WEFTEC® 2003

Refractory COD removal in the Chemical Industry:

Technico-Economic comparison of Advanced Oxidation Processes

M. Coste*, C. Batbedat*, C Feliers*, D. Olejnik*, J.Cigana *, P. Cervantes*

*: Anjou-Recherche / Vivendi Water

Chemin de la Digue, B.P. 76, 78603 Maisons-Laffitte, France

ABSTRACT

Advanced Oxidation Process (AOP) comparison was carried out on an effluent coming from
cosmetic industry. Refractory COD was obtained after a complete biological treatment by
membrane bioreactor (BIOSEP). Six oxidation treatments were compared for COD
elimination: Fenton reagent, direct photolysis with a low pressure UV lamp, O3 at acidic pH
and pH 6.25, UV/O3, UV/H2O2, O3/H2O2, O3/UV/H2O2. The results obtained seem to indicate
that removal of COD of this specific solution was mainly due to radical mechanism.
However, even if free radical are necessary, the best results are obtained with processes
involving UV irradiation. Moreover, Fenton reagent offered a maximum COD elimination of
80% whatever the peroxide concentration, presenting a fraction that may be more refractory
toward free radical oxidation perhaps due to scavengers competition. Therefore, the
components of the treated solution were removable by radical mechanism but a fraction
needed UV photolysis to be completely oxidized. A comparison of the operating cost of these
processes shows that for a COD removal of more than 80% the catalytic UV treatment of this
low load industrial wastewater was the most efficient and the most cost effective.

KEY WORDS

Refractory COD, Industrial wastewater, Advanced oxidation processes, ozone, Fenton, UV

INTRODUCTION

Refractory COD is one of the most frequent and crucial problems found in industrial
wastewater treatment and more specifically in the Chemical Industry. Biological processes are
often unable to degrade complex compounds that can be found in such effluents, resulting in
an refractory chemical organic demand (COD) fraction after treatment. Advanced oxidation
processes (AOP) are one of the many possible treatments available (incineration, evapo-
concentration, separation, etc..) for this refractory COD (Luck and al., 2000 – G Leboucher
and al. 2000). The objective of this study is to suggest a methodology of side-by-side testing
of AOP treatment in order to select the optimum treatment process. Although the
methodology involved in general, this article will concentrate on the results generated from
the comparison of several AOP processes on refractory COD removal of an effluent from
Cosmetic Industry. An economic comparison of the different processes tested has been
conducted.

Copyright ©2003 Water Environment Federation. All Rights Reserved.

 WEFTEC® 2003

These tests were conducted on bench scale plant in laboratory. The tested AOP processes
were:
ƒ FENTON: Hydrogen peroxide in presence of Ferrous iron
ƒ Hydrogen peroxide with UV
ƒ Ozone in presence of UV and/or hydrogen peroxide

FENTON
In FENTON process the oxidative action of hydrogen peroxide is enhanced by adding ferrous
iron (Fe(II)), to generate the OH° radical.
Fe2+ + H2O2 → Fe3+ + OH- + OH°
The organic components RH reacts with the radicals
RH + OH° → R° + H2O
R° + H2O2 → ROH + OH°
Ferric iron formed is then reduced to produce other OH° radicals
Fe3+ + R° → R+ + Fe2+
The optimum pH of the reaction is between 3 and 6. The temperature, peroxide and iron rate
are the parameter to be optimized. The main drawback of this process is the generation of
sludges due to the precipitation of ferric hydroxide.

Hydrogen peroxide/UV (Luck and al., 1999)

Hydrogen peroxide, at wavelength below 360 nm, generates OH° radicals by homolysis of the
O-O bond.
Initiation reaction : H2O2 + hv → 2OH°
Propagation reactions : OH° + H2O2 → HO2° + H2O
H2O2 + H2O → HO2- + H3O+
OH° + HO2- → O2- + H2O

The presence of organic or mineral components, absorbing in the same range of wavelength,
in the solution influences the decomposition kinetics of H2O2.

Ozone/UV or H2O2 (HOIGNE J. and al, 1998)

There are 3 methods to increase the production rate of OH° in presence of ozone :
1. pH elevation
The OH- initiate the in chain mechanism of radicals generation.
3. Addition of hydrogen peroxide
The presence of H2O2 in. The reactions involved are the following ones

The stoichiometry of the global reaction of decomposition of O3 by H2O2 is in theory of 2

moles of dissolved O3 for 1 mole of H2O2 (meaning 0.35 gram of H2O2 per gram of O3) to
reach an optimal consumption of the oxidizing agents.

Copyright ©2003 Water Environment Federation. All Rights Reserved.

 WEFTEC® 2003

At basic pH the decomposition of O3 by H2O2 is accelerated due to the higher fraction of

H2O2 under HO2- form. However, when pH is above 8.2, the efficiency of the process
decreases dramatically. The optimum pH is situated between 7.5 et 8.

3. UV irradiation
UV irradiation allows the transformation of ozone in hydrogen peroxide. Photolysis of O3 by
UV radiation below 300nm produces dioxygen and 1D oxygen atom. This O(1D) reacts with
H2O to produce H2O2. This produced H2O2 can be photolysed or decomposed by O3, as
followed.

2OH°
UV
H2O2
O3
OH° + O2- + O2

MATERIALS AND METHODOLOGY

ANALYSIS
The characterization of the treated wastewater coming from the industrial site (outlet of
membrane bioreactor treatment) is presented in the following table :
COD pH BOD5 Absorbance 254 nm TOC
190 mg/L 6.25 11 mg/L 0.55 62 mg/L

P-PO4 N-NH4 NO3 Cl Alkalinity

(mg CaCO3/L)
1 mg/L 10.2 mg/L 2.6 mg/L 263 mg/L 62 mg/L
Table 1 : Characterization of the wastewater
It can be seen from the Table 1 that BOD5 level are extremely low implying that the
remaining COD is essentially refractory COD.

FENTON PROCESS
The FENTON reaction is realized in a 1L stirred reactor. The pH is ajusted at 3.0 with H2SO4.
Fe(II) is added as FeSO4, before hydrogen peroxide. Tests are conducted in a thermostated
bath (Ambiant temperature and 50°C).
At the end of the test COD, COT, H2O2 residual and the quantity of sludge produced are
analyzed.

UV IRRADIATION
Before performing any tests, the UV output of the lamp was determined by actinometric
measurement, using Potassium Ferrioxalate and photolysis of hydrogen peroxide (6). The
output was 2.9 W at 254 nm, presenting an efficiency of the lamp close to 20% (UV output/
Electrical Power).
• The photolysis tests are carried out in a batch reactor, magnetically stirred, of 1 L capacity
. The lamp is installed in a quartz shield.

Copyright ©2003 Water Environment Federation. All Rights Reserved.

 WEFTEC® 2003

• The recirculation flow is 40 L/h.

• A heat exchanger avoids the heating of the solution

OZONE

Venting
outlet
1,2 m

O2

Ozone OZONE
H2O
Analyser Generator
O3

Volumetric
counter
Ozone analyser

Thermal destruction

Recirculation pump

Figure 1: Ozonation bench scale plant

The volume of the ozonation column is 5 L and 1.2 m high. The ozone is continuously
injected, and the solution is recycled (120 l/h).
Ozone concentration is analyzed continuously in gaseous phase (inlet and outlet) and a
volumetric counter determines the volume of gas passed during the test to calculate the
quantity of ozone transfered. Samplings are taken during the ozonation test. At the end of the
test, COT, COD and residual H2O2 are analyzed
The definition and the procedure used for data handling are the following:
- Treatment Rate (TR)
g of ozone transfered
TR=
g of initial COD

- Molar ratio R
ozone moles
R=
hydrogen peroxide moles

- Transfert rate of ozone (Tf): %

ozone inlet − ozone outlet

Tf = : *100
ozone inlet

Copyright ©2003 Water Environment Federation. All Rights Reserved.

 WEFTEC® 2003

Selection of a Selection of a
Transfer Rate molar Ratio

TEST

Residual peroxyde
measurement

measurement
Transfer

Efficiency determination
vs real TR and R

Figure 2: OPERATING MODE

RESULTS AND DISCUSSION

FENTON
The operation conditions were: A temperature of 50°C during 2h at a pH of 3,5 and a mass
ratio H2O2/Fe of 5. The following figure presents COD removal vs [H2O2] with a constant
H2O2/Fe ratio.

100

90

80

70
COD removal (%)

60

50

40

30

20

10

0
0 0,05 0,1 0,15 0,2 0,25
[H 2O 2] (m ol/L)

Figure 3: Influence of H2O2 concentration

Copyright ©2003 Water Environment Federation. All Rights Reserved.

 WEFTEC® 2003

As can be observed in Figure 2, a COD removal of 80% is reached at 0,05M of H2O2.

When [H2O2] is above 0,05Mmol/L a plateau for COD removal is reached. The presence of
radicals traps could explain this plateau. At this level of COD concentration the components
responsible of COD are not the most reactive with OH°.
In order to achieve a 70% COD removal rate, 3 g of H2O2 per g of COD, need to be added.

The next figure presents COD removal vs time with several ratio H2O2/Fe and for an
equivalent initial concentration of H2O2 (0,01 mol/L).

80 180

160
70

140
60

120
COD removal (%)

50

[H2O2] (mg/L)
100

40
80

30
60
R = 5; COD

20 R = 10; COD
40
R = 25; COD

10 20 R = 5; H2O2

R = 10; H2O2

0 0 R = 25; H2O2
0 1 2 3 4 5 6
time (h)

Figure 4: Fe influence on the kinetic of the reaction

Iron influences the kinetic of the reaction while the final COD removal efficiency is driven by
the H2O2 concentration. A high load of iron allows a rapid reaction but produces higher
amounts of sludges. There is an optimum solution to be found.

Figure 5: Sludge production with H2O2/Fe ratio of 5, 10 and 25

Copyright ©2003 Water Environment Federation. All Rights Reserved.

 WEFTEC® 2003

Weight ratio H2O2/Fe Volumic quantity of sludges

(%)
5 1.6
10 1.2
25 0.8
Table 2: Sludges produced

UV Photolysis

This figure presents COD removal and absorbance for different wavelengths vs time.
40 3

35
2,5

30

2
COD removal (%)

25

Absorbance
COD removal (%)
20 1,5 Absorbance at 254 nm
Maximum absorbance

15
1

10

0,5
5

0 0
0 100 200 300 400 500
time (min)

Figure 6: UV photolysis

Simple UV irradiation has an action on the pollutants: 35% of COD removal is reached in 6h
of irradiation.
The maximum of absorbance is around 196nm. This seams to indicate that a better results can
be reached with another lamp than with the low pressure lamp (max at 254nm).

Copyright ©2003 Water Environment Federation. All Rights Reserved.

 WEFTEC® 2003

UV/H2O2 PHOTOLYSIS

This figure presents COD removal and [H2O2] vs time.

100 400

90
350
80
300
70
COD removal (%)

250

[H2O2] (mg/L)
60

50 200

40
150
30
100 COD removal (%)
20
H2O2 (gl/l)
50
10

0 0
0 1 2 3 4 5
temps (h)

Figure 7: UV /H2O2 (0.01 mol/L) process

Radicals are quite efficient because a COD removal of 95% is reached in 2h. A combination
of the action of radicals and direct photolysis can be seen through this data.

OZONATION

All the tests are conducted in the same operation condition: the gaseous flow rate is 20.5 L/h
and the ozone concentration in this gaseous phase is 100g/Nm3.
The two following figures respectively present the COD removal and the ozone transfer
during time for several ozonation processes (O3, O3/H2O2, O3/UV, O3/UV/H2O2).

Copyright ©2003 Water Environment Federation. All Rights Reserved.

 WEFTEC® 2003

100,0
COT = 8,2 mg/L

90,0 COT = 8 mg/L

80,0
COT = 30 mg/L

70,0
COT = 33 mg/L
COD removal (%)

60,0
COT = 29 mg/L

50,0
O3
O3 + UV
40,0 O3 + H2O2 =0,001 mol/L
O3 + H2O2 =0,006 mol/L
30,0 O3 acid pH 2,26
O3/UV/H2O2 = 0,001 mol/L
20,0
COT = 60,4 mg/L
10,0

0,0
0 10 20 30 40 50 60 70 80 90
time (min)

Figure 8: COD removal comparison for each AOP’s vs time

100

90

80
O3
70 O3 + UV

O3 + H2O2 =0,001 mol/L

60
Transfer (%)

O3 + H2O2 =0,006 mol/L

50 O3 acid pH 2,26

O3/UV/H2O2 = 0,001 mol/L

40

30

20

10

0
0 10 20 30 40 50 60 70 80 90
time (min)

Figure 9: Ozone transfer in each of the AOP’s vs time

Copyright ©2003 Water Environment Federation. All Rights Reserved.

 WEFTEC® 2003

The first test is realized at low pH to estimate the effect of molecular ozone (the main form
present in this range of pH) on the organics. It appears that its effect on COD removal is very
low. At this effluent pH (6,3), O3 and radicals are present in solution that’s why the efficiency
of the treatment is higher for COD removal. When O3 is combined with H2O2 or UV results
are always better except in the case with 0,006 mol/L of H2O2. This could be caused by an
excess of H2O2 which is decomposing the produced radicals which are not available anymore
for organics oxidation. The ozone is then consumed, explaining the good transfer rate, but not
to react with organics, explaining the loss of efficiency in COD removal. With a lower H2O2
concentration the efficiency is higher, because there is no more excess of H2O2 to decompose
the radicals produced. Otherwise, as the decreasing of the transfer rate during the test shows
it, the H2O2 is quickly consumed and the end of the test is probably simple ozonation. The
addition of UV to ozone gives very good results due to a continuous production of radicals,
without any excess of chemicals to decompose them, and the direct effect of photolysis. The
following figure condenses these observations.

1 0 0 ,0

9 0 ,0

8 0 ,0

7 0 ,0
COD removal (%)

O3
6 0 ,0
O3 + UV
5 0 ,0
O 3 + H 2 O 2 = 0 ,0 0 1 m o l/L
4 0 ,0

O 3 + H 2 O 2 = 0 ,0 0 6 m o l/L
3 0 ,0

2 0 ,0 O 3 a cid p H 2 ,2 6

1 0 ,0 O 3 /U V /H 2 O 2 = 0 ,0 0 1
m o l/L

0 ,0
0 ,0 1 ,0 2 ,0 3 ,0 4 ,0 5 ,0
O 3 tra n s fe re d g O 3 /g D C O

Figure 10: COD removal vs ozone transfered

Copyright ©2003 Water Environment Federation. All Rights Reserved.

 WEFTEC® 2003

COST ESTIMATE

This cost estimate includes only operation cost du to chemicals (H2O2, FeSO4, H2SO4, NaOH
(to re-adjust the pH after the acid tests), oxygen (ozone generation, considering a 10% rate of
ozone production from O2 and 90% of ozone transfered on a full scale plant) and energy (UV,
considering a lamp efficiency of 30%, and ozone generation)

Comparative operating costs

1,80
1,60
1,40
Cost (€/m3

1,20
1,00 70% COD removal
0,80 80% COD removal
0,60
0,40
0,20
0,00
2
N

2
)

)
M

2O
O

2O
TO

01

06

H
H
N

,0

,0

V/
V/
FE

(0

(0

U
U

3/
2

2
2O

2O

O
H

H
3/

3/
O

Figure 11: Operation cost estimate

The comparison of the operating cost of these processes shows that for a COD removal of
more than 80% the catalytic UV treatment of such low loaded industrial wastewater is the
most cost effective

CONCLUSIONS

Comparison of AOP processes was carried out for the treatment of the effluent coming from a
cosmetic industry and biologically treated by a membrane bioreactor. The composition of the
effluent was: 190 mg.L-1 COD, 62 mg.L-1 TOC, 11 mg.L-1 BOD5, pH 6.25 and absorbance at
254 nm = 0.55. Six oxidation treatments were compared for COD elimination: Fenton reagent
,direct photolysis with a low pressure UV lamp, ozonation at acidic pH and pH 6.25,
UV/H2O2, ozone/H2O2, UV/Ozone, UV/Ozone/H2O2.
The results seems to prove that removal of COD of this specific solution was mainly due to
radical mechanism. However, the best results were reached with processes involving UV
irradiation. Moreover, Fenton reagent shown a maximum COD removal of 80% whatever the
peroxide concentration (up to 0.05 mol.L-1, H2O2/Fe(II) = 5), presenting a fraction that may be
more refractory toward free radical oxidation perhaps due to scavengers competition.
Therefore, the components of the solution treated are removable by radical mechanism but a
fraction needed UV photolysis to be completely oxidized. In these case, the catalytic UV
treatment of such low loaded industrial wastewater seems to be the most efficient process.

Copyright ©2003 Water Environment Federation. All Rights Reserved.

 WEFTEC® 2003

REFERENCES

F. Luck, C. de Traversay, M. Djafer (2000) The contribution of advanced oxidation

technologies towards a sustainable environment, Proceedings of the 3rd Asia Pacific
Conference on sustainable energy and environmental technologies, ASCSEET 2000, Hong
Kong (China)

G. Leboucher, Y. Moreau-Le Golvan, F. Ducray, A. Wagner, L. Schaegis, «Industrial

application of biofiltration followed by advanced oxidation for landfill leachate treatment :
Montreuil-sur-Barse (France) case study», IOA/Wasser – Berlin 2000 – Oct 23/26, 2000

F. Luck, C. de Traversay, “Used of advanced oxidation for the removal of pollutants from
drinking waters or wastewaters”, Seminar on applications of ozone for drinking and
wastewater treatments, IWSA/IOA/AIDIS/ Buenos Aires, Sept. 17, 1999

HOIGNE J. et al., “Chemistry of Aqueous ozone and transformation of pollutants by

ozonation and advanced oxidation processes”, The Handbook of Environmental chemistry,
1998

Copyright ©2003 Water Environment Federation. All Rights Reserved.