Catalysis Today 249 (2015) 5962

Contents lists available at ScienceDirect

Catalysis Today
journal homepage: www.elsevier.com/locate/cattod

Carbonized polyacrylonitrile bers for the catalytic ozonation of

oxalic acid
Alexandra G. Goncalves , Jssica Moreira, Juliana P.S. Sousa, Jos L. Figueiredo,
Manuel F.R. Pereira, Jos J.M. rfo
Laboratory of Catalysis and Materials Associate Laboratory LSRE/LCM, Faculty of Engineering, University of Porto, Rua Dr. Roberto Frias, 4200-465 Porto,
Portugal

a r t i c l e

i n f o

Article history:
Received 13 August 2014
Received in revised form
17 December 2014
Accepted 28 December 2014
Available online 25 January 2015
Keywords:
Catalytic ozonation
Polyacrylonitrile bers
Nitrogen-containing groups
Surface chemistry
Oxalic acid

a b s t r a c t
Textile polyacrylonitrile (PAN) bers, which contain nitrogen, were carbonized at different temperatures
and activated during 10 h in order to assess their performance as ozonation catalysts in the oxalic acid
degradation The ber carbonized at 800 C shows the best performance due to its basic character and,
mainly, the high concentration of nitrogen surface groups, particularly N-pyridinic (N6) groups. The
carbonization of PAN bers is a simple and effective method to prepare materials with good catalytic
activity in ozonation processes
2015 Elsevier B.V. All rights reserved.

1. Introduction
Catalytic ozonation has emerged as a powerful treatment of
water pollutants, even for refractory organic compounds. Carbon
materials, such as activated carbon [115], carbon xerogels [16] and
carbon nanotubes [1722], have shown to be promising materials
as ozonation catalysts. Some studies have reported that their textural properties and, mainly, the chemical surface properties play
an important role in their performance as ozonation catalysts. In
the case of the surface chemistry, it is well known that basic carbon materials favor the formation of hydroxyl radicals, which are
species capable to quickly react with organic pollutants in solution
[2,18]. Furthermore, the basicity of the carbon materials favors the
adsorption of those organic compounds and their oxidation by surface reactions [2,4,18]. The basic character of the carbon materials
is originated by a high density of electrons on the basal planes.
Therefore, this basicity could be promoted removing oxygenated
electron-withdrawing groups and/or introducing some convenient
functional groups, such as different nitrogenated groups (pyridines,

Corresponding author: Tel.: +351 225 084 871; fax: +351 225 081 449.
E-mail addresses: agg@fe.up.pt (A.G. Goncalves), ext12035@fe.up.pt (J. Moreira),
juliana.sousa@fe.up.pt (J.P.S. Sousa), jlg@fe.up.pt (J.L. Figueiredo), fpereira@fe.up.pt
(M.F.R. Pereira), jjmo@fe.up.pt (J.J.M. rfo).
http://dx.doi.org/10.1016/j.cattod.2014.12.045
0920-5861/ 2015 Elsevier B.V. All rights reserved.

pyridones and pyrroles), which contribute to the increase of the

electronic density on the material surface [23].
Nitrogen-containing groups can be introduced after preparation
of carbon materials or during the material synthesis by several
treatments. Among them, nitric acid oxidation, reaction of surface carboxyl groups with diamine compounds, and treatment
at high temperatures with ammonia, ammonia-air, or ammoniasteam mixtures are commonly used [24]. Various types of carbon
materials with incorporated nitrogen (carbon xerogels [25], carbon
nanotubes [26] and activated carbons [27,28]) have been tested in
catalytic ozonation. These studies have reported that the presence
of nitrogenated groups on the surface of carbon materials improves
their catalytic performance.
Adequate carbon materials having nitrogenated groups can
be synthesized by carbonization of nitrogen-containing precursors, like polymers with nitrogen in their structure. This is the
case of polyacrylonitrile (PAN) bers, produced by polymerization from acrylonitrile and vinyl acetate monomers, which contain
high amounts of nitrogen [29]. In this approach, chemical treatments for the introduction of nitrogenated groups are not required,
decreasing the cost and time consumed for the synthesis of carbon
materials enriched with nitrogen.
Therefore, this work aims to carbonize textile polyacrylonitrile
bers at different temperatures and expose then to 10 h of activation in order to test them as catalysts in the ozonation of oxalic

60

A.G. Goncalves et al. / Catalysis Today 249 (2015) 5962

acid, which is a common nal oxidation product of several organic

pollutants and is refractory to single ozonation.

Table 1
Textural properties of the prepared samples.
Sample

2. Experimental
2.1. Preparation of materials
The materials to be used as ozonation catalysts were prepared
from textile PAN bers (FISIVON) supplied by FISIPE, Portugal,
which were subsequently knitted in the facilities of the Technological Centre for the Textile and Clothing Industries of Portugal
(CITEVE).
Pre-treatment of the bers was carried out by heating 5 g of the
original bers up to 300 C in a tubular reactor. A heating rate of
1 C min1 under a constant air/N2 ow of 85 cm3 min1 was used,
and the nal temperature was maintained for 2 h. The bers were
then carbonized by raising the temperature at 5 C min1 up to the
desired temperature (800, 850 or 900 C), which was maintained for
1 h [30]. The study of the effect of the ber activation in its performance as ozonation catalyst was carried out using the carbonized
sample that presented the highest catalytic activity (i.e. carbonized
at 800 C). Thus, further activation was obtained by raising the temperature again at 15 C min1 to 800 C, but then switching the N2
ow to a CO2 ow for a specic time [29]. The reference given to
each sample indicates carbon ber (FI) as well as the respective
carbonization temperature (800900 C) and activation time (0 or
10 h).
2.2. Characterization
The textural characterization of the materials, namely the
BrunauerEmmettTeller (BET) surface area, was based on the N2
adsorption isotherms, determined at 196 C with a Quantachrome
NOVA 4200e apparatus.
The surface chemistry of the prepared samples was characterized by temperature programmed desorption (TPD), in order to
quantify the oxygenated groups [31], and by X-ray photoelectron
spectroscopy (XPS). TPD analyses were performed in an AMI-200
(Altamira Instruments) apparatus. Helium was used as carrier gas
(25 cm3 min1 ) and the temperature was programmed from room
temperature to 1100 C at a heating rate of 5 C min1 The amounts
of CO and CO2 released from the samples were monitored with
a Dymaxion mass spectrometer (Ametek Process Instruments).
XPS analyses were performed in a Kratos AXIS Ultra HSA using
a monochromatic Al K X-ray source (1486.7 eV), operating at
15 kV (90 W), in FAT mode (Fixed Analyser Transmission), with
a pass energy of 40 eV for regions of interest and 80 eV for survey.
Elemental analysis was performed in a Carlo Erba instrument,
model EA 1108.
2.3. Kinetic experiments
The ozonation experiments were carried out in a laboratory
scale reactor (ca. 1 L) equipped with agitation and a circulation
jacket. Ozone was produced from pure oxygen in a BMT 802X ozone
generator. The concentration of ozone in the gas phase was monitored with a BMT 964 ozone analyzer. Ozone leaving the reactor
was removed in a series of gas washing bottles lled with potassium iodide (KI) solution. In each ozonation experiment the reactor
was lled with 700 mL of a solution containing 1 mM of oxalic
acid, at natural pH (approximately 3). Oxalic acid was purchased
from SigmaAldrich. All solutions were prepared with ultrapure
water with a resistivity of 18.2 m cm at room temperature. In
catalytic ozonation experiments, 100 mg of catalyst was introduced in the reactor. The experiments were performed at constant

FI
FI
FI
FI

800 C
800 C
850 C
900 C

0h
10 h
0h
0h

SBET (m2 g1 )

Vpore (cm3 g1 )

S =/

<0.5
117
<0.5
<0.5

0.050

20

pore

(m2 g1 )

gas ow rate (150 cm3 min1 ) and constant inlet ozone concentration (50 g m3 ). The stirring rate was maintained constant at
200 rpm, in order to keep the reactor content perfectly mixed. For
comparative purposes, adsorption experiments were performed
in the same system, under identical conditions. All experiments
were performed at room pressure and temperature. Samples for
analysis were collected at selected times using a syringe and
centrifuged. The experiments were carried out in duplicate and
the maximum relative deviation obtained was 1%. The analytical
measurements were also performed in duplicate with a maximum relative error of 0.5%. Concentration of oxalic acid was
followed by HPLC, according to the procedure described in reference [18].

3. Results and discussion

3.1. Characterization of catalysts
The textural properties of the materials are shown in Table 1.
The carbonized bers have negligible surface areas regardless of
the temperature used. However, activation has a strong impact on
the textural properties. In fact, the activated sample presents a BET
surface area of 117 m2 g1 .
XPS analysis was performed for all prepared materials in order
to obtain additional information about the nature of the N bonding. According to the literature [30,3234], the N1s XPS spectra
can be decomposed in four peaks (1) at about 398 eV, attributed to
pyridinic-N (N6); (2) between 400.0 and 400.9 eV, corresponding
to pyrrolic-N or pyridone-N (N5); (3) in the range 401.4401.7 eV,
attributed to quaternary nitrogen (N-Q); and (4) between 402 and
403 eV, related to certain forms of oxidized nitrogen (N-X). However, in the XPS N1s spectra of all prepared samples only two peaks
were found, as can be observed in Fig. 1. These peaks are associated
to the presence of pyridinic groups (N6) and quaternary nitrogen
(N-Q) on the bers surface. The nitrogen contents obtained from the
deconvolution of the XPS N1s spectra and from elemental analysis
are presented in Table 2.
The original ber has a large amount of nitrogen (24%) [30]. As
expected, increasing the carbonization temperature, the bulk nitrogen amount (obtained from elemental analysis) decreases, as well
as the total nitrogenated surface. Further activation also leads to a
reduction of both bulk and surface nitrogen.
The amounts of CO and CO2 released, obtained by integration of
the areas under TPD proles, the ratio CO/CO2 and the mass percentage of oxygen (mO %) on the surface of the carbon materials are
shown in Table 2. For all the samples, the amount of CO-releasing
groups is larger than the amount of groups which decompose into
CO2 , indicating their basic character, which is in agreement with
the respective pHpzc values (between 7.2 and 8.5 for FI 800 C 10 h
and FI 850 C 0 h, respectively), as reported in the literature [30]. As
the temperatures used in the preparation of samples are higher or
equal to 800 C, the oxygenated surface groups present are mainly
neutral and basic groups like carbonyls/quinones [31], which have
a positive effect on the ozonation process [2,4,18]. The increase of
carbonization temperatures or the activation lead to a decrease in
the amount of CO-releasing groups.

61

Intensity (a. u.)

Intensity (a. u.)

A.G. Goncalves et al. / Catalysis Today 249 (2015) 5962

FI_800C_0h

410

FI_850C_0h

405

400

395

410

405

400

395

400

395

B. E. (eV)

Intensity (a. u.)

Intensity (a. u.)

B. E. (eV)

FI_900C_0h

410

FI_800C_10h

405

400

395

410

405

B. E. (eV)

B. E. (eV)

Fig. 1. N1s XPS spectra for the prepared samples.

Table 2
Chemical characterization of the prepared samples.

FI
FI
FI
FI

800 C
800 C
850 C
900 C

N-Q

N6

Sample

0h
10 h
0h
0h

B.E. (eV)

(%, m/m)

B.E. (eV)

(%, m/m)

398.6
398.5
398.6
398.8

4.6
2.3
1.9
2.6

400.9
401.2
401.3
401.1

4.6
3.4
4.3
3.3

Ntotal , XPS (%, m/m)

Ntotal , EA (%, m/m)

CO (mol g1 )

CO2 (mol g1 )

CO/CO2

mO %

9.2
5.7
6.2
5.9

13.7 0.3
10.3 0.2
12.5 0.3
10.8 0.2

4687
1009
3400
1359

174
92
138
87

26.9
11.0
24.7
15.7

15.3
3.4
11.1
4.5

3.2. Kinetic experiments

In the present study, the ozonation of oxalic acid, at the natural
pH (approximately 3), in the presence of the carbonized or activated
bers was investigated. This carboxylic acid was selected because
it has been identied among the most common oxidation products from organic pollutants degradation and is refractory to single
ozonation in references [4,18,35]. The kinetic results are depicted
in Fig. 2. Besides the degradation prole, the oxalic acid removal
after 360 min of reaction was the value used to compare the catalysts. Ozonation of oxalic acid leads directly to mineralization, i.e.
there is no formation of organic intermediates.
The adsorption capacity of the samples towards oxalic acid was
determined. The results indicated that adsorption on the prepared

Fig. 2. Evolution of the dimensionless concentration of oxalic acid at natural pH

(3) during single ozonation, ozonation catalyzed by the prepared samples and
adsorption (C0 = 1 mM, catalyst = 0.14 g L1 ).

catalysts does not contribute to the removal of oxalic acid (see

Fig. 2), and can therefore be neglected. Thus, it can be considered
that the removal of oxalic acid in the presence of ozone and the
prepared samples mainly occurs by catalytic ozonation.
The inuence of carbonization temperature in the catalytic
activity of the carbonized PAN bers for ozonation of oxalic acid
can be evaluated by comparison of the catalytic performance of
the following samples: FI 800 C 0 h, FI 850 C 0 h and FI 900 C 0 h
(see Fig. 2). It can be veried that the activity of these materials is
not directly related to the temperature of carbonization, since the
worst performance was obtained with the ber carbonized at the
intermediate temperature (850 C). The non-activated materials
have similar textural properties, since the surface area is negligible for all of them, but different performances in the oxidation of
oxalic acid. In this context, the presence of pyridinic groups (N6)
seems to decisively inuences the catalysis: the higher the concentration of N6 groups on the surface of carbonized samples, the
better is its performance. However, this correlation is not strong
(see Fig. 3) probably because the basic oxygen-containing surface groups present also play a role in the catalytic mechanism
[2,4,18]. Nevertheless, the presence of pyridinic groups (N6) on
carbonized bers contributes positively to the oxalic acid removal,
since they are Lewis bases inducing basicity at the carbon surface
by increasing the electron density [23,36] and, therefore, increasing the number of active sites for ozonation. In order to assess the
inuence of the activation in the performance of the materials, the
kinetic results obtained with FI 800 C 0 h and FI 800 C 10 h were
compared. Fig. 2 shows that the activated sample (FI 800 C 10 h)
presents better performance for oxalic acid removal than the nonactivated sample (FI 800 C 0 h) in the beginning of reaction, which
may be mainly justied by its high specic surface area, where
oxalic acid and molecular ozone can adsorb and react. However,
this trend is inverted after 120 min and, thus, the non-activated
sample leads to a faster removal of oxalic acid from the solution.
These results indicate that the specic surface area of the studied

62

A.G. Goncalves et al. / Catalysis Today 249 (2015) 5962

100

References

Xoxalic acid (%)

95
R = 0.8906

90
85
80
75
1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

N6 groups (%)
Fig. 3. Oxalic acid conversion after 360 min of reaction vs. N6 content of carbonized
samples (FI 800 C 0 h, FI 850 C 0 h and FI 900 C 0 h).

materials has not the most important role in the ozonation of

oxalic acid, suggesting again that the reaction is mainly favored by
the presence of nitrogenated groups and basic oxygenated surface
groups (CO-releasing groups).
4. Conclusions
PAN bers containing high amounts of nitrogen were carbonized at different temperatures and further activated during
10 h. The prepared samples were tested as catalysts in the oxalic
acid degradation by catalytic ozonation.
The inuence of the surface chemistry was evidenced by comparison of carbonized bers, for which the surface area is negligible.
The presence of nitrogenated groups on the surface, particularly
pyridinic (N6) groups, favors the oxidation of oxalic acid.
The activation of the carbonized bers, which signicantly
increased their surface area, did not improve the performance,
stressing the important role of nitrogenated and basic oxigenated
surface groups on the reaction mechanism.
The ber carbonized at 800 C presented the best performance in
the removal of oxalic acid due to its basic character and the relative
high concentration of pyridinic (N6) groups on the surface.
The carbonization of PAN bers is a simple and effective method
to prepare materials with good catalytic performance in ozonation
processes.

[1] P.C.C. Faria, J.J.M. rfo, M.F.R. Pereira, Water Res. 39 (2005) 14611470.
[2] P.C.C. Faria, J.J.M. rfo, M.F.R. Pereira, Ind. Eng. Chem. Res. 45 (2006)
27152721.
[3] P.C.C. Faria, J.J.M. rfo, M.F.R. Pereira, Chemosphere 67 (2007) 809815.
[4] P.C.C. Faria, J.J.M. rfo, M.F.R. Pereira, Appl. Catal. B 79 (2008) 237243.
[5] P.C.C. Faria, J.J.M. rfo, M.F.R. Pereira, Appl. Catal. B 83 (2008) 150159.
[6] F.J. Beltrn, B. Acedo, F.J. Rivas, O. Gimeno, Ozone Sci. Eng. 27 (2005) 159169.
[7] F.J. Beltrn, J.F. Garca-Araya, I. Girldez, Appl. Catal. B 63 (2006) 249259.
[8] F.J. Beltrn, J.F. Garca-Araya, I. Girldez, F.J. Masa, Ind. Eng. Chem. Res. 45 (2006)
30153021.
[9] F.J. Beltran, P. Pocostales, P. Alvarez, A. Aguinaco, Ozone Sci. Eng. 31 (2009)
422427.
[10] F.J. Beltrn, P. Pocostales, P. Alvarez, A. Oropesa, J. Hazard. Mater. 163 (2009)
768776.
[11] F.J. Bltran, F.J. Rivas, L.A. Fernandez, P.M. Alvarez, R. Montero-de-Espinosa, Ind.
Eng. Chem. Res. 41 (2002) 65106517.
[12] F.J. Beltrn, J. Rivas, P. lvarez, R. Montero-de-Espinosa, Ozone Sci. Eng. 24
(2002) 227237.
[13] J. Rivera-Utrilla, M. Sanchez-Polo, Water Resour. Res. 39 (2003) 12321244.
[14] J. Rivera-Utrilla, M. Sanchez-Polo, Langmuir 20 (2004) 92179222.
[15] J. Rivera-Utrilla, M. Snchez-Polo, Appl. Catal. B 39 (2002) 319329.
[16] C.A. Orge, J.J.M. rfo, M.F.R. Pereira, Appl. Catal. B 126 (2012) 2228.
[17] A. Goncalves, J.J.M. rfo, M.F.R. Pereira, Appl. Catal. B 140141 (2013)
8291.
[18] A.G. Goncalves, J.L. Figueiredo, J.J.M. rfo, M.F.R. Pereira, Carbon 48 (2010)
43694381.
[19] A.G. Goncalves, J.J.M. rfo, M.F.R. Pereira, J. Hazard. Mater. 239240 (2012)
167174.
[20] A.G. Goncalves, J.J.M. rfo, M.F.R. Pereira, Catal. Commun. 35 (2013) 8287.
[21] Z.-Q. Liu, J. Ma, Y.-H. Cui, B.-P. Zhang, Appl. Catal. B 92 (2009) 301306.
[22] Z.-Q. Liu, J. Ma, Y.-H. Cui, L. Zhao, B.-P. Zhang, Appl. Catal. B 101 (2010)
7480.
[23] H.-P. Boehm, Catalytic properties of nitrogen-containing carbons, in: P. Serp,
J.L. Figueiredo (Eds.), Carbon Materials for Catalysis, John Wiley & Sons, Inc,
2008, pp. 219265.
[24] J. Lahaye, G. Nans, A. Bagreev, V. Strelko, Carbon 37 (1999) 585590.
[25] R.P. Rocha, J. Restivo, J.P.S. Sousa, J.J.M. rfo, M.F.R. Pereira, J.L. Figueiredo,
Catal. Today 241 (2015) 7379.
[26] J. Restivo, R.P. Rocha, A.M.T. Silva, J.J.M. rfo, M.F.R. Pereira, J.L. Figueiredo,
Chin. J. Catal. 35 (2014) 896905.
[27] H. Cao, L. Xing, G. Wu, Y. Xie, S. Shi, Y. Zhang, D. Minakata, J.C. Crittenden, Appl.
Catal. B 146 (2014) 169176.
[28] L. Xing, Y. Xie, H. Cao, D. Minakata, Y. Zhang, J.C. Crittenden, Chem. Eng. J. 245
(2014) 7179.
[29] P.J.M. Carrott, J.M.V. Nabais, M.M.L. Ribeiro Carrott, J.A. Pajares, Carbon 39
(2001) 15431555.
[30] J.P.S. Sousa, A.M.T. Silva, M.F.R. Pereira, J.L. Figueiredo, Sep. Sci. Technol. 45
(2010) 15461554.
[31] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. rfo, Carbon 37 (1999)
13791389.
[32] H.F. Gorgulho, F. Goncalves, M.F.R. Pereira, J.L. Figueiredo, Carbon 47 (2009)
20322039.
[33] R.J.J. Jansen, H. van Bekkum, Carbon 33 (1995) 10211027.
[34] F. Kapteijn, J.A. Moulijn, S. Matzner, H.P. Boehm, Carbon 37 (1999) 11431150.
[35] C.A. Orge, J.J.M. rfo, M.F.R. Pereira, Appl. Catal. B 102 (2011) 539546.
[36] S. Matzner, H.P. Boehm, Carbon 36 (1998) 16971703.