Polymer Degradation and Stability 98 (2013) 1655e1661

Contents lists available at SciVerse ScienceDirect

Polymer Degradation and Stability

journal homepage: www.elsevier.com/locate/polydegstab

Monitoring the degradation of stabilization systems in polypropylene

during accelerated aging tests by liquid chromatography combined
with atmospheric pressure chemical ionization mass spectrometry
Susanne Beißmann a, *, Martin Stiftinger a, Klemens Grabmayer b, Gernot Wallner b,
David Nitsche c, Wolfgang Buchberger a
a
Johannes Kepler-University Linz, Institute of Analytical Chemistry, Altenbergerstraße 69, 4040 Linz, Austria
b
Johannes Kepler-University Linz, Institute of Polymeric Materials and Testing, Altenbergerstraße 69, 4040 Linz, Austria
c
AGRU Kunststofftechnik GmbH, Ing.-Pesendorfer-Str. 31, 4050 Bad Hall, Austria

a r t i c l e i n f o a b s t r a c t

Article history: Degradation pathways of three commonly used antioxidants were successfully studied by using accel-
Received 1 March 2013 erated aging tests for polymers. Additionally, thermal stability and resistance to discoloration of seven
Received in revised form stabilizers were investigated by aging pure stabilizers dissolved in the polymer-mimicking solvent
29 May 2013
squalane. Methods based on high-performance liquid chromatography hyphenated with highly sensitive
Accepted 12 June 2013
Available online 21 June 2013
tandem mass spectrometric detection (HPLC-MS) were developed for structural elucidation of degra-
dation products. Subsequent quantification was done using UV-detection. While Irganox 1330, Irganox
3114 and Cyanox 1790 showed a similar degradation mechanism with highly colored decomposition
Keywords:
Phenolic antioxidants
products, no corresponding oxidized species could be found for other stabilizers and less discoloration
Degradation pathway was observed. For Irganox 1010, hydrolysis was the preferred degradation mechanism, leading to
Discoloration products with an increased solubility in water. Therefore this stabilizer is less suitable for materials
Oxidation intended for water applications. In the aged materials previously unknown degradation mechanisms
Hydrolysis were observed for Irganox 1010 and Irgafos 168 which also contribute to the inhibition of autoxidation of
the polymer.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction known that degradation products are one cause of yellowing in

The implementation of novel polymeric materials in solar-
thermal systems has great potential as they provide substantial
advantages in comparison to traditional materials [1e5]. A key
challenge is to maintain required durability for extended lifetimes,
since these materials are prone to degradation caused by reactions
with oxygen or UV-light [6,7]. Such reactions lead to undesirable
changes in the properties of the polymer. To minimize decompo-
sition during processing, storage and use, different kinds of stabi-
lizers are added to the polymer prior to the compounding
procedure. The protection of the polymer depends on the presence
of these additives in sufficient concentrations. Therefore the iden-
tification and quantification of stabilizers and their possible
decomposition products are of major importance for quality con-
trol, troubleshooting and life test studies. In addition, resistance to
discoloration is a specific requirement for polymer additives, as it is
* Corresponding author. Tel.: þ43 732 2468 8720.
E-mail address: susanne.beissmann@jku.at (S. Beißmann).
polymeric materials [8]. However, both identification and quanti-
fication can be very challenging for several reasons. Firstly, the
complexity of additives and additive formulations has increased
over the last decade. Additionally, the amount added is often small
and can be further decreased due to degradation. Furthermore,
assignment of decomposition products to the corresponding sta-
bilizers is difficult.
Nowadays there are various different analytical tools available
for analyzing polymer additives, which generally require the
extraction from the particular matrix as a first step. Of the different
procedures to analyze polymer additives, chromatographic methods
play a dominant role. Besides the application of gas chromatography
(GC) and pyrolysis coupled to gas chromatography (Py-GC) [9e11],
high performance liquid chromatography (HPLC) in combination
with ultraviolet (UV) detection or mass spectrometric (MS) detec-
tion is commonly applied [12e16]. Much work has been put into the
development of methods for the analysis of different kinds of sta-
bilizers, but less attention has been paid to the extension of these
methods for identification of degradation products [14,16e19].

0141-3910/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.polymdegradstab.2013.06.015
1656 S. Beißmann et al. / Polymer Degradation and Stability 98 (2013) 1655e1661
The present work is a detailed study on the formation of
degradation products of commonly used antioxidants in order to
provide a better insight into the stabilization mechanism of some
commonly used polymer additives. It is known that primary and
secondary antioxidants can transform when stabilizing the poly-
mer against oxidation or radical attack. Characterization of degra-
dation pathways is of great importance since it is not yet fully clear
which additive package provides the best performance for a given
application. Additionally, structure elucidation of degradation
products provides information about their contribution to polymer
discoloration. HPLC combined with MS and sophisticated tandem
mass spectrometry (MS/MS) was selected as an appropriate
analytical method to gain insight into the structure of the various
degradation products.
2. Material and methods
2.1. Chemicals
The following polymer stabilizers were included in this work:
Irganox 1010, Irganox 1330, Irganox 1135, Irganox 3114, Irganox
1076, Cyanox 1790 and Irgafos 168. Their chemical structures are
given in Fig. 1. Irganox-types were obtained from Ciba (Basel,
Switzerland), Irgafos 168 from Ciba (Lampertheim, Germany) and
Cyanox 1790 from Cytec Industries (Woodland Park, USA). Squalane
was obtained from Chrompack (Middelburg, Netherlands). Toluene
was purchased from VWR (Fontenay-sous-Bois, France), acetoni-
trile from VWR (Leuven, Belgium). 18 MU water obtained from a
Fig. 1. Chemical structures of the antioxidants included in the present work.
Milli-Q water purification system (Millipore, Bedford, MA, USA)
was used throughout the work. All chemicals had a purity of >99%.
2.2. Instrumentation
All MS-measurements were performed on an Agilent MSD SL
ion trap mass spectrometer (Agilent Technologies, Waldbronn,
Germany) equipped with an atmospheric pressure chemical ioni-
zation (APCI) source. Chromatography was performed on an Agilent
Series 1100 HPLC system (Agilent, Palo Alto, CA, USA), equipped
with vacuum degasser, quaternary pump, autosampler, and UVevis
diode array detector. The separation column was a Kinetex C18
(50  3.0 mm, 2.6 mm particle size) from Phenomenex (Phenom-
enex, Aschaffenburg, Germany). HPLC-MS using positive and
negative ion mode was used for the separation and identification of
stabilizers and degradation products.
2.3. Sample preparation
A total of 10 mg of a modified polypropylene random copolymer
model material was dissolved in 0.5 ml toluene in a closed vial.
Tributyl phosphite was added to avoid oxidation of the stabilizers.
Dissolution at 130  C required 1 h. After precipitation of the poly-
mer with 0.5 ml acetonitrile, the sample was centrifuged. A defined
volume of the supernatant fluid was evaporated to dryness using a
stream of nitrogen. The residue was redissolved with an equal
volume of acetonitrile. The resulting solution was used without any
further treatment for chromatographic analysis.
2.4. Chromatographic conditions
2.4.1. HPLCeUVeMS
A binary gradient with acetonitrile/water at a flow rate of
1.2 mL min1 was used (Table 1). The injection volume was 10 mL.
The column was maintained at 40  C. The wavelength of the UV-
detection was 210 nm for all analytes. APCI was selected as the
ion source for MS detection as it provides very good performance
for substances with low polarity. Both negative and positive ion
modes were utilized. The following optimized parameters were
employed: nebulizer gas pressure 60 psi, drying gas flow rate
11 L min1, drying gas temperature 350  C, vaporizer temperature
400  C, capillary voltage 3500 V/þ3500 V (positive/negative),
corona needle current þ4000 nA/20,000 nA (positive/negative).
2.5. Accelerating aging tests
All aging experiments were performed with a Binder FD 53
material test chamber. Different environments (air, deionized wa-
ter) and different temperatures (95  Ce135  C) were used in the
tests. In order to be capable of carrying out aging tests in liquids at
higher temperatures, an autoclave was constructed with a capacity
of 0.42 L. Maximum pressure during aging in the autoclaves was
Table 1
Binary gradient for HPLC separation of stabilizers and their degradation products.
Time (min) Acetonitrile (%) Water (%)
0 60 40
4 60 40
7 75 25
10 87 13
14 92 8
17 92 8
23 100 0
 S. Beißmann et al. / Polymer Degradation and Stability 98 (2013) 1655e1661
below 8 bar. The specific aging conditions are representative for
large-area hot water storage tank applications.
3. Results and discussion
3.1. Analysis of exposed polyolefin materials
To elucidate degradation pathways of antioxidants in a practical
time scale, polymers were subjected to accelerated-aging tests,
which intensify the conditions responsible for degradation.
A commercial anti-oxidant system, consisting of Irganox 1330,
Irganox 1010 and Irgafos 168 in a polypropylene sample was
examined in regards to migration and chemical reaction during
aging in the test chamber. Aged polymer samples were withdrawn
at regular intervals and stabilizers were extracted as described in
Section 2.3. The type and concentration of additives and their
degradation products were monitored by HPLC with UV-detection
and an ion trap as an MS detector (equipped with APCI). In Fig. 2
the chromatographic separation of polymer extracts is shown.
The UV-chromatogram obtained for the unexposed sample is rep-
resented by a solid line. Chromatograms for the same material
exposed to air or water at an elevated temperature (115  C) for 6
months are shown as dashed and dotted lines respectively. To allow
clear assignment of the degradation products to the respective
additive, aging of individual additives was performed as discussed
in Section 3.2.
For structure elucidation of chromatographically separated
species, the following approach was utilized. A first hint of the
nature of the molecule in regard to its polarity is obtained via the
retention time. Following this, APCI-MS detection reveals the mo-
lecular mass of the compound. Together with understanding of
theoretical degradation pathways [20] these data were sufficient
for structure elucidation of most peaks in the chromatogram. In
some cases, a more advanced approach, namely tandem mass
spectrometry, was needed. Characteristic fragmentation patterns
are obtained, which provide additional information about the
chemical structure. All identified degradation products for the
different stabilizers based on MS measurements are listed in
Table 2.
3.1.1. Irganox 1330
The amount of intact stabilizer (Fig. 2, Peak 38, see Table 2 for
peak numbering) decreases strongly when exposed to air and
vanishes completely when exposed to water. However, several
additional peaks occurred which were not monitored in the extract
of the untreated material. MS-fragmentation patterns of three more
Fig. 2. UV-chromatograms before and after aging of a polypropylene sample at 115  C
for 6 months; full line: untreated sample; dotted: exposure to deionized water;
dashed: exposure to air (peak numbers see Table 2).
1657
highly retained compounds (Peaks 40e42) were quite similar to
that of the intact molecule; the only difference was a shift of the
molecular ion signal by 2, 4 or 6 Da to lower mass-to-charge ratios.
A closer look at the degradation reactions of sterically hindered
phenols offers a simple explanation. The protective effect of steri-
cally hindered phenols against autoxidation of the polymer stems
from the transfer of the phenolic hydrogen to a radical species
originating from the polymer [20,21]. In further reactions, an alpha-
hydrogen can be donated or disproportionation of two molecules
can take place leading to quinone structures. Accordingly, the mass
shifts were caused by the losses of hydrogen. Degradation of Irga-
nox 1330 during aging in a polymer-mimicking solvent was also
investigated by Barret et al. [17]. Transformation to an oxidized
species was found to be the main reaction mechanism, but exper-
iments in the presence of a polymeric matrix are inevitable in order
to check the relevance of observed degradation reactions under
realistic conditions. The results of the present work indicate that, in
all aged polymer materials containing Irganox 1330, the intact
stabilizer (1a in Fig. 3) is converted to quinoid structures 1b, 1c and
1d (see Fig. 3) according to the mechanism mentioned above. The
conversion rate is highly dependent on the temperature and the
aging environment used, with significantly faster degradation at
higher temperatures and in water as a surrounding medium. The
data obtained from different exposure durations allowed the
monitoring of transformation of the intact molecule over time.
Fig. 4 shows an air-exposed sample at 135  C which demonstrates
that the concentrations of decomposition products first increase
steadily, but then decrease via further degradation or migration
into the surrounding aging fluid. This loss is one of the reasons why
the sum of the intact stabilizer and the three oxidation products
does not equal 100%. Another explanation might be that the con-
centration of the oxidized species was calculated using the
response factor of the intact stabilizer which is not strictly correct
in MS detection. Unfortunately, the exact response factors are un-
known because no standards are commercially available. In addi-
tion to the oxidized species, several other degradation products
from Irganox 1330 with methyl, tert.-butyl and di-tert. butylphenol
groups partially lost could be identified (Table 2). These undesired
splitting-off reactions reduce the antioxidant efficiency of the sta-
bilizers without contributing to the protection of the polymer.
Detection with APCI-MS showed an interesting phenomenon for
Irganox 1330. In the negative ion mode, the intact molecule and the
degradation products were detected as deprotonated molecule
ions. The only exception was the three-times oxidized product,
where a radical ion was formed during MS-ionization as it lacks of a
hydroxyl group. In the positive ion mode, the behavior of some
compounds was different from the expected ionization mechanism.
Protonated molecule ions were only observed for molecules with
already existing oxidized hydroxyl groups. For all other identified
molecules including the intact stabilizer, the measured mass to
charge ratios were 2, 4 or 6 Da lower than anticipated. The possi-
bility of oxidation of the molecule during ionization in the positive
ion mode offers an explanation why lower mass to charge ratios
than in the negative ion mode were observed.
3.1.2. Irganox 1010
Transformation to quinone structures played only a minor role
in the case of Irganox 1010. Instead, hydrolysis of the ester bonds
was the preferred degradation mechanism, which is an unwanted
side reaction as it reduces the antioxidant efficiency without add-
ing to the protection of the polymer as already indicated above. This
is aggravated by the fact that the formed products with free hy-
droxyl groups show an increased solubility in water. Irganox 1010 is
less suitable for materials intended for water applications as it is
prone to hydrolysis. Even though subsequent transformation
1658 S. Beißmann et al. / Polymer Degradation and Stability 98 (2013) 1655e1661

Table 2
Results obtained by LC-APCI-MS including retention times and measured mass to charge ratios with corresponding molecular formulas of the investigated stabilizers and the
identified degradation products after accelerated aging.

Peak nr. tR (min) Measured Measured Molecular Identified substance

m/z positive m/z negative formula

1 0.5 N/A 248.9 C16H26O2 2,6-Di-tert-butyl-4-(1-hydroxyethyl)phenol

2 0.6 N/A 277.0 C17H26O3 3-(3,5-Di-tert-butyl-4-hydroxyphenyl)propanoic acid
3 0.8 N/A 233.2 C16H26O 2,6-Di-tert-butyl-4-ethylphenol
4 0.9 248.9 247.0 C16H24O2 2,6-Di-tert-butyl-4-(1-hydroxyethylidene)cyclohexa-2,5-dienone
5 1.4 N/A 291.0 C18H28O3 Methyl 3-(3,5-di-tert-butyl-4-hydroxyphenyl)propanoate
6 1.6 N/A 219.9 C15H24O 2,6-Di-tert-butyl-4-methylphenol
7 5.8 N/A 655.3 C39H60O8 Irganox 1010, two ester bonds hydrolized
8 7.3 702.6a 683.3 C40H60O9 Irganox 1010, one ester bond hydrolized, 2,6-di-tert-butyl-4-ethylphenol split off
9 7.4 822.6a 803.4 C48H68O10 Irganox 1010, one ester bond hydrolized, two tert-butyl groups split off
10 7.7 716.5a 697.3 C41H62O9 Irganox 1010, one ester bond hydrolized, 2,6-di-tert-butyl-4-methylphenol split off
11 9.5 700.5a 681.4 C40H58O9 Irganox 1010, one hydroxyl group oxidized, one ester bond hydrolized,
2,6-di-tert-butyl-4-ethylphenol split off
12 9.6 878.7a 859.6 C52H76O10 Irganox 1010 one ester bond hydrolized, one tert-butyl group split off
13 9.9 1026.6a 1007.7 C61H84O12 Irganox 1010, three tert-butyl groups split off
14 10.3 948.8a 929.7 C56H82O11 Irganox 1010, one methyl group split off, 2,6-di-tert-butyl-4-ethylphenol split off
15 11.2 1082.7a 1063.7 C65H92O12 Irganox 1010, two tert-butyl groups split off
16 11.4 1006.7a 987.6 C59H88O12 Irganox 1010 with additional oxygen, 2,6-di-tert-butylphenol split off
17 11.6 934.7a 915.6 C56H84O10 Irganox 1010 one ester bond hydrolized
18 12.1 962.7a 943.5 C57H84O11 Irganox 1010, 2,6-di-tert-butyl-4-ethylphenol split off
19 12.2 960.5a 941.6 C57H82O11 Irganox 1010, one hydroxyl group oxidized, 2,6-di-tert-butyl-4-ethylphenol split off
20 12.4 976.7a 957.6 C58H86O11 Irganox 1010, 2.6-di-tert-butyl-4-methylphenol split off
21 13.2 1152.8a 1133.7 C70H102O12 Irganox 1010, three methyl groups split off
22 13.8 1138.8a 1119.9 C69H100O12 Irganox 1010, one tert-butyl group split off
23 14.2 950.7a 931.4 C56H84O11 Irganox 1010 one ester bond hydrolized with additional oxygen
24 14.3 1209.0a 1189.9 C73H106O13 Irganox 1010, one hydroxyl group oxidized, with additional oxygen
25 15.3 1226.9a 1207.8 C73H108O14 Irganox 1010 with two additional oxygens
26 15.8 1210.9a 1191.8 C73H108O13 Irganox 1010 with additional oxygen
27 16.5 1195a 1175.7 C73H108O12 Irganox 1010
28 16.9 1192.8a 1173.8 C73H106O12 Irganox 1010, one hydroxyl group oxidized
29 12.2 553.4b 555.2 C39H56O2 Irganox 1330, 2,6-di-tert-butyl-4-methylphenol split off
30 12.6 567.4b 569.2 C40H58O2 Irganox 1330, 2,6-di-tert-butylphenol split off
31 13.3 713.5b 717.4 C50H70O3 Irganox 1330 one tert-butyl group split off
32 13.5 789.9 788.6 C54H76O4 Irganox 1330, one hydroxyl group oxidized, with additional oxygen
33 13.8 555.6 553.9 C39H54O2 Irganox 1330, one hydroxyl group oxidized, 2,6-di-tert-butyl-4-methylphenol split off
34 14.1 569.6 567.4 C40H56O2 Irganox 1330, one hydroxyl group oxidized, 2,6-di-tert-butylphenol split off
35 14.6 717.5 715.3 C50H68O3 Irganox 1330, one hydroxyl group oxidized, one tert-butyl group split off
36 15.4 787.9 785.7 C54H74O4 Irganox 1330 two hydroxyl groups oxidized, with additional oxygen
37 15.8 731.8 729.5 C51H70O3 Irganox 1330, one hydroxyl group oxidized, three methyl groups split off
38 16.2 769.5b 773.4 C54H78O3 Irganox 1330
39 17.4 785.9 783.6 C54H74O4 Irganox 1330 three hydroxyl groups oxidized, with additional oxygen
40 19.2 773.7 771.4 C54H76O3 Irganox 1330, one hydroxyl group oxidized
41 21.4 771.7 769.4 C54H74O3 Irganox 1330 two hydroxyl groups oxidized
42 22.9 769.7 768.7c C54H72O3 Irganox 1330 three hydroxyl groups oxidized
43 1.3 191.1 N/A C14H22 1.3-di-tert-butylbenzene
44 10.2 459.5 457.1 C28H43O3P Irgafos 168, 2,6-di-tert-benzene group split off
45 17.9 663.6 N/A C42H63O4P Irgafos 168 oxidized
46 19.5 679.5 N/A C42H63O5P Irgafos 168 with two additional oxygens
47 22.6 647.5 N/A C42H63O3P Irgafos 168
48 1.7 527.3a 508.2 C29H39N3O5 Cyanox 1790, 2-tert-butyl-3,4,6-trimethylphenol split off
49 5.4 698.5 714.3 C42H57N3O7 Cyanox 1790 with one additional oxygen
50 7.3 698.5/731.5a 712.3 C42H55N3O7 Cyanox 1790, one hydroxyl group oxidized, with one additional oxygen
51 7.9 703.4a 684.3 C42H55N3O6 Cyanox 1790, methyl group split off
52 8.6 698.6/717.4a 698.3 C42H57N3O6 Cyanox 1790
53 13.1 698.5 696.2 C42H55N3O6 Cyanox 1790, one hydroxyl group oxidized
54 5.4 N/A 269.2 C18H38O Octadecan-1-ol
55 14.7 N/A 473.2 C31H54O3 Irganox 1076, one tert-butyl group split off
56 15 529.3 545.2 C35H62O4 Irganox 1076 with additional oxygen
57 19.5 529.4b 529.4 C35H62O3 Irganox 1076
58 20.4 529.4 527.3 C35H60O3 Irganox 1076, hydroxyl group oxidized
59 4.6 583.4a 564.3 C33H47N3O5 Irganox 3114, 2,6-di-tert-butyl-4-methylphenol split off
60 9.5 745.4a 508.2 C44H61N3O6 Irganox 3114, one hydroxyl group oxidized, one tert-butyl group split off
61 10.1 815.5a 578.3 C48H67N3O7 Irganox 3114 with one additional oxygen, one hydroxyl group oxidized
62 10.9 833.6a 596.3 C48H69N3O8 Irganox 3114 with 2 additional oxygens
63 11.7 801.5a 564.1 C48H69N3O6 Irganox 3114
64 12.3 782.7 562.1 C48H67N3O6 Irganox 3114, one hydroxyl group oxidized
65 8.4 375.3b 375.1 C24H40O3 Irganox 1135, one methyl split off
66 9.1 389.3b 389.2 C25H42O3 Irganox 1135
67 9.7 389.3 387.1 C25H40O3 Irganox 1135, one hydroxyl group oxidized
a
Detected as [M þ NH4]þ.
b
Detected as fully oxidized species (all hydroxyl groups oxidized).
c
Radical ion.
 S. Beißmann et al. / Polymer Degradation and Stability 98 (2013) 1655e1661 1659

Fig. 3. Structural formulas of the sterically hindered phenolic antioxidant Irganox 1330 (1a) and its oxidation products (1be1d).

products still have antioxidant capacity, they are rendered useless Besides the molecular ion signal with a mass to charge ratio of
as they are easily leached out from the polymeric material. 679.5, fragment ion signals with partially cleaved oxygen and tert.-
Nevertheless it is known, that sterically hindered phenols are not butyl groups are observed. This may be because the molecule is
only effective hydrogen donors, but may also undergo numerous fragmented to some extent in the ion source. Consequently, it may
further chemical reactions that inhibit autoxidation of the polymer be assumed that this degradation product is less stable than the
[20]. In the aged polymer samples containing Irganox 1010 a pre- intact molecule and corresponding phosphate, where in-source
viously unknown degradation mechanism was observed (Fig. 5). fragmentation played a minor role. This particular degradation
Two peaks with similar MS spectra but higher molecular weight product was only observed in polymer samples subjected to
than the intact molecule were found. The mass differences of 16 accelerated aging, but not in materials after polymer processing.
and 32 respectively, which can be explained by additional oxygen Thus the formation takes either not place at high process temper-
chemically linked to the molecules. Following Fig. 5 it is assumed atures or the molecule is not stable enough and is transformed to
that the formation is due to the reaction of a phenoxyl radical with another species immediately. The exact position of the oxygen in-
a hydroxyl or alkoxy radical, which are formed in a preceding side the molecule is still unknown.
decomposition of a hydroperoxide. The hydroperoxide is formed
through hydrogen abstraction of a peroxy radical from the poly- 3.1.4. Kinetics
olefin polymer backbone [20]. These degradation products prove The kinetics of degradation of the two primary antioxidants,
that Irganox 1010 not only acts as a hydrogen donor, but also as a Irganox 1010 and Irganox 1330, was investigated using Arrhenius
radical scavenger, thus further contributing to avoid polymer modeling. The experimental results indicated the validity of a first
degradation. Similar to Irganox 1330, several additional deal- order kinetic. Therefore the logarithmical concentration was
kylation products were observed (Table 2). plotted against time for each individual temperature and aging
media. The slope of the straight line is equivalent to the rate
3.1.3. Irgafos 168
Up to now less attention has been paid to the degradation
mechanism of hydroperoxide decomposers (HD) such as Irgafos
168. The main stabilization mechanism involves the reaction with
hydroperoxides which are reduced to thermally stable alcohols,
while the HD is oxidized stoichiometrically to the corresponding
phosphate [20,22e25].
In the current work a previously unknown degradation product
was identified for Irgafos 168. The chromatographic separation is
shown in Fig. 2 (Peaks 45e47). It is assumed that the primary
oxidation product (phosphate) participated in a further oxidation
reaction by reducing an alkoxy radical generated from a hydro-
peroxide and taking up an additional oxygen. The corresponding
MS-spectrum for the degradation product is shown in Fig. 6.

Fig. 4. Dependence of concentration of intact stabilizer and degradation products (see

Fig. 3) upon time for Irganox 1330 after aging in air at 135  C. Fig. 5. The proposed degradation mechanism for Irganox 1010.
1660 S. Beißmann et al. / Polymer Degradation and Stability 98 (2013) 1655e1661
Fig. 6. MS-spectrum of a degradation product of Irgafos 168.
constant. By measuring the rate constant at different temperatures
a graph can be constructed to determine the activation energy
of the degradation reaction. The obtained logarithmical rate con-
stants (ln k, days1) were plotted against the associated inverse
temperatures (1/T). For water exposure a linear fit with the
slope EA/R was obtained for both stabilizers. The equation ln
k ¼ 25.339  10,856/T (R2 ¼ 0.99996) describes the time temper-
ature dependence of Irganox 1330 and ln k ¼ 24.843e9224/T
(R2 ¼ 0.9997) that of Irganox 1010. For determination of the acti-
vation energy the slope was multiplied with the gas constant R
(8.314 J K1 mol1). The value obtained was 90.2 kJ mol1 for
Irganox 1330 and 76.7 kJ mol1 for Irganox 1010. No linear Arrhe-
nius plot was observed for the primary antioxidants extracted from
the polypropylene exposed to air. A possible reason may be the fact,
that evaporation of the stabilizers takes place during air exposure at
elevated temperatures, which does not take place in the case of
water aging inside closed autoclaves.
3.2. Aging of pure additives in squalane
After exposure of plastic materials several additional peaks
occur in chromatograms due to degradation of the additives. For
plastic formulations consisting of more than one stabilizer the
assignment of these additional peaks to the respective additive
faces some difficulties. Furthermore, the concentrations of intact
stabilizers added are often low and, subsequently, the concentra-
tion of degradation products is frequently below the detection limit
of the analytical method employed. Aging pure additives separately
in adequate quantities helps to overcome these limitations.
Therefore the thermal stability of pure antioxidants dissolved in
squalane was investigated. The apolar solvent squalane was used as
it is a liquid model for polypropylene [17,26]. 10 mg of the stabi-
lizers Irganox 1010, Irganox 1330, Irganox 3114, Irganox 1076,
Irganox 1135, Cyanox 1790 and Irgafos 168 were added separately
to 1 mL of the polymer-mimicking solvent squalane. The solutions
were heated in an ambient atmosphere to 130  C. At various times
samples were taken and the stabilizers along with their degrada-
tion products were extracted by shaking with acetonitrile
(squalane:acetonitrile ¼ 1:10). After separation of the layers, the
upper phase (containing intact stabilizers and degradation prod-
ucts) is 10-fold diluted and analyzed. Within a very short period
(five days) slight yellowing of the initial colorless solutions was
observed by visual inspection for those containing Irganox 1330
and Cyanox 1790. Several weeks later, all solutions except the one
with Irgafos 168 showed discoloration, with increasing color in-
tensity in the following order: Irgafos 168 < Irganox 1076 < Irganox
Fig. 7. Chromatographic separation of Irganox 1010 and its degradation products after
aging in squalane for 3 weeks at 130  C (Peak numbers see Table 2).
1010, Irganox 1135 < Irganox 3114 < Cyanox 1790, Irganox 1330.
For comparison purposes, the same procedure was conducted for
the pure solvent without the addition of stabilizers. Only a very
minor visible change of color was observed within the same
exposure time. Therefore, it can be presumed that the discoloration
depends on the structure and concentration of the degradation
products, since all phenolic antioxidants were colorless in their
original form. The three compounds with the strongest color
change have a similar chemical structure. The only difference is that
the center of the molecule is isocyanuric acid for Irganox 3114 and
Cyanox 1790, and mesitylene for Irganox 1330. Therefore, it can be
concluded that transformation to an oxidized species (which is the
main degradation mechanism for these substances) leads to a
bathochromic shift and subsequent absorption of light in the visible
range. The solution containing Irganox 1010 showed only a slight
discoloration. The chromatogram depicted in Fig. 7 for the sepa-
ration of Irganox 1010 and its degradation products provides an
explanation. Besides the intact molecule (Peak 27) mainly hydro-
lyzed products (Peaks 9, 11, 12, 17) and dealkylation products (Peaks
4, 18, 19, 20, 21, 22) are observed. In comparison, the intensities of
the peaks assigned to oxidized degradation products (Peaks 11, 19,
28) were quite low. It can thus be concluded that the concentration
is too small to bring about a significant change in color. Generally,
discoloration may be a useful degradation indicator. However, one
has to keep in mind that it does not provide information about the
extent of degradation, and besides stabilizers other components
may contribute to polymer discoloration. The aim was not to use
discoloration as a degradation indicator but rather to show which
additives lead to strong discoloration as in some applications color
formation is unwanted and could be prevented in the first place by
applying the right additive formulation.
Transformation to products with additional oxygen (Peaks 25,
26), as already observed in the presence of polymeric matrix, were
also found with squalane as the aging media. Chemical bonding of
oxygen to the stabilizers has also been observed for all other pri-
mary antioxidants.
4. Conclusions
The potential of highly sensitive APCI-MS/MS detection in
combination with HPLC could be clearly demonstrated for detec-
tion and identification of various common stabilizers and degra-
dation products. Within this work, a large number of degradation
products could be elucidated providing additional knowledge
about thermal stability and resistance to discoloration of several
commonly used antioxidants. Studies using extracts from polymers
 S. Beißmann et al. / Polymer Degradation and Stability 98 (2013) 1655e1661
subjected to accelerated aging tests visualized a general difference
in degradation pathways between different primary antioxidants.
The results showed that Irganox 1330 is better suited for materials
intended for water application, as its oxidation products show a
good extraction resistance while hydrolyzed products in the case of
Irganox 1010 are easily leached out when in contact with water.
A previously unknown degradation mechanism was discovered for
Irganox 1010 and a thus far unknown degradation product
could be identified for the secondary antioxidant Irgafos 168.
Identification of these new degradation pathways contributes to
the general understanding of stabilization mechanisms in poly-
meric matrices. Investigations of seven antioxidants after aging in
the polypropylene-mimicking solvent squalane led to additional
insights. Formation of highly colored degradation products was
observed for Irganox 1330, Irganox 3114 and Cyanox 1790. This is
due to their similar structure leading to oxidation being the main
degradation mechanism. Other stabilizers, like Irganox 1010, Irga-
nox 1076 and Irganox 1135 showed a higher resistance to discol-
oration. The developed analytical methods for polymer stabilizers
and newly acquired knowledge of individual complex degradation
pathways in different environments are of significant importance
for applying the right additive formulation for a given application.
Additionally, antioxidants can only be improved if degradation
pathways are fully studied to analyze their weaknesses.
Acknowledgments
This research work was performed in work package AP-02 “Test
Methods” of the fundamental research project SolPol-1 and work
package AP-05 “Durable Compounds” of the cooperative research
project SolPol-2 on Solar-thermal Systems based on Polymeric Mate-
rials (www.solpol.at). This Project is funded by the Austrian Climate
and Energy Fund (KLI.EN) within the program “Neue Energien
2020”. The authors wish to express their acknowledgments to the
Scientific Partners JKU-Institute of Polymeric Materials and Testing
(JKU-IPMT), JKU-Institute of Analytical Chemistry (JKU-IAC) and the
Company Partners AGRU Kunststofftechnik GmbH (AGRU) und APC
Advanced Polymer Compounds (APC).
References
[1] Kohl MD. Polymeric materials for solar thermal applications. Weinheim:
Wiley-VCH; 2012.
[2] Kahlen S, Wallner GM, Lang RW. Aging behavior of polymeric solar absorber
materials e part 2: commodity plastics. Sol Energy 2010;84:1577e86.
[3] Wallner GM, Lang RW. Polymeric materials for solar energy applications. Sol
Energy 2005;79:571e2.
[4] Wallner GM, Weigl C, Leitgeb R, Lang RW. Polymer films for solar energy
applications e thermoanalytical and mechanical characterisation of ageing
behaviour. Polym Degrad Stab 2004;85:1065e70.
1661
[5] Wallner GM, Lang RW. Kunststoffe e Neue Möglichkeiten in der Solarthermie.
Erneu Energ 2006;2/2006:10e2.
[6] Gijsman P, Dozeman A. Comparison of the UV-degradation chemistry of
unstabilized and HALS-stabilized polyethylene and polypropylene. Polym
Degrad Stab 1996;53:45e50.
[7] Attwood J, Philip M, Hulme A, Williams G, Shipton P. The effects of ageing by
ultraviolet degradation of recycled polyolefin blends. Polym Degrad Stab
2006;91:3407e15.
[8] Pospisil J, Habicher WD, Pilar J, Nespurek S, Kuthan J, Piringer GO, et al.
Discoloration of polymers by phenolic antioxidants. Polym Degrad Stab
2002;77:531e8.
[9] Herrera M, Matuschek G, Kettrup A. Fast identification of polymer additives by
pyrolysis-gas chromatography/mass spectrometry. J Anal Appl Pyrolysis
2003;70:35e42.
[10] Jansson KD, Zawodny CP, Wampler TP. Determination of polymer additives
using analytical pyrolysis. J Anal Appl Pyrolysis 2007;79:353e61.
[11] Wang FCY. Polymer additive analysis by pyrolysis-gas chromatography IV.
Antioxid J Chromatogr A 2000;891:325e36.
[12] Choi SS, Jang JH. Analysis of UV absorbers and stabilizers in polypropylene by
liquid chromatography/atmospheric pressure chemical ionization-mass
spectrometry. Polym Test 2011;30:673e7.
[13] Himmelsbach M, Buchberger W, Reingruber E. Determination of polymer
additives by liquid chromatography coupled with mass spectrometry.
A comparison of atmospheric pressure photoionization (APPI), atmospheric
pressure chemical ionization (APCI), and electrospray ionization (ESI). Polym
Degrad Stab 2009;94:1213e9.
[14] Reingruber E, Himmelsbach M, Sauer C, Buchberger W. Identification of
degradation products of antioxidants in polyolefins by liquid chromatography
combined with atmospheric pressure photoionisation mass spectrometry.
Polym Degrad Stab 2010;95:740e5.
[15] Dopico-Garcia MS, Noguerol-Cal R, Castro-Lopez MM, Cela-Perez MC, Pinon-
Giz E, Lopez-Vilarino JM, et al. Determination of polyolefin additives by
reversed-phase liquid chromatography. Cent Eur J Chem 2012;10:585e610.
[16] Reingruber E, Buchberger W. Analysis of polyolefin stabilizers and their
degradation products. J Sep Sci 2010;33:3463e75.
[17] Barret J, Gijsman P, Swagten J, Lange RFM. A molecular study towards the
interaction of phenolic anti-oxidants, aromatic amines and HALS stabilizers in
a thermo-oxidative ageing process. Polym Degrad Stab 2002;76:441e8.
[18] Bertoldo M, Ciardelli F. Water extraction and degradation of a sterically hin-
dered phenolic antioxidant in polypropylene films. Polymer 2004;45:8751e9.
[19] El Mansouri H, Yagoubi N, Ferrier D. Extraction of polypropylene additives and
their analysis by HPLC. Chromatographia 1998;48:491e6.
[20] Zweifel H, Maier RD, Schiller M. Plastics additives handbook. 6th ed. Cincin-
nati, Ohio: Hanser Publications; 2009.
[21] Pospisil J. Chemical and photochemical behavior of phenolic antioxidants in
polymer stabilization e a state-of-the-art report. 2. Polym Degrad Stab
1993;39:103e15.
[22] Drake WO, Pauquet JR, Zingg J, Zweifel H. Stabilization of polypropylene
during melt processing influence. Abstr Pap Am Chem Soc 1993;206:229.
POLY.
[23] Zweifel H. Effect of stabilization of polypropylene during processing and its
influence on long-term behavior under thermal stress. Adv Chem Ser
1996;249:375e96.
[24] Konig T, Schwetlick K, Kudelka I, Pospisil J. Phosphorus-containing antioxi-
dants. 5. Mechanism of hydroperoxide decomposition by 4,8-di-tert-butyl-
2,10-dimethyl-6-phenoxy-dibenzo[D, G]-1,3,6,2-dioxathiaphosphocin. Polym
Degrad Stab 1986;15:151e60.
[25] Walling C, Rabinowitz R. The reaction of trialkyl phosphites with thiyl and
alkoxy radicals. J Am Chem Soc 1959;81:1243e9.
[26] Soebianto YS, Katsumura Y, Ishigure K, Kubo J, Koizumi T, Shigekuni H, et al.
Model experiment on the protection effect in polymers e radiolysis of liquid
squalane in the presence and absence of additives. Polym Degrad Stab
1993;42:29e40.