Journal of Thermal Analysis and
An International Forum for Thermal
J Therm Anal Calorim (2012)
Functionalisation of polypropylene non-
woven fabrics (NWFs)
Viktória Vargha, Avashnee Chetty, Zsolt
Sulyok, Judith Mihály, Zsófia Keresztes,
András Tóth, István Sajó, László Korecz,
Your article is protected by copyright and
all rights are held exclusively by Akadémiai
Kiadó, Budapest, Hungary. This e-offprint is
for personal use only and shall not be self-
archived in electronic repositories. If you
wish to self-archive your work, please use the
accepted author’s version for posting to your
own website or your institution’s repository.
You may further deposit the accepted author’s
version on a funder’s repository at a funder’s
request, provided it is not made publicly
available until 12 months after publication.
Functionalisation of polypropylene non-woven fabrics (NWFs)
Functionalisation by oxyfluorination as a first step for graft polymerisation
Vikto ´ria Vargha•Avashnee Chetty•Zsolt Sulyok•Judith Miha ´ly•
Zso ´fia Keresztes•Andra ´s To ´th•Istva ´n Sajo ´ •La ´szlo ´ Korecz•
Rajesh Anandjiwala•Lydia Boguslavsky
Received: 26 April 2011/Accepted: 20 September 2011/Published online: 21 October 2011
? Akade ´miai Kiado ´, Budapest, Hungary 2011
polypropylene non-woven fabric (PP NWF) samples of
different morphologies and pore sizes. The modified sur-
faces were characterised by Attenuated Total Reflectance
Fourier Transform InfraRed (ATR-FTIR)-spectroscopy,
FTIR imaging microscopy, X-Ray Photoelectron Spec-
troscopy (XPS), Electron Spin Resonance (ESR) spec-
troscopy, Differential Scanning Calorimetry (DSC), X-Ray
Surface oxyfluorination had been carried out on
Diffraction (XRD) analysis, Scanning Electron Microscopy
(SEM), dynamic rheometry and Thermo-Gravimetry (TG).
ATR-FTIR and XPS techniques revealed the presence of –
CF, –CF2, –CHF and –C(O)F groups. The formed –C(O)F
groups mostly got hydrolysed to –COOH groups. The C=O
groups of alpha-haloester, and the C=C stretching of the
formed –CF=C(OH)– groups could also be detected. Long-
lived radicals could be detected on the functionalised
surfaces as middle-chain peroxy radicals by ESR spec-
troscopy. SEM micrographs showed slight roughening of
the oxyfluorinated surfaces. Oxyfluorination had no sig-
nificant effect on the crystalline structure and phase com-
position of the PP NWF samples supported by DSC and
XRD measurements. The molecular mass of the samples
were unaffected by the oxyfluorination treatment as proved
by oscillating rheometry. The surface modification, how-
ever, significantly affected the thermal decomposition but
not affected the thermo-oxidative decomposition of PP
NWFs. Different morphologies and pore sizes of PP NWF
samples resulted in reproducibility of the findings, although
did not substantially affect surface functionalisation.
ATR-FTIR ? FTIR-imaging ? XPS ? ESR ? SEM ? DSC ?
WAXS ? Dynamic rheometry ? TG
Oxyfluorination ? PP non-woven fabrics ?
Commodity polyolefins, namely, polyethylene and poly-
propylene (PP) are widely used in industry because of their
low cost, ease of processing, good mechanical properties,
and excellent chemical resistance. These polymers, how-
ever, have low surface tension, poor adhesion and insuffi-
cient chemical functionality which limit their use in certain
V. Vargha (&) ? Z. Sulyok
Budapest University of Technology and Economics, Department
of Physical Chemistry and Material Science, M} uegyetem rkp. 3.
H/1, Budapest 1111, Hungary
CSIR Materials Science and Manufacturing, Polymers and
Composites, P. O. Box 395, Pretoria 0001, South Africa
J. Miha ´ly ? Z. Keresztes ? I. Sajo ´
Chemical Research Center of the Hungarian Academy
of Sciences, Institute of Nanochemistry and Catalysis,
Pusztaszeri u ´t 59-67, Budapest 1025, Hungary
A. To ´th
Chemical Research Center of the Hungarian Academy of
Sciences, Institute of Materials and Environmental Chemistry,
Pusztaszeri u ´t 59-67, Budapest 1025, Hungary
Chemical Research Center of the Hungarian Academy of
Sciences, Institute of Structural Chemistry, Pusztaszeri u ´t 59-67,
Budapest 1025, Hungary
R. Anandjiwala ? L. Boguslavsky
CSIR Materials Science and Manufacturing, Polymers &
Composites Competence Area, POBox 1124, Summerstrand,
Port Elizabeth 6000, South Africa
J Therm Anal Calorim (2012) 109:1019–1032
Author's personal copy
applications. There are a number of pre-treatment tech-
niques commonly used to activate polymer surfaces, such
as plasma modification, gamma-irradiation, UV-irradia-
tion, ozonation, surface-graft polymerisation, chemical
reaction and flame treatment.
In recent years, surface fluorination has proved to be an
effective pre-treatment technique for increasing the wet-
tability, adhesion and barrier properties of hydrocarbon
polymers [1–5]. Fluorination is attractive over other func-
tionalising methods since it is less invasive, e.g. compared
to radiation, is a dry technology, can be used to modify
articles of any shape, and can be conducted at relatively
low temperatures in a low vacuum. In addition, fluorine gas
is highly reactive and can penetrate polymer surfaces to
large depths up to about 10 lm, while the bulk properties
mainly remain unchanged .
materials are exposed to diluted fluorine gas mixture,
which reacts with the polymer surface via a free-radical
chain reaction mechanism (see Eqs. 1 and 2), resulting in
the formation of a very thin, partially fluorinated polymer
surface layer .
R ? H þ F2! R?þ HF þ F?
R?þ F2! R ? F þ F?
The thickness of the fluorinated layer is dependent on
fluorine gas diffusion through the sample , which is
controlled by the reaction variables, such as polymer type
and structure, fluorine partial pressure, fluorine concentra-
tion, reaction time and reaction temperature. The impor-
tance of diluting fluorine gas with an oxygen-free inert gas
(such as nitrogen and helium) to prevent fragmentation of
the carbon backbone has been reported in the literature
[7–9]. It is however, still believed by many research groups
that oxidation always accompanies fluorination, since
commercial fluorine is known to contain oxygen as an
impurity [10, 11].
During oxyfluorination, hydrocarbon polymers are
simultaneously fluorinated and functionalised by signifi-
cant amounts of oxygen (see Eq. 3) in either in the fluo-
rinating reaction mixture or by the use of oxygen directly
as a reactive diluting gas .
R?+ O2! R ? O ? O?
The existences of acid fluoride groups and peroxides as
well as crosslinking of the surface during oxyfluorination
have been reported [7, 12]. The ratio of the fluorocarbon
radicals to hydrocarbon radicals is higher during fluorina-
tion than in oxyfluorination, with the former having been
preferred because of its longer life-time. The rate of the
fluorination reaction in the presence of oxygen is much
slower than that of a fluorination reaction in which nitrogen
is used as the diluting gas. This is to be expected since
oxygen, which is very reactive towards radicals, inhibits
the reaction by reacting with the radicals that are formed
during fluorination (Eq. 3) to form peroxy radicals which
are much less reactive than is the R*radical. Jeong et al.
utilised the peroxy radicals generated by oxyfluorination to
assist in graft polymerisation for modifying the surfaces of
low-density polyethylene (LDPE) films .
The possibility of etching the PP surface during oxy-
fluorination has also been suggested . The effects of
oxygen/fluorine gas mixtures on a PP surface include
improved wettability, a significant increase in polarity and
a roughening of the surface resulting from the exothermic
nature of the reaction [2, 14]. Sanderson et al. have found
that PP had a higher fluorination rate than polyethylene
Direct fluorination may also result in the formation of
peroxy radicals because of the presence of some per cent
oxygen in nitrogen as diluting gas and the trace of oxygen
in commercial fluorine gas. There remains uncertainty,
however, about the mechanism by which oxygen is intro-
duced into the surfaces of fluorinated polymers and about
the nature of the functionalisation. Kharitonow and co-
workers [15–18] have confirmed the formation and
termination of long-lived peroxy radicals and short-lived
radicals by the ‘direct fluorination’ of various polymers.
However, according to Sanderson and co-workers, no
obtained for PP samples fluorinated by 10:90 F2:N2gas
mixture, and even after long fluorination times, function-
alisation was limited to the formation of only –CHF and
In this study, oxyfluorination of NWFs based on PP have
been investigated as a pre-treatment step to impregnate
reactive functional groups on the surface of the PP for post-
graft polymerisation. The aim of this study is to modify the
surface of five PP open-porous NWFs with different mean
flow pores and densities by an oxyfluorination pre-treat-
ment method. We are reporting on five samples to see the
reproducibility of the oxyfluorination treatment, as well as
to see the effect of pore size and NWF density on the extent
of fluorination, since the more porous structure should
enable better diffusion and penetration of the gas into the
scaffold. The chemical and physical characterisations of
the oxyfluorinated NWFs compared to virgin PP are the
subject of this article. Characterisation techniques include
attenuated total reflectance fourier transform infrared
(ATR-FTIR)-spectroscopy, FTIR imaging microscopy,
X-ray photoelectron spectroscopy (XPS), electron spin
resonance (ESR) spectroscopy, differential scanning calo-
rimetry (DSC), X-ray diffraction (XRD) analysis, scanning
electron microscopy (SEM), dynamic rheometry and
thermo-gravimetry (TG). The
1020 V. Vargha et al.
Author's personal copy
prepared in this study have been used in a subsequent study
for the graft.
PP non-woven fabrics (NWFs) have been prepared in-
house at the CSIR Materials Science and Manufacturing in
Port Elizabeth by a needle-punching technology . The
characteristics of the non-woven samples are summarised
in Tables 1 and 2.
The PP NWFs were oxyfluorinated at Pelchem Pty Ltd.
(South Africa) by a proprietary method whereby the NWFs
are loaded into a reaction vessel (RV). The RV is purged
and partially evacuated, and then a 20/80 vol/vol F2:N2gas
mixture is introduced for a certain time-period. The gas
mixture is then cycle purged from the RV to complete the
fluorination process. Owing to the partial vacuum, and the
presence of some air in the reactor, and with the 4–5 vol/%
oxygen impurity present in nitrogen, the process can be
referred to as oxyfluorination.
Methods of characterisation
Attenuated total reflectance infrared (ATR-FTIR)
Infrared spectra were collected by a Varian Scimitar 2000
FTIR spectrometer using MCT (mercury–cadmium–tellur)
detector equipped with a single reflection ATR unit
(SPECAC ‘Golden Gate’) with diamond ATR element
(active surface: 0.6 9 0.6 mm2). All spectra were recorded
by co-addition of 128 individual spectra with 4 cm-1
Homogeneity/inhomogeneity had been studied by creating
chemical images of selected surface groups/species. FTIR
images were collected on individual oxyfluorinated fibres
by means of a Varian FTS-7000 spectrometer coupled to a
microscope configuration (UMA-600) with an FPA (Focal
Plane Array) multidetector (Stingray) system consisting of
a 64 9 64 MCT detector units. For the measurements, the
reflexion technique was selected, using 8 cm-1resolution
and 16 scans. To each detector unit (pixel) corresponds on
average, *5.5 lm lateral resolution.
X-ray photoelectron spectroscopy (XPS)
XPS spectra were taken by a Kratos XSAM800 spec-
trometer, using Mg Ka radiation, 225 W X-ray power, and
fixed analyser transmission (FAT) mode. The survey and
detailed spectra were recorded with 80 and 40 eV pass
energies, respectively. The survey spectra were taken in the
kinetic energy range of 100–1,300 eV with a resolution of
0.5 eV, while the resolution of the detailed spectra of the
selected ranges was 0.1 eV. Four sweeps were applied to
improve the signal-to-background ratio. The spectra were
referenced to the C1s line (binding energy, BE = 285 eV).
Data acquisition and processing were performed by the
Kratos Vision 2 program.
Table 1 PP non-woven fabrics (NWFs) used for the experiments
NWFPreparation Depth of penetration/mm
Fibre linear density
pPP-1 Needle-punched6 6.7 125
pPP-2 Needle-punched10 2.2150
pPP-4 Needle-punched ? calenderedn.a. 2.2 n.a.
pPP-5 Needle-punched 106.7 260
NWF pPP refers to pure (virgin or untreated) polypropylene
Table 2 Pore size of PP non-woven fabrics (NWFs) used for the
experiments determined by PMI Capillary Flow Porometer according
to Test method 6212005-134 corresponding to ASTM E 1294
NWF Minimum/lm MFPa/lmMaximum/lm
pPP-17 97 384
aThe mean flow pore (MFP) diameter is such that fifty percent of
flow is through pores larger than MFP diameter and the rest of the
flow is through smaller pores. The mean flow pore diameter is a
measure of permeability
Polypropylene non-woven fabrics (NWFs) 1021
Author's personal copy
Electron spin resonance (ESR) spectroscopy
ESR spectra were obtained on a Bruker Elexsys 500 spec-
trometer operating at X-band frequencies (*9–10 GHz).
Spectra were recorded as first derivatives of microwave
absorption of a total 10–20 scans, at ambient tempera-
ture, using 2 mW microwave power, modulation fre-
quency = 100 kHz, modulation amplitude = 1 G.
Scanning electron microscopy (SEM)
Surface morphology was monitored with a JEOL JSM
6380 LA scanning electron microscope. The samples were
coated with gold to make imaging possible. For attaining
the best image quality, the acceleration voltage was varied
between 5 and 15 kV. The detection modes are indicated in
the bottom-right corner of each image.
Differential scanning calorimetry (DSC)
For DSC measurements a Perkin Elmer DSC 7 was used.
About 5 mg of the sample was heated from -60 to
?250 ?C with a heating rate of 10 ?C min-1followed by a
cooling step and a second heating both with the same
cooling and heating rate. Stream of purging nitrogen gas
was 40 mL min-1. The course of measurements is repre-
sented by Fig. 1.
Wide-angle X-ray scattering (WAXS)
WAXS scans were obtained in a Philips model PW 3710
based PW 1050 Bragg–Brentano parafocusing goniometer
using CuKa radiation (k = 0.15418 nm), graphite mono-
chromator and proportional counter.
For rheological measurements an Anton Paar MCR 301
dynamic rheometer was used. Change in complex viscosity
with frequency was tested in the frequency range of
600–0.1 s-1with a plate–plate arrangement and a slit
distance of 1 mm at 180 ± 0.2 ?C. The calculated shear
rate interval is 29.2–0.005 s-1.
Thermogravimetric analysis (TG)
Thermal behaviour and decomposition of PP and oxyflu-
orinated PP NWFs were tested with a Perkin Elmer TG 6.
Approx. 10 mg sample was heated from 30 to 700 ?C in
purging nitrogen gas with a temperature rate of 10 ?C/min.
Thermo-oxidative degradation was tested in purging
Results and discussion
Chemical composition of the surface of NWFs detected
by ATR-FTIR spectroscopy and FTIR imaging
The results of ATR-FTIR spectroscopy for untreated and
oxyfluorinated PP non-wovens (pPP4 and oPP4) are rep-
resented in Fig. 2.
Fluorination resulted in the appearance of the absorption
bands of newly formed groups compared to untreated PP.
The broad band in the range of 1,000–1,300 cm-1can be
assigned to –CF and –CF2groups. The bands maxima at
about 1,191 and 1,079 cm-1correspond to the symmetric
stretching vibrations of the –CF2 group and the C–F
stretching vibration in –CF, –CHF, and –C(O)F groups,
respectively [20, 21]. The absorption bands in the range of
1,900–1,600 cm-1refer to the formed C=O groups after
1,727 cm-1can be assigned to the C=O stretching vibra-
tion of the formed –COOH groups. The formation of the
latter may be the result of hydrolysis of the –COF groups
on the effect of moisture. As a result of hydrolysis a broad
absorption band can be detected in the fluorinated samples
in the vicinity of 3,346 cm-1characteristic to the associ-
ated –OH groups. The weak band at 1,850 cm-1can be
attributed to carbonyl vibration in the unhydrolysed
–C(O)F groups. At 1,780 cm-1the C=O stretching vibra-
tion of alpha-haloester, at 1,627 cm-1the C=C stretching
of the formed –CF=C(OH)– groups could be detected .
Owing to the penetration depth of the ATR-FTIR method
(around 10 lm), reacted products included into the oxy-
fluorinated layer are also observed.
The integrated area under the C=O absorption bands of
COOH groups, as well as of F-containing groups are
90 807060 50
10 °C/min 10 °C/min
Fig. 1 Program of DSC measurements
1022V. Vargha et al.
Author's personal copy
relatively higher in oPP-4 than in oPP-5. This is supported
also by XPS measurements.
The IR images of the oxyfluorinated samples are shown
in Fig. 3.
Taking into consideration the geometry of the fibre
sample, we can assume that the fluorine-containing species
are distributed rather inhomogeneously penetrating deeply
into the fibre interior, and the oxygen-containing species
resulted as a consequence of hydrolysis are on the fibre
Surface analysis by XPS
The surface compositions obtained by XPS for the five
oxyfluorinated samples are summarised in Table 3.
Pore size did not have any effect on the degree of
functionalisation. Since fluorination is a gas surface treat-
ment, it is assumed that a more porous NWFs will enable
better diffusion of the gas into the structure. As a result of
oxyfluorination, significant amounts of fluorine and oxygen
could be detected on the surface of the samples. The con-
centration of fluorine atoms was found to be in the range of
21.5–31 at% (28–39 mass%) and that of oxygen atoms
between 20 and 23 at% (22–25 mass/%). Figure 4 shows
typical F 1s, O 1s and C 1s spectra for the oxyfluorinated PP
A.u. = 0.05
1600 14001200 1000800
Fig. 2 ATR-FTIR spectra of
NWF before (pPP-4) and after
Difference spectrum is shown in
νC = O
Row = 7 Col = 40
1700 1600 1500 1400 1300 1200 1100 1000
Fig. 3 Optical (a) and chemical
(c, d, e) images of the
oxyfluorinated samples focusing
on a single fibre (oPP-4). The
chemical images are constructed
from areas under selected band
intensities reflecting surface
distribution of characteristic
surface species. b is a
characteristic spectrum detected
by a selected detector element
(pixel) corresponding to a
sample area around
5.5 9 5.5 lm2collected in
Table 3 Surface composition of the oxyfluorinated NWF samples as
revealed by XPS
SampleF/at%F/mass%O/at% O/mass%C/at% C/mass%
oPP-121.5 28.322.7 25.1 55.846.5
oPP-226.7 34.6 20.622.452.7 43.1
oPP-330.738.9 20.622.0 48.739.1
oPP-427.4 35.023.225.0 49.440.0
oPP-526.5 34.419.921.7 53.6 44.0
Polypropylene non-woven fabrics (NWFs) 1023
Author's personal copy
samples. The F 1s peak is symmetric, centred at 687.0 eV,
testifying to the presence of C–F bonds . The O 1s peak
is asymmetric, suggesting the presence of O in bonding
modes like C–O at 533.2 eV and C=O at 531.6 eV .
The C 1s peak can be resolved into several components.
Neglecting the minor secondary shifts, the following
assignments can be made: C–C and C–H at 285 eV, C–O at
296.4 eV, C=O, O–C–O, and C–F states at 287.9 eV and
O=C–O at about 289.4 eV .
In case of direct static fluorination hydrogen atoms are
substituted by fluorine, double and conjugated bonds are
saturated with fluorine. Crosslinking (formation of C–C
bonds) and destruction of C–C bonds may also occur. In
oxyfluorination, i.e. treatment of polymeric materials with
COOH groups can be inserted onto the polymer chains
Both ATR-FTIR spectroscopy and XPS analysis support
the presence of oxygen-containing groups on the surface of
our functionalised NWF samples. Despite the applied
process of fluorination, meaning that a mixture of 20/80
vol/vol fluorine/nitrogen gas was used, the surface of our
treated samples is abundant with oxygen-containing
groups. Oxygen was present during the fluorination process
since the reactor was only partially evacuated. In addition,
some oxygen may have been entrapped in the loose
structure of NWFs, on the surface of the fluorinating ves-
sel, and it may also have been present as an impurity (cca
4–5 vol/%) in the nitrogen gas. Another noteworthy
observation is the fact, that –C(O)F groups could still be
detected on all our treated surfaces. These groups should
have hydrolysed on long time exposure to atmospheric
conditions. Sanderson et.al. investigated the kinetics of
hydrolysis of the –C(O)F groups. They have found that
the hydrolysis of oxyfluorinated PP was initially fast, but
levelled off after about 4 h. After 40 h of exposure to
atmospheric conditions, some unhydrolysed groups still
remained. In comparison with other surface modification
techniques, fluorine atoms penetrate the surface of PP to
F 1s:1(0F–PP–4) 0 1s:2(0F–PP–4)C 1s:3(0F–PP–4)
Intensity/CPS Intensity/CPS Intensity/CPS
Binding energy/eV Binding energy/eV
690 688 686 684 682536 534 532 530 528
292 290 288286 284 282 280
Fig. 4 Typical F 1s, O 1s, and
C 1s peaks of oxyfluorinated
non-woven PP fibres
Fig. 5 Experimental and simulated ESR spectra of oxyfluorinated
sample 2 (oPP-2)
100020003000 4000 5000
Fig. 6 The change in radical concentration with time at 70 ?C (the
temperature of grafting)
1024 V. Vargha et al.
Author's personal copy
relatively great depths, the extent of which depends upon
treatment conditions. The plateau in the hydrolysis curve
may indicate the inability of atmospheric moisture to reach
the unreacted groups deep in the polymer sample, espe-
cially since the barrier properties of the fluorinated and
oxyfluorinated layers are well known .
Virgin samples Oxyfluorinated samples
Fig. 7 SEM micrographs of
virgin and oxyfluorinated PP
Polypropylene non-woven fabrics (NWFs)1025
Author's personal copy
Detection of long-lived radicals on the surface
of oxyfluorinated NWF by ESR-spectroscopy
Weak asymmetric singlet absorption signals were observed
for all of the oxyfluorinated NWFs, indicating the presence
of long-lived free radicals in the polymer structure. The
structureless asymmetric line shape can be interpreted in
several different ways. The best fit was obtained in case of
simple g anisotropy g\=2.0022, gk= 2.0054. These values
correspond to midchain peroxy radical . Experimental
and simulated spectra of oxyfluorinated sample 2 (oPP-2) is
represented by Fig. 5.
In polymers treated with fluorine–oxygen mixtures, a
controlled amount of long-lived peroxy radicals are gen-
erated . Peroxy radicals terminate faster than fluoro-
radicals . The radicals formed in fluorinated polymers
are long-lived ones; their amount is decreased by a factor
of 2 in several hours at room temperature—from 1 to 15 h
depending on the polymer nature .
Table 4 Glass transition temperatures of virgin and oxyfluorinated
oPP-4 n.d.5.3 n.d.
pPP-5 n.d.-5.8 n.d.
oPP-5n.d. 0.1 n.d.
n.d. not detectable
Average fibre diameter/μm
Fig. 8 The average diameter of the fibres of pPP and oPP NWF
First heating of pPP-4
Fig. 9 The splitting of the enthalpy peak of pPP-4 during the first
DSC heating step
0 50 100
–500 50 100
150 200 250
Fig. 10 Results of DSC analysis of virgin and oxyfluorinated NWF
sample 2 (a 1st cooling, b 2nd heating)
1026 V. Vargha et al.
Author's personal copy
Although the oxyfluorinated NWFs were tested several
weeks after the oxyfluorination process, long-lived radicals
could still be detected on the surfaces of the NWFs
as midchain peroxy radicals. This may indicate that
the radicals were entrapped in the NWFs during the
oxyfluorination process, and slowly diffuse to the surface
with time. Also long-lived peroxy radicals will be very
useful for inducing graft polymerisation for attachment of
poly (N-isopropylacryalmide) onto the NWFs.
Further measurements have been carried out at 70 ?C, to
follow the change in radical concentration with time at the
temperature of grafting (Fig. 6). The concentration scale is
not calibrated, although the samples are normalised to
1 mg, so that they can be compared. The absolute con-
centrations are very small, and the signal/noise ratio is *5,
which is rather low. The background noise is rather high.
On sample oPP-2 was the easiest to measure the time-
dependence of radical concentration at 70 ?C. It can be
seen that after some increase, the concentration of the
radicals does not change. Samples oPP-3 and oPP-5 did not
show an increase in radical concentration with time at
Table 5 Data of crystallisation of PP-NWFs from 1st DSC cooling
pPP-2 119.0 113.3108.3-98.2
pPP-3 118.1 113.5109.3-92.4
oPP-5 126.1119.0 113.4-93.1
Table 6 Melting characteristics of the crystalline phase of PP-NWFs
from 2nd DSC heating scan
oPP-1169.3165.0 98.6 67
pPP-2 160.5 168.197.9 66
oPP-2162.9168.8 86.0 58
pPP-3 163.2170.2121.6 82
oPP-3 166.5170.5 91.7 62
pPP-4 163.9168.5 98.266
Fig. 11 Typical WAXS
diffractogram of NWFs pPP-4
1 10 100
Fig. 12 The change in complex viscosity of virgin and oxyfluori-
nated sample with frequency at 180 ± 0.2 ?C
Polypropylene non-woven fabrics (NWFs) 1027
Author's personal copy
Change of morphology detected by SEM
The micrographs on the virgin and oxyfluorinated PP NWF
fibres detected by SEM are represented by Fig. 7. A slight
roughening can be detected on the surface of the fibres as a
result of surface treatment. This coincides with the findings
of other researchers, who have also reported a significant
increase in roughening of the surface resulting from the
exothermic nature of the oxyfluorination reaction [2, 14].
The possibility of etching of the PP surface during oxy-
fluorination has also been suggested .
From the SEM micrographs the average diameters of the
virgin and oxyfluorinated fibres was determined. The
results are represented by Fig. 8. As expected the 6.6 dtex
linear density fibres had a larger fibre diameter compared to
the lower 2.2 dtex linear density fibres. It was also
observed that the diameter of the oxyfluorinated fibres is
higher, than the untreated ones as a result of surface
Effect of oxyfluorination on the crystallinity of NWF
The relaxational and thermal phase transitions of the virgin
and oxyfluorinated PP NWF samples were determined by
DSC measurement. The Tgwas determined with the Perkin
Elmer Pyris sotware. Tgwas given as half Cp extrapolated.
The glass transition of the samples could not always be
detected by DSC analysis because of the small enthalpy
change (Table 4). It has been found that the Tgwas between
-5.8 and ?5.4 ?C. It is difficult to draw any conclusion on
the effect of oxyfluorination on the glass transition tem-
perature of PP NWFs from DSC measurement.
Crystallisation and melting of the crystalline phase were
investigated in the temperature range of 100–200 ?C.
During the first heating step of all the samples the melting
of the crystalline phase was detected by the appearance of
split enthalpy peaks for all the samples as demonstrated by
Fig. 9. The reason of the noisy pattern is the contraction of
highly extended material therefore its contact to the pan
Table 7 Onset and end temperatures of thermal decomposition of oxyfluorinated PP NWF samples (oPPs) and the corresponding masses of
oxyfluorinated (oPPs) and untreated, virgin PP (pPPs) determined from the TG curves
400 500 600 700
First derivative/% min–1
Fig. 13 TG and derivative TG thermograms of virgin and oxyfluo-
rinated PP NWF sample 1 in purging nitrogen
1028 V. Vargha et al.
Author's personal copy
changes. However, the first heating run could give real
information about the changes in the properties. More
appropriate experimental technique would be necessary to
get real data from the first heating run, such as temperature-
modulated DSC . Thermal prehistory was eliminated
by the first heating step, and crystallisation and melting
were investigated during the cooling and second heating
step of DSC measurements. After the melting of the fibres,
the system is not homogeneous, the fluorinated part can be
regarded as separated from the main phase. Purpose of this
measurement was the calculation of crystalline portion
before and after oxyfluorination. The results are repre-
sented on NWF sample 2 by Fig. 10, and the evaluation of
the data on crystallisation and melting are summarised in
Tables 5 and 6.
The oxyfluorinated NWFs crystallise at somewhat
higher temperatures than the untreated NWFs (Table 5).
The endotherm peak on the DSC curve of the second
heating represents the melting of the crystalline phase of
the PP samples (Table 6). The asymmetry of the peaks may
be due to the formation of more stable crystalline forms
during the heating procedure. This phenomenon had been
reported previously for isotactic PP [26, 27]. Considering
the melting characteristics of the crystalline phase of NWF
samples from Table 6, the melting peak temperature of the
oxyfluorinated PP samples is somewhat higher than that of
the untreated ones. There is no significant difference in the
end temperature of melting of the untreated and function-
alised PP samples. The enthalpy of melting is somewhat
lower for the oxyfluorinated PP non-wovens compared
with the virgin, untreated ones, referring to slight decrease
in the ratio of the crystalline phase in relation to the
amorphous after oxyfluorination (Table 6). In case of NWF
sample 3, there is a significant (20%) decrease in the
crystalline phase after oxyfluorination, and the elucidation
of this needs further investigation. The crystalline portion
was calculated according to Monasse and Haudin .
The effect of oxyfluorination on crystallinity has also
been investigated by WAXS. The WAXS diffractograms
for NWF sample 4 are shown by Fig. 11, from which the
Table 8 Characteristic data of thermal decomposition of oxyfluorinated PP NWF samples (oPPs) and the corresponding untreated, virgin PP
(pPPs) determined from the first derivative of the TG curves
1 Onset 129498.8 98.0 283 98.298.9
Shoulder––– 436 48.254.2
Onset 2––– 432 52.9 60.5
Peak 44139.940.3 453 19.7 9.7
End 4680.0 0.0474 0.00.0
2 Onset133898.695.5 29497.4 99.3
Shoulder––– 42558.7 61.0
Onset 2––– 433 50.154.7
Peak 47628.5 25.845222.215.3
End445 0.01.24761.2 0.0
3 Onset 1316 97.396.5 30097.197.5
Shoulder––– 42349.3 75.8
Onset 2––– 417 56.479.7
Peak 45233.43.0 442 19.9 53.4
End481 0.0 0.0463 0.010.2
4 Onset 1 29999.297.1281 97.4 99.3
Shoulder––– 456 50.369.2
Onset 2––– 43252.072.6
Peak 44836.032.943451.1 15.2
End4790.00.0 480 0.0 0.0
5 Onset 1 347 97.495.4305 97.697.9
Shoulder––– 431 51.072.8
Onset 2––– 41768.1 85.7
Peak431 35.127.5 453 19.5 23.2
End 4760.0 0.2 474 0.20.0
Polypropylene non-woven fabrics (NWFs) 1029
Author's personal copy
differences of the absolute intensities can be read. Owing
to sample geometry neither the absolute nor the relative
intensities can be compared. The diffraction peaks how-
ever, for the treated and untreated NWFs were the same
indicating that the oxyfluorination treatment had no effect
on the crystalline structure or crystalline phases in PP. The
ratio of amorphous and crystalline phases has been calcu-
lated by profile fitting. The fitting of the amorphous
quantity is not better than ±5%. Differences may arise
from the deviations of sample thickness, sample size etc.
WAXS measurements support that oxyfluorination does
not have significant effect on the crystallinity and phase
composition of PP NWFs.
Based on the results of both of DSC and XRD mea-
surements, we can state that oxyfluorination had no sig-
nificanteffect on the crystallinity
amorphous phase compositions of PP NWF samples.
Effect of oxyfluorination on the molecular mass of PP
The effect of oxyfluorination on the change of molecular
mass of PP NWFs was determined using dynamic rheom-
etry. The results are shown by Fig. 12, indicating that the
melt viscosity of PP is unaffected by the oxyfluorination
treatment, implying that the molecular mass of the PP
NWFs does not significantly change after oxyfluorination.
Thermal behaviour detected by TG
The effects of oxyfluorination on the thermal and thermo-
oxidative stabilities and decompositions of PP NWFs were
studied by TG analysis. The curves in purging nitrogen
refer to the thermal stability of the samples and are rep-
resented on NWF sample 1 by Fig. 13. There are two
important differences in the thermal decompositions of
virgin and oxyfluorinated PP samples. The onset of thermal
decomposition of the oxyfluorinated PP samples is lower
than that of the untreated ones, and the mass loss is higher
than that in case of pure PP because of higher moisture
absorption and partial decomposition of functional groups
Table 9 Onset and end temperatures of thermo-oxidative decomposition of oxyfluorinated PP NWF samples (oPPs) and the corresponding
masses of oxyfluorinated (oPPs) and untreated, virgin PP (pPPs) determined from the TG curves
1 Onset 22799.5 97.522797.8 99.0
End 361 1.924.9 3740.5 1.6
2 Onset 22399.9 98.522498.899.5
End3563.7 34.4369 5.03.4
3 Onset22699.8 0.0 24099.6 97.4
End 3692.6 35.4 3884.6 2.1
4 Onset218 99.899.223099.0 98.3
End362 3.332.6 3822.6 3.0
5 Onset 23399.9 98.223298.799.2
End 360 5.5 24.7 372 4.15.3
150 250350 200 300
First derivative/% min–1
Fig. 14 TG and derivative TG thermograms of virgin and oxyfluo-
rinated PP NWF sample 1 in purging oxygen
1030 V. Vargha et al.
Author's personal copy
on the surface. Owing to the presence of functional groups
on the surface of oxyfluorinated samples, the mechanism of
thermal decomposition is also different from that of the
virgin PP non-woven ones. As the derivative curves show,
oxyfluorinated NWFs decompose in two steps compared to
the pure PP NWFs which decompose in one step. The
evaluation of the curves and derivative curves detected in
purging nitrogen are summarised in Tables 7 and 8.
The TG and derivative TG curves of the untreated and
oxyfluorinated NWF sample 1 in purging oxygen are
shown in Fig. 14. The evaluations of the corresponding TG
and derivative TG curves of all the NWFs are summarised
in Tables 9 and 10.
The thermo-oxidative decompositions of the untreated
and oxyfluorinated PP NWF samples are very similar. The
onset of thermo-oxidative decomposition is almost the
same, namely in the range of 218–240 ?C. In the case of
samples 3 and 4, the thermo-oxidative decompositions of
the oxyfluorinated NWFs take place at higher temperature,
than for the untreated PP ones. Thermo-oxidative decom-
position takes place in two steps for each sample, the first
steps are roughly in the same temperature range beginning
with the onset and ending in the vicinity of 300 ?C. The
second step, however, is at much higher range of temper-
ature for the oxyfluorinated samples.
Surface functionalisation of PP NWF samples of different
morphologies and pore sizes has been carried out by
oxyfluorination. The functionalised surfaces have been
characterised for functional groups by ATR-FTIR spec-
troscopy, XPS analysis and ESR-spectroscopy. ATR-FTIR
and XPS techniques revealed the presence of –CF, –CF2,
–CHF and –C(O)F groups. The formed –C(O)F groups
mostly get hydrolysed to –COOH groups. The C=O
stretching vibration of alpha-haloester, and the C=C
Table 10 Characteristic data of thermo-oxidative decomposition of oxyfluorinated PP NWF samples (oPPs) and the corresponding untreated,
virgin PP (pPPs) determined from the 1st derivative of the TG curves
oPP mass/% Temperatue/
1 Onset 121199.798.1 21798.1 99.5
Onset 2––– 324 58.5 67.1
Peak35611.7 31.6 3725.0 1.8
End 3671.9 16.9375 0.5 1.7
2 Onset 1 210100.099.0 20899.0 100.0
Shoulder-–– 27784.2 87.5
Onset 2––– 30871.2 71.0
Peak34520.649.4 36615.6 3.5
End360 3.829.0 370 5.13.4
3 Onset 120899.9 99.8 23299.898.6
Shoulder-–– 296 79.3 79.5
Onset 2––– 325 65.859.4
Peak35418.1 49.0 38511.4 2.3
End 370 2.733.1 389 4.92.2
4 Onset 120899.999.2 219 99.299.5
Peak357 12.238.7379 8.5 2.9
End3653.430.2 382 2.9 2.9
5 Onset 1 229 99.498.5227 98.799.6
Shoulder 299 64.0 73.3–––
Onset 2321 43.461.0–––
Peak343 19.144.7370 7.0 5.3
End358 6.027.0 373 5.2 4.2
Polypropylene non-woven fabrics (NWFs)1031
Author's personal copy
stretching of the formed –CF=C(OH)– groups could also be Download full-text
detected. Long-lived radicals could be detected on the
functionalised surfaces, as middle-chain peroxy radicals by
ESR-spectroscopy. Three samples (oPP-1, oPP-2 and oPP-
4 showed a slight increase in radical concentration with
time at 70 ?C (temperature of grafting). SEM micrographs
revealed slight roughening of the oxyfluorinated surfaces.
Oxyfluorination had no significant effect on the crystalline
structure and crystalline/amorphous phase composition of
the PP NWF samples supported by DSC and XRD mea-
surements. Oxyfluorinaton did not significantly effect the
molecular mass of PP as confirmed by dynamic rheometry.
Surface modification, however, significantly affected the
thermal decomposition and mechanism of PP NWFs.
Thermo-oxidative decomposition of untreated and oxyflu-
orinated PP NWFs were very similar. Morphology and
pore size of PP NWF samples had no significant effect on
Technology Foundation ZA-9/2006, the National Science Foundation
(South Africa), and the CSIR (South Africa) for financial support The
authors are also grateful to Pelchem Pty Ltd (in South Africa) for the
oxyfluroination treatments on all the NWFs. The authors thank also
Ja ´nos Kova ´cs for dynamic rheological assistance, and Jo ´zsef Ha ´ri for
The authors thank the Hungarian Science and
1. du Toit FJ, Sanderson RD, Engelbrecht WJ, Wagener JB. The
effect of surface fluorination on the wettability of high density
polyethylene. J Fluor Chem. 1995;74(1):43–8.
2. du Toit FJ, Sanderson RD. Surface fluorination of polypropylene:
1. characterisation of surface properties. J Fluor Chem. 1999;
3. du Toit FJ, Sanderson RD. Surface fluorination of polypropylene:
2. adhesion properties. J Fluor Chem. 1999;98(2):115–9.
4. Mohr JM, Paul DR, Tam Y, Mlsna TE, Lagow RJ. Surface
fluorination of composite membranes. Part II. Characterization of
the fluorinated layer. J Membr Sci. 1991;55(1–2):149–71.
5. Brass I, Brewis DM, Sutherland I, Wiktorowicz R. The effect of
fluorination on adhesion to polyethylene. Int J Adhes Adhes.
6. Park SJ, Song SY, Shin JS, Rhee JM. Effect of surface oxyflu-
orination on the dyeability of polyethylene film. J Colloid Inter-
face Sci. 2005;283(1):190–5.
7. Sanderson R, du Toit F, Carstens P, Wagener J. Fluorination rates
of polyolefins as a function of structure and gas atmosphere.
J Therm Anal Calorim. 1994;41(2):563–81.
8. Lagow RJ, Margrave JL. Direct fluorination: a ‘‘new’’ approach
to fluorine chemistry. progress in inorganic chemistry. Hoboken:
John Wiley & Sons, Inc; 2007.
9. Shimada J, Hoshino M. Surface fluorination of transparent
polymer film. J Appl Polym Sci. 1975;19(5):1439–48.
10. Hayes LJ. Surface energy of fluorinated surfaces. J Fluor Chem.
11. Leroux JD, Paul DR, Arendt MF, Yuan Y, Cabasso I. Surface
fluorination of poly(phenylene oxide) composite membranes. 2.
Characterization of the fluorinated layer. J Membr Sci. 1994;
12. Volkmann T, Widdecke H. Oxifluorination of polyethylene.
Kunstst Ger Plast. 1989;79(8):743–4.
13. Jeong E, Bae T-S, Yun S-M, Woo S-W, Lee Y-S. Surface
characteristics of low-density polyethylene films modified by
oxyfluorination-assisted graft polymerization. Colloids Surf A.
14. du Toit FJ. Surface modification of polymers using elemental
flourine. Ph.D. Thesis, University of Stellenbosch; 1995.
15. Kharitonov AP. Practical applications of the direct fluorination of
polymers. J Fluor Chem. 2000;103(2):123–7.
16. Kharitonov AP, Taege R, Ferrier G, Teplyakov VV, Syrtsova
DA, Koops G-H. Direct fluorination—useful tool to enhance
commercial properties of polymer articles. J Fluor Chem.
17. Tressaud A, Durand E, Labruge `re C, Kharitonov AP, Kharito-
nova LN. Modification of surface properties of carbon-based and
polymeric materials through fluorination routes: from funda-
mental research to industrial applications. J Fluor Chem. 2007;
18. Kharitonov AP, Kharitonova LN. Surface modification of poly-
mers by direct fluorination: a convenient approach to improve
commercial properties of polymeric articles. Pure Appl Chem.
19. Boguslavsky L. High efficiency particulate air (HEPA) Filters
from polyester and polypropylene -Nonwovens,. International
Conference FILTREX 2010; October; Cologne, Germany 2010.
20. Socrates G. Infrared and Raman characteristic group frequencies :
tables and charts. 3 ed. ed. Wiley: Chichester; 2001.
21. Kharitonov AP. Direct fluorination of polymers—from funda-
mental research to industrial applications. Prog Org Coat.
22. Mihaly J, Sterkel S, Ortner HM, Kocsis L, Hajba L, Furdyga E,
et al. FTIR and FT-Raman spectroscopic study on polymer based
high pressure digestion vessels. Croatica Chemica Acta. 2006;
23. Beamson G, Briggs D. High resolution XPS of organic polymers.
The scienta ESCA300 database. Chichester: Wiley; 1992.
24. Schlick S, Mcgarvey BR. Motion of midchain peroxy-radicals in
poly(tetrafluoroethylene). J Phys Chem. 1983;87(2):352–3.
25. Genovese A, Shanks R. Crystallization and melting of isotactic
polypropylene in response to temperature modulation. J Therm
Anal Calorim. 2004;75(1):233–48.
26. Horvath Z, Sajo IE, Stoll K, Menyhard A, Varga J. The effect of
molecular mass on the polymorphism and crystalline structure of
isotactic polypropylene. eXPRESS Polym Lett. 2010;4(2):
27. Nurul Huda M, Dragaun H, Bauer S, Muschik H, Skalicky P. A
study of the crystallinity index of polypropylene fibres. Colloid
Polym Sci. 1985;263(9):730–7.
28. Monasse B, Haudin JM. Growth transition and morphology
change in polypropylene. Colloid Polym Sci. 1985;263(10):
1032 V. Vargha et al.
Author's personal copy