Polymer Degradation and Stability 92 (2007) 2042e2050

www.elsevier.com/locate/polydegstab

Influence of water on the photodegradation of poly(ethylene oxide)

F. Hassouna, S. Morlat-Thérias*, G. Mailhot, J.L. Gardette
Laboratoire de Photochimie Moléculaire et Macromoléculaire (UMR CNRS-UBP 6505), Université Blaise Pascal, 63177 AUBIERE Cedex, France
Received 21 June 2007; accepted 30 July 2007
Available online 9 August 2007

Abstract

The UV-light degradation of polyethylene oxide (PEO) in aqueous solution was investigated operating under long wavelengths (l > 300 nm)
at 20  C in different pH conditions varying from 2.3 to 12.0 and at two different concentrations. Thermo-oxidation experiments on PEO aqueous
solution at 50  C are also reported and compared to photo-oxidation results. The formation of oxidation products was followed by infrared anal-
ysis of deposits obtained by evaporation of aliquots of irradiated polymer solution. Photo-oxidation led to formates and esters but a third product
was also identified, formic acid ions formed by partial hydrolysis of formates. The degradation of PEO in water led to the acidification of the
aquatic medium. Size exclusion chromatography (SEC) was used to monitor the changes in molar weight and intrinsic viscosity with irradiation
time. It was shown that the photo-oxidation produced a dramatic decrease of the average molar weights which is more important in acidic me-
dium. Total organic carbon (TOC) measurements of the aged aqueous solutions showed that the mineralization of PEO could not be achieved in
these photo-oxidative conditions.
Ó 2007 Elsevier Ltd. All rights reserved.

Keywords: Photodegradation; Water; Poly(ethylene oxide); SEC

1. Introduction different branches of industry, cosmetics, pharmacy and medi-

Water-soluble polymers are produced in large volumes
and are widely used in a variety of industrial and consumer
products. These polymers find use in polymeric surfactant,
pigment dispersants, mining oil drilling, detergents, etc. After
use, depending on their characteristics and particular applica-
tions, water-soluble polymers are discarded into solid waste
disposal systems or as dilute aqueous solutions and they
are potential contributors to environmental problems [1].
Their photochemical degradation in aqueous solution gives
insight of their impact on environment in outdoor conditions.
The knowledge of their degradation is then of prime impor-
tance to evaluate their fate and their long term behaviour.
Among polyethers, PEO with PVME are the only one that
are readily soluble in water [2]. PEO is biocompatible, biode-
gradable, non-ionic and a water-soluble polymer of consider-
able industrial significance. PEO finds applications in many
* Corresponding author. Tel.: þ33 (0)73 40 71 43; fax: þ33 (0)73 40 77 00.
E-mail address: sandrine.therias@univ-bpclermont.fr (S. Morlat-Thérias).
cine. For example it is typically used as a surfactant, lubricant,
textile size, hydraulic fluid, and water-soluble packaging film
[3e6].
Poly(ethylene oxide) (PEO) has a simple chemical structure.
It is built from flexible linear macrochains consisting of ele-
ments with different electronegativity, carbon and oxygen
(eCH2eCH2eOe). In the crystalline solid state, PEO exists
mainly in the form of a regular helix. The characterisation of
the spatial configuration of PEO shows an arrangement with
trans (OeCH2), gauche (CH2eCH2) and trans (OeCH2) con-
formations [7,8]. In aqueous solution PEO conformation shifts
to a random coil with statically placed residual helix segments
[2,4,9,10], the helix structure being conserved. According to
experimental and theoretical studies realised on waterePEO
systems [4,11e13], hydrogen bonding plays an important
role in the interactions between PEO and water. PEO in water
adopts the local conformations (eOeCeCeOe dihedral se-
quences) tgt, tgg, ttt,.etc. (with t ¼ trans and g ¼ gauche).
The conformations tgt and ttg are favoured under low concen-
tration and low temperature conditions whereas the population

0141-3910/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymdegradstab.2007.07.016
 F. Hassouna et al. / Polymer Degradation and Stability 92 (2007) 2042e2050
of the hydrophobic conformers increases with concentration
and temperature. These two conformers exhibit a substantial
hydrophilic character while the rest are essentially hydropho-
bic. Hydrogen bonds play an important role in the preferential
conformation tgt characterised by OeCeCeOe dihedral an-
gle of 85.4 .
This study of photo-oxidation of PEO is part of a general
study of water-soluble polymers carried out in our laboratory
[14,15]. The mechanism of photo-oxidation of PEO in the
solid state has already been studied under irradiation at long
wavelengths (l > 300 nm, 35  C) in presence of oxygen
[14]. This previous study has shown that PEO is very sensitive
to oxidation. Hydroperoxides (POOH) are formed as primary
products within a classical chain oxidation reaction that
involves hydrogen abstraction which occurs on methylene
groups in a-position of the oxygen atom. These primary hy-
droperoxides readily decompose to form macroalkoxy radi-
 
cals PO and hydroxyl radical OH. In photo-oxidative
conditions, the main route of decomposition of macroalkoxy
radicals is a b-scission, which generates formates in a 5/1
ratio against chain esters. In contrast, in conditions of
thermo-oxidation, the situation is different. The hydroxy rad-

icals OH formed by decomposition of the hydroperoxides
mainly react with the macroalkoxy radicals in a cage reaction,
producing esters. A same amount of esters and formates is
then observed.
A first approach of the photodegradation of PEO in aqueous
solution has been carried out [15]. Irradiation of PEO aqueous
solutions leads to the formation of macromolecular esters, for-
mate end groups and a third photoproduct was identified,
HCOO. The main difference observed between the photo-
products formed in photo-oxidation conditions either in solid
state or in aqueous solution concerns the formation of formic
acid ions (HCOO) released in solution. A strong influence of
the pH of aqueous solution was observed on the photochemi-
cal behaviour of PEO, which could not have been explained.
The chemical structure of the polymer cannot explain the in-
fluence of pH on the mechanism of oxidation, and the impor-
tance of radical or ionic reactions occurring in water was still
questionable.
The present paper is mainly focused on the influence of pH
of the aquatic medium on the mechanism of photo-oxidation
of PEO; a specific attention is being paid to the analysis of
photoproducts and to the characterisation of the rate of
photo-oxidation. Photo-oxidation experiments were carried
out in different pH conditions. The modifications of the chem-
ical structure of the polymer were followed by IR spectrom-
etry and by SEC analysis in order to characterise the
changes of molar weight distribution and the evolution of
the viscosity of the solution. The aqueous solutions were
also characterised by measuring the variations of pH during
irradiation. The influence of the pH on the stability of the
photoproducts was also studied. Moreover, an evaluation of
the possibility to achieve the mineralization of PEO in
photo-oxidation conditions was also carried out by measuring
the Total Organic Carbon (TOC) concentration of the irradi-
ated solution.
2043
2. Experimental
2.1. Materials
Poly(ethylene oxide) was obtained from Scientific Polymer
Products and used without further purification. Average molar
weight of the polymer was 100 000 g mol1. PEO solutions
of 5 wt.% and 0.5 wt.% (respectively, 50 gm mL1 and
5 mg mL1) were prepared by dissolving PEO into a distilled
and deionized water with mild stirring overnight at room tem-
perature. The pH value was measured with an ORION pH meter
to 0.1 pH unit. Dissolution of PEO in deionized ultrapure
water (at concentration of 50 mg mL1 or 5 mg mL1) results
in a solution with a pH value of 8.4. Solutions in acidic or basic
medium (pH ¼ 2.3 or 3.1 and 12.0) were, respectively, obtained
by adding drops of diluted HClO4 or NaOH (HClO4 was chosen
to acidify the aqueous solution because ClO 4 is photochemi-
cally stable contrary to Cl). No buffer was used in order to
avoid making the system too complex by adding too many
species.
2.2. Irradiation
Irradiations of PEO in solid state (as films) were carried out
in a SEPAP 14/24 unit. This apparatus has been designed for
the study of polymer photodegradation in artificial ageing cor-
responding to medium accelerated conditions [14,16].
Irradiation of PEO aqueous solutions was carried out in an
elliptical chamber which is equipped with one medium-
pressure mercury lamp (Mazda MA 400) in a vertical position
at one focal axis of the chamber [15]. Wavelengths below
300 nm are filtered by a glass envelope. The aqueous solution
of PEO was irradiated in a Pyrex reactor cooled by water circu-
lation. The temperature of the solution was maintained at 15  C
in order to avoid water evaporation. This reactor was placed at
the second focal axis of the chamber. The solutions were
continuously stirred with a magnetic stirrer and a Teflon bar.
Thermo-oxidation experiments were carried out at 50  C in
a 100-mL round bottom flask equipped with reflux condenser.
This temperature was chosen to keep the aqueous solution at
a temperature below the cloud point temperature
(CPT ¼ 100  C) [17,18]. The reactor was therefore placed
into an oil bath.
2.3. Chemical characterisations
PEO solutions of 5 wt.% concentration were chosen in order
to follow the chemical modifications of the irradiated solutions
by IR analysis of deposits on CaF2 windows. Thin deposits of
oxidized samples of PEO obtained by evaporation of aliquots
of irradiated aqueous solution of PEO were analysed in transmis-
sion mode before infrared analysis. The samples were dried at
room temperature for 24 h. In the manuscript, the figures of
FTIR spectra are limited to the range 4000e1000 cm1 due
to the opacity of the CaF2 plate between 1000 cm1 and
400 cm1. Infrared spectra were recorded with a Nicolet
FTIR 760 spectrometer, working with OMNIC software.
2044 F. Hassouna et al. / Polymer Degradation and Stability 92 (2007) 2042e2050
Spectra were obtained using 32 scan summations and 4 cm1
resolution. In order to take into account the variations of thick-
ness resulting from the mode of preparation of the samples,
a calibration of the photo-oxidized samples was performed at
1965 cm1, which corresponds to a combination band of
(COC) stretching modes and (CH2) rocking modes. The
thickness dependence of the absorbance obeyed the equation:
e ¼ absorbance at 1965 cm1  1=0:00405 [14].
Chemical treatments with SF4 and NH3 carried out to iden-
tify the photoproducts were performed following the proce-
dure previously described [14,19e21].
2.4. SEC
The changes of molar weight (Mw, Mn), molar weight
distribution (polydispersity index (Mw/Mn) and the intrinsic
viscosity (IV) were measured by size exclusion chromatogra-
phy (SEC) with double distilled deionized water as eluent at
a flow rate of 1 mL min1. A linear column was used for phys-
ical separation (TSK/Ref: GMPWXL). The refractive index
was monitored continuously with a differential refractometer
(Viscotek VE 3580) and the intrinsic viscosity with a viscom-
eter (Viscotek TriSec Model 270).
2.5. Ion chromatography
Ion chromatographic analyses were performed in a clean
room (class 10 000). Aqueous solutions were injected directly
without filtration to avoid further contamination. Working con-
ditions are similar to those detailed by Jaffrezo et al. [22]. A
DX 320 chromatograph equipped with an AS11 (4 mm) col-
umn for the anions was used in order to detect HCOO or
HCOOH in aqueous solutions after ageing.
2.6. Total organic carbon
Total Organic Carbon (TOC) measurements of aqueous
solution were obtained with a Total Organic Carbon (TOC)
analyzer Shimadzu model TOC-5050A. The Shimadzu
TOC-5050A is able to perform Total Carbon (TC) analyses
by combustion and Inorganic Carbon (IC) analyses by oxida-
tion. TOC measurements were obtained by calculating the
difference between TC and IC measurements.
3. Results and discussion
3.1. Photo-oxidation of PEO in water (concentration
50 mg mL1, pH ¼ 8.4)
3.1.1. IR analysis
The concentration of PEO aqueous solution was
50 mg mL1. As recalled above, this ‘‘high’’ concentration
was chosen because it permits obtaining well calibrated de-
posits around 16 mm that can be used to observe the formation
of photoproducts by infrared analysis. The pH of the solution
at this concentration was 8.4.
0.20
1270 h
1090 h
990 h
0.15
850 h
700 h
Absorbance
575 h
0.10
370 h
150 h
0h
0.05
0.00
1900 1800 1700 1600 1500
Wavenumber (cm-1)
Fig. 1. Infrared spectra deposits obtained from irradiated aqueous solutions of
PEO ([PEO] ¼ 50 mg mL1, pHinitial ¼ 8.4).
Fig. 1 shows the changes in the carbonyl region of spectra
obtained from aqueous solutions for several irradiation times
ranging from 0 h to 1272 h. Monitoring each spectrum neces-
sitated that a sample was prepared, following the procedure
described in the experimental part. In the carbonyl region,
three absorption bands develop at 1750 cm1 and 1720 cm1,
and at 1590 cm1. The two absorption bands at 1750 cm1
and 1720 cm1 have already been observed in the case of irra-
diation of PEO in solid state [14]. On the basis of chemical
derivatisation (NH3, SF4 and formic acid), these absorption
bands have been attributed to formate end groups
(1720 cm1) and macromolecular ester groups (1750 cm1).
The intense band that develops at 1590 cm1 has also been ob-
served and attributed to formic acid ions [15] (Scheme 1). It is
recalled that this band is not observed in the case of the photo-
oxidation of PEO in the solid state (film).
If one considers the same absorption coefficients for the
two C]O stretches at 1750 cm1 and 1720 cm1, it can be
deduced that the photo-oxidation of PEO in aqueous solution
results in the formation of four formate groups to one ester
function (ratio: 4/1).
3.1.2. Evolution of pH
The variations of the pH of PEO aqueous solution during
irradiation are reported in Fig. 2. For a PEO concentration
of 50 mg mL1, the initial pH value of the water solution is
8.4. Irradiation of this aqueous solution of PEO leads to a dra-
matic acidification of the solution, with a pH that decreases
from 8.4 to 4.5 after 1200 h of irradiation.
The dramatic decrease of the pH of the PEO solution re-
veals an acidification of the medium that could be correlated
with the release of formic acid ions (HCOO). In order to con-
firm this hypothesis, formic acid and formic acid ions were
measured and quantified by Ionic Chromatography (IC)
(Fig. 2).
An intense peak corresponding to formic acid ions appeared
in the IC chromatogram of the irradiated solutions. As shown in
 F. Hassouna et al. / Polymer Degradation and Stability 92 (2007) 2042e2050 2045

O CH2 CH2

O2 /h ν O2 / h ν
1 O2 / h ν 3
2

O O
O C C O C CH2 C O-
H H2 H
O
Formate Ester Formic acid ion
IR : 1720 cm-1 IR : 1750 cm-1 IR : 1590 cm-1

Scheme 1. Photoproducts formed by photo-oxidation of PEO in aqueous solutions.

Fig. 2, the concentration of formic acid and formic acid ions in-
creased from 0.075 mol L1 to 0.503 mol L1 after 660 h of
irradiation.
This result confirms that formic acid formed is responsible
for the acidification of the medium. This is an important result
in terms of toxicity because acidification of the aquatic me-
dium could contribute to a drastic perturbation of the
ecosystems.
3.2. Influence of the pH of the aquatic medium
In order to understand the effect of pH on the photo-oxida-
tion of PEO in aqueous solution, irradiations were carried out
at different pHs: 2.3, 3.1 and 12.0 and compared with the re-
sults reported above in the case of pH ¼ 8.4.
3.2.1. Analysis by infrared spectroscopy
3.2.1.1. pH ¼ 12.0. The modifications of the infrared spectra
of PEO provoked by irradiations of solutions at pH ¼ 12 are
different from those reported above in the case of pH ¼ 8.4. Af-
ter 300 h of irradiation, two maxima developed at 1610 cm1
and 1665 cm1 and a shoulder around 1600 cm1. After
500 h, a wide band centred over 1600 cm1 developed. This
pH
maximum was progressively shifted towards 1590 cm1. After
700 h of irradiation, two new absorption bands were observed
at 1720 cm1 and 1750 cm1. This clearly indicates that differ-
ent photoproducts are formed, in comparison with the results
given above. In parallel, the pH of the solution decreased dur-
ing irradiation and reached a value around 6.0 after 400 h and
4.6 after 1200 h.
The modifications of the chemical structure of the photo-
products can result in the decrease of the pH of the medium dur-
ing the photo-oxidation. Absorption bands observed at
1610 cm1 and 1665 cm1 could correspond to carboxylate
acid ions, which can explain the first phase with the decrease
of pH from 12.0 to 6.0. In the second phase, from pH  6, for-
mates (1720 cm1) and esters (1750 cm1) (with the ratio of
formate/ester ¼ 4/1), as well as formic acid ions (1590 cm1)
appear.
3.2.1.2. pH ¼ 3.1. Fig. 3 shows the changes in the carbonyl re-
gion of the IR spectra for deposits obtained from PEO aqueous
solutions acidified to pH 3.1 for irradiation times ranging from
0 h to 790 h. In the first stage of irradiation, an absorption
band with a maximum at 1720 cm1 appears, which indicates
the formation of formats. A shoulder around 1750 cm1 due to
790 h
700 h

[HCOOH] and [HCOO-] (mol L-1) 9 0.20 625 h

0.5 575 h
[HCOOH] and [HCOO-] (mol L-1)

500 h
8 410 h
0.4 0.15
360 h
310 h
7
Absorbance

0.3 240 h
pH

0.10 190 h
120 h
6
0.2 72 h
0.05 24 h
5 0h
0.1

0.00
0.0 4
0 200 400 600 800 1000 1200
Irradiation time (h) 1900 1800 1700 1600 1500
Wavenumber (cm-1)
Fig. 2. Variations versus irradiation time of the concentration of formic acid
and formic acid ions and of the pH of PEO aqueous solution Fig. 3. FTIR spectra of PEO deposit from an acidic aqueous solution at
([PEO] ¼ 50 mg mL1, pHinitial ¼ 8.4). pH ¼ 3.1 during photo-oxidation, [PEO] ¼ 50 mg mL1.
2046 F. Hassouna et al. / Polymer Degradation and Stability 92 (2007) 2042e2050

the formation of esters is also observed. After 400 h of irradi- (a) pH = 12.0
ation, this absorption band widens and the absorption maxi- 100000
pH = 8.4
mum is shifted towards 1735 cm1 with two shoulders at pH = 3.1
1750 cm1 and 1770 cm1. These two absorption maxima 80000 pH = 2.3
(1735 cm1 and 1770 cm1) indicate the formation of second-

Mw (g mol-1)
ary products resulting from the degradation of the photoprod-
60000
ucts in acidic medium. Considering the frequencies, lactones
with five and six atoms are good candidates [23]. Opposite
40000
to the case of the solutions with pH 8.4 or 12.0, the absorption
band at 1590 cm1 corresponding to the formic acid ions is not
observed. In fact, at pH ¼ 3.1, the equilibrium between the 20000
ionic and the acidic form is shifted to the acidic form and for-
mic acid ions are no more detected. It is recalled that pKa 0
(HCOOH/HCOO) ¼ 3.78, HCOOH is not detected in the an- 0 300 600 900 1200 1500 1800
alysed deposit because it migrates out of the polymer [15]. Irradiation time (h)
The variations of the pH during irradiation only reveal
a slight decrease from 3.1 to 2.8, which however indicates (b) 1.0
pH = 12.0
that a large amount of Hþ coming from formic acid is released pH = 8.4
into the solution after 800 h of irradiation. 0.8 pH = 3.0
pH = 2.3
3.2.1.3. pH ¼ 2.3. From the beginning of the irradiation,
0.6
a broad absorption band appears in the carbonyl region
(1900e1500 cm1) with a maximum at 1720 cm1 (formate)
IV

and a shoulder around 1750 cm1 (ester) (ratio of formate/ 0.4

ester ¼ 4/1). After 120 h of irradiation, two maxima develop
at 1735 cm1 and 1770 cm1. As reported above in the case
0.2
of irradiations at pH ¼ 3.1, formic acid ions are not detected
as the equilibrium is shifted to the acidic form. In addition,
strong modifications of the IR spectra between 1600 cm1 0.0
and 1200 cm1 after 100 h of irradiation were noticed. The ab- 0 300 600 900 1200 1500

sorption bands are no more of those characteristics of PEO, Irradiation time (h)
which means that the PEO polymer is highly degraded in Fig. 4. (a) Evolution of PEO molar weight (Mw) as a function of irradiation
such an acidic medium (pH ¼ 2.3). time and in different pH conditions (12.0, 8.4, 3.1 and 2.3),
[PEO] ¼ 50 mg mL1 and (b) evolution of PEO intrinsic viscosity of the solu-
tion as a function of irradiation time.
3.2.2. Analysis by size exclusion chromatography (SEC)
Fig. 4a shows the variations of the molar weights of PEO At pH ¼ 12, after 24 h of induction period, the average mo-
during irradiation of aqueous solution at different initial pH. lar weights and intrinsic viscosity starts to decrease. The rate
At pH ¼ 8.4, the changes caused by irradiation of the PEO of degradation of PEO is faster in very basic medium
aqueous solution are characterised by a shift of the molar (pH ¼ 12.0).
weight distribution curves towards lower molar weights. The SEC analysis of PEO aqueous solution at pH ¼ 3.1 shows
average molar weight decreases by about 50% after 700 h of also the decrease of the average molar weights and the intrin-
irradiation and 82% after 1664 h. This decrease in the average sic viscosity. However, the rate of degradation of PEO is faster
molar weight can be attributed to the chain scissions resulting and without induction period at this pH. Indeed, the average
from the photo-oxidation of PEO. This indicates that a dra- molar weight decreases by about 82% only after 145 h of
matic degradation process takes place during the oxidation
of the polymer. Table 1
Evolution of PEO polydispersity index of the solution as a function of irradi-
The intrinsic viscosity (IV) drops during irradiation as
ation time (pH ¼ 8.4)
shown in Fig. 4b. This decrease in viscosity and in molar
Irradiation time (h) IP
weight is correlated to the chain scissions due to the formation
of formate end groups. 0 3.8
150 3.2
Table 1 reports the evolution of the polydispersity index of
370 2.6
PEO during photo-oxidation in aqueous solution. The evolu- 535 2.5
tion of the polydispersity index versus irradiation time shows 660 1.9
a decrease of polydispersity during irradiation corresponding 850 1.7
to more uniform mass distribution of degraded PEO. This re- 920 1.6
1664 1.2
veals that the mechanism of chain scissions is predominant.
 F. Hassouna et al. / Polymer Degradation and Stability 92 (2007) 2042e2050
irradiation and by 94% after 310 h. These results confirm that
chain scission is the main way of degradation implied in the
mechanism during photo-oxidation in acidic medium.
At pH ¼ 2.3, a spectacular decrease of average molar
weights and of intrinsic viscosity is observed. Indeed, the av-
erage molar weight decreases by about 86% only after 24 h of
irradiation and by 99% after 120 h. After 120 h, the molar
weights are too low and out of range of the column separation,
then the polymer can be considered as highly degraded. Possi-
ble candidates could be oligomers, formic acid and lactones.
The correlation of the results obtained by SEC with those ob-
tained with IR indicates that after 120 h of irradiation, the
polymer is completely degraded. The absorption bands at
1735 cm1 and 1770 cm1 correspond to secondary products
resulting from the degradation or from the transformation of
formate and ester at pH lower than 3.
All the results obtained by SEC analysis (Mw, IV, IP) show
that the degradation of PEO in aqueous solution leads to
a mechanism of chain scissions.
3.3. Stability of the photoproducts
In order to study the influence of pH on the photo-oxidation
of PEO, it is important to characterise the stability of PEO
photoproducts in water. Fig. 5 compares the IR spectrum of
a photo-oxidized PEO film (after 42 h of irradiation) and the
spectrum of a deposit of the same film dissolved in water
(pH ¼ 8).
It is recalled that photo-oxidation of PEO as a film in solid
state leads to the formation of two products, formate
(1720 cm1) and ester (1750 cm1). The IR spectrum of a de-
posit obtained after dissolution of the photo-oxidized film of
PEO in water shows a third absorption band at 1590 cm1
and a decrease of intensity of the absorption band at
1720 cm1 corresponding to formate. This band can be attrib-
uted to formic acid ions, which suggests that a partial hydrolysis
of formate occurs in water, leading to the formation of HCOO.
0.30
a pH = 2.6
b pH = 4.0
0.25
0.20
Absorbance
0.15
0.10
0.05
0.00
1900 1800 1700 1600 1500
Wavenumber (cm-1)
Fig. 5. FTIR spectra of PEO after 40 h of photo-oxidation in solid state (a) as
a film and (b) a deposit after dissolution of this film in water (pH ¼ 8).
2047
This confirms that the absorption band at 1595 cm1 ob-
served for the photo-oxidation of PEO in water at pH 12.0
and 8.4 comes from a partial hydrolysis of formate in water.
The pH of the solution can be adjusted to different values.
A PEO film photo-oxidized for the same time (42 h) was dis-
solved in water and the pH was adjusted to values ranging
from 2.6 to 11.4. The IR spectra of deposits obtained from
these solutions are shown in Fig. 6.
In a basic medium, when the pH of the solution (in which
the PEO film is dissolved) is adjusted to 9.6, the intensity of
the absorption band corresponding to formates (1720 cm1)
decreases whereas that corresponding to formic acid ions
(1590 cm1) increases. In a more basic medium, pH ¼ 11.4,
the absorption bands of formates and esters disappear and
two new bands centred at 1610 cm1 and 1665 cm1 appear.
As reported above, these two bands have been already ob-
served during the first time of irradiation of PEO in basic
aqueous solution at pH ¼ 12.0.
In an acidic medium, at pH ¼ 4.0, the IR spectrum of de-
posit shows the decrease of intensity of the bands at
1720 cm1 and 1750 cm1 (formate and ester) and a band at
1590 cm1 with a shoulder at around 1620 cm1 is observed.
In a more acidic medium, pH ¼ 2.6, the absorption bands at
1720 cm1 and 1590 cm1 almost disappear. Indeed, the pH
being lower than pKa (HCOOH/HCOO), the disappearance
of formate groups by hydrolysis leads to formic acid.
These results thus show the instability of formate and ester
groups in acidic and basic media. Formate groups are hydro-
lysed in an acidic medium leading to the formation of formic
acid, which however are not detected by IR analysis because it
evaporates during the drying of the deposit. In a basic medium
formic acid ions are formed [15]. Ester groups can be hydro-
lysed in basic medium to form carboxylate ions (1665 cm1 or
1620 cm1).
Scheme 2 summarizes the comments made above concern-
ing the reactivity of ester and formate groups under different
pH conditions. Formates are hydrolysed in acidic medium
0.5
pH = 7.9
0.4
pH = 9.6
pH = 11.4
0.3
Absorbance
0.2
0.1
0.0
1900 1800 1700 1600 1500
Wavenumber (cm-1)
Fig. 6. Infrared spectra of deposits obtained after dissolution in water at differ-
ent pHs of a PEO solid film photo-oxidized for 42 h.
2048 F. Hassouna et al. / Polymer Degradation and Stability 92 (2007) 2042e2050

O CH2 CH2

O2 / h ν O2 / h ν

O CH2 O C CH2
O C
Formate H Ester O
IR : 1720 cm-1 IR : 1750 cm-1

Basic medium basic medium

Acid medium

O O -
O CH2
C
C O- C OH
H O
H
Formic acid ion Formic acid Carboxylate ion
(volatile)

Scheme 2. Stability of formates and esters in aqueous solution at different pHs.

into formic acid which is likely to migrate out of the deposit formates undergo hydrolysis. This explains in turn the pres-
and into formic acid ions (1590 cm1) in basic medium. The ence of HCOO detected in the medium.
ester is hydrolysed into carboxylate ions (1665 cm1) at
very basic medium (pH > 11). 3.5. Influence of polymer concentration

3.4. Photo-oxidation rate The concentration of PEO in the solution can influence both
the polymer concentration and the rate of degradation of the
The kinetics of photo-oxidation of PEO irradiated in aque- polymer. The comparison of the behaviour of PEO aqueous
ous solutions at different pH can be compared by following the solutions of concentration 50 mg mL1 and 5 mg mL1 at dif-
increase of the infrared absorbance at 1720 cm1 as a function ferent pH conditions (12.0, 8.4, 3.1 and 2.3). The change of pH
of irradiation time (Fig. 7). during the irradiation was also followed.
According to the initial pH of the solution, the rate of accu- It is important to underline that the thickness of the deposits
mulation of formates is different. The absorbance increases obtained from solutions at 5 mg mL1 was limited to 2 mm
very quickly in acidic media whereas in basic media whereas the thickness for deposit from PEO solutions at a con-
(pH ¼ 12.0 and 8.4) formates are not detected by infrared centration of 50 mg mL1 was 16 mm. Comparison of the rates
analysis of the deposits before 500 h and 700 h for the solu- of photo-oxidation based on the infrared analysis would be
tions of pH ¼ 12.0 and 8.4, respectively. This can be explained meaningless because no reliable results would be obtained
by the fact that in basic medium (pH > 6e7), macromolecular by considering the correction factor of 8 that has to be used
for comparing the absorbances.
0.16 At pH ¼ 8.4, within the first 1000 h of irradiation, three ab-
pH = 12.0
sorption bands develop at 1750 cm1 (formate), 1720 cm1
0.14 pH = 8.4 (ester), and at 1590 cm1 (formic acid ion). The formate/ester
pH = 3.1 ratio is equal to 4/1. The pH of the solution was observed to
Absorbance (1720 cm-1)

0.12
dramatically decrease, from the value of 8 to 2.9 after
0.10 1000 h of irradiation. After 1000 h, the intensity of the absorp-
tion at 1720 cm1 band decreases. The proportion of formate/
0.08
ester is then 1/1. This decrease can be related to the evolution
0.06 of the pH of the solution during irradiation. At the same time,
the band at 1590 cm1 corresponding to formic acid ions is
0.04 progressively shifted towards 1605 cm1. This can be ex-
plained by the protonation of HCOO ions into formic acid
0.02
(pH < pKa (HCOOH/HCOO) ¼ 3.8).
0.00 The comparison with the IR spectra of deposits obtained for
0 200 400 600 800 1000 1200 1400 PEO concentration (50 mg mL1) shown in Fig. 1 indicates
Irradiation time (h) that the photoproducts formed are the same with the similar
Fig. 7. Evolution of absorbance at 1720 cm1 as a function of irradiation time proportion during the first 1000 h of irradiation. However, af-
for PEO aqueous solutions and in different pH conditions, ter 1000 h of irradiation, as the pH of the diluted solution is
[PEO] ¼ 50 mg mL1. lower (pH ¼ 2.9) than that of the concentrated solution
 F. Hassouna et al. / Polymer Degradation and Stability 92 (2007) 2042e2050 2049

(pH ¼ 4.0), the partial hydrolysis of formate is then more im- Table 2
portant in the diluted solution than in the concentrated one and Formic acid concentration during irradiation (pH ¼ 2.3, [PEO] ¼ 5 mg mL1)
explains the discrepancy observed. Time of irradiation (h) Concentration of formic acid (mol L1)
In parallel, the solutions were characterised by SEC during 0 5.2  103
irradiation. Fig. 8 compares the solutions of PEO at concentra- 410 0.239
tion of 5 mg mL1 or 50 mg mL1. This figure clearly shows 840 0.343
a higher degradation rate in the case of a diluted solution at
5 mg mL1. This behaviour can be directly correlated with
was observed during 850 h of irradiation. This means that
the change of the pH of the solution.
even low concentration PEO aqueous solutions cannot be min-
The degradation of PEO could be higher in diluted solution
eralized during the photo-oxidation.
than a more concentrated solution because the mobility of
polymer chains is higher in a less viscous solution [24]. More-
over, this difference could be explained by the fact that for
3.7. Thermo-oxidation
a concentration of 50 mg mL1, PEO aqueous solutions pres-
ent a notable diffusion of light. This diffusion effect could con-
Thermo-oxidation at 50  C of PEO aqueous solution at
tribute to the decrease of the PEO phototransformation rate.
various pHs was carried out.
In acidic medium, pH ¼ 2.3, in the carbonyl region, a new
At pH 8.4 and 12.0, no significant evolution was detected
absorption band at 1718 cm1 appears in IR spectra from the
after 900 h at 50  C contrarily to acidic media (pH ¼ 3.1
beginning of the irradiation and a shoulder around 1750 cm1
and 2.3).
corresponding to macromolecular esters. On the basis of NH3
At pH ¼ 3.1, the evolution of IR spectra shows the forma-
and SF4 chemical treatments the absorption band at
tion of two absorption bands at 1720 cm1 and 1750 cm1
1718 cm1 can be attributed to the formation of carboxylic
corresponding to the two previously identified oxidation prod-
acids.
ucts, respectively, formate and ester. However, the relative pro-
The absence of macromolecular formates in these condi-
portion of ester is much higher in the case of thermo-oxidation
tions of concentration and pH ([PEO] ¼ 0.5 mg mL1,
than in the case of photo-oxidation, and one observed the for-
pH ¼ 2.3) can be explained by the complete hydrolysis of for-
mation of esters and formates to the same extent whereas
mate groups to formic acid. As shown in Table 2, the quanti-
photo-oxidised solutions develop four formate end groups
fication of formic acid during irradiation by ionic
for one ester function. The same observation was made in
chromatography analysis confirmed this hypothesis.
the case of PEO in solid state [14].
In parallel, the pH of the solution decreased, from 3.1 to 2.4
3.6. Evolution of the total organic carbon (TOC) after 500 h of heating, and stabilized at 2.4. After 400 h, the
absorption band of formate (1720 cm1) became a shoulder
The measurements of the quantity of carbon present in so- while the ester band (1750 cm1) was still increasing. The be-
lution during irradiation of PEO aqueous solutions in different haviour of formates results from their hydrolysis that becomes
pH conditions (pH ¼ 2.3, 8.4 and 12.0) were carried out by more important when pH is under 3.0.
TOC analysis and the results are about 5 mg mL1. The ob- Fig. 9 shows the evolution of the molar weight of PEO in
tained results show that no change of the carbon quantity aqueous solution at pH ¼ 3.1 in conditions of thermo-oxidation.

120000
120000
[PEO] = 50 mg mL-1

100000 [PEO] = 5 mg mL-1 100000

80000 80000
Mw (g mol-1)
Mw (g mol-1)

60000 60000

40000 40000

20000 20000

0 0
0 300 600 900 1200 1500 1800 0 100 200 300 400 500 600 700
Irradiation time (h) Heating time (h)

Fig. 8. Evolution of molar weight (Mw) as a function of irradiation time of Fig. 9. Evolution of PEO molar weight (Mw) as a function of heating time and
PEO aqueous solutions (pH ¼ 8.4, [PEO] ¼ 50 mg mL1 and 5 mg mL1). irradiation time at pH ¼ 3.1.
2050 F. Hassouna et al. / Polymer Degradation and Stability 92 (2007) 2042e2050
0.35
pH = 2.3
0.30 pH = 3.1
Absorbance (1750 cm-1)
0.25
0.20
0.15
0.10
0.05
0.00
0 100 200 300 400 500 600 700
Heating time (h)
Fig. 10. Variations of the absorbance at 1750 cm1 as a function of heating
time; pH ¼ 2.3 and 3.1.
A decrease of the molar weight is observed; the average molar
weight decreases by about 50% only after 240 h of heating and
by 90% after 700 h. This result shows that chain scissions oc-
cur during thermo-oxidation in acidic medium.
Fig. 10 compares the rates of oxidation of PEO measured at
1750 cm1 (ester) at two different pHs (3.1 and 2.3). It can be
observed that the rate of formation of esters (rate of oxidation)
is faster when the solution is more acidic. The same behaviour
has been observed for ageing in condition of photo-oxidation.
4. Conclusion
It can be recalled that photo-oxidation of PEO in solid state
occurs through a conventional radical mechanism [14] involv-
ing the formation of macromolecular formate end groups and
esters. The photo-oxidation of PEO in aqueous solution
(concentration 50 mg mL1 or 5 mg mL1, pHinitial ¼ 8.4)
leads also to the formation of formate end groups and esters
but a third product has also been identified, formic acid ions
or formic acid depending on the pH of the medium. Formic
acid ions or formic acid might come from partial acidic or ba-
sic hydrolysis of formate end groups of the polymer matrix.
The release of formic acid is mainly responsible for the acid-
ification of the medium. For very advanced time of irradiation
in acidic medium, secondary products appear.
The oxidation products depend on the pH of the medium,
due to the instability of formate and ester in acidic and basic
media. Formates are hydrolysed in acidic medium with the re-
lease of formic acid, in basic medium with the production of
formic acid ions. Ester groups are also hydrolysed in basic
and acidic media to give carboxylate ions/carboxylic acids.
SEC analysis of PEO aqueous solutions shows that in all
pH conditions the photo-oxidation of PEO aqueous solution
leads to the decrease of molar weights, which means that the
degradation of PEO implies a chain scissions mechanism.
SEC analysis also shows that the rate of degradation is
much higher in acidic medium, because of the partial hydroly-
sis of formate.
The influence of the medium, aqueous solution in regard to
solid state, lies in the stability of the photoproducts which is
different depending also on the pH of the medium in aqueous
solution but the mechanism of oxidation of PEO is not modi-
fied. Indeed, the differences observed are connected to hydro-
lysis reactions of photoproducts in aqueous solution.
The experiments of thermo-oxidation, at 50  C, of PEO
aqueous solution do not show the degradation of PEO in solu-
tion of initial pH 12.0 and 8.4 while in acidic medium, the for-
mation of oxidation products is observed.
The comparison of results obtained either in thermo-oxidation
or in photo-oxidation of PEO aqueous solution shows a different
ratio of formate/ester as already observed in solid state. Thermo-
oxidation of PEO aqueous solution produces equal amounts of
800 900
esters and formates whereas photo-oxidation generates formates
in a ratio of 4/1 against chain esters.
The next step of our study on PEO deals with the influ-
ence of Fe(III) on the photodegradation of PEO aqueous
solutions.
Acknowledgements
The authors are grateful to Marius Parazols, Ph.D. student
at the ‘‘Laboratoire de Photochimie¨Moléculaire et Macromo-
léculaire’’ for his kind collaboration in ion chromatographic
experiments.
References
[1] Swift G. Polym Degrad Stab 1998;59:19e24.
[2] Molyneux P. Water-soluble synthetic polymers: properties and behavior,
vol. 1. Boca Raton: CRC Press; 1983 [Chapter 2].
[3] Al-Saigh ZY. Polymer 1999;40:3479e85.
[4] Aray Y, Marquez M, Rodriguez J, Vega D, Simon-Manso Y, Coll S, et al.
J Phys Chem B 2004;108:2418e24.
[5] Scheirs J, Bigger SW, Delatycki O. Polymer 1991;32(11):2014e9.
[6] Bjordal CG, Nilsson T, Daniel G. Int Biodeterior Biodegrad 1999;43:63.
[7] da Costa VM, Fiske TG, Coleman LB. J Phys Chem 1994;101:2746e51.
[8] Miyazawa T, Fukushima K, Ideguchi Y. J Chem Phys 1962;37(12):
2764e76.
[9] Begum R, Matsuura H. J Chem Soc Faraday Trans 1997;93(21):
3839e48.
[10] Takahashi Y, Tadokora H. Macromolecules 1973;6:672e5.
[11] Bedrov D, Pekny M, Smith GD. J Phys Chem B 1998;102:996e1001.
[12] Dormidontova EE. Macromolecules 2002;35:987e1001.
[13] Smith GD, Bedrov D, Borodin O. J Am Chem Soc 2000;122:9548e9.
[14] Morlat S, Gardette JL. Polymer 2001;42:6071e9.
[15] Morlat S, Gardette JL. Polymer 2003;44:7891e7.
[16] Philippart JL, Sinturel C, Gardette JL. Polym Degrad Stab 1997;58:
261e8.
[17] Pang P, Englezos P. Fluid Phase Equilib 2002;194:1059e66.
[18] Bailey Jr FE, Callard RW. J Appl Polym Sci 1959;1:56e62.
[19] Wilhelm C, Gardette JL. J Appl Polym Sci 1994;51:1411e20.
[20] Morlat S, Cezard N, Loubineux B, Philippart JL, Gardette JL. Polym
Degrad Stab 2001;72(2):199e208.
[21] Gauvin P, Philippart JL, Lemaire J. Makromol Chem 1985;186:1167e80.
[22] Jaffrezo JL, Calas N, Bouchet M. Atmos Environ 1998;32:2705e8.
[23] Lin-Vin D, Colthup NB, Fateley WG, Grasselli JG. The handbook of
infrared Raman characteristic of organic molecules. San Diego: Aca-
demic Press; 1991.
[24] Vijayalakshmi SP, Senapati D, Madras G. Polym Degrad Stab
2005;87:521e6.