NUKLEONIKA 2007;52(2):69−75 ORIGINAL PAPER

Azo dye degradation by high-energy Erzsébet Takács,

László Wojnárovits,
irradiation: kinetics and mechanism Tamás Pálfi

of destruction

Abstract. The kinetics and mechanism of dye destruction is discussed on the example of Apollofix Red (AR-28) and
Reactive Black 5 (RB-5) radiolysis in dilute aqueous solution. The dose dependence of colour disappearance is linear
when the reactive intermediate reacts with the colour bearing part of the molecule causing destruction of the conjugation
here with nearly 100% efficiency. In this case, spectrophotometry can be used to follow-up dye decomposition. Such
a linear dependence was observed when hydrated electrons or hydrogen atoms reacted with the dye. However, in
the case of hydroxyl radical reactions some coloured products form with absorption spectra very similar to those of the
starting dye molecules. For that reason, spectrophotometric measurements give false results about the concentration of
intact dye molecules. Analysis made by the HPLC (high-pressure liquid chromatography) method reveals logarithmic
time dependence in agreement with a theoretical model developed.

Key words: azo dyes • decolouration • reaction mechanism • kinetics • Apollofix Red • Reactive Black 5

Introduction

Azo dyes are produced in very large quantities and used

as reactive dyes by the cotton industry. They are very
stable and resistant to chemical processes. Therefore,
their removal from industrial effluents by conventional
methods (mechanical screening, sedimentation,
biological treatment, biofilters) alone is usually not very
effective. This problem can be solved by combining the
conventional techniques with other methods. Radiation
chemistry is one of the most promising advanced
oxidation processes (AOP’s) [3, 8, 9] for environmental
remediation. In AOP’s hydroxyl radicals are the main
oxidants for the degradation of organic pollutants [13].
The disappearance of the dye is generally followed by
spectrophotometry at the absorbance maximum between
450 and 600 nm.
The dye disappearance curves are very often
described by applying some kinetic equations [4, 6, 7,
10, 15, 17−19] in order to describe the dependence of
the destruction on such parameters as dose, dose rate,
E. Takács , L. Wojnárovits, T. Pálfi additives, etc. in radiolysis experiments, illumination
Institute of Isotopes, Hungarian Academy of Sciences, intensity, illumination time, etc. in photo oxidation
H-1525 Budapest, P. O. Box 77, Hungary, experiments [15].
Tel./Fax: +36 1 392 2548, For the present mechanistic and kinetic studies, the
E-mail: [email protected] experimental results obtained during radiolysis of
Apollofix Red (AR-28) and Reactive Black 5 (RB-5)
Received: 11 January 2007 are discussed. Some of the experimental results were
Accepted: 2 April 2007 already published [16, 20, 21].
70 E. Takács, L. Wojnárovits, T. Pálfi

Scheme 1. Structures of Reactive Black 5 (RB-5) and Apollofix Red (AR-28) dyes.

The colours of these dyes are determined by the The chromatographic system consisted of a Jasco
conjugated phenylsulphonic acid-azo group-H-acid PU-2089Plus quaternary gradient pump, a Jasco MD-
parts of the molecules [1] (see Scheme 1). Any change 2015Plus diode-array Multiwavelenght Detector and
in these parts leads to change of colour, i.e. change of a Nucleosil 100 C18 column (Teknokroma®) (pore size
the absorption spectrum in the visible range. Destruction 5 µm, length 15 cm, diameter 0.4 cm). Separations were
of the extensive conjugation leads to loss of colour, i.e. made using 50 mmol⋅dm−3 aqueous solution of TBAHS
disappearance of absorption in the visible range. at pH 6.1 and methanol. All mobile phases were filtered
using a Millipore 0.45 µm filter and were degassed by
ultrasonication before use. The separations were made
Experimental at room temperature using injection volume of 10 µl,
and flow-rate of 0.8 ml/min.
AR-28 was obtained from Taiheung Corporation Standard radiation chemical techniques were
(Kyunggido, South Korea), RB-5 was purchased from applied to separate the water radiolysis intermediates
Aldrich, the dyes were purified by recrystallization from and follow their reactions with the dye. The radiolysis
methanol/ethanol mixtures. Perchloric acid was used of water supplies hydrated electron (e−aq), hydroxyl
to set pH of the solutions when the •H atom reactions radical (•OH) and hydrogen atom (•H) reactive inter-
were studied. Methanol (HPLC grade) and tetrabutyl- mediates with G values (species (100 eV)−1) shown in
ammonium-hydrogen sulphate (TBAHS) were obtained Eq. (1). In order to reduce the number of reactive
from Sigma-Aldrich, tert-butanol from Spektrum-3D. intermediates, the reactions of •OH radicals were
The gamma radiolytic experiments were carried out studied in N2O saturated solutions (pH 5−6), in such
using a 60Co irradiation facility with a dose rate of solutions the hydrated electrons are converted to •OH
3.0 kGy⋅h−1 as determined by Fricke dosimetry (G(Fe2+) radicals (Eq. (2)). When the e −aq reactions were
= 15.6 (100 eV)−1). The solutions were bubbled with investigated the solutions were bubbled with N2 for
N2 or N2O for 5 min prior to irradiation in 5 ml Pyrex deoxygenation and contained 0.5 mol⋅dm−3 tert-butanol
glass ampoules. 1 cm Suprasil quartz optical cells were in order to transform the •OH radicals to less reactive
•
used for taking the UV-VIS absorption spectra before CH2(CH3)2COH radicals (Eq. (3)) (pH 5−6). H• atoms
and after the reaction by using a Jasco 550 UV-VIS react with a low rate coefficient with tert-butanol
spectrophotometer. (Eq. (4)). The H• reactions were studied in t-butanol
Azo dye degradation by high-energy irradiation: kinetics and mechanism of destruction 71

containing N 2 purged pH 1.1 solutions, where e −aq (4) H + (CH3)3COH → H2O + •CH2(CH3)2COH
was converted to H• atom (Eq. (5)) [2]. k4 = 1 × 106 mol−1⋅dm3⋅s−1

(1) H2O − /\/ → e−aq, •OH, •H (5) e−aq + H+ → •H k5 = 2.3 × 1010 mol−1⋅dm3⋅s−1
G(e−aq) = 2.6; G(•OH) = 2.7; G(•H) = 0.6

(2) e−aq + N2O + H2O → N2 + OH− + •OH Results and discussion

k2 = 9.1 × 109 mol−1⋅dm3⋅s−1
Degradation of AR-28 and RB-5
(3) •OH + (CH3)3COH → H2O + •CH2(CH3)2COH
k3 = 6 × 108 mol−1⋅dm3⋅s−1 Figure 1 shows the comparison of the decolouration of
AR-28 and RB-5 in the reaction with •OH (1A, 1B)

Fig. 1. Decolouration plots of AR-28 (0.085 mmol⋅dm−3) and RB-5 (0.1 mmol⋅dm−3) obtained when the hydroxyl radicals
reacted with the dye (A, B), or when the hydrated electrons were the reaction partners (C, D).
72 E. Takács, L. Wojnárovits, T. Pálfi

and with the azo group subparts of the molecule [12,

14]. The addition of •OH radical to an aromatic ring on
the conjugated part of the molecule results in a cyclo-
hexadienyl type radical. This radical may disappear in
radical-radical combination reaction where dimer
molecule is produced (in this reaction the conjugation
may be destroyed). Radical disproportionation leads
to compounds that contain cyclohexadiene ring and
another compound with extensive conjugation in
the molecule characteristic of the dye. However, the
latter molecule contains an extra OH group in
the conjugated part of the molecule (see Scheme 2).
For that reason, its absorption spectrum is slightly
different from that of the starting molecule. This shift
in the spectrum certainly affects the molar extinction
coefficient, and, therefore, any kinetic analysis based
Fig. 2. Comparison of the absorption spectra of AR-28 and on the absorbance measurements should be handled
one of its main transformation products formed in •OH radical with a certain caution. The correct analysis should be
reaction. The spectra were recorded by using diode-array based on chemical separation and individual compound
detection after HPLC separation. analysis (most conveniently by the HPLC method).
The hydrated electron (and also the H• atom) is
and e−aq (1C, 1D), respectively. When •OH radicals suggested to react with the azo group (with rate coeffi-
reacted with the dye, during the decolouration the shape cients close to the diffusion controlled limit [5, 11])
of the spectra in the visible range changed: the visible destroying the conjugation with nearly 100% efficiency
bands shifted to longer wavelengths. After separation [20, 21]. The other parts of the molecule and the products
of the main product formed from AR-28 by HPLC, we formed during decay have smaller reactivity with e−aq
found that its spectrum is very similar to the spectrum of and H• atom than the azo group.
the starting dye. The only difference is the shift of the
visible band to longer wavelength (Fig. 2) and this causes
the change in the shape of the UV-VIS spectra (Fig. 1A). Kinetic analysis
The initial G values (extrapolated to zero dose) of dye
disappearance were 2.92 for RB 5 and 2.5 for AR-28. Under steady state conditions, the • OH, H •, e −aq
When e−aq reacted with the dyes (Figs. 1C and 1D), intermediates that destroy the dye molecules may react
the intensity of the visible absorption decreased with the practically exclusively with the intact dye molecules
radiation dose without any major change in the shape decolourizing them. The time dependence of dye
of the spectrum in the visible range for AR-28. In the disappearance can be described by the differential
case of RB-5 there is no change in the spectrum above equation:
ca. 550 nm. The initial G values (extrapolated to zero
d[Dye]
dose) of dye disappearance were 1.91 for RB 5 and 2.6 (6) − = k[I][Dye]
for AR-28. dt
In the equation [Dye] represents the dye concentration,
[I] is the intermediate radical concentration, and k is
Mechanism of degradation the second order rate coefficient.
During steady-state radiolysis, the intermediate
As it was discussed in our former papers [20, 21] •OH radicals form with a constant rate (rI) and they decay
radical has very high reactivity with the aromatic ring with the same rate (k[I] [Dye]):

Scheme 2. Degradation of AR-28 by •OH radical.

Azo dye degradation by high-energy irradiation: kinetics and mechanism of destruction 73

Fig. 3. Dose dependence of decolouration of AR-28 and Fig. 4. Dose dependence of decolouration of AR-28 and
RB-5 caused by hydrated electron. RB-5 caused by •H atom.

the time dependence curve at higher depletion deviates

d[I]
(7) = rI − k[I][Dye] ≈ 0 from linearity.
dt The reaction of •OH radical with the dye is different
from the reactions of e−aq and H•. The •OH radical reacts
At the beginning of the process, when the dye concen- with practically diffusion limited rate coefficient (k ≈
tration is high, the intermediate radicals quickly react 1010 mol−1⋅dm3⋅s−1) with the aromatic rings and probably
with the solute molecules, and therefore, the also with the –N=N– double bond. The dye molecules
intermediate radical concentration is low. At the end generally have several places for the attack, e.g. AR-28
of the process, [Dye] is low and k[I] is high. Substituting has five aromatic rings and one –N=N– double bond.
[I] ≈ rI/k [Dye] into Eq. (6), after integration, we obtain The reaction with the –N=N– double bond or with the
linear time dependence (since the irradiation is usually rings involved in the extensive conjugation leads
carried with a constant dose rate, there is linear dose to the destruction of the conjugation, so may decolourize
dependence): the molecule. However, the new products have also
high, practically the same reactivity towards the •OH
[Dye] rI
(8) = 1− t = 1 − kobs t radicals as the starting compound. That is probably true
[Dye]0 [Dye]0 for the product of the product, i.e. for the secondary
product, tertiary product, etc. For that reason, the rate
Such a linear dependence was observed several times equations should be written in the following form:
during decolouration at the beginning part of the
d[Dye]
process. We found linear initial dose dependence, when (9) − = k[ • OH][Dye]
in radiolysis the hydrated electrons or the hydrogen dt
atoms reacted with the dyes (Figs. 3 and 4). As we d[ • OH]
mentioned before, these intermediates were found to (10) = rOH − k[Dye][ • OH]
react principally with the azo group causing destruction dt
of the intensive conjugation in the molecule through − k[Product][ • OH] ≈ 0
the azo bond. In AR-28 there is only one azo group
with high reactivity, and here the linearity is maintained Since − as it was mentioned before − the reaction of
•
till about 80% destruction. In the RB-5 there are two OH with both the dye and the products is practically
azo groups. When e−aq or •H destroys one of them, due diffusion limited; we use for both reactions the same k
to the decreased conjugation, the longest wavelength rate coefficient. Applying the simplification:
absorption disappears. However, the other azo group
may also react with high rate coefficient with e−aq or (11) [Product] ≈ [Dye]0 – [Dye]
•
H and, therefore, the linear range is shorter. The
formation rates of reactive intermediates at a dose rate also taking [•OH] ≈ rOH/k[Dye]0 expressed from formulae
of 3 kGy⋅h−1, calculated from the slopes of the straight (6) and (7) (I = •OH), the differential equation has the
lines are rI,measured ≈ (2.0−2.5) × 10−7 mol⋅s−1 for both e−aq form:
and H•. The value is just slightly lower than the value
d[Dye] [Dye]rOH
calculated based on the radiation chemical yield of the (12) − =
intermediates: rI,calculated ≈ 2.8 × 10−7 mol⋅s−1. dt [Dye]0
At higher doses, however, the dye is depleted; After integration, we obtain logarithmic time depend-
therefore, the e −aq and H • intermediates of water ence (Eq. (13)) with rate parameter of kobs = rOH/[Dye]0.
radiolysis react mainly with the products of radiolysis
[Dye] r
and only a small part of the intermediates reacts with the (13) ln = − OH t = − kobs t
azo moiety causing decolouration. For that reason, [Dye]0 [Dye]0
74 E. Takács, L. Wojnárovits, T. Pálfi

products formed are less reactive towards e−aq and •H

atom. The dose dependence of colour disappearance
is linear when e −aq or •H atom reacts with the dye
molecules. In case of •OH reactions there is logarithmic
dose dependence of dye disappearance. In the paper
a kinetic model was proposed which described the
experimental results.

Acknowledgment We express our thanks to the Hungarian

Science Foundation (OTKA K 60 096). Support of the
International Atomic Energy Agency (contract No. 302-
F2-HUN-12015) is also acknowledged.

References

1. Bredereck K, Schumacher C (1993) Structure reactivity

Fig. 5. Dose dependence of decolouration caused by hydroxyl correlations of azo dyes based on H-acid. I NMR chemical
radicals, as measured by spectrophotometric evaluation, or shift values, pKa values, dyestuff aggregation and dyeing
by HPLC separation and evaluation. behaviour. Dyes Pigments 21:23−43
2. Buxton G, Greenstock CL, Helman WP, Ross AB (1988)
Critical review of rate constants for reactions of hydrated
electrons, hydrogen atoms and hydroxyl radicals (•OH/
By expanding the logarithmic term in series and keeping • −
O ) in aqueous solution. J Phys Chem Ref Data
only the first part Eq. (13) is transformed into Eq. (8) 17:513−886 (updated version: www.rcdc.nd.edu)
which means that at low conversion the dose depend- 3. Cooper WJ, Curry RD, O’Shea KE (eds) (1999) Environ-
ence is again linear. mental application of ionizing radiation. Wiley, New York
Figure 5 shows the ln [Dye]/[Dye]0 vs. dose plot 4. El-Assy NB, El-Wakeel E-SI, Fattah AAA (1991) The
obtained when the dyes were degraded by •OH radicals degradation of triazo dye Chlorantine Fast Green BLL
(N2O saturated solution). Good linear logarithmic plot in aqueous solution by gamma radiation-III. Appl Radiat
was obtained when the dye concentrations in the Isot 42:89−96
5. Flamigni L, Monti S (1985) Primary processes in the
irradiated solutions were determined by the HPLC
reduction of azo dyes in alcohols studied by pulse
method. The figure also shows the dose dependences radiolysis. J Phys Chem 89:3702−3707
obtained by spectrophotometric evaluation at the 6. Galindo C, Jacques P, Kalt A (2000) Photodegradation
absorbance maximum in AR-28 and RB-5 solutions. of aminoazobenzene Acid Orange 52 by three advanced
The strong positive deviations from the linearity, as it oxidation processes: UV/H2O2, UV/TiO2 and VIS/TiO2.
was mentioned before, may be attributed to reformation Comparative mechanistic and kinetic investigations. J
of the somewhat modified dye molecules. Photochem Photobiol 130:35−47
In Eq. (13) kobs = rOH/[Dye]0, i.e. it is the ratio of the 7. Galindo C, Kalt A (1999) UV/H2O2 oxidation of azo dyes
intermediate formation rate and the initial dye concen- in aqueous media: evidence of a structure-degradability
tration. The slope of the curve obtained by HPLC in relationship. Dyes Pigments 42:199−207
8. Han B, Ko JI, Kim Y et al. (2003) Electron beam
Fig. 5 is kobs = 6.4 kGy−1, or using the dose rate applied
treatment of textile dyeing wastewater. In: Proc of Symp
(3 kGy⋅h−1), 5.4 × 10−2 s−1. Then, the formation rate of on Radiation Technology in Emerging Industrial
•
OH radicals based on the dye disappearance is Applications, 6−10 November 2000, Beijing, China.
rOH,measured = 4.6 × 10−7 mol⋅s−1, the value agrees well with IAEA, Vienna, pp 194−199
the one calculated from the dose rate and the 9. Makarov IE, Ponomarev AV, Han B (2004)
known radiation chemical yield •OH radicals, rOH,calculated Demonstration plant for electron-beam treatment of
≈ 4.7 × 10−7 mol⋅s−1. The high value of rOH,measured indicates Taegu dye industry complex wastewater. In: Emerging
that the •OH radicals react with practically 100% applications of radiation processing. IAEA, Vienna.
efficiency with the dye molecules transforming them to IAEA-TECDOC-1386, pp 138−152
new, coloured, or non-coloured products. 10. Neamtu M, Siminiceanu I, Yediler A, Kettrup A (2002)
Kinetics of decolorization and mineralization of reactive
azo dyes in aqueous solution by the UV/H2O2 oxidation.
Dyes Pigments 53:93−99
Conclusions 11. Neta P, Levanon H (1977) Spectrophotometric study of
the radicals produced by the reduction of syn- and anti-
AR-28 and RB-5 can be effectively decomposed by •OH azobenzene. J Phys Chem 81:2288−2292
and e−aq intermediates of water radiolysis. •OH radicals 12. Panjakar MS, Mohan H (1993) Investigations of
add to the aromatic rings and as products of this transients produced on reactions of •OH radicals with
reaction hydroxylated dyes and probably quinine type azobenzene in aqueous solutions. Indian J Chem Sect A
molecules and dimers are formed. These products have 32:25−27
also high reactivity towards •OH radicals. e−aq and •H 13. Pera-Titus M, Garcia-Molina V, Banos MA, Gimenez J,
Esplugas S (2004) Degradation of chlorophenols by
atom attacks the azo group, instantly destroying
means of advanced oxidation processes: a general review.
the electron conjugation in the dye molecules and the Appl Catal, B 47:219−256
Azo dye degradation by high-energy irradiation: kinetics and mechanism of destruction 75

14. Roder M, Wojnárovits L, Földiák G, Emmi SS, Beggiato 18. Tezcanli-Guyer G, Ince NH (2004) Individual and
G, D’Angelantonio M (1999) Addition and elimination combined effects of ultrasound, ozone and UV
kinetics in OH radical induced oxidation of phenol and irradiation: a case study with textile dyes. Ultrason
cresols in acidic and alkaline solutions. Radiat Phys Chem Sonochem 42:603−609
54:475−479 19. Wang M, Yang R, Wang W, Shen Z, Bian S, Zhu Z (2006)
15. Shu H-Y, Chang M-C (2006) Development of rate Radiation-induced decomposition and discoloration of
expression for predicting decolorization of C.I. Acid Black reactive dyes in the presence of H2O2. Radiat Phys Chem
1 in a UV/H2O2 process. Dyes Pigments 70:31−37 75:286−291
16. Solpan D, Güven O, Takács E, Wojnárovits L, Dajka K 20. Wojnárovits L, Pálfi T, Takács E (2005) Reactions of azo
(2003) High-energy irradiation treatment of aqueous dyes in aqueous solution: Apollofix Red. Res Chem
solutions of azo dyes: steady state gamma radiolysis Intermed 31:679−690
experiments. Radiat Phys Chem 67:531−534 21. Wojnárovits L, Pálfi T, Takács E, Emmi SS (2005)
17. Tezcanli-Guyer G, Ince NH (2003) Degradation and Reactivity differences of hydroxyl radicals and hydrated
toxicity reduction of textile dyestuff by ultrasound. electrons in distracting azo dyes. Radiat Phys Chem
Ultrason Sonochem 10:235−240 74:239−246