Self-Diagnostic Polymers—Inline Detection of Thermal Degradation of Unsaturated Poly(ester imide)s

Article abstract of DOI:10.1002/adma.202100068

Monitoring polymer degradation is an important quest, particularly relevant for industry. Although many indirect methodologies for assessing polymer degradation exist, only few are applicable for an inline-monitoring via optic detection-systems. An inline

Full text of DOI:10.1002/adma.202100068

ReseaRch aRticle
www.advmat.de
Self-Diagnostic Polymers—Inline Detection of Thermal
Degradation of Unsaturated Poly(ester imide)s
Alexander Funtan, Philipp Michael, Simon Rost, Jürgen Omeis, Klaus Lienert,
and Wolfgang H. Binder*
methodologies are used to quantify degra-
Monitoring polymer degradation is an important quest, particularly rel-
evant for industry. Although many indirect methodologies for assessing
polymer degradation exist, only few are applicable for an inline-monitoring
via optic detection-systems. An inline-monitoring system is introduced
for the thermal degradation of crosslinked poly(ester imide)s (PEIs) by
embedding trifluoroacetyl functionalized stilbene molecules, serving as
chemosensors to track the release of generated alcoholic byproducts.
Nucleophilic addition of an alcohol to the sensors trifluoroacetyl functionality
triggers hemiacetal formation which is accompanied by significant changes in
optical properties, in turn allowing monitoring of sensor activation by direct
spectroscopy. Fluorescence spectroscopy offers an easy detection tool for the
inline thermal monitoring of PEI-degradation.
dation, either via classical ex situ methods,
such as changes in molecular weights,
mechanical properties, or changes in their
[3,4]
dielectric behavior.
However, an exami-
nation of the polymer condition during
operation, especially for installed parts,
proves to be difficult as many common ana-
lytical methods to this endeavor, e.g., ther-
mogravimetric methodologies, pyrolytic
methods, gel-permeation chromatography
(GPC), matrix-assisted laser-desorption
time-of-flight mass spectrometry (MALDI-
1
TOF-MS), or
H NMR spectroscopy,
are often not practical for inline-meas-
[
3,5]
urements.
Consequently the field of
promising
sensor technology offers
a
approach to overcome the problem of real-time investigations of
polymer materials during operation. The continuous detection
of polymer degradation requires a trigger, directly associated to
the degradation process. Thus sensing systems based on ther-
1
. Introduction
Polymers, indispensable materials in society, are prone to per-
manent degradation by temperature, mechanical force, or
[6,7]
[8]
[9,10]
light, resulting in deterioration or loss of the original proper-
mochromic,
mechanochromic, electrostatic,
and mass-
ties by depolymerization- or cross-linking processes.[
1–3]
Since
sensitive approaches, or via molecularly imprinted polymers
[11]
[12,13]
the primordial degradation of polymeric materials may induce
serious impact by failure of structural materials or devices,
there is increased interest in monitoring the level and nature of
degradation, with a focus on its time-dependent profile. Based
on the wide structural diversity of different polymers, various
(MIPs)
have been developed to detect and quantify polymer
degradation or damage. MIPs frequently have been used to
track specific template molecules, enabling analyte detection
[
14,15]
by, e.g., voltammetry
within epoxy- and polycarbonate
[14]
resins
such as creatinine
or the detection of amino containing compounds
[16]
[17]
or dansylphenylalanine
via optical
[18]
spectroscopy.
Chemosensors in particular allow the detec-
A. Funtan, Dr. P. Michael, Prof. W. H. Binder
Macromolecular Chemistry
Institute of Chemistry
Faculty of Natural Science II
Martin Luther University Halle-Wittenberg
Von-Danckelmann-Platz 4, 06120 Halle (Saale), Germany
E-mail: wolfgang.binder@chemie.uni-halle.de
Dr. S. Rost, Dr. K. Lienert
ELANTAS Europe GmbH
Großmannstraße 105, 20539 Hamburg, Germany
Dr. J. Omeis
ALTANA AG
Abelstraße 43, 46483 Wesel, Germany
tion of degradation products released during polymer decom-
position by activation, constituting a promising sensor class to
[19–21]
track polymer degradation.
Thus, several optical sensing
principles enable a fast and easy visualization by UV–vis and
[
10,22,23–25]
fluorescence spectroscopy, or even by the naked eye.
Conjugated polymers
[6,13,25]
such as carbazole-based conju-
gated polymers display reversible on/off fluorescence in the
presence of electron-rich respectively electron-deficient arene
[26]
[27]
vapors,
detecting explosives such as 2,4,6-trinitrotoluene.
Single-stranded DNA mixed with a cationic polythiophene
derivate was used to transduce protein binding events into an
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adma.202100068.
[28]
optical signal.
Other chromophoric sensing principles rely
on 1,2-naphtoquinone-4-sulfonic acid sodium salts embedded
©
2021 The Authors. Advanced Materials published by Wiley-VCH
[24]
within a modified PDMS matrix or on trifluoroacetyl func-
tionalized azobenzenes within polyacrylates and epoxy resins,
GmbH. This is an open access article under the terms of the Creative
Commons Attribution-NonCommercial-NoDerivs License, which per-
mits use and distribution in any medium, provided the original work is
properly cited, the use is non-commercial and no modifications or adap-
tations are made.
[20,29]
serving as chromogenic receptor for amines.
We here use stilbene-based sensor-molecules to continu-
ously track the thermal degradation process of a poly(ester
DOI: 10.1002/adma.202100068
imide)-polystyrene (PEI) resin via a fluorescence based
Adv. Mater. 2021, 33, 2100068
2100068 (1 of 9)
© 2021 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
Figure 1. A) Trifluoroacetyl functionalized stilbene dyes Stil-3 and Stil-4 are used as sensor molecules, allowing hemiacetal formation by reaction with
released neopentyl glycol (NPG) during degradation of a poly(ester imide) PEI triggering detectable optical changes by UV–vis and fluorescence spectros-
copy (R = –C H ). B) Chemical identity of the products generated during thermal degradation of the PEI-resins, with NPG being the main detected analyte.
6
13
response (Figure 1). PEIs, widely used in the electrical industry
neopentyl glycol (NPG) released during their thermal aging via
a nucleophilic addition, in turn triggering hemiacetal formation
(Figure 1A,B). This sensor activation can easily be followed by
a shift of λabs,max and λem,max to smaller wavelengths via UV–vis
and fluorescence spectroscopy as depicted in Figure 1A. Moni-
toring the signals of both, the virgin and the activated sensor
molecule, enables to calculate the thermal degradation process
of the poly(ester imide)-polystyrene resins.
as electrical insulating materials for copper wires and coils
in electric motors,[
30,31]
are known to degrade thermally by
releasing alcohols and amines by cleavage of imide- and ester
[2,32]
bonds via a main-chain scission mechanism.
PEIs feature high temperature and mechanical resistance,
Although
[
30,33]
they are thermally degraded over a period of several years
making a continuous monitoring of their degradation an
attractive feature by detecting the release of alcoholic degra-
dation products. Conceptually (see Figure 1A) we have used
substituted stilbenes as sensors, as their molecular design
2
. Results and Discussion
can be linked to the optical properties of choice for a specific
sensor-application,[
19,20,34]
reversibly altering their emission
2.1. Sensor Activation in Solution
[23,35]
from non-fluorescent to fluorescent.
The designed trif-
luoroacetyl-functionalized stilbene dyes Stil-3 and Stil-4, when
embedded into a PEI matrix, can react with the target molecule
We aimed to develop a sensor system for the detection of
thermal polymer degradation within a solid poly(ester imide)
Adv. Mater. 2021, 33, 2100068
2100068 (2 of 9)
© 2021 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
Figure 2. A) Stil-3 and Stil-4 in dioxane under UV light (λ = 366 nm) before and after activation with methanol. B,C) Sensor activation of Stil-3
−
1
(
7.5 µmol L , λ = 442 nm, λem = 557 nm) after addition of different analytes monitored by fluorescence decay of the trifluoroacetyl emission
ex
−
1
maxima. D,E) Sensor activation of Stil-4 (7.5 µmol L , λ = 406 nm, λem = 496 nm) after addition of different analytes monitored by fluores-
cence increase of the hemiacetal emission maxima.
ex
PEI resin crosslinked with styrene. Prior to investigations of
solid samples by fluorescence spectroscopy, preliminary experi-
ments in solution using dioxane were performed to evaluate
in the presence of amines due to their higher nucleophilicity
compared to alcohols. Furthermore a strong dependence on the
nature of the alcohol was observed, which essentially results
from the steric demands of the analyte. Fastest sensor response
was found for methanol and 1,4-butanediol, the slowest con-
version was observed for tert-butanol, displaying the highest
steric demand of all investigated analytes. It is important to
note that neopentyl glycol shows a good reaction with both sen-
sors molecules, Stil-3 and Stil-4, despite its two methyl groups,
constituting a significant steric hindrance. The faster response
of Stil-4 into the optical detectable hemi-form, resulting from
the weaker electron-donor properties of the distyryl substitu-
tion compared to Stil-3, is highlighted together with the higher
quantum yield of its hemiacetal form, potentially advantageous
for the subsequent solid-state reactions, where diffusion is
restricted and detection thus reduced due to additional matrix
the suitability of both sensor molecules, Stil-3 (ε
=
λ=440nm
−1 −1
−1
−1
2
6 500 L mol cm ) and Stil-4 (ε
= 42 700 L mol cm ),
λ=438nm
especially with regard to response and specificity. Linear absorp-
tion and fluorescence behavior were observed for concentrations
−3
−1
up to 9 × 10 µmol mL (Figures S1 and S2 in the Supporting
Information), as response of Stil-3 and Stil-4 toward various
analytes, especially amines and alcohols via nucleophilic addi-
tion to the trifluoroacetyl group and the subsequent change in
their fluorescence spectra. Results are summarized in Figure 2
and Figures S3 and S4 (Supporting Information) exemplarily
show the shift of emission spectra associated with the sensor
response. For kinetic measurements a dye concentration of
–
3
−1
7
.5 × 10 µmol mL and analyte concentrations between 0.5
−1
[34]
and 1.0 mmol mL were used, except for n-butylamine (c = 8.4
and 16.8 nmol mL ) due to its fast response time.
interference. As the response of Stil-4 toward n-butylamine
−1
was fast, it prevented a reliable monitoring of the kinetics even
−
1
Both sensor molecules, Stil-3 and Stil-4, can be well activated
by addition of n-butylamine as well as by alcohols in reasonable
response-times of seconds. Whereas Stil-3 shows a decrease
of the fluorescence intensity after activation Stil-4 shows an
increase in fluorescence of its generated hemi-form (compare
Figures S3 and S4 in the Supporting Information). A focus was
placed on neopentyl glycol, later serving as the main analyte
during thermal PEI degradation. Sensor response is accelerated
though low concentrations (8.4 and 16.8 nmol L ) were used.
Both sensors show an only low conversion triggered by mois-
ture through nucleophilic addition of water molecules, unavoid-
able for long-term measurements over 70 d but less prominent
in the later application due to restricted diffusion processes in
the solid-state. Overall, both Stil-3 and Stil-4 showed excellent
and promising response in solution with an only low response
to water.
Adv. Mater. 2021, 33, 2100068
2100068 (3 of 9)
© 2021 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
Figure 3. A–C) Specimen of PS/DVB: A) with embedded Stil-3 and Stil-4 dye under UV light (λ = 366 nm) before and after 15 d storage in n-butylamine
−
1
−1
solution, B) with Stil-3 (0.25 mg g resin, λ = 414 nm) and C) with Stil-4 (0.25 mg g resin, λex = 409 nm) embedded in PS/DVB after storage
ex
−
1
in n-butylamine solution (1.5 mmol mL ) monitored by fluorescence spectroscopy. D) Summarized results showing fluorescence intensities plotted
versus time together with corresponding control experiments.
2
.2. Sensor Activation in a PS/DVB Resin by Amines
a second emission maximum at 475 nm, which is not the case
for the hemiaminal form of Stil-3 due to its low quantum yield.
Overall, the findings from solid-state measurements correlate
to results from solution experiments. Reference specimen of
PS/DVB containing Stil-3 and Stil-4 stored in water lack the ini-
tial drop of fluorescence intensity, and the overall change is in the
order of −1% and 2%, hence being significantly smaller. Addi-
tionally, a sample of PS/DVB without dye stored in n-butylamine
solution, shows only negligible changes of its fluorescence inten-
sity. Therefore matrix/solvent interference can be excluded and
measurable changes are attributed to sensor activation by the
n-butylamine analyte with an excellent stability against water.
As a model resin to probe application in crosslinked resin sys-
tems a poly(styrene divinylbenzene) resin PS/DVB was syn-
thesized by free radical crosslinking of styrene with 4 mol%
of divinylbenzene, 1 mol% of azobisisobutyronitrile (AIBN)
−4
−4
containing 2.5 × 10 to 5.0 × 10 w% of Stil-3 or Stil-4. Activa-
tion was triggered by exposing the specimen to aqueous n-but-
ylamine solution, providing a fast response. Subsequently we
focused on solid-state fluorescence as it proves more suitable
compared to solid-state UV−vis spectroscopy. The corresponding
fluorescence spectra after sensor activation are depicted in
Figure 3b,c. As PS/DVB itself is non-fluorescent (compare
Figure S5 in the Supporting Information) it provides a simple
model system, proving the embedding of the sensor molecules
in a solid polymer and its successful activation and detection in
their virgin and activated forms. To exclude cross reaction or sol-
vent/matrix interference proper control experiments were con-
ducted by exposing specimen of PS/DVB with or without Stil-3
or Stil-4 to either water or an aqueous n-butylamine solution.
Results are presented in Figure 3d. Sensor-containing specimen
stored in n-butylamine solution show a distinct decrease of
the original emission maximum at 535 and at 557 nm, corre-
sponding to the virgin trifluoroacetyl form, indicating the forma-
tion of the activated hemiaminal. As the fluorescence quantum
yield of the virgin trifluoroacetyl moiety is higher for Stil-3 com-
pared to Stil-4,[ fluorescence intensity is significantly higher
for Stil-4 albeit at the same concentration. Fluorescence intensity
decreases with a maximum value of 24% for Stil-4 at a concen-
2.3. Sensor Activation in PEI Resin by Amines
The PEI matrix, a network of an unsaturated poly(ester imde)
crosslinked with styrene, shows thermal degradation into
amines and alcohols with neopentyl glycol being one of the
main volatile analytes, according to a main chain scission reac-
[
32]
tion in the first pyrolysis step. To assess the thermal stability
of the used PEI, a sample was measured by TGA and it was
found that PEI smoothly decomposes between 350 and 450 °C
(compare Figure S6 in the Supporting Information). PEI deg-
radation experiments were conducted at 200 and 220 °C, below
its decomposition temperature and more representative for
application, by inserting a resin sample into a glass-tube, placed
in a sand-bath for continuously heating. This chosen tempera-
ture is also inducing an accelerated kinetics, representative for
a long-term aging of the resin at the usually lower tempera-
tures of such electrical devices of ≈100 °C. Gaseous degradation
34]
tration of c = 0.5 mg g− down to 27% for Stil-3 at c = 0.25 mg g .
Formation of the hemiaminal form of Stil-4 can be observed by
1
−1
Adv. Mater. 2021, 33, 2100068
2100068 (4 of 9)
© 2021 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
products were flushed into an equipped adsorber column by
applying a constant flow of oxygen (4 mL min ) and all con-
densable degradation products were collected after 500, 1000,
total fluorescence decrease is significantly higher than for PS/
DVB, explained by a better diffusion of the aqueous solutions
within PEI, facilitating sensor activation in bulk material.
−1
[37]
3
000, and 5000 h by thoroughly washing the adsorption mate-
Therefore a fluorescence decrease of 21% for Stil-3 and 23%
for Stil-4 was observed for reference specimen stored in water,
which was not observable for PS/DVB specimen. Additional con-
trol experiments demonstrate that water has no influence on the
intrinsic fluorescence of the PEI matrix (compare Figure S10 in
the Supporting Information) and storage in n-butylamine solu-
tion only slightly increases fluorescence intensity. The major
drawback of the PEI resin is the lack of a second emission max-
imum of Stil-4 which corresponds to the activated sensor.
rial with acetone. The mass loss of the specimen was recorded
by weighing the PEI specimen after each degradation step and
is depicted in Figure S7 in the Supporting Information. Within
the first 500 h the mass loss of PEI at degradation tempera-
tures of 200 and 220 °C is attributed to evaporation of mono-
meric residues, additives and residual moisture from the resin
dominating the actual polymer degradation. After 500 h a sig-
nificantly slower degradation is observed for PEI at 200 °C com-
pared to 220 °C, explained by the progressive crosslinking- and
cyclization processes in PEI, leading to a slower mass loss for
the sample at 220 °C after 3000 h.[
3,36]
Presence of neopentyl
2.4. Sensor Activation in PEI Resin by Neopentyl Glycol
glycol during degradation was proven by analysis of the des-
orbed degradation products using H NMR spectroscopy and
1
Both, Stil-3 and Stil-4 were embedded in cross-linked PEI and
specimen were exposed to neopentyl glycol to access sensor
activation. Since thermal stress leads to aging of PEI and thus
to the release of neopentyl glycol, two approaches were pur-
sued. First, PEI samples (a) containing sensor molecules were
stored in neopentyl glycol at room temperature in vacuum to
exclude PEI aging. Second, PEI samples (b) with embedded
sensor molecules were placed in a lid of a petri dish where the
lower part was covered with neopentyl glycol. Subsequently,
the petri dish was placed in a sand bath at 180 °C, thus being
exposed to the gaseous neopentyl glycol atmosphere, at the
same time reducing direct thermal load on PEI due to the dis-
tance from the sand bath. Additionally PEI samples (c) without
sensor molecule were equally treated as samples (b), serving as
control experiment. Results for Stil-3 are shown in Figure 5d,
results for Stil-4 are shown in Figure S11 in the Supporting
Information. As expected, for sample (a) containing Stil-3, a
HPLC/ESI-TOF MS and calculated values are presented in
Table S1 (see Supporting Information, recorded spectra are
exemplarily shown in Figures S8 and S9). PEI was fabricated
as a two component systems by mixing two formulations in
a ratio of 50:50 w%, containing the poly(ester imide), styrene
and radical initiator. To the mixture either Stil-3 or Stil-4 (1 w%)
were added and the specimen were cured at 80 °C for 50 min,
followed by 5 min at 140 °C.
Similar to PS/DVB resin, sensor activation was first probed
under accelerated conditions by storing the specimen in aqueous
n-butylamine solution and monitoring emission decay by fluores-
cence spectroscopy as depicted in Figure 4b,c. Results are sum-
marized in Figure 4d. Specimen containing sensor molecules
again show a significant drop in emission intensities of 56%,
with the response of Stil-4 being faster compared with Stil-3,
coinciding with the observation from solution experiments. The
Figure 4. A) Specimen of PEI with embedded Stil-3 and Stil-4 dye under UV light (λ = 366 nm) before and after 22 d storage in n-butylamine solu-
−
1
−1
tion. B,C) Sensor activation of Stil-3 (1 mg g resin, λex = 447 nm) (B) and Stil-4 (1 mg g resin, λex = 410 nm) (C) embedded in PEI after storage
−
1
in n-butylamine solution (1.5 mmol mL ) monitored by fluorescence spectroscopy. D) Summarized results showing fluorescence intensities plotted
versus time together with corresponding control experiments.
Adv. Mater. 2021, 33, 2100068
2100068 (5 of 9)
© 2021 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
Figure 5. A) Specimen of PEI under UV light (λ = 366 nm) before and after 62 d storage in neopentyl glycol atmosphere at 180 °C. B) Sensor activation
−
1
of Stil-3 (1 mg g resin, λex = 447 nm) embedded in PEI after storage in neopentyl glycol atmosphere at 180 °C monitored by fluorescence spectroscopy.
C,D) Control experiment under the same conditions as listed for (B) without neopentyl glycol summarized results showing fluorescence intensities
plotted versus time together with corresponding control experiments.
total decrease of the fluorescence intensity of 8% was observed,
indicating its successful activation by neopentyl glycol.
Sample (b) with embedded Stil-3, indicates a decrease of
fluorescence by 26%, proving a faster sensor activation in gas-
eous neopentyl glycol atmosphere (Figure 5b). For sample (c)
containing the sensor molecule Stil-3 devoid of neopentyl glycol
a decrease of fluorescence by 4%, thus distinctly smaller com-
pared to the activation experiment (b), was observed (Figure 5c).
Additional control experiments of PEI without sensor verify
that intrinsic fluorescence properties of the matrix are unaf-
fected by experimental conditions, therefore thermal influences
on PEI matrix are excluded and suitability of Stil-3 within
PEI was demonstrated.
total this inline detection system allows the continuous moni-
toring of PEI degradation as it happens in, e.g., all commercial
electric engines, where wire-enamels based on PEI are used for
insulation—a big step toward their reliable function.
4. Experimental Section
Reagents and Chemicals: All chemicals were purchased from
commercial suppliers and, except the following, used without further
purification. Poly(ester imide) resin PEI was provided by ELANTAS
GmbH. Styrene was destabilized by washing with a solution of
NaOH (10 wt%) and drying over Na SO4 before usage. Adsorber
2
material Amberlite XAD-4 was washed thoroughly with distilled water
and acetone before the adsorber column was charged. AIBN was
recrystallized from methanol at 50 °C before usage.
Instruments and Methods: UV–vis absorbance spectra of dissolved
sensor molecules were recorded on a Perkin Elmer UV-VIS Lambda
3
. Conclusion
365 spectrometer. For solution experiments typically a stock-solution of
Two different sensor molecules, Stil-3 and Stil-4, were success-
fully embedded within a poly(styrene divinylbenzene) resin PS/
DVB. Sensor response and detection were proven under accel-
erated conditions by conversion of the virgin trifluoroacetyl
form into the hemiaminal after exposure to n-butylamine solu-
tion, demonstrating the usability of the chosen detection system
in the solid-state. Function of the sensor system was demon-
strated for poly(ester imide) PEI specimen in both, solution
and thermal experiments, thus providing a powerful tool for
tracking the thermal long-term degradation of poly(ester imide)
impregnating resins for coils in electric engines. In future
work, we will further investigate the in situ activation of Stil-3
and Stil-4 under engine-working conditions, in which PEI sam-
ples release the required analyte neopentyl glycol by thermal
degradation and consequently trigger hemiacetal formation. In
either 1 mg Stil-3 or Stil-4 in 1 mL dioxane was prepared. The required
amount of stock-solution was transferred into a cuvette and was filled up
to 3 mL using dioxane. For solid-state measurements, the spectrometer
was equipped with a Lambda 365 Integrating sphere. Fluorescence spectra
of both, dissolved and embedded sensor molecules, were recorded on
an Agilent Technologies Cary Eclipse Fluorescence Spectrophotometer,
detector voltage was set to 500 V. For solution experiments, first
the analyte was transferred into a cuvette and dissolved in dioxane.
Afterward the required amount of the stock-solution containing Stil-3
respectively Stil-4 was added and the fluorescence measurement was
1
13
started immediately after sensor addition. H NMR-, C NMR-, and
1
9
F NMR spectra were recorded either on an Agilent Varian Gemini
00 (400 MHz) or on an Agilent Varian Unity Inova (500 MHz) spectrometer
2
at 27 °C. Chemical shifts (δ) were given in parts per million (ppm)
1
and were referred to the solvent signal of CDCl (7.26 ppm ( H NMR),
7
3
13
7.0 ppm ( C NMR)). Values of coupling constants J are given in
hertz (Hz). HPLC measurements were performed on a VWR Hitachi
Adv. Mater. 2021, 33, 2100068
2100068 (6 of 9)
© 2021 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
Chromaster on an Atlantis T3 (5.0 µm, 4.6 mm x 250 mm) C18 reversed
(3 × 50 mL). The combined organic phase was washed with distilled
water (2 × 75 mL) and dried over Mg SO . The obtained oil was purified
−1
column, using a flow rate of 0.8 mL min , an injection volume of 10 µL
2
4
and UV–vis detection. For sample preparation HPLC-solvent was used
by flash chromatography on silica gel using hexane/ethyl acetate
−1
[39]
with sample concentrations of 1 mg mL . ESI-TOF MS measurements
were performed on a Bruker Daltonics microTOF spectrometer, using
(19:1 = v/v) as the eluent, yielding 2 as a yellow liquid.
1
H NMR (400 MHz, CDCl , δ): 0.89 (t, J = 6.7 Hz, 6H, H-1), 1.30 (m, 12H,
3
−1
direct injection with a flow rate of 180 µL h in either positive or negative
H-2 + H-3 + H-4), 1.54 (m, 4H, H-5), 3.21 (m, 4H, H-6), 6.40 (m, 2H, H-9),
1
3
mode. For sample preparation HPLC-grade solvent was used with
7.42 (m, 2H, H-10); C NMR (100 MHz, CDCl , δ): 14.2 (C-1),
3
−1
sample concentrations of 0.5 mg mL and without addition of salt. For
22.8 (C-2) 27.0 (C-4), 27.2 (C-5), 31.9 (C-3), 51.2 (C-6), 75.5 (C-11), 114.2 (C-9),
137.8 (C-10), 147.8 (C-8).
−1
HPLC/ESI-TOF MS coupling a sample concentration of 1 mg mL was
−1
used. 10 µL were injected on the column and a flow rate of 0.5 mL min
Synthesis of 4-Bromo-4′-(dihexylamino)stilbene 3: A mixture of 2
(9.00 g, 23.40 mmol), 4-bromostyrene (3.78 ml, 5.31 g, 29.25 mmol,
1.25 eq.), palladium diacetate (57.78 mg, 0.261 mmol, 0.011 eq.), tri-o-
tolylphosphine (150.84 mg, 0.495 mmol, 0.021 eq.), and triethylamine
(19.26 mL, 138.06 mmol, 5.90 eq.) was dissolved in dry DMF (30 mL)
and refluxed at 115 °C under nitrogen. To the cooled solution water
and chloroform were added and the aqueous layer was extracted with
chloroform (3 × 50 mL). The combined organic layer was washed with
distilled water (3 × 75 mL) and dried over Mg SO The crude product
was used. A TripleToF 6600–1 mass spectrometer (AB Sciex) was used
for high-resolution mass spectrometry, which was equipped with an ESI-
DuoSpray-Ion-Source (operated in positive ion mode) and was controlled
by Analyst 1.7.1 TF software (AB Sciex). The ESI source operation
parameters were as follows: ion spray voltage: 4500 V, nebulizing gas:
6
0 p.s.i., source temperature: 450 °C, drying gas: 70 p.s.i., curtain gas:
1
55 p.s.i. Data acquisition was performed in the MS -ToF mode, scanned
from 100 to 2500 Da with an accumulation time of 50 ms.
2
4.
TGA measurements were performed on a TGA Netzsch T210, using
sample amounts of 5 mg. Measuring was done under a flow of nitrogen
was purified by column chromatography on silica gel using hexane/
chloroform (2:1 = v/v) as the eluent and afterward was recrystallized
−
1
−1
[39]
(
(
20 mL min ) and a heating rate of 5 K min . Thin-layer chromatography
TLC) was performed on either Merck silica gel 60 or Macherey–Nagel
from methanol yielding 3 as a yellow solid.
1
H NMR (400 MHz, CDCl , δ): 0.91 (m, 6H, H-1), 1.32 (m, 12H, H-2
3
ALUGRAM ALOX N/UV254 sheets. Spots on TLC plates were visualized
by UV light (254 or 366 nm) or by using the oxidizing agent “blue stain”
consisting of (NH ) Mo O ∙4H O (2.5 g) and Ce(SO ) ∙4H O (1 g)
+ H-3 + H-4), 1.59 (m, 4H, H-5), 3.28 (m, 4H, H-6), 6.61 (d, J = 8.9 Hz,
2H, H-9), 6.77–7.02 (m, 2H, H-11 + H-12), 7.31–7.43 (m, 6H, H-10 + H-14
1
3
4
6
7
24
2
4 2
2
+ H-15); C NMR (100 MHz, CDCl , δ): 14.2 (C-1), 22.8 (C-2), 27.0 (C-4),
3
dissolved in distilled water (90 mL) and concentrated H SO (6 mL).
27.4 (C-5), 31.9 (C-3), 51.2 (C-6), 111.8 (C-9), 120.0 (C-17), 122.4 (C-11),
124.2 (C-10), 127.5 (C-12), 128.0 (C-13), 129.8 (C-15), 131.7 (C-16), 137.5
(C-14), 148.2 (C-8).
2
4
For column chromatography either Merck Kieselgel 60 (230–400 mesh)
or neutral aluminum oxide (Brockmann I Typ 507, 150 mesh) from
Sigma Aldrich was used. For degradation experiments, PEI was cured
in a teflon-mould (50 mm × 10 mm × 10 mm), yielding specimen with
a weight of 5 g. PEI specimen were stored in a glass-tube which was
placed in a sand-bath for continuously heating the samples at 200 and
Synthesis of 4-Trifluoracetyl-4′-(dihexylamino)stilbene Stil-3: 3 (250.0 mg,
0.55 mmol) was dissolved in 5 mL of dry tetrahydrofuran and cooled
to −78 °C using methanol/liquid nitrogen as the cooling agent. To the
stirred solution was added a 1.6 ꢀ solution of butyllithium in hexane
(387.5 µL, 0.65 mmol, 1.10 eq.) and stirring was continued for 30 min.
Then, ethyl trifluoroacetate (70.0 µL, 0.85 mmol, 1.03 eq.) was added
and the solution was stirred for another 60 min. The solution was
warmed up to room temperature, 1 mL of methanol was added and,
subsequently, 20 mL of diethylether. The orange solution was washed
once with 3 mL of 1 ꢀ hydrochloric acid, 3 mL of a sodium bicarbonate
solution and twice with 15 mL of distilled water. After drying over
Mg SO and evaporation to dryness, the orange oil was purified by flash
2
20 °C. The glass-tube was equipped with an adsorber column filled
with Amberlite XAD-4 (50 g). By applying a constant flow of oxygen
−
1
(4 mL min ), gaseous degradation products were flushed into the
adsorber column and adsorbed degradation products were collected
after 500, 1000, 3000, and 5000 h by washing the adsorption material
thoroughly with acetone. Quantification of neopentyl glycole was done
1
by H NMR spectroscopy, using 1,3,5-trioxane as external standard.
Synthesis of 2,2,2-Trifluoro-1-(4-vinylphenyl)ethan-1-one 1:
A two-
2
4
chromatography on silica gel using hexane/dichloromethane (v/v = 4:1)
[39]
as the eluent, yielding Stil-3 as orange crystals.
1
H NMR (400 MHz, CDCl , δ): 0.91 (t, J = 6.8 Hz, 6H, H-1), 1.33 (m,
3
12H, H-2 + H-3 + H-4), 1.60 (m, 4H, H-5), 3.30 (m, 4H, H-6), 6.64 (d, J =
8.9 Hz, 2H, H-9), 6.91 (d, J = 16.2 Hz, 2H, H-12), 7.22 (d, J = 16.2 Hz, 2H,
H-13), 7.42 (d, J = 8.8 Hz, 2H, H-10), 7.57 (d, J = 8.5 Hz, 2H, H-15), 8.01
1
3
(d, J = 7.9 Hz, 2H, H-16); C NMR (100 MHz, CDCl , δ): 14.2 (C-1), 22.9
3
(C-2), 27.0 (C-4), 27.4 (C-5), 31.9 (C-3), 51.2 (C-6), 111.5 (C-9), 121.3 (C-7),
125.9 (C-12), 128.6 (C-13), 130.9 (C-15), 134.0 (C-16), 146.1 (C-14), 148.9
1
9
(C-8), 179.8 (C-18); F NMR (470 MHz, CDCl , δ): −71.1 (s, 3F); UV–vis
3
(dioxane): λmax(ε) = 440 nm (26 500). HR-MS (See Figure S29 in the
Supporting Information, M–H (460,2817) and M–H 0–H (478,2933))
+
+
2
MgSO . The solvent was removed under vacuum and the crude product
was purified by column chromatography using pentane / diethyl ether
Synthesis of 1-[4-(2-[4-[2-(4-Dihexylaminophenyl)-vinyl]-phenyl]-vinyl)-
phenyl]-2,2,2,-trifluoroethanone Stil-4: A two-neck flasked was heated
under vacuum and flushed with nitrogen three times. Then 3 (2.00 g,
4.52 mmol, 1.15 eq.), 1 (788.64 mg, 3.94 mmol, 1.00 eq.), and dichlorobis-
(triphenylphosphine)-palladium(II) (74.66 mg, 0.10 mmol, 0.027 eq.)
were dissolved in triethylamine (15.00 mL, 108.00 mmol, 27.4 eq.) and
dry DMF (16.00 mL). The reaction mixture was stirred at 115 °C for
16 h. The solution was allowed to cool down and water (50 mL) was
added. The aqueous layer was extracted with DCM (3 × 50 mL) and the
combined organic layers were washed with distilled water (3 × 100 mL). After
4
1
13
19
(
v/v = 50:1) and isolated as a colorless oil (for H-, -C, and F NMR
spectra of all synthesized compounds see Figures S14 to S26 in the
[38]
Supporting Information).
1
H NMR (400 MHz, CDCl , δ): 5.49 (d, J = 10.9 Hz, 1H, H-1a), 5.96 (d,
3
J = 17.6 Hz, 1H, H-1b), 6.79 (dd, J = 17.6 Hz, J = 10.9 Hz, 1H, H-2), 7.56
1
3
(
(
(
m, 2H, H-4), 8.05 (m, 2H, H-5); C NMR (100 MHz, CDCl , δ): 116.9
3
C-1), 118.7 (C-8), 126.9 (C-6), 129.2 (C-4), 130.7 (C-5), 135.6 (C-2), 144.7
C-3), 180.1 (C-7); 1 F NMR (470 MHz, CDCl , δ): -71.4 (s, 3F).
9
3
Synthesis of p-Iodo-N,N-dihexylaniline 2: A mixture of p-iodoaniline
the organic layer was dried with MgSO the solvent was removed under
4
3.79 g, 17.31 mmol), 1-bromohexane (8.50 mL, 10.00 g, 60.58 mmol,
.50 eq.), N-ethyldiisopropylamine (10.34 mL, 7.83 g, 60.58 mmol),
vacuum and the crude product was purified by column chromatography on
neutral Al O using hexane/ethyl acetate (v/v = 2:1) as the eluent, yielding
2
3
[34]
Stil-4 as red crystals.
H NMR (400 MHz, CDCl , δ): 0.91 (m, 6H, H-1), 1.33 (m, 12H, H-2 +
H-3 + H-4), 1.60 (m, 4H, H-5), 3.29 (m, 4H, H-6), 6.62 (d, J = 8.9 Hz, 2H,
1
3
00 mL of distilled water and the product was extracted with chloroform
Adv. Mater. 2021, 33, 2100068
2100068 (7 of 9)
© 2021 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
H-9), 6.90–7.31 (m, 4H, H-12 + H-13 + H-18 + H-19), 7.40 (d, J = 8.8 Hz,
[3] N. C. Billingham, in Encyclopedia of Polymer Science and Technology,
John Wiley & Sons, New York 2002, p. 1.
[4] J. R. White, A. Turnbull, J. Mater. Sci. 1994, 29, 584.
[5] I. C. McNeill, in Comprehensive Polymer Science and Supplements
2
H, H-10), 7.51 (m, 4H, H-15 + H-16), 7.64 (d, J = 8.5 Hz, 2H, H-21),
1
3
1
9
8
.07 (d, J = 7.9 Hz, 2H, H-22); C NMR (100 MHz, CDCl , δ); F NMR
3
(
(
(
470 MHz, CDCl , δ): −71.2 (s, 3H); UV–vis (dioxane): λ (ε) = 438 nm
3
m
a
x
+
42 700). HR-MS (See Figure S30 in the Supporting Information, M–H
(Eds: G. Allen, J. C. Bevington), Pergamon, Amsterdam, The
+
562,3280) and M–H 0–H (580,3387)).
2
Netherlands 1989, p. 451.
Preparation of Model Resin PS/DVB and Sensor Embedding: The
required amount of dye Stil-3 or Stil-4 (dye concentrations between
[
6] M. Leclerc, Adv. Mater. 1999, 11, 1491.
[
7] W. J. Yoo, J. K. Seo, K. W. Jang, J. Y. Heo, J. S. Moon, J.-Y. Park,
−
1
−1
0
.125 mg g PS/DVB up to 0.5 mg g PS/DVB were used) was dissolved
B. G. Park, B. Lee, Opt. Rev. 2011, 18, 144.
in the reaction mixture of styrene, divinylbenzene (4 mol% referred to
styrene), and AIBN (1 mol% referred to styrene). The reaction mixture
was transferred into a teflon mould (25 mm × 10 mm × 2 mm) and
was stored in an oven at 78 °C for 3 h. During the curing process the
teflon mould was covered with a watch glass to reduce the evaporation
of styrene. After 3 h the polymer was obtained as a transparent, glass-
like specimen.
[
8] a) O. Rifaie-Graham, E. A. Apebende, L. K. Bast, N. Bruns, Adv.
Mater. 2018, 30, 1705483; b) C. Calvino, L. Neumann, C. Weder,
S. Schrettl, J. Polym. Sci., Part A: Polym. Chem. 2017, 55, 640.
[
9] a) C. Viets, S. Kaysser, K. Schulte, Composites, Part B 2014, 65, 80;
b) H. Zhao, M. Zhang, M. Zhu, S. Xu, Y. Cao, J.-H. Yin, Appl. Sci.
2
019, 9, 2390.
[
[
10] J. W. Grate, Chem. Rev. 2008, 108, 726.
11] a) J.-i. Hahm, Sensors 2011, 11, 3327; b) R. Yang, J. Zhao, Y. Liu,
Polym. Degrad. Stab. 2013, 98, 2466.
[12] a) J. J. BelBruno, Chem. Rev. 2019, 119, 94; b) O. S. Ahmad,
T. S. Bedwell, C. Esen, A. Garcia-Cruz, S. A. Piletsky, Trends
Biotechnol. 2019, 37, 294; c) M. P. Tiwari, A. Prasad, Anal. Chim.
Acta 2015, 853, 1; d) S. Suriyanarayanan, P. J. Cywinski, A. J. Moro,
G. J. Mohr, W. Kutner, Top. Curr. Chem. 2012, 325, 165.
Preparation of PEI Specimen and Sensor Embedding: PEI is a two-
component system and was prepared according to the manufacturing
protocol, by mixing both components in a 50/50 (w%) ratio. The
required amount of dye Stil-3 or Stil-4 (dye concentrations between
−1
−1
0
.125 mg g PEI resin up to 2 mg g PEI resin were used) was dissolved
in the mixture, which subsequently was transferred into a teflon mould
25 mm × 10 mm × 2 mm) and curing was done by storing the teflon
(
mould within a petri-dish in an preheated oven at 80 °C for 50 min, followed
by 5 min at 140 °C. After cooling down, the specimen was removed from
the mould, having a transparent, yellowish/brownish appearance.
[
[
13] H. Kim, Y. Kim, J. Y. Chang, Macromol. Chem. Phys. 2014, 215, 1274.
14] S. Dadkhah, E. Ziaei, A. Mehdinia, T. Baradaran Kayyal, A. Jabbari,
Microchim. Acta 2016, 183, 1933.
[
15] a) W.-R. Zhao, T.-F. Kang, L.-P. Lu, F.-X. Shen, S.-Y. Cheng,
J. Electroanal. Chem. 2017, 786, 102; b) D. Lakshmi, A. Bossi,
M. J. Whitcombe, I. Chianella, S. A. Fowler, S. Subrahmanyam,
E. V. Piletska, S. A. Piletsky, Anal. Chem. 2009, 81, 3576.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
[
16] S. Subrahmanyam, S. A. Piletsky, E. V. Piletska, B. Chen, K. Karim,
A. P. F. Turner, Biosens. Bioelectron. 2001, 16, 631.
[
17] D. Kriz, O. Ramstroem, A. Svensson, K. Mosbach, Anal. Chem.
995, 67, 2142.
18] O. Y. F. Henry, D. C. Cullen, S. A. Piletsky, Anal. Bioanal. Chem.
005, 382, 947.
Acknowledgements
1
[
The authors would like to thank ALTANA AG for their financial support
as well as the provision of the investigated poly(ester imide). The
authors thank the EU-BAT-4-ever project within EU-HORIZON 2020 for
financial support.
2
[19] G. J. Mohr, U. E. Spichiger-Keller, Anal. Chim. Acta 1997, 351, 189.
[20] G. J. Mohr, N. Tirelli, C. Lohse, U. E. Spichiger-Keller, Adv. Mater.
1998, 10, 1353.
Open access funding enabled and organized by Projekt DEAL.
[
21] H. N. Kim, Z. Guo, W. Zhu, J. Yoon, H. Tian, Chem. Soc. Rev. 2011,
0, 79.
22] a) P. Michael, W. H. Binder, Angew. Chem., Int. Ed. 2015, 54,
3918; b) T. A. Dickinson, J. White, J. S. Kauer, D. R. Walt, Nature
996, 382, 697; c) O. S. Wolfbeis, J. Mater. Chem. 2005, 15, 2657;
d) P. Anbukarasu, D. Sauvageau, A. L. Elias, Biotechnol. J. 2017, 12,
700050.
23] M. Martinez-Abadia, R. Gimenez, M. B. Ros, Adv. Mater. 2018, 30,
704161.
4
[
Conflict of Interest
The authors declare no conflict of interest.
1
1
1
[
Data Availability Statement
1
Research data are not shared.
[24] N. Jornet-Martínez, Y. Moliner-Martínez, R. Herráez-Hernández,
C. Molins-Legua, J. Verdú-Andrés, P. Campíns-Falcó, Sens. Actuators,
B 2016, 223, 333.
[
[
25] S. Rochat, T. M. Swager, ACS Appl. Mater. Interfaces 2013, 5, 4488.
26] X. Liu, Y. Xu, D. Jiang, J. Am. Chem. Soc. 2012, 134, 8738.
Keywords
chemosensors, optical sensors, poly(ester imide)s, stilbenes, thermal
polymer degradation
[27] J.-S. Yang, T. M. Swager, J. Am. Chem. Soc. 1998, 120, 5321.
[28] H.-A. Ho, M. Bera-Aberem, M. Leclerc, Chem. – Eur. J. 2005, 11,
1
718.
Received: January 5, 2021
Revised: February 2, 2021
Published online: March 30, 2021
[29] S. Ghosh, R. Manna, ChemistrySelect 2016, 1, 6558.
[30] K.-W. Lienert, in Progress in Polyimide Chemistry II (Ed: H. R. Kricheldorf),
Springer, Berlin/Heidelberg, Germany 1999, p. 45.
[
31] A. Anton, K.-W. Lienert, G. Hegemann, Macromol. Mater. Eng. 2008,
93, 331.
2
[
32] a) L.-H. Perng, J. Polym. Res. 2000, 7, 185; b) M. Nedjar, J. Appl.
Polym. Sci. 2011, 121, 2886.
[
1] A. Göpferich, Biomaterials 1996, 17, 103.
[
2] W. Guo, T.-H. Chuang, S.-T. Huang, W.-T. Leu, S.-H. Hsiao, J. Polym.
Res. 2007, 14, 401.
[33] a) T. J. Murray, Macromol. Mater. Eng. 2008, 293, 350; b) S. Maiti,
S. Das, J. Appl. Polym. Sci. 1981, 26, 957.
Adv. Mater. 2021, 33, 2100068
2100068 (8 of 9)
© 2021 The Authors. Advanced Materials published by Wiley-VCH GmbH

www.advancedsciencenews.com
www.advmat.de
[
[
34] G. J. Mohr, U.-W. Grummt, J. Fluoresc. 2006, 16, 185.
35] a) P. Xue, C. Zhang, K. Wang, M. Liang, T. Zhang, Dyes Pigm.
019, 163, 516; b) A. Pucci, F. Di Cuia, F. Signori, G. Ruggeri,
[37] a) S. Marais, M. Metayer, T. Q. Nguyen, M. Labbe, J. M. Saiter, Eur.
Polym. J. 2000, 36, 453; b) J. Reichlin, E. Bormashenko, A. Sheshnev,
R. Pogreb, Proc. SPIE 2000, 4129, 305.
2
J. Mater. Chem. 2007, 17, 783; c) F. Cellini, S. Khapli, S. D. Peterson,
M. Porfiri, Appl. Phys. Lett. 2014, 105, 061907.
[38] F. Scheidt, M. Schäfer, J. C. Sarie, C. G. Daniliuc, J. J. Molloy,
R. Gilmour, Angew. Chem., Int. Ed. 2018, 57, 16431.
[
36] M. Yang, L. Chen, C.-S. Zhao, H.-Z. Huang, J.-S. Wang, Y.-Z. Wang,
Polym. Adv. Technol. 2009, 20, 378.
[39] G. J. Mohr, F. Lehmann, U.-W. Grummt, U. E. Spichiger-Keller, Anal.
Chim. Acta 1997, 344, 215.
Adv. Mater. 2021, 33, 2100068
2100068 (9 of 9)
© 2021 The Authors. Advanced Materials published by Wiley-VCH GmbH