This article appeared in a journal published by Elsevier.

The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
 Author's personal copy

Analytica Chimica Acta 643 (2009) 108–116

Contents lists available at ScienceDirect

Analytica Chimica Acta

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

Characterization of synthetic polymers and speck impurities in cellulose pulp:

A comparison between pyrolysis-gas chromatography–mass spectrometry and
Fourier transform infrared spectroscopy
F.O. Silvério a , L.C.A. Barbosa a,∗ , C.R.A. Maltha a , D. Piló-Veloso b
a
Chemistry Department, Federal University of Viçosa, 36570-000, Viçosa, MG, Brazil
b
Chemistry Department, Federal University of Minas Gerais, 31270-901, Belo Horizonte, MG, Brazil

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

Article history: Pyrolysis-gas chromatography–mass spectrometry (Py-GC/MS) and Fourier transform infrared (FT-IR)
Received 8 December 2008 spectroscopy were used for characterizing specks in cellulose pulp, polymeric materials and pitch formed
Received in revised form 4 March 2009 during the cellulose extraction and paper production in the Brazilian mill. Three samples were analyzed
Accepted 31 March 2009
and the pyrograms and infrared spectra obtained were compared. The results showed that the analytical
Available online 8 April 2009
pyrolysis more effectively differentiated between impurities (dirt specks) when compared to the infrared
spectroscopy.
Keywords:
© 2009 Elsevier B.V. All rights reserved.
Pyrolysis-gas chromatography–mass
spectrometry
Fourier transform infrared spectroscopy
Specks in pulp
Pitch

1. Introduction On the other hand, pyrolysis-gas chromatography coupled with

During pulp and paper production cellulose fibers can be
impregnated with polymeric materials coming from rubbers used
in the equipments causing impurity specks in pulps [1]. These
imperfections represent a serious problem for the exigent con-
sumer market, since they are responsible for pulp depreciation and
disqualification, leading to a decline in production and implying
great losses for the manufacturers [2,3]. In addition, these poly-
meric materials can also be mixed with sticky deposits of wood
lipophilic extractive on machinery, commonly known as pitch [4].
During the last years studies have been carried out in order to
discover the origin of these specks [1,2,5–8]. Infrared spectroscopy
is the most used tool for investigating the chemical composition of
these dirt specks [9,10]. Despite being frequently used in routine
analyses, as in paper and cellulose industries, this technique has
some limitations for the identification of speck in cellulose pulp.
The methodology is based on the comparison between infrared
spectra of unknown samples and known patterns, which are ana-
lyzed under the same conditions. However, the procedure is not
always effective because a large number of spectra of pulp specks
are very similar to the ones of different polymeric materials, gen-
erating doubts about the results.
mass spectrometry (Py-GC/MS) was employed in the works of Burns
and Doolan [11–13] to identify samples of automotive painting
that are indistinguishable by Fourier transform infrared (FT-IR)
examinations. The results obtained allowed to distinguish types
of paintings presenting very similar infrared spectra. In that work,
the methodology employed was also based on comparison of both
specky and reference pyrograms obtained under same conditions.
The main advantage of the analytic pyrolysis is the possibility to
confirm the results by mass spectrometry [12,13].
This technique stands out for the possibility of analyzing very
small samples (approximately 100 ␮g). Other advantages include
a short period of analysis, generally about 1 h [14–18], little previ-
ous sample manipulation and the possibility of analyzing insoluble
materials [17,19].
In view of the need of a faster and more reliable analytical tech-
nique for detailed characterization of speck samples in pulp mills, in
this work we compared data obtained from Py-GC/MS and infrared
spectroscopy of dirt specks (impurities) in cellulose pulp, pitch and
polymeric materials used in the industrial process.
2. Experimental
2.1. Sample description

∗ Corresponding author. Tel.: +55 031 3899 3068; fax: +55 031 3899 3065. Three specks in cellulose fiber samples were obtained from
E-mail address: lcab@ufv.br (L.C.A. Barbosa). bleached kraft pulp prepared across the OD(EP)DP sequence in

0003-2670/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.aca.2009.03.049
 Author's personal copy

F.O. Silvério et al. / Analytica Chimica Acta 643 (2009) 108–116 109

Table 1 a Brazilian mill. A black pitch sample was collected from the
Specification of each polymeric material collected at different localization at the
cellulose-drying machine of the same company.
pulp mill.
Specks found in cellulose fibers separated from the fiber under
Polymeric materials Specification a microscope using tweezers were designated as I-1, I-2 and I-3.
R-1 Rubber collected in the depuration sieve The three samples of specks (I-1, I-2 and I-3) and one pitch sam-
R-2 Valve of the virgin butterfly ple were initially analyzed by infrared spectroscopy. At the same
R-3 Rubber collected in the depuration sieve time, nine polymeric items coming from different parts of the same
R-4 Front tire of the forklift
mill and designated as R-1 to R-9 (Table 1) were collected and
R-5 Hard seal of the Kvaerner press
R-6 Depuration residue 1 analyzed by FT-IR. All specks were placed in special quartz tubes
R-7 Back tire of the forklift and analyzed by automatic Py-GC/MS without any prior treatment.
R-8 Inflatable seal of the Kvaerner press The same procedure was accomplished with the nine polymeric
R-9 Depuration residue 2
materials.

Fig. 1. IR spectra and pyrograms of polymeric material R-3, after each bleaching stage. (1)-Sulphur dioxide, (2)-hex-1-ene, (3)-benzene, (4)-toluene, (5)-non-1-ene, (6)-dec-
1-ene, (7)-undec-1-ene, (8)-dodec-1-ene, (9)-tridec-1-ene, (10)-tetradec-1-ene, (11)-pentadec-1-ene, (12)-hexadec-1-ene, (13)-heptadec-1-ene, (14)-octadec-1-ene, (15)-ni,
(16)-ni.
 Author's personal copy

110 F.O. Silvério et al. / Analytica Chimica Acta 643 (2009) 108–116

Fig. 1. (Continued ).

3. Instrumentation 3.2. Pyrolysis-gas chromatography–mass spectrometry

(Py-GC/MS)
3.1. Fourier transform infrared (FT-IR) spectroscopy
Analytical pyrolysis was performed by using a pyrolyser Pyr
The infrared spectra of samples prepared in KBr disks were A-4 model, Shimadzu. The sample was put in a platinum cup
recorded in the region of 4000–500 cm−1 using a PerkinElmer Spec- then placed inside a 2 mm × 40 mm quartz tube. The pyrolysis
trum 1000 spectrophotometer. was carried out at 600 ◦ C for 10 s. The pyrolysis chamber was

Table 2
Compounds released from the pyrolysis of several polymeric materials.

Polymeric materials Main pyrolysis fragments in order of retention time

R-1 (1)-2-chlorobuta-1,3-diene, (2)-benzene, (3)-1,4-dichlorobuta-1,3-diene, (4)-2-chlorobuta-1,3-diene dimer, (5)-butyl benzoate,

(6)-ni, (7)-ni, (8)-butyl phtalate, (9)-octadecanoic acid, (10)-nia
R-2 (1)-hex-1-ene, (2)-hept-1-ene, (3)-oct-1-ne, (4)-non-1-ene, (5)-dec-1-ene, (6)-undec-1-ene, (7)-dodec-1-ene, (8)-tridec-1-ene,
(9)-tetradec-1-ene, (10)-pentadec-1-ene, (11)-hexadec-1-ene, (12)-heptadec-1-ene, (13)-ni, (14)-octadec-1-ene,
(15)-nonadec-1-ene, (16)-eicos-1-ene, (17)-heneicos-1-ene, (18)-docos-1-ene, (19)-tricos-1-ene, (20)-tetracos-1-ene, (21)-ni
(22)-ni, (23)-ni, (24)-ni, (25)-ni, (26)-ni, (27)-ni, (28)-ni, (29)-ni, (30)-ni, (31)-ni, (32)-ni, (33)-ni
R-3 (1)-Sulphur dioxide, (2)-hex-1-ene, (3)-benzene, (4)-toluene, (5)-non-1-ene, (6)-dec-1-ene, (7)-undec-1-ene, (8)-dodec-1-ene,
(9)-tridec-1-ene, (10)-tetradec-1-ene, (11)-pentadec-1-ene, (12)-hexadec-1-ene, (13)-heptadec-1-ene, (14)-octadec-1-ene,
(15)-ni, (16)-ni, (17)-ni, (18)-ni, (19)-ni, (20)-ni
R-4 (1)-Buta-1,3-diene, (2)-isoprene, (3)-3-methylpent-2-ene, (4)-isoprene dimer
R-5 (1)-Buta-1,3-diene, (2)-isoprene, (3)-3-methylpent-2-ene, (4)-isoprene dimer
R-6 (1)-But-2-ene, (2)-2-methylbut-2-ene, (3)-2-methylpent-1-ene, (4)-hex-1-ene, (5)-hept-1-ene, (6)-non-1-ene, (7)-dec-1-ene,
(8)-undec-1-ene, (9)-dodec-1-ene, (10)-tridec-1-ene, (11)-tetradec-1-ene, (12)-pentadec-1-ene, (13)-hexadec-1-ene,
(14)-heptadec-1-ene, (15)-octadec-1-ene, (16)-nonadec-1-ene
R-7 (1)-Pent-1-ene, (2)-hex-1-ene, (3)-hept-1-ene, (4)-oct-1-ene, (5)-non-1-ene, (6)-dec-1-ene, (7)-undec-1-ene, (8)-dodec-1-ene,
(9)-tridec-1-ene, (10)-tetradec-1-ene, (11)-pentadec-1-ene, (12)-hexadec-1-ene, (13)-heptadec-1-ene, (14)-octadec-1-ene,
(15)-nonadec-1-ene, (16)-eicos-1-ene, (17)-heneicos-1-ene, (18)-docos-1-ene, (19)-tricos-1-ene, (20)-tetracos-1-ene,
(21)-pentacos-1-ene, (22)-hexacos-1-ene, (23)-heptacos-1-ene, (24)-octacos-1-ene, (25)-nonacos-1-ene, (26)-triacont-1-ene,
(27)-hentriacont-1-ene, (28)-docotriacont-1-ene, (29)-tritriacont-1-ene, (30)-tetratriacont-1-ene, (31)-pentatriacont-1-ene,
(32)-hexatriacont-1-ene, (33)-heptatriacont-1-ene, (34)-octatriacont-1-ene, (35)-nonatriacont-1-ene, (36)-tetracont-1-ene,
(37)-hentetracont-1-ene, (38)-docotetracont-1-ene, (39)-tritetracont-1-ene, (40)-tetratetracont-1-ene
R-8 (1)-Sulphur dioxide, (2)-buta-1,3-diene, (3)-2-methylbuta-1,3-diene, (4)-benzene, (5)-2-ethylhex-1-ene,
(6)-4-ethenylcyclohex-1-ene, (7)-octyl benzoate
R-9 (1)-Sulphur dioxide, (2)-chlorotrimethylsilane, (3)-benzene, (4)-toluene, (5)-1,2-dimethylbenzene, (6)-octyl phtalate
a
The number in brackets corresponds to the number in each pyrogram.
 Author's personal copy

F.O. Silvério et al. / Analytica Chimica Acta 643 (2009) 108–116 111

kept at 100 ◦ C and purged with helium to transfer the pyrolysis
products as quickly as possible to the GC column. The pyrolyser
was connected to a GC/MS apparatus (Shimadzu, model PQ5050,)
equipped with an on-column injector and a fused silica capillary
column (DB-5, 30 m × 0.25 mm ID, 0.25 ␮m film thickness). The
chromatograph oven temperature was programmed to vary from
40 ◦ C (5 min) to 300 ◦ C at a rate of 5 ◦ C min−1 . The final temper-
ature was held for 20 min. The injector and GC/MS interface was
kept at 200 ◦ C. Compounds were identified by comparing their
mass spectra with the GC/MS spectral library (Willey 333,000),
with data from the literature [1,6] and when necessary by mass
fragmentography.
3.3. Evaluation of polymeric materials chemical transformations
Considering the possibility of some polymeric materials react-
ing or undergoing structural chemical changes during the bleaching
process, the polymers were submitted to the same conditions of the
bleaching process of cellulose fiber. The bleaching was carried out
in the laboratory employing the OD(EP)DP sequence, under con-

Fig. 2. IR spectra and pyrograms of polymeric materials used in the industrial process.
 Author's personal copy

112 F.O. Silvério et al. / Analytica Chimica Acta 643 (2009) 108–116

Fig. 2. (Continued )

ditions specifically pre-established, to produce pulps with a final
brightness of 90% and 88% ISO, respectively. Following each stage,
laboratory sheets were prepared, according to NBR ISO 5269-1
standards, and specks samples were collected from the cellulose
pulp.
4. Results and discussion
The IR spectra of all the polymeric materials were similar. In
the same way, the pyrograms did not show any significant change,
indicating that no chemical modification with the bleaching con-
ditions was observed. For illustrative purposes, only the results of
polymer material R-3 are presented. However the same behaviour
was found for the other polymeric materials during the process
(Fig. 1).

4.1. Influence of each bleaching stage in the polymeric materials
To verify any chemical modification of the polymeric materials
after each bleaching stage, nine polymeric materials were submit-
ted to the same conditions of the bleaching process used for the
cellulose fiber. After each bleached stage, the polymeric materials
were analyzed by FT-IR spectroscopy and Py-GC/MS.
4.2. Comparison among FT-IR and Py-GC/MS in analysis of speck
samples
Three speck samples of the industrial process (specks I-1, I-2
and I-3) were chosen to demonstrate how the Py-GC/MS tech-
nique can be used to analyze specks in cellulose pulp as routine
process. The analysis was initially accomplished by infrared spec-
 Author's personal copy

F.O. Silvério et al. / Analytica Chimica Acta 643 (2009) 108–116 113

Fig. 2. (Continued ).

troscopy and then by Py-GG/MS. The IR spectra of the specks
were compared and associated with spectra of polymeric mate-
rials (R-1 to R-9) used in the industrial process (Fig. 2 and
Table 2).
4.2.1. Analysis of speck sample I-1
Fig. 3 shows the IR spectrum and the pyrogram of speck I-1.
The IR spectrum of speck I-1 was partially similar to the spec-
trum of polymeric material R-2 (Fig. 2). Although the fingerprint
region (1500–800 cm−1 ) of IR spectrum presents some differences,
the spectrum does not allow great distinctions from the other
polymeric materials, which can generate confusions in the inter-
pretation of the spectra, resulting in doubts and wrong decisions in
the mill.
Therefore, the identification of such sample required a more in-
depth analysis. Speck I-1 was analyzed by Py-GC/MS, and the results
obtained from Py-GC/MS showed that the pyrogram of speck I-
1 is greatly different from that of polymeric material R-2 (Fig. 2).
Sulphur dioxide, buta-1,3-diene, benzene, toluene, ethylbenzene
and styrene were the main compounds released upon Py-GC/MS
of speck I-1, while saturated hydrocarbons were the most abun-
dant compounds in the pyrogram of the R-2 (Table 2). Although
sulphur dioxide, benzene and toluene were also identified in the
pyrogram of R-3. Its chemical composition is very different to that
of I-1 (Table 2).
In accordance with the IR spectra, the pyrogram of I-1 did not
show any similarity of the pyrograms of the other polymeric mate-
rials.
4.2.2. Analysis of speck sample I-2
The IR spectrum and the pyrogram of the speck I-2 are shown
in Fig. 4. The IR spectrum of speck I-2 presented some similarities
 Author's personal copy
114 F.O. Silvério et al. / Analytica Chimica Acta 643 (2009) 108–116
Fig. 3. IR spectrum and pyrogram of speck I-1. (1)-Sulphur dioxide, (2)-buta-1,3-
diene, (3)-benzene, (4)-toluene, (5)-ethylbenzene, (6)-styrene.
with the spectrum of polymeric material R-1 (Fig. 2), although some
differences were observed in the fingerprint region. However, those
differences do not allow total distinctions among the spectra of the
other polymeric materials (Fig. 2).
Therefore, speck I-2 was submitted to pyrolysis and its pyro-
gram was shown to be different from the pyrogram of polymeric
material R-1 (Fig. 2 and Table 2). The compounds released upon
Py-GC/MS of specks I-1 and I-2 were the same, although present
in different amounts. However, the pyrolysis data indicates that
I-2 was very different to other nine polymeric items (R-1 to
R-9).
These findings enabled us to verify that the analysis carried
out using only infrared data, is inaccurate and can lead to mis-
placed conclusions, as reported by previous works [1]. Therefore,
Py-GC/MS emerges as an alternative method to be used in routine
analysis.
4.2.3. Analysis of speck I-3
The speck sample I-3 was analyzed by FT-IR spectroscopy and Py-
GC/MS (Fig. 5), and the spectrum was compared with the spectra
of polymeric materials (Fig. 2).
Comparison with the pyrograms of polymeric materials R-1 to
R-9 (Fig. 2), showed that polymeric materials R-4 and R-5 pro-
duced similar pyrograms as speck sample I-3, having as main
constituents isoprene and dipentene (isoprene dimer), whereas the
rest of polymeric materials (R-1 to R-3, R-6 to R-9) gave chemi-
cal composition and pyrograms completely different (see Table 2).
Although the IR spectrum of polymeric material R-7 was similar to
Fig. 4. IR spectrum and pyrogram of speck I-2. (1)-Sulphur dioxide, (2)-buta-1,3-
diene, (3)-benzene, (4)-toluene, (5)-ethylbenzene, (6)-styrene.
the IR spectrum of speck I-3, the pyrograms from R-7 to I-3 were
totally different. The pyrogram of the R-7 presented 40 compounds
while two compounds were identified in the pyrogram of the
speck I-3.
The results showed that the polymeric materials R-4 and R-
5 have the similar chemical composition, because after pyrolysis
only buta-1,3-diene, isoprene and isoprene dimer were produced
(see Table 2). Comparing the mass spectra, we observed that these
compounds were also found in the pyrogram of the analyzed I-3.
However, as the amount of the specky sample (I-3) was very small,
its pyrogram did not show the sign of the isoprene dimer. This result
showed how Py-GC/MS allows a more accurate confirmation, and
that R-4 and R-5 polymeric materials (Fig. 2) are responsible for the
contamination.
The results obtained by Py-GC/MS technique allowed to obtain
more information from the pyrograms, therefore reducing the pos-
sibility of errors in the evaluation of the real causes of imperfections
found in cellulose pulps.
4.3. Comparison among FT-IR and Py-GC/MS in analysis of pitch
sample
A pitch sample was collected from the cellulose-drying machine
and was initially analyzed by FT-IR spectroscopy. The obtained spec-
trum (Fig. 6) was compared with the spectrum of the polymeric
materials (R-1 to R-9) used in the industrial process. The IR spectrum
of polymer R-3 was similar to the spectrum of pitch, suggesting that
some particles of R-3 could be present in this pitch sample. The
 Author's personal copy
F.O. Silvério et al. / Analytica Chimica Acta 643 (2009) 108–116
Fig. 5. IR spectrum and pyrogram of speck I-3. (1)-Buta-1,3-diene, (2)-isoprene.
pyrolysis data indicated that the source of the pitch sample is the
wood extractives.
However, when carrying out the Py-GC/MS analysis, we found
that the pyrograms were completely different (Fig. 7), indicating
the absence of polymeric materials R-3 in this particular pitch
sample. The pyrogram of the pitch showed 34 compounds and
only 16 compounds were identified in the pyrogram of the sample
R-3. However, 10 compounds (non-1-ene, dec-1-ene, undec-1-
ene, dodec-1-ene, tridec-1-ene, tetradec-1-ene, pentadec-1-ene,
hexadec-1-ene, heptadec-1-ene, octadec-1-ene) were detected in
both programs.
Fig. 6. IR spectrum of a pitch sample and of the polymeric material R-3.
115
Fig. 7. Pyrogram of the pitch sample. (1)-But-2-ene, (2)-cyclopentanone, (3)-
non-1-ene, (4)-dec-1-ene, (5)-undec-1-ene, (6)-dodec-1-ene, (7)-tridec-1-ene,
(8)-tetradec-1-ene, (9)-pentadec-1-ene, (10)-hexadec-1-ene, (11)-heptadec-1-ene,
(12)-2,6,10,14-tetramethylhexadec-2-ene, (13)-octadec-1-ene, (14)-nonadec-1-ene,
(15)-hexadecanoic acid, (16)-eicos-1-ene, (17)-heneicos-1-ene, (18)-docos-1-ene,
(19)-tricos-1-ene, (20)-ni, (21)-tetracos-1-ene, (22)-ni, (23)-ni, (24)-ni, (25)-ni, (26)-
ni, (27)-ni, (28)-ni, (29)-ni, (30)-ni, (31)-ni, (32)-ni, (33)-ni, (34)-ni.
5. Conclusions
Py-GC/MS has proven a very useful technique for the analysis of
polymeric materials and small specks of impurities found in cellu-
lose pulp, as it has allowed the distinction of pairs of samples with
similar IR spectra.
The Py-GC/MS methodology employed in this work is relatively
simple, fast and cheap, being a more reliable procedure than FT-IR
spectroscopy for the identification of pulp specks.
The main advantage of Py-GC/MS over other usual methods is
the possibility of analyzing small samples without any prior treat-
ment.
The exact identification is important for tracing contamination
along the process and eliminating its cause. Therefore, Py-GC/MS
could be a complementary tool to FT-IR for the classification of
specks in pulps.
The results of this investigation should have a direct impact in
paper and cellulose mills as Py-GC/MS could be a complementary
tool for the correct classification of specks and thus to trace them
back to their source, allowing the development of strategies for their
control.
This result showed therefore how Py-GC/MS allows a more accu-
rate confirmation, and that polymeric materials R-4 and R-5 (Fig. 5)
are responsible for the contamination.
Acknowledgements
Thanks are due to Conselho Nacional de Desenvolvimento Cien-
tífico e Tecnológico (CNPq) for financial support and research
fellowships (LCAB) and a Post-Doctoral Research (FOS); Fundação
de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG)
for financial support. We are also grateful to Ishmael Roslan
(Department of Chemistry, Oxford University) for corrections and
suggestions presented in the manuscript.
References
[1] J.C. Del Río, M. Hernando, P. Landín, A. Gutiérrez, J. Romero, J. Anal. Appl. Pyrol.
68–69 (2003) 251–268.
[2] M.P. Cruz, L.C.A. Barbosa, C.R.A. Maltha, J.L. Gomide, A.F. Milanês, Quím. Nova
29 (2006) 459–466.
[3] F.O. Silvério, L.C.A. Barbosa, C.R.A. Maltha, P.H. Fidêncio, M.P. Cruz, D. Piló-Veloso,
A.F. Milanez, Bioresour. Technol. 99 (2008) 4878–4886.
 Author's personal copy
116 F.O. Silvério et al. / Analytica Chimica Acta 643 (2009) 108–116
[4] E.L. Back, L.H. Allen, Pitch Control, Wood Resin and Deresination, Tappi Press,
Atlanta, 2000, pp. 1–35.
[5] F.J. González-Vila, A. Gutiérrez, F. Martin, T. Verdejo, J. Anal. Appl. Pyrol. 40–41
(1997) 501–510.
[6] J.C. Del Río, A. Gutiérrez, E.J. Gonzáles-Vila, F. Martin, J. Anal. Appl. Pyrol. 49
(1999) 165–177.
[7] J.C. Del Río, A. Gutiérrez, F.J. González-Vila, J. Chromatogr. A 830 (1999) 227–232.
[8] F.O. Silvério, L.C.A. Barbosa, C.R.A. Maltha, G.P. Pinho, P.H. Fidêncio, D. Pilo-
Veloso, L.C. Souza, Appita J. (2009), in press.
[9] N. Dunlop-Jones, L.H. Allen, Tappi J. 71 (1988) 109–113.
[10] E. Sjöström, R. Alen, Analytical Methods in Wood Chemistry, Pulping and Paper-
making, Springer Series in Wood Science, Germany, 1998, pp. 1–316.
[11] D.T. Burns, K.P. Doolan, Anal. Chim. Acta 422 (2000) 217–230.
[12] D.T. Burns, K.P. Doolan, Anal. Chim. Acta 539 (2005) 145–155.
[13] D.T. Burns, K.P. Doolan, Anal. Chim. Acta 539 (2005) 157–164.
[14] H.L. Hardell, J. Anal. Appl. Pyrol. 27 (1993) 73–85.
[15] Y. Ishida, H. Ohtani, T. Kato, S. Tsuge, T. Yano, Tappi J. 77 (3) (1994) 177–183.
[16] C. Smith, Pyrolysis measurements, in: S. Frank (Ed.), Handbook of Instrumen-
tal Techniques for Analytical Chemistry, Prentice Hall PTR, Upper Saddle River,
1997, pp. 1–511.
[17] F.O. Silvério, L.C.A. Barbosa, D. Piló-Veloso, Quím. Nova 31 (6) (2008) 1543–1552.
[18] C.F. Lima, L.C.A. Barbosa, C.R. Marcelo, F.O. Silvério, J.L. Colodette, Bioresource 3
(3) (2008) 701–712.
[19] G. Montaudo, M.S. Montaudo, in: G. Montaudo, R.P. Lattimer (Eds.), Mass Spec-
trometry of Polymers, CRC Press, Boca Raton, FL, 2002, pp. 41–111.