Polymer Testing 30 (2011) 673–677

Contents lists available at ScienceDirect

Polymer Testing
journal homepage: www.elsevier.com/locate/polytest

Analysis Method

Analysis of UV absorbers and stabilizers in polypropylene by liquid

chromatography/atmospheric pressure chemical ionization-mass
spectrometry
Sung-Seen Choi*, Joong-Hee Jang
Department of Chemistry, Sejong University, 98 Gunja-dong, Gwangjin-gu, Seoul 143-747, South Korea

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

Article history: Polymer degradation by outdoor weathering, heat and ultraviolet (UV) light causes
Received 19 April 2011 chemical changes in polymer structure and affects the mechanical properties and service
Accepted 31 May 2011 lifetime of the polymer. Both UV absorbers (UVAs) and stabilizers (UVSs) are used to
enhance the lifetime of polymeric materials. Tinuvin 234, Tinuvin 329, and Tinuvin 360 are
Keywords: commonly used as UVAs, while Tinuvin 770 and Tinuvin 123 are commonly used as UVSs.
UV absorbers
A sample mixture of the three UVAs and two UVSs was separated using high performance
UV stabilizers
liquid chromatography (HPLC) and the separated species were quantified using atmo-
LC/APCI-MS
Separation spheric pressure chemical ionization-mass spectrometry (APCI-MS). Separation was per-
Quantitation formed using a C18-column and gradient elution system of methylene chloride/ethanol
(80/20) and methanol. Polypropylene (PP) composite containing the five additives was
aged at 90
C for 5–20 days in a convection oven. The additives remaining in the sample
after the thermal aging were extracted and analyzed using LC/APCI-MS to measure the
amounts of consumed additives. By using this method, the degree of consumption of UVAs
and UVSs in a polymer composite by thermal aging can be estimated.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction polymer matrix at low concentrations [3,4], a sensitive

method for quantification is required. UVAs are mostly
Various kinds of additives have to be admixed with based on benzotriazole, while UVSs are mostly based on
commercial polymers to improve their properties and hindered amine light stabilizers (HALS). UVAs slow down
make them suitable for particular applications [1]. Some the degradation process by preferentially absorbing
additives are used to enhance the service lifetime of poly- harmful UV radiation and dissipating it as thermal energy
meric materials [2]. Among these additives, ultraviolet [5]. UVSs provide stability against photo-oxidation and
absorbers (UVAs) and stabilizers (UVSs) are essential in protect the material from UV light damage [6,7].
polyolefins to minimize degradation by photochemical Tinuvin 329, Tinuvin 234, and Tinuvin 360 are
reactions. Qualitative and quantitative analyses of these commonly used as UVAs, while Tinuvin 770 and Tinuvin
additives are very important to understand polymer 123 are commonly used as UVSs [8]. Analyses of Tinuvin
degradation and in quality-assurance testing [2]. In general, 234 [9,10], Tinuvin 770 [1,9,11–13], and Tinuvin 123 [8,14]
polymeric materials are protected from UV light by adding have been reported. Himmelsbach and coworkers [10]
UVAs and UVSs. Since these additives are present in the analyzed polymer additives including UVAs with liquid
chromatography/mass spectrometry (LC/MS), and analyt-
ical results using atmospheric pressure chemical ionization
* Corresponding author. Tel.: þ82 2 3408 3815. (APCI), atmospheric pressure photoionization (APPI) and
E-mail address: [email protected] (S.-S. Choi). electrospray ionization (ESI) as a detector were compared.

0142-9418/$ – see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymertesting.2011.05.011
674 S.-S. Choi, J.-H. Jang / Polymer Testing 30 (2011) 673–677

However, they did not analyze HALS compounds. HALSs are Schemes 1 and 2, respectively. HPLC grade solvents of
not easily separated from the other additives with LC acetone, methylene chloride, ethanol and methanol were
because of the numerous basic amino groups of HALS [15]. purchased from JT Baker Co.
Thus, determination of the separation condition of mixed A model sample mixture of the three UVAs and two
additives is a very important task. UV detection is a sensi- UVSs was prepared for determination of the separation
tive method for determination of UVAs, but is not a suitable conditions. An APCI-MS and a UV detector were employed
method for determination of UVSs [15]. Trones and as detection systems. A liquid chromatograph used a binary
coworkers [16,17] analyzed a mixture of HALS compounds pump and an injection valve with a 10 mL sample loop. A
using temperature-programmed LC and light scattering sample of 10 mL was introduced by means of a Rheodyne
detection. As a result of their various applications, UVAs valve. A ZORBAX Eclipse XDB-C18 column (4.6 mm
and UVSs can enter the environment by different pathways.  150 mm, 5 mm, Agilent Technologies Co.) was used at an
Some UVAs and UVSs can bio-accumulate in fish and operating temperature of 30
C. Gradient elution systems
human breast milk [14,18]. Because some of these additives as well as isocratic ones were employed to determine the
are considered potentially toxic and could damage humans, optimal separation conditions of the mixture sample.
qualitative and quantitative analyses of these additives are Methanol, methylene chloride, ethanol, acetonitrile and 1%
very important to identify the amount consumed. formic acid in water were employed as mobile phases. The
Liquid chromatography/atmospheric pressure chemical flow rate also varied.
ionization-mass spectrometry (LC/APCI-MS) is a suitable An LC 10AD instrument and a diode array detector SPD-
method for determination of UVAs as well as UVSs since APCI M10A from Shimadzu Co. were used for LC-UV. The detec-
is a sensitive method to detect less polar materials. In the tion wavelength was set at 340 nm. MS detection was ach-
present work, a sample mixture of the commonly used three ieved using a single quadrupole spectrometer equipped with
UVAs and two UVSs was separated using high performance an APCI ionization source. An LC/APCI-MS of a HPLC 1200
liquid chromatography (HPLC) and was quantified using instrument coupled to a single quadrupole LC-MS 6130 mass
detection by APCI-MS and UV. The five additives were mixed spectrometer from Agilent Technologies Inc. was used. The
with polypropylene (PP) and the PP sample was aged in following instrumental parameters were used for the LC/
a convection oven. Amounts of the additives remaining in the APCI-MS analysis in the positive ion mode: nebulizer pres-
PP samples before and after the thermal aging were analyzed sure 60 psig; capillary voltage 4 kV; fragmentor voltage
using LC/APCI-MS to estimate the amounts of the additives 75 V; corona current 4.0 A; drying gas (N2) flow rate 12.0 L/
consumed during the thermal aging. min; drying gas temperature 350
C; quadrupole tempera-
ture 100
C; vaporizer temperature 325
C.
2. Experimental Polypropylene (PP) composite was prepared by mixing
the three UVAs and two UVSs with PP using a HAAKE
Three UVAs consisting of 2-(2H-benzotriazol-2-yl)-4- MiniLab micro-compounder (Thermo Fisher Scientific Inc.).
(1,1,3,3-tetramethylbutyl)phenol (Tinuvin 329), 2-(2H- Concentration of each additive was 1000 ppm. The mixer
benzotriazol-2-yl)-4,6-bis(1-methyl-1-phenylethyl)phenol works as a co-rotating screw extruder. The following
(Tinuvin 234), and 2,2-methylenebis[6-(2H-benzotriazol- instrumental parameters were used for preparation of the
2-yl)-4-(1,1,3,3-tetramethylbutyl)phenol] (Tinuvin 360), PP composite: motor power 400 W; motor speed 60 rpm;
together with two UVSs consisting of bis(2,2,6,6- torque 1 N m; temperature 180
C; pressure 150 bar. The
tetramethyl-4-piperidyl)sebacate (Tinuvin 770) and bis(1- mixing process was as follows. First, the instrument was
octyloxy-2,2,6,6-tetramethyl-4-piperidinyl)sebacate (Tinu- preheated to 180
C. Second, 2 g of PP and 1000 ppm
vin 123) were purchased from Aldrich Chemical Co. additives were inserted into the inlet of the instrument.
Chemical structures of the UVAs and UVSs are shown in Third, the mixture was cycled for 3 min and then was

N
N N OH

N N
N
N N HO HO
N

HO
N N
N

Tinuvin 329 Tinuvin 234 Tinuvin 360

Scheme 1. Chemical structures of the UV absorbers.

 S.-S. Choi, J.-H. Jang / Polymer Testing 30 (2011) 673–677 675

O NH

O
O

HN O

Absorbance
Tinuvin 770

N
O

O
O

O
N 2 4 6 8 10 12 14
O Tinuvin 123 Retention time (min)

Scheme 2. Chemical structures of the UV stabilizers. Fig. 1. LC-UV chromatogram of the mixture of three UVAs and two UVSs.

of polymer additives including UVAs and UVSs using LC/MS,

discharged from the mixer. A sheet was made with
but did not report separation of a mixture containing HALS
dimensions of 140 mm  140 mm and 1.2 mm thickness by
such as Tinuvin 770. In the present work, we successfully
pressing at 180
C for 2 min in a compression mold.
analyzed the mixture containing UVAs and HALSs using LC/
A PP composite sample (10 mm  40 mm) was aged at
APCI-MS as shown in Fig. 2.
90
C for 5, 10, 15, and 20 days in a convection oven.
Limits of quantitation (LOQ) and limits of linearity (LOL)
Samples of 0.3 g were taken before and after thermal aging
of the UVAs and UVSs were examined. The LOQs were
to extract the organic additives remaining in the PP
determined by reducing the concentrations. Fig. 3 shows
composite with 3.0 mL of acetone. The extracts were
calibration curves of the UVAs and UVSs. The dynamic
filtered by a nylon 66 syringe filter (13 mm, 0.45 mm,
ranges were marked with their curve fitting equations
Whatman) and were analyzed using LC/APCI-MS. Amounts
(dotted lines in Fig. 3). The curve fitting equations were
of the additives consumed during thermal aging were
y ¼ 3.55  107x  1.64  107 (Tinuvin 770),
determined by comparing the PP composites before and
y ¼ 3.67  107x  5.30  104 (Tinuvin 234),
after the thermal aging.
y ¼ 2.24  107x þ 2.40  104 (Tinuvin 123),
y ¼ 1.74  107x þ 6.06  104 (Tinuvin 329), and
3. Results and discussion y ¼ 2.55  106x þ 1.64  103 (Tinuvin 360). The correlation
coefficients were about 0.999. The LOQs and LOLs are
The optimal elution condition was a gradient elution summarized in Table 2. Order of the sensitivity (the slope in
system using solvent A (methylene chloride/ethanol ¼ 80/ Fig. 3) was Tinuvin 770 > Tinuvin 234 > Tinuvin
20) and solvent B (methanol). The flow rate was 1.3 mL/min. 123 > Tinuvin 329 > Tinuvin 360. Order of the LOQ was
Detailed elution conditions are summarized in Table 1. Fig. 1 Tinuvin 770 > Tinuvin 234, Tinuvin 329 > Tinuvin
shows the HPLC chromatograms of the sample mixture with 360 > Tinuvin 123, while that of the LOL was Tinuvin
UV absorption detection. The three UVAs were observed in
abundance but the two UVSs were not detected. Fig. 2 shows
the LC/APCI-MS TIC chromatograms of the sample mixture.
All of the UVSs and UVAs were observed in abundance. The
200000
peaks at 3.02, 3.61, 8.05, 8.46, and 9.31 min were Tinuvin 329,
Tinuvin 234, Tinuvin 360, Tinuvin 123, and Tinuvin 770,
respectively. Duderstadt and Fischer [15] analyzed mixtures 150000
Abundance

Table 1
100000
Optimized gradient elution system.

Time (min) Solvent A (%) Solvent B (%)

50000
0 0 100
4 0 100
8 80 20
10 80 20 0
12 100 0 2 4 6 8 10 12 14
15 100 0
Retention time (min)
Solvent A : mixed solvent of methylene chloride and ethanol (80 : 20,
volume ratio). Fig. 2. LC/APCI-MS TIC chromatograms of three UVAs and two UVSs
Solvent B : methanol. extracted from the unaged PP sample (solid line) and from the PP composite
Flow rate : 1.3 mL/min. aged at 90
C for 20 days (dotted line).
676 S.-S. Choi, J.-H. Jang / Polymer Testing 30 (2011) 673–677

7
8.0x10 consumption of the UVAs and UVSs in the PP composite as
a function of the aging time. The consumptions linearly
increased as the aging time elapsed. The aging was con-
7
6.0x10 ducted in the absence of light so no reactions with the UVAs
and UVSs were expected. However, the additives are
Peak area

7
consumed by migration in the sample to the PP composite
4.0x10 surface and evaporation. The order of consumption was
Tinuvin 329 > Tinuvin 770 > Tinuvin 234 > Tinuvin
7 360 > Tinuvin 123. This can be explained by the molecular
2.0x10
weights. In general, organic additives with higher molec-
ular weight are lost more slowly than those with lower
0.0 weight [19,20]. Molecular weights of Tinuvin 329, Tinuvin
770, Tinuvin 234, Tinuvin 360, and Tinuvin 123 are 323,
0.0 0.5 1.0 1.5 2.0
480, 447, 658, and 736, respectively.
Concentration (mM)
The linear curve fitting equations of Fig. 5 were
y ¼ 1.85x þ 32.75, y ¼ 2.27x þ 15.27, y ¼ 2.47x þ 10.51,
Fig. 3. Calibration curves of mixed sample containing three UVAs and two y ¼ 2.43x þ 3.26, and y ¼ 2.30x þ 13.93 for Tinuvin 329,
UVSs using LC/APCI-MS. The squares, circles, up-triangles, down-triangles Tinuvin 234, Tinuvin 360, Tinuvin 123, and Tinuvin 770,
and diamonds indicate the Tinuvin 329, Tinuvin 234, Tinuvin 360, Tinuvin
respectively. The correlation coefficients of the linear curve
123, and Tinuvin 770, respectively. The dotted lines are the dynamic ranges.
fitting equations were larger than 0.990. Using the linear
curve fitting equations, the half-lifetimes can be obtained
770 > Tinuvin 123 > Tinuvin 329, Tinuvin 360 > Tinuvin
by substituting the “y” of the curve fitting equation with
234. Himmelsbach and coworkers [10] measured LOQs of
“50”. The half-lifetimes of Tinuvin 329, Tinuvin 234, Tinu-
Tinuvin 234 using ESI, APCI, and APPI as the detector, and
vin 360, Tinuvin 123, and Tinuvin 770 at 90
C are 9.32,
reported that APPI had better LOQ than the others and APCI
15.3, 16.0, 19.2, and 15.7 days, respectively. The additive
has better LOQ than ESI.
with higher molecular weight had longer half-lifetime than
The PP composite containing the three UVAs and two
UVSs was aged at 90
C. The UVAs and UVSs remaining in
the PP composite were extracted with acetone, and
amounts of the additives consumed by the thermal aging Tinuvin 329
were analyzed using LC/APCI-MS. Fig. 2 shows LC/APCI-MS
TIC chromatograms of the extracts before and after the
thermal aging at 90
C for 20 days. Amounts of all the
additives remaining in the PP composite were significantly
reduced by thermal aging. Consumptions of the UVAs and
UVSs were obtained by dividing the difference in amounts Tinuvin 234
of the additives remaining in the PP composite before and
after the thermal aging by the initial content (equation (1)).


DCð%Þ ¼ 100  Cini  Caged Cini (1)
Abundance

where DC is the consumption by thermal aging, Cini is the Tinuvin 360

initial concentration of the additive before thermal aging,
and Caged is the concentration of the additive after thermal
aging.
Peak areas of the analytes were calculated using an
extracted ion chromatogram (EIC) as shown in Fig. 4. All of
Tinuvin 123
the UVAs and UVSs generated [M þ H]þ ions in APCI. The
[M þ H]þ ions of Tinuvin 329, Tinuvin 234, Tinuvin 360,
Tinuvin 123, and Tinuvin 770 are m/z 324, 448, 659, 737,
and 481, respectively. Fig. 5 shows variation of the

Tinuvin 770
Table 2
Limit of quantitation (LOQ) and limit of linearity (LOL) of three UVAs and
two UVSs by LC/APCI-MS and LC-UV.

Compound LOQ (mM) LOL (mM)

Tinuvin 329 2.0  103 0.6 2 4 6 8 10 12
Tinuvin 234 2.0  103 0.3
Tinuvin 360 1.0  103 0.6 Retention time(min)
Tinuvin 123 2.0  104 1.0
Tinuvin 770 3.0  103 2.6 Fig. 4. EIC chromatograms of the [M þ H]þ ions of Tinuvin 329, Tinuvin 234,
Tinuvin 360, Tinuvin 123, and Tinuvin 770, respectively.
 S.-S. Choi, J.-H. Jang / Polymer Testing 30 (2011) 673–677 677

70 References

[1] R. Noguerol-Cal, J.M. Lopez-Vilarino, M.V. Gonzalez-Rodriguez, L.F.

60 Barral-Losada, Development of an ultraperformance liquid chro-
matography method for improved determination of additives in
polymeric materials, J. Sep. Sci. 30 (2007) 2452.
Consumption (%)

50
[2] C. Block, L. Wynants, M. Kelchtermans, R. De Boer, F. Compernolle,
Identification of polymer additives by liquid chromatography-mass
spectrometry, Polym. Deg. Stab 91 (2006) 3163.
40
[3] H.E. Mansouri, N. Yagoubi, D. Ferrier, Extraction of polypropylene
additives and their analysis by HPLC, Chromatographia 48 (1998)
30 491.
[4] K. Gaudin, H. Ho-Sung, J. Bleton, J. Joseph-Charles, P. Dallet, P. Puig,
J.-P. Dubost, Determination of N,N-ethylenebisstearamide additive
20 in polymer by normal phase liquid chromatography with evapora-
tive light scattering detection, J. Chromatogr. A 1167 (2007) 27.
[5] S.-S. Choi, M.J. Song, Analysis of cyanoacrylate ultraviolet absorbers
10 using liquid chromatography/atmospheric pressure chemical ioni-
5 10 15 20 zation mass spectrometry: influence of fragmentor voltage and
Aging time (day) solvent on ionization and fragmentation behaviors, Rapid Commun.
Mass. Spectrom. 22 (2008) 2580.
Fig. 5. Variation of consumption rates of the UVAs and UVSs in the PP [6] N. Haider, S. Karlsson, Loss of Chimassorb 944 from LDPE and
identification of additive degradation products after exposure to
composites aged at 90
C with the aging time. The squares, circles, up-
water, air and compost, Polym. Deg. Stab 74 (2001) 103.
triangles, down-triangles, and diamonds indicate the Tinuvin 329, Tinuvin
[7] Y. Taguchi, Y. Ishida, S. Tsuge, H. Ohtani, K. Kimura, T. Yoshikawa, H.
234, Tinuvin 360, Tinuvin 123, and Tinuvin 770, respectively.
Matsubara, Structural change of a polymeric hindered amine light
stabilizer in polypropylene during UV-irradiation studied by reac-
tive thermal desorption- gas chromatography, Polym. Deg. Stab 83
that with lower one. Of the five additives, the additive (2004) 221.
[8] P. Sotonyi, B. Merkely, M. Hubay, J. Jaray, E. Zima, P. Soos, A. Kovacs,
having the longest half-lifetime was Tinuvin 123 (MW 736),
I. Szenrmariay, Comparative study on cardiotoxic effect of Tinuvin
whereas that having the shortest one was Tinuvin 329 (MW 770: a light stabilizer of medical plastics in rat model, Toxico. Sci. 77
323). By using accelerated thermal aging results with good (2004) 368.
linear relationships at high temperatures, it is possible to [9] R. Buchalla, T.H. Begley, Characterization of gamma-irradiated
polyethylene terephthalate by liquid-chromatography–mass-spec-
predict the service lifetimes of the additives in polymeric trometry (LC–MS) with atmospheric-pressure chemical ionization
materials [20]. (APCI), Rad. Phys. Chem. 75 (2006) 129.
[10] M. Himmelsbach, W. Buchberger, E. Reingruber, Determination of
polymer additives by liquid chromatography coupled with mass
4. Conclusions spectrometry. A comparison of atmospheric pressure photoioniza-
tion (APPI), atmospheric pressure chemical ionization (APCI), and
In general, polymeric materials are protected from UV electrospray ionization (ESI), Polym. Deg. Stab 94 (2009) 1213.
[11] W. He, R. Shanks, G. Amarasinghe, Analysis of additives in polymers
light by adding both UVAs and UVSs to ensure extended by thin- layer chromatography coupled with Fourier transform-
durability. UVAs are mostly based on benzotriazole, while infrared microscopy, Vib. Spectrosc. 30 (2002) 147.
UVSs are mostly based on HALS. HALSs are not easily [12] R. Mani, R.P. Singh, S. Chakrapani, S. Sivaram, Synthesis, character-
ization and performance evaluation of hindered amine light stabi-
separated from the other additives with LC because of lizer end functionalized poly(ethylene-alt-propene) copolymer,
numerous basic amino groups. A mixture sample of three Polymer 38 (1997) 1739.
commonly used UVAs (Tinuvin 329, Tinuvin 234, and [13] L. Coulier, E. Kaal, T. Hankemeier, Identification and quantification
of (polymeric) hindered-amine light stabilizers in polymers using
Tinuvin 360) and two popular UVSs (Tinuvin 123 and
pyrolysis–gas chromatography–mass spectrometry and liquid
Tinuvin 770) was separated using HPLC and was quantified chromatography–ultraviolet absorbance detection–evaporative
using APCI-MS. The five additives were successfully sepa- light scattering detection, J. Chromatogr. A. 1062 (2005) 227.
rated by using a C18-column and gradient elution system of [14] V. Jackson-Lewis, G. Liberatore, Effects of a unilateral stereotaxic
injection of Tinuvin 123 into the substantia nigra on the nigros-
methylene chloride/ethanol (80/20) and methanol. The PP triatal dopaminergic pathway in the rat, Brain Res. 866 (2000) 197.
composite containing the five additives was aged in [15] R.E. Duderstadt, S.M. Fischer, Effect of organic mobile phase
a convection oven, and amounts of the UVAs and UVSs composition on signal responses for selected polyalkene additive
compounds by liquid chromatography–mass spectrometry, J.
consumed by the thermal aging were analyzed using LC/ Chromatogr. A 1193 (2008) 70.
APCI-MS. The order of consumption was Tinuvin [16] R. Trones, T. Andersen, T. Greibrokk, D.R. Hegna, Hindered amine
329 > Tinuvin 770 > Tinuvin 234 > Tinuvin 360 > Tinuvin stabilizers investigated by the use of packed capillary temperature-
programmed liquid chromatography I. Poly((6-((1,1,3,3-tetramet-
123. This trend was due to the molecular weights. hylbutyl)-amino)-1,3,5-triazine-2,4- diyl)(2,2,6,6-tetramethyl-4-pi-
Consumption variations of the UVAs and UVSs in the PP peridyl)imino)-1,6-hexanediyl ((2,2,6,6-tetramethyl- 4-piperidyl)
composite with the aging time showed very good linear imino)), J. Chromatogr. A 874 (2000) 65.
[17] R. Trones, T. Andersen, D.R. Hegna, T. Greibrokk, Hindered amine
relationships. The half-lifetimes of Tinuvin 329, Tinuvin stabilizers investigated by the use of packed capillary temperature
234, Tinuvin 360, Tinuvin 123, and Tinuvin 770 at 90
C are programmed liquid chromatography II: poly-(N-b-hydroxyethyl-
9.32, 15.3, 16.0, 19.2, and 15.7 days, respectively. 2,2,6,6-tetramethyl-4- hydroxypiperidyl succinate), J. Chromatogr.
A 902 (2000) 421.
[18] A. Zenker, H. Schmutz, K. Fent, Simultaneous trace determination of
Acknowledgement nine organic UV-absorbing compounds (UV filters) in environ-
mental samples, J. Chromatogr. A 1202 (2008) 64.
This work was supported by the Korea Science and [19] S.-S. Choi, Migration behaviors of wax to the surface in rubber
vulcanizates, J. Appl. Polym. Sci. 73 (1999) 2587.
Engineering Foundation (KOSEF) grant funded by the Korea [20] S.-S. Choi, Correlation between migration behaviors of antiozonants
government (2009-0072952). and temperature, J. Appl. Polym. Sci. 80 (2001) 1566.