. Calcd for C45H50N2O: C,
85.13; H, 7.94; N, 4.41;. Found: C, 85.02; H, 8.06; N, 4.19.
N-Butyl-N-(2-ethylhexyl)-4-ethynylbenzeneamine (3b)
3b was obtained from 2b according to the procedure
described for 3a. The residue was purified by column chro-
matography (PE: EE ¼ 40:1) giving the product (80%) as
light yellow oil.
1H NMR (200 MHz, CDCl3) d 7.36 (d, J ¼ 8.9 Hz, 2H), 6.60
(d, J ¼ 8.9 Hz, 2H), 3.34 (t, J ¼ 7.4 Hz, 2H), 3.20 (d, J ¼ 7.3
Hz, 2H), 2.98 (s, 1H), 2.02 – 1.71 (m, 1H), 1.69 – 1.21 (m,
12H), 1.12 – 0.63 (m, 9H). 13C NMR (50 MHz, CDCl3) d
148.55, 133.29, 111.57, 107.65, 85.09, 74.52, 55.20, 51.54,
37.59, 30.75, 28.82, 28.56, 24.01, 23.25, 20.36, 14.16, 14.03,
10.85. Anal. Calcd for C20H31N: C, 84.15; H, 10.95; N, 4.91.
Found: C, 84.33; H,10.75; N, 4.83.
2,6-Bis((4-(dibutylamino)phenyl)ethynyl)anthracene-9,10-
dione (B3AN) (5d)
5d was prepared from 3a and 4d according to the procedure
described for 5a. Purification by column chromatography
3696
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
(PE: CHCl3 ¼ 2:5) gave the product 5d as dark red solid
with a yield of 35%. Mp: 243-244 ꢄC. 1H NMR (200 MHz,
CDCl3) d 8.34 (d, J ¼ 1.6 Hz, 2H), 8.23 (d, J ¼ 8.1 Hz, 2H),
7.80 (dd, J ¼ 8.1, 1.7 Hz, 2H), 7.38 (d, J ¼ 8.8 Hz, 4H), 6.58
(d, J ¼ 8.9 Hz, 4H), 3.32 – 3.24 (t, J ¼ 7.3 Hz, 8H), 1.57 (m,
8H), 1.45 – 1.16 (m, 8H), 0.95 (t, J ¼ 7.2 Hz, 12H). 13C NMR
(50 MHz, CDCl3) d182.27, 148.72, 135.72, 133.57, 133.25,
131.30, 130.99, 129.51, 127.33, 111.18, 107.44, 97.16, 86.88,
50.70, 29.34, 20.31, 13.99. Anal. Calcd for C46H50N2O2: C,
83.34; H, 7.60; N, 4.23; Found: C, 83.48; H, 7.45; N, 4.11.
sion spectra were corrected using a correction function
obtained from secondary fluorescence standards.41
For fluorescence decay measurements with subnanosecond
time resolution, samples were excited at 395 nm with a pulsed
laser diode (Picoquant, LDH-P-C-400B) and detected with a
Hamamatsu photomultiplier tube. Interference and longpass
filters were used to select the wavelength range for detection.
Molar extinction coefficients, e, were measured in dichloro-
methane using 0.1 mm cuvettes. The emission quantum
yields were obtained using fluorescein in basic ethanol (Uf ¼
0.97)42 and coumarin 515 in acetonitrile (Uf ¼ 0.67)43 as
quantum counters. Samples and quantum counter solution
were excited at the same wavelength, ensuring an absorption
of smaller than 0.15 at and beyond the excitation wavelength.
2,7-Bis((4-(butyl(2-ethylhexyl)amino)phenyl)ethynyl)-9H-
fluoren-9-one (BB3FL) (5e)
5e was prepared from 3b and 4b according to the procedure
described for 5a. Purification by column chromatography
(PE: CHCl3 ¼ 2:3) gave the product 5e as red crystals with a
yield of 58%. Mp: 67-68 ꢄC. 1H NMR (200 MHz, CDCl3) d
7.73 (d, J ¼ 1.4 Hz, 2H), 7.56 (dd, J ¼ 7.8, 1.3 Hz, 2H), 7.42
(d, J ¼ 7.8 Hz, 2H), 7.34 (d, J ¼ 8.7 Hz, 4H), 6.58 (d, J ¼ 8.9
Hz, 4H), 3.30 (t, J ¼ 7.3 Hz, 4H), 3.17 (d, J ¼ 7.2 Hz, 4H),
1.94 – 1.65 (m, 2H), 1.65 – 1.12 (m, 24H), 1.09 – 0.75 (m,
18H). 13C NMR (50 MHz, CDCl3) d 192.85, 148.49, 142.46,
137.15, 134.42, 132.90, 126.96, 125.34, 120.29, 111.65,
108.11, 93.17, 86.73, 55.15, 51.50, 37.57, 30.68, 28.75,
28.53, 23.95, 23.18, 20.31, 14.10, 13.98, 10.80. Anal. Calcd
for C53H66N2O: C, 85.20; H, 8.90; N, 3.75; Found: C, 85.17; H,
8.62; N, 3.79.
Characterization
1H NMR (200 MHz) and 13C NMR (50 MHz) spectra were
measured with a BRUKER ACE 200 FT-NMR-spectrometer.
The chemical shift (s ¼ singlet, bs ¼ broad singlet, d ¼ dou-
blet, t ¼ triplet, m ¼ multiplet) is stated in ppm using the
nondeuterated solvent as internal standard. Solvents with a
grade of deuteration of at least 99.5% were used. Melting
points were measured on a Zeiss axioscope microscope with
a Leitz heating block and remained uncorrected.
GC-MS runs were performed on a Thermo Scientific DSQ II
using a BGB 5 column (l ¼ 30 m, d ¼ 0.32 mm, 1.0 lm film;
achiral). Elemental microanalysis was carried out with an EA
1108 CHNS-O analyzer from Carlo Erba at the microanalyti-
cal laboratory of the Institute for Physical Chemistry at the
University of Vienna.
2,6-Bis((4-(butyl(2-ethylhexyl)amino)phenyl)ethynyl)an-
thracene-9,10-dione (BB3AN) (5f)
5f was prepared from 3b and 4d according to the procedure
described for 5a. Purification by column chromatography
(PE: CHCl3 ¼ 1:3) gave the product 5f as dark red solid with
a yield of 40%. Mp: 184-185 ꢄC. 1H NMR (200 MHz, CDCl3)
d 8.34 (d, J ¼ 1.5 Hz, 2H), 8.23 (d, J ¼ 8.1 Hz, 2H), 7.80 (dd,
J ¼ 8.2, 1.6 Hz, 2H), 7.38 (d, J ¼ 8.8 Hz, 4H), 6.60 (d, J ¼ 8.9
Hz, 4H), 3.31 (t, J ¼ 7.3 Hz, 4H), 3.18 (d, J ¼ 7.2 Hz, 4H),
1.87 – 1.66 (m, 2H), 1.66 – 1.14 (m, 24H), 1.08 – 0.76 (m, J
¼ 10.4, 7.2 Hz, 18H). 13C NMR (50 MHz, CDCl3) d 182.28,
148.91, 135.74, 133.58, 133.30, 131.30, 131.00, 129.52,
127.34, 111.65, 107.46, 97.14, 86.89, 55.14, 51.49, 37.58,
30.67, 28.74, 28.53, 23.94, 23.17, 20.30, 14.09, 13.96, 10.79.
Anal. Calcd for C54H66N2O2: C, 83.68; H, 8.58; N, 3.61; Found:
C, 83.51; H, 8.63; N, 3.49.
A Ti: sapphire laser system (90 fs pulse duration, 1 kHz repeti-
tion rate) was used for the open aperture z-scan analysis. A
detailed description of the setup and the fitting equations used
can be found elsewhere.44 Rhodamine B in MeOH were used as
reference standards to verify the reliability of the experimental
setup.45 All PIs were prepared as 1.0 ꢁ 10ꢂ2 M solutions in
spectroscopic grade THF. The PI solutions were measured in a
0.2 mm thick flow cell in a nonrecycling volumetric flow of
4 mL/h. The excited volume is therefore refreshed approxi-
mately every 100 pulses, which approximately corresponds to
10 times for each z-position, which was found to be sufficient.
The measurements were carried out at different pulse ener-
gies in the range of 15-240 nJ. At higher energies a signal of
the pure solvent appears and the solvent will contribute to
the effective nonlinear absorption and even thermal effects
are more likely to influence the measurement. Care had to
be taken to collect the whole transmitted laser energy using
a big diameter and short focal length lens. Additionally, a
proper Gaussian beam profile in time and space is essential
for the analysis.
Photophysics
Cyclohexane (>99.5%, Sigma-Aldrich), butyl ether (99 þ %,
Acros), ethyl ether (99.9% spectrophotometric grade, Sigma-
Aldrich), butyl acetate (ꢅ99.0%, Fluka), propyl acetate (99%,
Alfa Aesar) and dichloromethane (>99.9%, Sigma-Aldrich)
were used as received. Both in absorption as well as emis-
sion measurements the pure solvents were measured sepa-
rately and the corresponding spectra were used as
background.
TPIP Structuring Tests
For optical absorption spectroscopy, a Cary 50 UV/vis spec-
trophotometer from Varian was used, and steady-state lumi-
nescence spectra were measured with a Horiba Fluorolog-3
(JobinYvon) or a Cary Eclipse (Varian) instrument. The emis-
Laser Device
For the direct laser writing of 3D structures, a titanium: sap-
phire (Ti.Sa) laser providing NIR pulses at 780 nm wavelength
with a pulse duration of 100 fs is used. The system operates at
SEVERAL AROMATIC KETONE-BASED TWO-PHOTON INITIATORS, LI ET AL.
3697

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
2 Moon, J. H.; Yang, S. Chem Rev 2009, 110, 547–574.
a repetition rate of 80 MHz. Direct laser writing with this sys-
tem is usually carried out at a laser power below 10 mW
(measured after passing the microscope objective). The laser is
focused by a 100ꢁ oil immersion microscope objective (NA ¼
1.4), and the sample is mounted on a high-precision piezoelec-
tric XYZ scanning stage with a 200 nm positioning accuracy.
3 Park, S.-H.; Yang, D.-Y.; Lee, K.-S. Laser Photon Rev 2009, 3,
1–11.
4 Maruo, S.; Nakamura, O.; Kawata, S. Opt Lett 1997, 22,
132–134.
5 Belfield, K. D.; Schafer, K. J.; Liu, Y.; Liu, J.; Ren, X.; Stryland,
General Procedure
E. W. V. J Phys Org Chem 2000, 13, 837–849.
For all samples the same fabrication process was imple-
mented: The optical material was drop-cast onto a glass sub-
strate. Subsequently, the samples were exposed to the laser
beam, and the focus was scanned across the photosensitive
material, which leads to an embedded 3D structure inside
the material volume. After laser writing, the unexposed ma-
terial was removed by development of the structure in etha-
nol (rinsing). The resulting structures, particularly their
structural dimensions, integrity, and surface quality, were
studied by means of SEM.
6 Ren, Y.; Yu, X.-Q.; Zhang, D.-J.; Wang, D.; Zhang, M.-L.; Xu,
G.-B.; Zhao, X.; Tian, Y.-P.; Shao, Z.-S.; Jiang, M.-H. J Mater
Chem 2002, 12, 3431–3437.
7 Belfield, K. D.; Schafer, K. J.; Liu, Y.; Liu, J.; Ren, X.; Van
Stryland, E. W. J Phys Org Chem 2000, 13, 837–849.
8 Belfield, K. D.; Ren, X.; Van Stryland, E. W.; Hagan, D. J.;
Dubikovsky, V.; Miesak, E. J. J Am Chem Soc 2000, 122,
1217–1218.
9 Schafer, K. J.; Hales, J. M.; Balu, M.; Belfield, K. D.; Van Stry-
land, E. W.; Hagan, D. J. J Photochem Photobiol A 2004, 162,
497–502.
CONCLUSIONS
A series of aromatic ketone-based two-photon initiators con-
taining dialkylamino groups as donors and triple bonds as con-
jugation bridges were successfully synthesized and evaluated.
Due to the long conjugation length and good coplanarity, the
evaluated initiators showed large two-photon cross section
(rTPA) values, with strongly solvent dependent fluorescence
lifetimes and quantum yields. With suitably strong absorption
at the desired wavelength of 400 nm, the 2-, 7-substituted fluo-
renone-based PI B3FL had the highest rTPA value (440 GM)
among the studied PIs. In terms of the two-photon polymeriza-
tion structuring test, PIs based on fluorenone as electron with-
drawing functional group with good coplanarity showed
broader processing windows than that possessing benzophe-
none or anthraquinone moieties. Significantly better results
were obtained compared to commercially available initiators
and also a slight improvement in relative to well-known highly
active initiators from the literature. The introduction of
branched-alkyl chains significantly improved the solubility of
the fluorenone-based PIs, but also slightly decreased the proc-
essing windows due to the lower migration ability of the PI
with the increase of its molecular size. B3FL turned out to be
the best performing initiator among the presented initiators,
having the broadest ideal structuring process windows and
good solubility in the resin. While high reactivity could be
obtained in TPIP with PI concentrations as low as 0.1% wt, the
novel PIs are surprisingly stable under one photon condition
and nearly no photo initiation activity was found in classical
photo DSC experiments.
10 Pond, S. J. K.; Rumi, M.; Levin, M. D.; Parker, T. C.; Bel-
jonne, D.; Day, M. W.; Bre´das, J.-L.; Marder, S. R.; Perry, J. W.
J Phys Chem A 2002, 106, 11470–11480.
11 Martineau, C.; Lemercier, G.; Andraud, C.; Wang, I.; Bouriau,
M.; Baldeck, P. L. Synth Met 2003, 138, 353–356.
12 Cumpston, B. H.; Ananthavel, S. P.; Barlow, S.; Dyer, D. L.;
Ehrlich, J. E.; Erskine, L. L.; Heikal, A. A.; Kuebler, S. M.; Lee, I.
Y. S.; McCord-Maughon, D.; Qin, J.; Rockel, H.; Rumi, M.; Wu,
¨
X. L.; Marder, S. R.; Perry, J. W. Nature 1999, 398, 51–54.
13 Xing, J.-F.; Chen, W.-Q.; Dong, X.-Z.; Tanaka, T.; Fang, X.-Y.;
Duan, X.-M.; Kawata, S. J. Photochem Photobiol A 2007, 189,
398–404.
14 He, G. S.; Tan, L.-S.; Zheng, Q.; Prasad, P. N. Chem Rev
2008, 108, 1245–1330.
15 Nguyen, K. A.; Rogers, J. E.; Slagle, J. E.; Day, P. N.;
Kannan, R.; Tan, L.-S.; Fleitz, P. A.; Pachter, R. J Phys Chem A
2006, 110, 13172–13182.
16 Ramakrishna, G.; Goodson, T. J Phys Chem A 2007, 111,
993–1000.
17 Marder, S. R.; Torruellas, W. E.; Blanchard-Desce, M.; Ricci,
V.; Stegeman, G. I.; Gilmour, S.; Bredas, J.-L.; Li, J.; Bublitz, G.
U.; Boxer, S. G. Science 1997, 276, 1233–1236.
˚
18 Norman, P.; Luo, Y.; Agren, H. Opt Commun 1999, 168,
297–303.
19 Kamada, K.; Ohta, K.; Iwase, Y.; Kondo, K. Chem Phys Lett
2003, 372, 386–393.
The authors acknowledge the financial support by the China
Scholarship Council (CSC), Swiss SNF, Austrian Science Fund
(FWF), and the Austrian Research Agency (FFG) under Con-
tracts N703 (STRUCMAT ISOTEC) and P18623 N-17.
20 Iwase, Y.; Kamada, K.; Ohta, K.; Kondo, K. J Mater Chem
2003, 13, 1575–1581.
21 Pucher, N.; Rosspeintner, A.; Satzinger, V.; Schmidt, V.;
Gescheidt, G.; Stampfl, J. R.; Liska, R. Macromolecules 2009,
42, 6519–6528.
22 Rumi, M.; Ehrlich, J. E.; Heikal, A. A.; Perry, J. W.; Barlow,
REFERENCES AND NOTES
S.; Hu, Z.; McCord-Maughon, D.; Parker, T. C.; Rockel, H.;
¨
Thayumanavan, S.; Marder, S. R.; Beljonne, D.; Bre´das, J.-L. J
Am Chem Soc 2000, 122, 9500–9510.
1 Rumi, M.; Barlow, S.; Wang, J.; Perry, J. W.; Marder, S. R.
Adv Polym Sci 2008, 213, 1–95.
3698
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
23 Rosspeintner, A.; Griesser, M.; Pucher, N.; Iskra, K.; Liska,
35 Ipaktschi, J.; Hosseinzadeh, R.; Schlaf, P.; Dreiseidler, E.;
R.; Gescheidt, G. Macromolecules 2009, 42, 8034–8038.
Goddard, R. Helv Chim Acta 1998, 81, 1821–1834.
24 Liu, J.; Gao, F.; Yang, L.; Wang, C.; Li, H.; Zhang, S. Chem
36 Bhatt, M. V. Tetrahedron 1964, 20, 803–821.
Lett 2010, 39, 324–325.
37 The majority of the observed fluorescence decays were
monoexponential. For 3,6-B3FL in BE and EE, as well as B3BP
in BE biexponential decays with a fast component have been
observed. These short components may tentatively be attrib-
uted to emission from the locally excited state.
25 Belfield, K. D.; Bondar, M. V.; Przhonska, O. V.; Schafer, K.
J. Photochem Photobiol Sci 2004, 3, 138–141.
26 Belfield, K. D.; Bondar, M. V.; Yanez, C. O.; Hernandez, F. E.;
Przhonska, O. V. J Mater Chem 2009, 19, 7498–7502.
38 Fujiwara, T.; Lee, J.-K.; Zgierski, M. Z.; Lim, E. C. Phys Chem
27 Corredor, C. C.; Belfield, K. D.; Bondar, M. V.; Przhonska, O.
Chem Phys 2009, 11, 2475–2479.
V.; Yao, S. J Photochem Photobiol A 2006, 184, 105–112.
39 Sheik-Bahae, M.; Said, A. A.; Wei, T. H.; Wu, Y. Y.; Hagan,
D. J.; Soileau, M. J.; Van Stryland, E. W. Proc SPIE-Int Soc Opt
Eng 1990, 1148, 41–51.
28 Estrada, L. A.; Neckers, D. C.
J Org Chem 2009, 74,
8484–8487.
29 Takalo, H.; Kankare, J.; Hanninen, E. Acta Chem Scand Ser
40 Heller, C.; Pucher, N.; Seidl, B.; Kalinyaprak-Icten, K.; Ullrich,
G.; Kuna, L.; Satzinger, V.; Schmidt, V.; Lichtenegger, H. C.;
Stampfl, J.; Liska, R. J Polym Sci Part A: Polym Chem 2007, 45,
3280–3291.
B 1988, B42, 448–454.
30 Suh, S. C.; Suh, M. C.; Shim, S. C. Macromol Chem Phys
1999, 200, 1991–1997.
31 Michel, P.; Gennet, D.; Rassat, A. Tetrahedron Lett 1999, 40,
41 Gardecki, J. A.; Maroncelli, M. Appl Spectrosc 1998, 52,
8575–8578.
1179–1189.
32 Gouloumis, A.; Gonzalez-Rodriguez, D.; Vazquez, P.; Torres,
T.; Liu, S.; Echegoyen, L.; Ramey, J.; Hug Gordon, L.; Guldi
Dirk, M. J Am Chem Soc 2006, 128, 12674–12684.
42 Martin, M. M. Chem Phys Lett 1975, 35, 105–111.
43 Jones, G.; Jackson, W. R.; Choi, C. Y.; Bergmark, W. R. J
Phys Chem 1985, 89, 294–300.
33 Abashev, G. G.; Lebedev, K. Y.; Osorgina, I. V.; Shklyaeva,
44 Ajami, A.; Husinsky, W.; Liska, R.; Pucher, N. J Opt Soc Am
E. V. Russ J Org Chem 2006, 42, 1873–1876.
B: Opt Phys 2010, 27, 2290–2297.
34 Novikov, A. N.; Slyusarchuk, V. T.; Chaikovskaya, L. G.
(Tomsk Polytechnic Institute, USSR; Tomsk Medical Institute).
SU 642288 A1 19790115, 1979.
45 Makarov, N. S.; Drobizhev, M.; Rebane, A. Opt Express
2008, 16, 4029–4047.
SEVERAL AROMATIC KETONE-BASED TWO-PHOTON INITIATORS, LI ET AL.
3699