PAPERwww.rsc.org/pps | Photochemical & Photobiological Sciences
Oxidation of polystyrene aerosols by VUV-photolysis and/or ozone†
Jos´ e Salas Vicente,‡aJuan L´ opez Gejo,§aSonja Rothenbacher,bSumalekshmy Sarojiniamma,¶a
Eliso Gogritchiani,?aMichael W¨ orner,**aGerhard Kasperband Andr´ e M. Braun*a
Received 10th February 2009, Accepted 8th April 2009
First published as an Advance Article on the web 27th April 2009
Aerosols of submicron polystyrene particles were oxidized by either vacuum-ultraviolet (VUV)
irradiation in the presence of molecular oxygen (O2) and/or by ozone (O3). Different degrees of
oxidation and oxidative degradation were reached by VUV-photolysis depending on radiant energy, O2
and H2O concentrations in the bulk gas mixture as well as on particle diameter. The same
functionalization was obtained by exposing the aerosol to O3, however, oxidation, in particular
oxidative degradation, was less efficient. The evolution of hydroxyl and carbonyl functions introduced
was quantified by ATR-FTIR spectroscopy of filtered particles, and oxidative degradation of the
polymer particles was confirmed by determining size and number of aerosol particles before and after
oxidation. Efficiency analyses are based on the results of an O3actinometry and on an evaluation of the
rate of absorbed photons by the aerosol particles in function of their size.
The preparation of functionalized polymer and polymer-
composite nanoparticles is of primary interest in materials re-
search, and corresponding development work already led to a
number of applications, e.g. in the biochemical and medical
domain including drug and enzyme carriers,1absorbents,2affin-
ity bioseparators,3as well as in the optical and optoelectrical
Functionalized (e.g. sulfonated) polymer particles are mostly
monomer or oligomer substrates bearing the corresponding
with such polymers.7However, the production of a larger range
of functionalized polymer or copolymer particles of defined
size, composition and morphology is usually realized in two
phases: (i) the production of the polymer or copolymer substrate
(particle or coat of nanoparticulate inorganic material) and (ii)
the functionalization of the native polymer. These processes are
normally performed in batch reactions, and to our knowledge,
there is no published work describing such functionalization
processes in a continuous regime.
aLehrstuhl f¨ ur Umweltmesstechnik, Engler-Bunte-Institut, Universit¨ at
Karlsruhe, Germany. E-mail: Andre.Braun@ciw.uni-karlsruhe.de
bBereich Gas-Partikel-Systeme, Institut f¨ ur Mechanische Verfahrenstechnik
und Mechanik, Universit¨ at Karlsruhe, 76128, Karlsruhe, Germany
‡Present address: UV-Consulting Peschl Espa˜ na S.L., 46980 Paterna
§Present address: Departamento de Qu´ ımica Org´ anica I, Universidad
Complutense de Madrid, 28040 Madrid, Spain.
¶Present address: School of Chemistry and Biochemistry, Georgia
Institute of Technology, Atlanta, GA 30332, USA.
?Present address: Reuter Chemische Apparatebau KG, 79108 Freiburg,
lare Aufarbeitung von Bioprodukten, Universit¨ at Karlsruhe, 76128 Karl-
A continuous process might be realized in a cascade of
reactors (i) to produce functionalized nanoparticles, and (ii) to
perform a primary as well as subsequent functionalizations of the
original polymer or polymer coated particles. The polymerization
of aerosol nanodroplets of liquid monomers or mixtures of
monomers in a continuous operating regime has long been
described8and claimed for a technical application,9yet the exam-
ples of the patent cited are not very informative. The experimental
results obtained so far by the authors confirm qualitatively that
(VUV-)photochemical polymerization of aerosol droplets, but the
development of such a process requires a detailed knowledge
of the gas dynamics in a continuously operating reactor and,
consequently, a very special reactor design.
The work presented here therefore is focused on the primary
functionalization of previously prepared and commercially avail-
able polystyrene nanoparticles. VUV-photochemical oxidation
was chosen to avoid any addition of chemicals to the reaction
system during primary functionalization,10–14except water (H2O)
and molecular oxygen (O2). Taking into account that VUV-
photolysis of O2 produces O3, the latter was used alternatively
for the thermal oxidation of the polymer nanoparticles.
Earlier work on the functionalization of polystyrene films
revealed that the VUV-photolysis of polystyrene leads to C–C-
bond homolysis (reaction (1)).12–14Due to the reduced mobility
within the polymer bulk, the C-centered radicals generated may
mainly recombine, but evidence for disproportionation and cross-
linking was found in the absence of O2. In the presence of
O2, peroxyl radicals are generated as key intermediates of the
subsequent thermal reactions leading to hydroxyl and carbonyl
and carboxyl groups (reactions (2), (3) and (4), respectively).
∑+ O2→ RO2
944 | Photochem. Photobiol. Sci., 2009, 8, 944–952This journal is © The Royal Society of Chemistry and Owner Societies 2009
∑→→ ROH →→ R¢COH →→ R¢COOH(3)
∑→→ RR¢C=O (4)
VUV-photolysis of the gaseous bulk system containing O2
generates atomic oxygen (O) (reaction (5)),15which is known to
react from both triplet and singlet states with organic substrates
by hydrogen abstraction (reaction (6)); the latter may also react
by insertion into a C–H-bond (reaction (7)).16,17However, O
predominantly adds to O2yielding ozone (O3) (reaction (8)).15
O2+ hn (VUV) → 2O(5)
O + RH → R
O + RH → ROH (7)
O + O2→ O3
In the presence of H2O, its VUV-photolysis may contribute
to the oxidation of the polymer. In fact, the VUV-photolysis of
H2O (reaction (9)) is used to initiate the oxidation of organic
compounds by intermediate hydroxyl (HO
abstraction (reaction (10)) and addition to p-systems (reaction
(11)). Both reactions generate C-centered radicals18that are
efficiently trapped by O2(reaction (2)). However, it was shown
recently that HO
inefficiently with solid polymer surfaces,19and similar reactions
in gas/solid heterogeneous systems cannot be competitive unless
∑) radicals via hydrogen
∑generated in an aqueous bulk phase reacts very
H2O+ hn (VUV) → H
∑+ RH → R
∑+ H2O (10)
∑+ RR¢C=CR¢¢R¢¢¢ → RR¢C
R¢COOH + HO
∑+ H2O (12)
∑is also known to react with carboxylates and carboxylic
acids by electron transfer and subsequent decarboxylation (reac-
tions (12) and (13), respectively).20
Under the experimental conditions applied, the concentration
of O3, generated by the VUV-photolysis of the bulk gas phase
on the overall reaction system are negligible. However, O3 was
found to react with solid polystyrene.12,14,21Alkanes are relatively
inert to O3,22,23and reaction products observed might either
be explained by reactions of atomic oxygen (vide supra) or by
reactive oxygen species generated in the presence of H2O (reaction
(14)) or hydrogen peroxide (H2O2, reaction (15)). In addition,
oxidation and oxidative degradation might be enhanced by the
generation (reaction (16)) and decay of ozonides (reaction (17))
and by reactions of O3with intermediates of VUV-photolyses (e.g.
reactions (18), (16) as well as (1) and (19)).
O3+ HO-→ HO2
∑+ O2→ HO2
The oxidation of aromatic moieties by O3is thought to involve
epoxidation and subsequent rearrangement to yield phenols.24
The aim of the present work was to check the feasibility
of a continuous reaction system to functionalize polystyrene
nanoparticles in the gas phase, before expanding the scope of
such a process to more complex particle substrates (e.g. nanocom-
posites) and designing and combining reactors for gas and liquid
phase reactions.25A differentiation between VUV-photochemical
initiated oxidation and ozonolysis was made in order to evaluate
the scope of application of the two processes.
Aqueous suspensions (approx. 10%) of polystyrene particles of
different sizes (50, 98 and 500 nm; Surfactant Free Sulfate White
Polystyrene Latex, Postnova Analitics, Germany), O24.5, N25.0
and synthetic air (Air Liquide, Germany) were used as purchased.
H2O was of tridistilled quality (UHQ-II).
250 ml of the aqueous suspensions of polystyrene particles of
defined diameter (see section 2.1.) were diluted under stirring in
80 ml of distilled water. The highly diluted suspensions were fed
into a Collison-Atomizer of defined diameter (Topas, ATM 220,
Germany)26in which they were dispersed into the bulk gas phase
by pressurized gas or gas mixtures (3 bar) (Fig. 1). Due to the high
dilution, submicron droplets contained mostly (approx. 96%) one
polystyrene particle. A deflector led bigger droplets back to the
reservoir. Droplets were evaporated after passing the deflector by
mixing the aerosol with heated dry gas mixtures.
2.3.VUV-photochemical oxidation of the polymer aerosol
The aerosol was led into an annular photochemical reactor that
was equipped with a cylindrical Xe2-excimer radiation source
emitting at 172 (±14) nm and positioned in the central axis of
the reactor. The radiation source (custom built) consisted of two
outside diameter: 3.0 cm) with an inner electrode (phase) made
of an aluminium foil and cooled with distilled water. The Xe2-
(ENI, Model HPG-2) with electrical powers (Pe) of 20 to 150 W
at 175 kHz. An additional Suprasil
3.8 cm) was positioned between the outer wall of the radiation
source and the aerosol, providing a gap for the outer electrode.
This outer electrode was made of an extensible net of stainless
R ?quartz tubes (length of discharge: 14 cm,
R ?tube (outside diameter:
This journal is © The Royal Society of Chemistry and Owner Societies 2009 Photochem. Photobiol. Sci., 2009, 8, 944–952 | 945
H: gas heating system, O3: ozonizer, O2, N2, Air: pressure bottles with respective gases. Functionalization unit: PR: photochemical reactor. Analytical
devices: UV-Vis spectrophotometer, SMPS: scanning mobility particle sizer, R: agglomeration reservoir, LPI: low pressure impactor, D: charcoal denuder,
MF: membrane filtration, P: vacuum pump.
steel (wire diameter: 0.1 mm) and was connected to the ground.
The gap between the two Suprasil
to avoid filter effects by O2. The reactor had an optical path (l)
of 7 mm, measured between the outer Suprasil
reactor wall. The outer (grounded) electrode could be covered
partially with a metallic sheet. This variation of the length (d) of
the irradiated annular volume allowed the control of the rate of
incident photons (P0,172). The maximum value of d was 140 mm.
The reactor temperature was not controlled but reached a stable
working temperature of approx. 60◦C.
R ?tubes was purged with N2
R ?tube and the
2.4.Oxidation of the polymer aerosol by ozone
The aerosol of the native particles was mixed at the entrance
of the photochemical reactor with O3(5 ¥ 10-6M in synthetic
air) produced by an ozonizer (Sander, Germany). The ozonizer
contained 7 water cooled elements of silent discharge that could
be operated at 7 to 7.5 kV with a maximum electric power (P¢e) of
80 W. P¢ecould be varied in % of the maximum value (see Fig. 10).
2.5. Analytic procedures
stein s-1]) was determined by O3actinometry27,28using synthetic
air (estimated limit of error: ±5%). In-line ozone analyses were
performed spectrophotometrically (Fig. 1, HP Spectrophotome-
ter, 8452 DAD, Suprasil
eO3,258: 3000 M-1cm-129)).
photochemical reactor increased with increasing Pe, but reached
a level of saturation at Pe≥ 150 W (Fig. 2). It is interesting to
note that addition of H2O did not affect [O3]. It may therefore be
The rate of incident photons (P0,172[ein-
R ?spectroscopic cell (optical path: 1 cm,
of O3 [M] in the gaseous mixture exiting the photochemical reactor
in function of the electric power of the Xe2-excimer radiation source.
Photolysis of dry synthetic air (?), addition of 3.6 mg L-1of H2O (?);
flux: 7 L min-1. Optical path: 7 mm; length of irradiated zone (d): 140 mm.
O3production by VUV-photolysis of synthetic air. Concentration
assumed that reaction (14) as well as reaction (20) are negligible
within this context.
O + H2O → 2HO
The Xe2-excimer radiation source was not operational at Pe<
20 W, and Pe ≥ 50 W were chosen to obtain stable [O3] for
analytic purposes. Under the experimental conditions described
in Fig. 2, [O3] = f(Pe) was found within a nonlinear domain.
Under conditions of quasi-constant absorption during the time of
actinometer photolysis, [Pr]Acdepends on Pa,land on t (eqn (21)30)
[Pr]Ac= P0,l(1 - 10-eAc,ll[Ac])UAc,lt = Pa,lUAc,lt = QaUAc,l
946 | Photochem. Photobiol. Sci., 2009, 8, 944–952This journal is © The Royal Society of Chemistry and Owner Societies 2009
∑ [Pr]Ac: concentration of product formed by the photolysis of
the actinometer during time t [M]
∑ UAc,l: quantum yield of actinometric reaction
∑ eAc,l: molar absorption coefficient of actinometer [M-1cm-1]
∑ l: optical path [cm]
∑ [Ac]: concentration of actinometer [M]
∑ Pa,l: rate of absorbed photons [einstein s-1]
∑ t: time of photolysis [s]
∑ Qa: absorbed photon energy [einstein].
of P0,172= f(Pe) and/or to too high values of Qa. The latter would
lead to levels of [O3] where the VUV-photolysis of O3has to be
taken into account and a quasi-steady state [O3] would eventually
be reached. Both factors can be controlled by working at lower
but constant values of Peand by diminishing and varying t. In a
t = t, (22)
where t is the residence time within the irradiated volume of the
reactor [s], the residence time can be changed by varying either
the flux of the gaseous reaction mixture (F) or the length of the
irradiated volume (d). The latter does not involve changes of the
flow characteristics and was implemented by inserting precisely
cut metallic sheets into the gap provided for the outer (grounded)
electrode (section 2.3.). A linear increase of [O3] in function of d
with a slope of 1.9 (±0.2) ¥ 10-7M mm-1was obtained for Pe=
50 W (Fig. 3).
mixture exiting the photochemical reactor as a function of d. F: 7 L min-1.
Pe: 50 W.
O3production by VUV-photolysis of O2. [O3] [M] in the gaseous
Based on eqn (21) and using the production of O3by VUV-
photolysis of O2as an actinometer in a continuous regime, P0,172
for a given reactor configuration is calculated with eqn (23)
∑ [O3] = [Pr]Ac
∑ fO3,172: apparent quantum yield of O3production.
Whereas UO3,172= 2, fO3,172takes into account the main decay
reactions of O3, and the experimental eqn (24) was proposed for a
most adequate fit.31
However, for [O3]/[O2] < 10-2, UO3,172= 2 was used. In earlier
work of VUV-photochemical oxidations in the gas phase, total
absorption by O2at 172 nm was found for l = 2.2 cm,32hence, for
l = 7 mm, a calculated absorption A172= 0.62 was applied. Since
A = log(1/T), eqn (23) may be rewritten
P0,172was related to Pe, and the corresponding values are given
in Fig. 4 for d = 4 and 140 mm, respectively. For reasons already
the graph. For d = 4 mm, a quasi-linear relation P0,172= f(Pe) was
found with a slope of 3.3 ¥ 10-9einstein s-1W-1corresponding to
a radiation efficiency of the Xe2-excimer radiation source of 3.8%.
d = 4 mm (?) and d = 140 mm (?). l: 7 mm, F: 7 L min-1.
VUV-photolysis of O2(synthetic air). P0,172in function of Pefor
filtered after passing the photochemical reactor unit (Teflon-
membrane filter, pore size: 0.2 mm (Pall GmbH, Germany).
An upstream denuder filled with charcoal was used to reduce
ozononlysis of the accumulating particles during sampling time.
Sampling for off-line analyses.
Aerosol particles were
tion of the polymer particle surfaces were monitored with a
Bruker Golden Gate
measurement). Fig. 5 shows the spectra of filtered polystyrene
nanoparticles before and after irradiation and indicates the
spectral domains used for quantification.
The evolution of the particle functionalization was determined
in introducing an oxidation index (Iox(fg)) that normalizes the
integrated absorption band of either one of the oxygen containing
functional groups (e.g. carbonyl) with reference to the absorption
band of carbon–carbon double bonds that was relatively little
or slightly inversely affected by the oxidation process (e.g. eqn
Hydroxylation and carbonyla-
R ?Diamond ATR Unit mounted on a Bruker
R ?55 FTIR spectrometer (resolution: 3 cm-1, 60 scans per
This journal is © The Royal Society of Chemistry and Owner Societies 2009 Photochem. Photobiol. Sci., 2009, 8, 944–952 | 947
groups (blue) and of reference peaks (yellow) for the calculation of oxidation indices Iox(fg)are shown.33
FTIR-spectra of polystyrene particles before and after VUV-photochemical oxidation. Wavenumber regions of oxygen containing functional
∑ Abs(t)C=O: absorbance of C=O (1615–1840 cm-1) at reaction
∑ Abs(t)ref: absorbance of reference peaks (1471–1522 cm-1) at
reaction time t
∑ Abs(0)C=O: absorbance of C=O (1615–1840 cm-1) for unmod-
∑ Abs(0)ref: absorbance of reference peaks (1471–1522 cm-1) for
VUV-photochemical oxidation, the aerosol was led through a
reservoir of 120 L to ensure equilibrium of the agglomeration–
dissociation processes. A scanning particle mobility sizer (SMPS)
TSI Inc., USA, Model 3071), operating at 0.3 L min-1of aerosol
flow and 3.0 L min-1of sheath air, and a condensation particle
counter (CPC; TSI Inc., USA, Model 3022).
Particle concentration and size distribution.
3. Results and discussion
As expected from the results of foregoing experiments with
polystyrene films,12–14the VUV-irradiation of polystyrene aerosol
particles led primarily to the oxidation of their surfaces,10where
carbonyl and hydroxyl functions were formed (Fig. 5), The
evolution of Iox(C=O)as a function of the concentration of O2in the
gaseous bulk phase and of Peof the Xe2-excimer lamp is shown in
The result confirms the mechanistic hypothesis already derived
C–C bond homolysis is the main reaction path leading to
the oxidation of polystyrene. At higher O2 concentration, the
increased absorption cross-section of the bulk gas phase leads to a
indices for comparable energy inputs. Alternatively, assuming that
photochemically generated HO
∑or O would primarily initiate the
oxidation, the short lifetimes of these intermediates would require
their generation close to the surface of the particles, and the same
explanation would hold. Given the fact that, for a given mass
of aerosol particles, the part of reflected radiation decreases with
decreasing diameter (vide infra), the result shown in Fig. 6b was
to be expected.
Iox(C=O) reached a limiting value of approx. 1.6 for different
incident radiant energies (Qe) depending on the particle diameter
(Fig. 7) and on the O2concentration in the gaseous bulk phase.
Based on these results, the rate of absorbed photons (Pa,172)
and the absorbed radiant energy (Qa) could be evaluated for
related experiments. For this purpose, P0,172was differentiated into
scattered (Ps,172), absorbed (Pa,172) and transmitted (Pt,172) rates of
P0,172= Pa,172,2r+ Ps,172,2r+ Pt,172,2r
∑ Pa,172,2r= P0,172(1 - e-rs172l)
∑ r: number density of particles [m-3]
∑ s172: absorption cross section [m2]
Ps,172,2r= P0,172(1 - e-rss,172l)34
∑ ss,172: scattering cross section [m2].
Assuming that Pt,lcould be neglected, Pa,172is calculated as the
Pa,172,2r= P0,172- Ps,172,2r= P0,172- (P0,172(1 - e-rss,172l))
and for very dilute suspensions, the exponential factor may be
Pa,172,2r= P0,172- Ps,172,2r= P0,172- P0,172rss,172l
= P0,172(1 - rss,172l)
Expressing r in terms of mass per unit volume,
948 | Photochem. Photobiol. Sci., 2009, 8, 944–952This journal is © The Royal Society of Chemistry and Owner Societies 2009
of O2in the gaseous bulk phase and (b) the aerosol particle diameter (2r).
The polystyrene aerosol was produced by spraying an approx. 0.3 g L-1
aqueous suspension of particles of defined diameter into a heated gas
stream (pressure at the atomizer: 3 bar); F: 7 L min-1; temperature (T):
(?), 20% (?), H2O: evaporating from the aerosol droplets, approx. 0.5%,
N2: difference to 100%. (b) Composition of gaseous bulk phase: O2: 20%,
H2O: evaporating from the aerosol droplets, approx. 0.5%, N2: difference
to 100%; 2r: 50 nm (?), 98 nm (?), 500 nm (?).
VUV-photochemical oxidation of an aerosol of polystyrene
∑ VR: volume of reactor [m3]
∑ d: density of polystyrene (1020 kg m-3)
∑ Sp: surface of projection of particles [m3]
eqn (30) may be written as
0 172, 0 172,
particles. Evolution of the oxidation index Iox(C=O) in function of the
aqueous suspension of particles of defined diameter into a heated gas
stream (pressure at the atomizer: 3 bar); F: 7 L min-1; T: approx. 60◦C;
composition of gaseous bulk phase: O2: 20%, H2O: evaporating from the
aerosol droplets, approx. 0.5%, N2: difference to 100%; 2r: 50 nm (?),
98 nm (?), 500 nm (?).
Ps,172,2rwas also related to the differential scattering cross section
VUV-photochemical oxidation of an aerosol of polystyrene
and after integrating C¢sover the whole sphere of observation,
172 20 172,
the combination of eqn (30) and (34) yields
a,172,s,1722 0 172,
Values of ss,172 could not be found in the literature and were
calculated using eqn (36)34(see Table 1).
∑ l0: wavelength of incident radiation [m]
∑ a: particle radius (r) [m]
cles used and the corresponding Ps,172,2rand Pa,172,2rcalculated (eqn (31)) for
experiments made with Pe= 50 W, F = 7 L min-1and d = 4 mm (P0,172=
5.70 ¥ 10-8einstein s-1)
Calculated sRay,172depending on the 2r of the polystyrene parti-
6.27 ¥ 10-15
4.01 ¥ 10-13
6.27 ¥ 10-9
3.23 ¥ 10-10
5.4 ¥ 10-9
3.23 ¥ 10-6
5.67 ¥ 10-
5.16 ¥ 10-8
This journal is © The Royal Society of Chemistry and Owner Societies 2009Photochem. Photobiol. Sci., 2009, 8, 944–952 | 949
∑ m = np/nmed
∑ np: refractive index of the particle (1.59, ref. 34)
nmed: refractive index of the surrounding medium (1.0003).
The calculated value of Ps,172,500 exceeds P0,172. The Rayleigh
approximation (eqn (36)) being in fact not applicable for 2r ≥
l,34Ps,172,500 is largely overestimated. Inversely, Pa,172,50 may be
taken as a reference to compare normalized values Pa,172,2r/Pa,172,50
and Iox(C=O),2r/Iox(C=O),50. Iox(C=O),100/Iox(C=O),50, based on experimental
results, exceeds the calculated Pa,172,100/Pa,172,50by a factor of 3. The
result is probably due to the vast manifold of oxidation reactions
initiated by the VUV-photolyis of polystyrene leading in most
cases to quantum yields of substrateoxidation (f-S) > 1. This kind
ozonolysis of polystyrene was found to be of minor importance
The maximum level for Iox(C=O)(Fig. 7) represents an average
surface concentration of carbonyl functions that apparently
cannot be exceeded because of an ongoing oxidative degradation
of the particles. In fact, Fig. 8 shows the diminutions of particle
diameter (2r, Fig. 8a) and concentration (% cp, Fig. 8b) depending
on Pe, but in the latter case more strongly on t. At short t
(F = 7 L min-1), the maximum Iox(C=O) of 1.6 was obtained
with a Pe ≥ 100 W. Under these experimental conditions, the
particles lost in average approximately 5% of their original
diameter, and their concentration diminished by 2%. Short t
seem to be favourable for a slow loss of mass that is primarily
controlled by an oxidative degradation of the particle surface.
At higher t, particle loss was much more pronounced than the
diminution of their size, and these results indicate that, parallel to
a continuous diminution of particle size, oxidation may lead to a
disruption of the particles, the smaller fragments being oxidized
at much higher rate. At high values of t and Pe, only 20% of the
original particle concentration (cP,0) were left. It is interesting to
note that the particle concentration diminished at highest rate
in the absence of O2, confirming the dominant impact of the
VUV-photolysis and the competitive rate of depolymerization of
VUV-photochemical oxidative functionalization of polystyrene
Styrene and unsaturated polymeric and/or oligomeric inter-
mediate products were already observed by GC/MS analysis
and fluorescence measurements (lexc > 350 nm), respectively,
when polystyrene films were irradiated in the absence of O2.12,14
Styrene as a product of a competing depolymerization was also
detected in the bulk gas phase during photooxychlorination
Under VUV-irradiation, photolysis of polystyrene was found
to be the primary chemical reaction that leads in the presence
of O2, via oxidation and oxygenation of the intermediates, to
hydroxylated and carbonylated products. This photochemically
initiated oxidative functionalization is assumed to be accelerated
by O3, produced by the VUV-photolysis of the gaseous bulk phase
containing O2(reactions (5) and (8)). Due to the relatively low
concentration of O3generated by the VUV-photolysis of O2, its
photolysis may be neglected, but its contribution to the overall
rate of polystyrene oxidation may involve the reaction manifold
described in section 1. In addition, little is known about the
efficiency of the oxidation of polystyrene exposed to O3, whereas
particles of a diameter (2r) of 100 nm. Evolution of (a) particle diameter
(2r) and (b) percent of initial particles (% cP,0)) measured at the exit
of the photochemical reactor depending on Pe and on t. The latter
is represented by F: 7 L min-1(?), 3 L min-1(?), 0.3 L min-1(?).
Polystyrene aerosol produced by spraying an approx. 0.3 g L-1aqueous
suspension of particles into a heated gas stream; d: 140 mm; pressure at
the atomizer: 3 bar; composition of gaseous bulk phase: O2: approx. 20%,
H2O: evaporating from the aerosol droplets, approx. 0.5%, N2: difference
to 100%; temperature: approx. 60◦C. Percent of initial particles measured
in the absence of O2is shown for comparison: F: 7 L min-1(¥).
VUV-photochemical oxidation of an aerosol of polystyrene
the UV/O3-technology was identified quite early as a convenient
means of polymer particle functionalization.37
In order to investigate the ozonolysis of polystyrene aerosols,
O3was produced from synthetic air by an ozonizer. The minimum
ozone concentration ([O3]) that could be maintained during
several hours was 5 ¥ 10-6M. For comparison, a steady [O3] of 8.6
radiation source (Pe= 150 W, d = 4 mm) irradiated synthetic air
containing 0.5% of H2O at F = 7 L min-1(Fig. 9).
The evolution of Iox(C=O)with the time of ozonolysis is depicted
in Fig. 10 for particles of 2r = 500 nm. As already observed
with polymer films, the increase of Iox(C=O) was found to be
quasi-linear, and in the case of the aerosol particles, a slope of
950 | Photochem. Photobiol. Sci., 2009, 8, 944–952 This journal is © The Royal Society of Chemistry and Owner Societies 2009
using synthetic air containing 0.5% of H2O, F: 7 L min-1. Measured O3
concentrations ([O3]) [M] in function of the electric powers of the ozonizer
(P¢e, see experimental details) and of the Xe2-excimer radiation source (Pe)
with d: 4 mm).
O3 production by ozonizer (?) and by VUV-photolysis (?)
synthetic air containing 5% of H2O by VUV-photolysis and/or by O3.
Polystyrene aerosol produced by spraying an approx. 0.3 g L-1aqueous
suspension of particles into a heated gas stream (pressure at the atomizer:
3 bar), F: 7 L min-1. (?) Evolution of the oxidation index Iox(C=O) in
of time of reaction with a steady [O3] = 5 ¥ 10-6M.
4.8 (±0.3) ¥ 10-5s-1was determined. It is also interesting to note
that the same value of Iox(C=O)was found when the polystyrene
particles were placed on a filter and ozonized for the same period
of time. Using polystyrene particles of the same diameter, the
VUV-photochemical experiment (t = 1.56 ¥ 10-2s) led to a
value of Iox(C=O) of 1.3 (Fig. 6b and 10). Functionalization by
the combination of a VUV-photochemically initiated oxidation
and ozonolysis of polystyrene is therefore by at least a factor 100
Oxidation of an aerosol of polystyrene particles (2r: 500 nm) in
be used advantageously for a strictly controlled functionalization
process without notable oxidative degradation.
VUV-photolysis of aerosols of polystyrene particles initiated
their oxidation to different degrees of surface functionalization
depending on radiant energy, O2 and H2O concentrations of
the bulk gas mixture and particle diameter. The evolution of
hydroxylation and carbonyl functions produced was quantified
by FTIR-spectroscopy of filtered particles after defining an index
of oxidation (Iox(fg)). Functionalization could not exceed a given
value independent of the particle diameter (e.g. Iox(C=O) ≤ 1.6),
most probably due to the oxidative degradation of the polymer
particles. Oxidative degradation of the polymer material, leading
to a diminution of particle size and concentration, occurred in
parallel, and optimum conditions for a given degree of oxidative
functionalization at minimum alteration of the morphologic
characteristics of the particles would have to be found. The same
functionalization, but with much less efficiency, was also obtained
by exposing the aerosol to O3. This thermal oxidation allows a
controlled functionalization to smaller values of Iox(fg) without
notable oxidative degradation. VUV-photolysis of O2producing
O3in a defined flux of synthetic air was used as an actinometer
to relate the electric power of the Xe2-excimer lamp (Pe) to its
emitted radiant power (P0,172), and, based on an approximation
of the scattering cross-section for the wavelength of the incident
radiation (172 nm), the rate of photons absorbed by the particles
was evaluated for particle diameters ≤ 100 nm.
The authors acknowledge financial support by the Deutsche
Forschungsgemeinschaft (DFG) and the Landesstiftung Baden-
1 J. Fu, J. Fiegel and J. Hanes, Synthesis and characterization of PEG
based ether-anhydride terpolymers: New polymers for controlled drug
delivery, Macromolecules, 2004, 37, 7174–7180.
2 See, for example: J.-Y. Kim, J. Wainaina, J.-H. Kim and J.-K. Shim,
Use of polymer nanoparticles as functional nano-absorbents for low-
molecular weight hydrophobic pollutants, J. Nanosci. Nanotechnol.,
2007, 7, 4000–4004.
3 S. Saliterman, Fundamentals of BioMEMS and Medical Microdevices,
SPIE Press, Bellingham WA, 98225–6705, USA, 2006.
4 See, for example: I. Willner and B. Willner, Functional nanoparticle ar-
chitectures for sensoric, optoelectronic and bioelectronic applications,
Pure Appl. Chem., 2002, 74, 1773–1783.
5 See, for example: R. Jordan, N. West, A. Ulman, Y.-M. Chou
and O. Nuyken, Nanocomposites by surface-initiated living cationic
polymerization of 2-oxazolines on functionalized gold nanoparticles,
Macromolecules, 2001, 34, 1606–1611.
6 See, for example: C. Pichot, Position paper: functional polymer latexes,
Polym. Adv. Technol., 1995, 6, 427–434.
7 See, for example: T. Dey, Polymer-coated magnetic nanoparticles: sur-
face modification and end-functionalization, J. Nanosci. Nanotechnol.,
2006, 6, 2479–2483.
8 (a) R. Pearson and G. Langer, Generation of aerosols by vapour-
phase polymerization of methyl acrylate, Nature, 1960, 187, 235; (b) R.
Partch, E. Matijevi, A. W. Hodgson and B. E. Aiken, Preparation
of polymer colloids by chemical reactions in aerosols. I. Poly,(p-
tertiarybutylstyrene), J. Polym. Sci., Polym. Chem. Ed., 1983, 21, 961;
(c) C. Esen and G. Schweiger, Preparation of Monodisperse Polymer
This journal is © The Royal Society of Chemistry and Owner Societies 2009Photochem. Photobiol. Sci., 2009, 8, 944–952 | 951
Particles by Photopolymerization, J. Colloid Interface Sci., 1996, 179, Download full-text
276–280; (d) H. Morita, K. Semba, Z. Bastl, J. Subrt and J. Pola,
N2 laser-induced formation of copolymeric ultrafine particles in a
gaseous tetraethenylgermane-carbon disulfide mixture, J. Photochem.
Photobiol., A, 2005, 171, 21–26.
9 D. L¨ osch, V. Seidl and M. R¨ ubenacker, Process for production
of polymer powders containing vinylformamide units, BASF AG,
Publ. Nr. WO/2007/054442, Intl. Appl. No.: PCT/EP2006/067866,
10 A. Hozumi, H. Inagaki and T. Kameyama, The hydrophilization of
polystyrene substrates by 172 nm vacuum ultraviolet light, J. Colloid
Interface Sci., 2004, 278, 383–392.
11 V. M. Graubner, R. Jordan, O. Nuyken, B. Schnyder, T. Lippert, R.
K¨ otz and A. Wokaun, Photochemical modification of cross-linked
poly(dimethylsiloxane) by irradiation at 172 nmn, Macromolecules,
2004, 37, 5936–5943.
12 J. L´ opez Gejo, Applications of the VUV-photochemically initiated
oxidation for waste gas treatment and surface functionalization, Dis-
sertation, Fakult¨ at f¨ ur Chemieingenieurwesen und Verfahrenstechnik,
Universit¨ at Karlsruhe, Karlsruhe, 2005.
13 J. L´ opez Gejo, H. Glieman, T. Schimmel and A. M. Braun, Vacuum-
ultraviolet photochemically initiated modification of polystyrene sur-
faces: chemical changes, Photochem. Photobiol., 2005, 81, 777–782.
14 J. L´ opez Gejo, N. P. Manoj, S. Sumalekshmy, H. Gliemann, T. Schim-
mel, M. W¨ orner and A. M. Braun, Vacuum-ultraviolet photochem-
ically initiated modification of polystyrene surfaces: morphological
changes and mechanistic investigations, Photochem. Photobiol. Sci.,
2006, 5, 948–954.
15 See, for example: V. V. Ivanov, N. A. Popov, O. V. Proshina, T. V.
Rakhimova, G. B. Rulev and V. B. Saenko, Study of the ozone
production and loss during oxygen photolysis in a VUV-ozonator
chamber, Tech. Phys. Lett., 2001, 27, 29–31.
16 C. S. Foote, J. S. Valentine and A. Greenberg, Active Oxygen in
Chemistry, Springer, New York, 1995.
17 R. Wrobel, W. Sander, E. Kraka and D. Cremer, Reactions of dimethyl
ether with atomic oxygen: a matrix isolation and a quantum chemical
study, J. Phys. Chem. A, 1999, 103, 3693–3705.
18 M. C. Gonzalez, E. Oliveros, M. W¨ orner and A. M. Braun, Vacuum
ultraviolet photolysis of aqueous reaction systems, J. Photochem.
Photobiol., C, 2004, 5, 225–246.
19 M.Zahorodna,E.Oliveros,M.W¨ orner,R.BogoczekandA.M.Braun,
Dissolution and mineralization of ion exchange resins: differentiation
between heterogeneous and homogeneous (photo-)Fenton processes,
Photochem. Photobiol. Sci., 2008, 7, 1480–1492.
20 N. Quici, M. I. Litter, A. M. Braun and E. Oliveros, Vacuum-UV
photolysis of aqueous solutions of citric and gallic acids, J. Photochem.
Photobiol., A, 2008, 197, 306–312.
21 W. Kern, Photochemical modification of polymer surfaces, Trends
Photochem. Photobiol., 2001, 7, 11–30.
22 D. Feldman, Polymeric Building Materials, Taylor & Francis, Rout-
ledge, New York, 1989.
23 G. A. Olah and´A. Moln´ ar, Hydrocarbon Chemistry, Wiley, New York,
24 V. D. Komissarov and I. N. Komissarova, Formation of phenol
during ozonolysis of benzene, Russ. Chem. Bull., 1973, 22, 656–
25 J. Salas, Vicente. Modificati´ on fotoqu´ ımica de submicropart´ ıculas
en aerosoles, Doctoral thesis, Universidad Polit´ ecnica de Valencia,
Universit¨ at Karlsruhe).
26 K. R. May, The Collison nebulizer: description, performance and
application, J. Aerosol Sci., 1973, 4, 235–238.
27 Zs. Laszl´ o, I. Ilisz, G. Peintler and A. Dombl, VUV intensity measure-
ment of a 172 nm Xe excimer lamp by means of oxygen actinometry,
Ozone: Sci. Eng., 1998, 20, 421–432.
28 H. J. Kuhn, S. E. Braslavsky and R. Schmidt, Chemical actinometry,
Pure Appl. Chem., 2004, 76, 2105–2146.
of-the-art measurements of ozone in the gas phase and in solution,
Ozone: Sci. Eng., 1988, 10, 353–365; (b) K. Rakness, G. Gordon, B.
Langlais, W. Masschelein, N. Matsumoto, Y. Richard, C. M. Robson
and I. Somiya, Guideline for measurement of ozone concentration in
the process gas from an ozone generator, Ozone: Sci. Eng., 1996, 18,
30 (a) A. M. Braun, M.-T. Maurette and E. Oliveros, Technologie
photochimique, Presses Polytechniques Romandes, Lausanne, 1986,
Chapter 6.11; (b) A. M. Braun, M.-T. Maurette and E. Oliveros,
Photochemical Technology, transl. D. F. Ollis and N. Serpone, Wiley,
Chichester, 1991, Chapter 6.11.
31 Z. Falkenstein, Ozone formation with (V)UV-enhanced dielectric
barrier discharges in dry and humid gas mixtures of O2, N2/O2, and
Ar/O2, Ozone: Sci. Eng., 1999, 21, 583–603.
32 A. M. Braun, I. Gassiot Pintori, H.-P. Popp, Y. Wakahata and M.
W¨ orner, Technical development of UV-C- and VUV-photochemically
induced oxidative degradation processes, Water Sci. Technol., 2004, 49,
33 J. Salas Vicente, J. L´ opez Gejo, S. Rothenbacher, S. Sarojiniamma, E.
Gogritchiani, M. A. Miranda, M. W¨ orner, E. Oliveros, G. Kasper and
A. M. Braun, VUV-photochemical functionalization of polystyrene
aerosols, in Basics and applications of photopolymerization reactions,
ed. J. P. Fouassier and X. Allonas, Research Signpost, Trivandrum,
34 A. J. Cox, A. J. DeWeerd and J. Linden, An experiment to measure
Mie and Rayleigh total scattering cross sections, Am. J. Phys., 2002, 70,
36 J. Salas Vicente, J. L´ opez Gejo, S. Rothenbacher, M. W¨ orner, E.
aerosol particles by photooxychlorination, Appl. Mater. Interface, in
37 G. V. Lubarsky, M. R. Davidson and R. H. Bradley, Characteri-
zation of polystyrene microspheres surface-modified using a novel
UV-ozone/fluidized- bed reactor, Appl. Surf. Sci., 2004, 227, 268–
952 | Photochem. Photobiol. Sci., 2009, 8, 944–952 This journal is © The Royal Society of Chemistry and Owner Societies 2009