2nd Mercosur Congress on Chemical Engineering

th
4 Mercosur Congress on Process Systems Engineering

TITANIUM CORROSION IN ALKALINE HYDROGEN PEROXIDE

BLEACHING ENVIRONMENTS
Elsa Rosa Ruiz1∗, Claudia Marcela Mendez2
1
Departamento de Ingeniería, FCEQyN, Universidad Nacional de Misiones.
2
Departamento de Fisicoquímica, FCEQyN, Universidad Nacional de Misiones.

Abstract. The pulp industry commonly uses chlorine-based solutions for pulp bleaching.
Environmental issues with chlorine and chlorine derivatives such as chlorine dioxide have pushed
the industry to examine alternative technologies. Hydrogen peroxide is an environmentally
attractive alternative and can be used for totally chlorine free bleaching or decrease chlorine based
bleach usage.
Pulp bleaching equipment is typically built from titanium to withstand the corrosive effects of
chlorine-based bleaches. In order for hydrogen peroxide to displace chlorine dioxide, titanium
corrosion resistance must extended to hot alkaline hydrogen peroxide.
Inhibitors and chelating agents are used in these pulp fabrication processes. Chelating agents such as
ethylenediaminetetraacetic acid (EDTA) and diaminetriethylenpentaacetic acid (DTPA) are added
to thwart accelerated hydrogen peroxide decomposition by complexing unwanted catalytic metal
ions of iron, copper, and others.
Potentiodynamic polarization and electrochemical impedance spectroscopy (EIS) were applied
during the dissolution of grade 2 titanium in solutions containing 3 g/l of H2O2. These experiments
were performed within pH= 9.5-11.5, temperature 68-80 ºC, frequency 10-3-105 Hz range, for 500
ppm of chelating agents (EDTA and DTPA) concentrations. The different time constants were
attained. The electrochemical impedance spectra were simulated according to a proposed equivalent
electric circuit and the calculated parameters were obtained, analyzed and discussed. This work
proves the influence of the EDTA and DTPA in the corrosion of Ti, the compatibility of emerging
peroxide bleaching processes with existing titanium bleach plant equipment.

Keywords: Peroxide bleaching processes, Titanium, Chelating agents.

1. Introduction

In the light of the continued pressure to reduce the use of chlorine dioxide for bleaching purposes in the
production of pulp and paper, due to the presumed environmentally hazardous by-product, dioxin, alternative
bleaching agents such as hydrogen peroxide are being considered. Titanium is one of materials which is selected
because of its typically high corrosion resistance in oxidizing chlorine environments and its wide use in the pulp
and paper industry. However, the technical literature reports high corrosion rates of titanium exposed to alkaline
hydrogen peroxide (Andreasson, 1995; Been and Tromans, 2000; Reichert, 1996; Schutz and Xiao, 1995; Stampella et al. ,
1999; Varjonen and Hakkarainen, 199; Wyllie II et al., 1995). The practical alloy corrosion performance windows are
0.2-0.4 % H2O2 solutions over the pH range of 9.5-12.5 and temperatures of 65-80ºC. The effects of various
natural ions Ca+2, in the in peroxide of the process are used, to which Mg+2 y SiO3= are added for them to act as
inhibitors, these were studied in a previous work (Ruiz, E.R. et al.,2003).
In this case it is important to mention that the pulp to be bleached by peroxide contains unwanted iron,
copper, nickel, manganese and molibdenum catalitic ions, in quantities depending on the type of wood.

∗
Laboratorio de Corrosión, FCEQyN, UNaM – Felix de Azara 1552, CP3300, Posadas, Misiones – Argentina.
E-mail: eruiz@fceqyn.unam.edu.ar
1
 2nd Mercosur Congress on Chemical Engineering
th
4 Mercosur Congress on Process Systems Engineering

Aminopolycarboxylic acids, such as ethylenediaminetetraacetic acid (EDTA) and diethylenediaminepentaacetic

acid (DTPA), are used extensively by the pulp and paper industry to form stable water-soluble chelates with
these metals (Bambrick, 1985; Poppius-Levlin, 1992). The chelating agents prevent the contact of metals and
hydrogen peroxide, since they reduce the catalytic decomposition of the H2O2. If of water cycles and total
chlorine-free processes that incorporate peroxide stages is closure the use of chelating agents can be expected to
increase significantly.
In this work, we studied the corrosion of Titanium by peroxide decomposition and by action of EDTA and
DTPA without the presence of catalytic metal ions.

2. Experimental method

The working electrodes were prepared from grade 2 comercially available titanium sheet of 2.74 cm2 of
exposed surface. The specimens were first polished with SiC papers of decreasing grain sizes down to grit 1200,
degreased with acetone in an ultrasonic cleaner, and finally washed with distilled water. All solutions were made
from H2O2 Interox 60-20 (Peróxidos do Brasil ltda.), EDTA and DTPA analytical grade reagents and distilled
water.
In these experiments the conditions were, pH: 9.5-11.5, temperature: 68 – 80ºC, and an hydrogen peroxide
solution of 3 g/l with this concentration and volume constants. It is important to remark that measures at pH=9,5
and T=68ºC were not considered as they appear as quite soft conditions for Ti. The initial concentration of
chelants EDTA and DTPA was 500 ppm selected for the corrosion rate of titanium to remain within acceptable
design values, 5 mpy (Stampella et al., 1999). The pH was regulated by the addition of sodium hydroxide. A pH
electrode continuously monitored the pH of the solution. If it were necessary, there sodium hydroxide adjustment
to the preset value. Samples were withdrawn at intervals and the amount of hydrogen peroxide was determined
by titration with 0.1 N sodium thiosulfate solution.
The electrode was tested in a conventional three-electrode electrochemical cell. A saturated sulphate
electrode (SSE; 640mV vs. SHE) as reference and a platinum counter electrode were employed. All
measurements were carried out in a Faraday cage. All the potentials given in this paper are referred to standard
hydrogen electrode.
The samples were catodized at 400 mV more negative than the open circuit potential, Eoc. At this potential,
Eoc, each specimen was maintained immersed during 1 h. The potentiodynamic anodic polarization
measurements were started 1 h after immersing the specimen, beginning at the Eoc, (Standard ASTM: G5-87).
The electrochemical impedance spectroscopy, EIS, measurements were carried out at Eoc and different
immersion times, computing after 1h at Eoc, that is: 0, 4, 24, 72 hs and 7, 14, 21 y 30 days. The impedance was
measured using an EIS 300 Gamry Instrument, an AC signal was superimposed and 10 measurement points were
spaced logarithmically over a frequency range from 0.001Hz ≥ ƒ ≤ 105 Hz. Photomicrographs of the samples
were obtained with a Microscopy Nikon, EPIPHOT-TME.

2
 2nd Mercosur Congress on Chemical Engineering
th
4 Mercosur Congress on Process Systems Engineering

3. Results and Discussion

In a peroxide-free alkaline solution, the prime oxidant is dissolved oxygen, and the general anodic corrosion
process of Ti is initiated by the formation of hydrated oxide film TiO2 ⋅ x(H 2 O ) (Eq. (1)), where x = 1 for

TiO2 ⋅ H 2 O and x = 2 for Ti (OH )4 (Benn and Tromans(2000)):

Step 1: Ti + ( x − 2)H 2O + 4OH − → TiO2 ⋅ x(H 2O ) + 4e − (1)

In peroxide solutions, dissociation of H 2 O2 in alkaline solutions leads to the formation of OOH −

(dissociation of H 2 O2 in solutions with OH- form OOH − y H2O) . Thus, the prime oxidant is OOH − , whose

reduction occurs according to:

OOH − + H 2O + 2e − ⇔ 3OH − E 0 = 0.878VSHE (2)

For H 2 O2 to react with the Ti oxide through redox reactions, the Ti redox potential must be below the redox

potential of Equation (2), where OOH − acts as an oxidant and is reduced, and above the redox potential of

Equation (3), where OOH − acts as a reductant and is oxidized:

O2 + H 2O + 2e − ⇔ OOH − + OH − E 0 = −0.076VSHE (3)

Diagrams of phase stability in the Ti-H2O2 were calculated at T= 74ºC and pH=10.5. Increasing the
temperature, T, displaces the passive region to lower pH values and broadens the region of HTiO3- (corrosion
region). Thus, within the stable water region in alkaline solutions, the passive film on Ti is likely to be TiO2.H2O
(a Ti+4 oxide), which dissolves at higher pH to form the soluble Ti+4 complex HTiO3-.
Consequently, with Equation (1), step 1, as the initial film formation step, Been and Tromans, proposed that

in step 2 an adsorbed metastable peroxide complex is formed through interaction between OOH − and the
hydrated Ti oxide film:

Step 2: TiO2 ⋅ x(H 2O ) + OOH − → (Ti[OH ]2 O2 )ads + (x − 1)H 2O + OH − (4)

The formation of more protective oxides has been observed in the literature at near-neutral,

low OOH − solutions. At sufficiently high pH and H 2 O2 concentrations (high OOH − ), it is proposed that

3
 2nd Mercosur Congress on Chemical Engineering
th
4 Mercosur Congress on Process Systems Engineering

further reactions between the adsorbed complex and OOH − occur to form the soluble complex, HTiO 3− , as

ilustrated en Step 3, Equation (3), where evolution of O2 contributes to the decomposition of H 2 O2 :

Step 3: (Ti[OH ]2 O2 )ads + OOH − → HTiO3− + H 2O + O2 (5)

In the absence of peroxide, the hydrated oxide film is more protective and dissolves more slowly to

form HTiO3− via Equation (6).

TiO2 ⋅ xH 2 O + OH − → HTiO3− + xH 2O (6)

Thus, it is proposed that the H 2 O2 enhances the dissolution rate of the hydrated film, leading directly to

enhanced Ti dissolution by the sequencial repetition of Step 1 (film formation), Step 2 (formation of adsorbed Ti
peroxide complex), and Step 3 (film dissolution). The overall result of three steps is:

Overall: ( )
Ti + 3OH − + 2 OOH − → HTiO3− + 2 H 2 O + O2 + 4e − (7)

Where it is seen that Ti dissolution should increase with increasing concentrations of OOH − and OH − , as
observed (Been and Tromans).
Once this mechanism is analyzed, the responses given by the potentiodynamic polarization tests were
resorted to. In Figure 1 polarization curves of Titanium, Ti, immersed in hydrogen peroxide solution are shown.
These Figures illustrate the presence of limiting current densities. This currents, with i increasing significantly
with increasing pH, principally, and T for Ti in H2O2 , is indicative of corrosion occurring in the presence of
increasingly poorly protective reaction product films. There are, in addition, small partial anodic currents
corresponding to oxidation of OOH- a O2 may have contributed to the polarization behavior observed, in these
cases, because such oxidation is thermodynamically possible above potentials of 0.1 VSHE. The increase in
anodic current density at the end of the limiting current, where potentials were well above 0.6 VSHE, are readily
attributable to oxidation of water to O2 (Pourbaix, 1974).

4
 2nd Mercosur Congress on Chemical Engineering
th
4 Mercosur Congress on Process Systems Engineering

H2O2

1.7
T=74ºC - pH=10.5
1.5 T=74ºC - pH=11.5
1.3 T=80ºC - pH=10.5

E(V/ENH)
1.1 T=80ºC - pH=11.5
0.9
0.7
0.5
0.3
0.1
-2 -1.5 -1 -0.5 0 0.5 1
i(A/m 2)
Fig.1. Anodic Polarization curves of titanium in alkaline hydrogen peroxide solution

About to the chelantes, a study (Rämö, J., 2003) the oxidative decomposition of DTPA and an alternative
complexing agent β -alaninediacetic acid (ADA) under hydrogen peroxide bleaching conditions was simulated.
ADA was found to be more degradable than DTPA (residual ligand: 71% to 94%, respectively), but no
difference was found in their ability to protect hydrogen peroxide from decomposition (residual hydrogen
peroxide: 40%). This same author studied the degradation of EDTA under conditions corresponding to bleaching
stages, but at a higher hydrogen peroxide concentration. The results of the two studies are not, therefore, directly
comparable. The percentages of the residual EDTA and hydrogen peroxide were presented. In the steady state
the residual EDTA averaged 94%. This indicates high chemical durability. The residual amount of hydrogen
peroxide was 74%.
As mentioned earlier, hydrogen peroxide anion is often considered to be an outstanding bleaching as well as
corroding species (Ti). It may break organic bonds other than chromophores of lignin including those of the
chelates, in which the weakest chemical bond is C-N with a bond energy of 276 kJ/mol. In all likelihood,
nucleophilic bond breaking by HOO- occurs at the weakest bond, C-N.
Figure 2 shows polarization curves of Ti, in H2O2 with EDTA and with DTPA. The limiting currents are
greater that in the case of the absence of chelants. The open circuit potential, Eoc, are more cathodic than those
of Titanium in peroxide. At 74ºC the medium increases its aggression as long as the pH increases.

H2O2 + EDTA H2O2 + DTPA

1,7 T=74ºC - pH=10.5 1,7

T=74ºC - pH=10.5
1,5 T=74ºC - pH=11.5 1,5
T=74ºC - pH=11.5
1,3 T=80ºC - pH=10.5 1,3
T=80ºC - pH=10.5
1,1 T=80ºC - pH=11.5 1,1
E(V/ENH)
E(V/ENH)

0,9 0,9
0,7 0,7
0,5 0,5
0,3 0,3
0,1 0,1
-0,1 -0,1
-1 -0,5 0 0,5 1 1,5 -1 -0,5 0 0,5 1 1,5 2
i(A/m 2) i(A/m 2)

(a) (b)
Fig.2. Anodic Polarization curves of titanium in alkaline hydrogen peroxide solution.
with (a) EDTA and (b) DTPA
5
 2nd Mercosur Congress on Chemical Engineering
th
4 Mercosur Congress on Process Systems Engineering

In Table 1 the electrochemical parameters what result of the respective polarizations are listed. The values in
red correspond to penetrations greater at 5 mpy, which is considered the maximum allowed for designs. The
velocities correspondidng to the aggregate of chelants are all above the allowed ones.

Tabla 1. Summary of electrochemical parameters obtained by potentiodynamics

polarization experiments in alkaline hydrogen peroxide solution with and without a chelating agent.

Temperature 74ºC 80ºC

Eca icorr Penetration Eca icorr Penetration
Chelating pH
V/ENH A/m2 mpy V/ENH A/m2 mpy
- 10.5 0.302 0.04 1.43 0.311 0.18 6.31
- 11.5 0.283 0.15 5.01 0.165 0.23 7.77
EDTA 10.5 0.161 0.92 31.52 0.077 1.46 50
EDTA 11.5 0.091 5.69 194.35 0.030 3.98 136
DPTA 10.5 -0.002 0.90 30.84 0.021 0.99 33.89
DTPA 11.5 -0.081 3.83 130.81 - - -

Typical sets of impedance spectra at different immersion times are depicted in Nyquist plots at the
corresponding corrosion potentials. The results at T=74ºC and pH=10.5 are shown in this paper, Figures 3-5.

Fig.3. Nyquist plots of EIS data of titanium exposed at alkaline hydrogen peroxide solution and diverse times.

6
 2nd Mercosur Congress on Chemical Engineering
th
4 Mercosur Congress on Process Systems Engineering

Fig.4. Nyquist plots of EIS data of titanium exposed at alkaline hydrogen peroxide + EDTA solution and diverse times.

Fig.5. Nyquist plots of EIS data of titanium exposed at alkaline hydrogen peroxide + DTPA solution and diverse times.

This frequency response of Ti was found to be remarkably dependent of the immersion time and change of
concentration of chelating agents. The impedance for Ti was characterized by two capacitive time constants.
Impedance parameters for immersion times of Ti in H2O2 with EDTA over the time range 0-4 hs are: Cdl= 38.6-
41.3 µF/cm2, Rct= 9 Ω. cm2, C2≈ 700 µF/cm2; at the others and at all times: Cdl≈ 3µF/cm2, Rct= 100-200 Ω.
cm2, C2≈ 5 µF/cm2, R2= 3-8 x 103 Ω. cm2 cm2, C3= 36-11 µF/cm2 R3= 2.3-10 x 104 Ω. Impedance parameters for
immersion times of Ti in H2O2 with DTPA resulted into: Cdl= 53.7-21.7 µF/cm2, Rct= 1 x 103-2 x 104 Ω. cm2,
C2≈ 15-27 µF/cm2, R2= 2 y 5 x 105 Ω. cm2. The corrosion process, observed at lower frequencies, could be

7
 2nd Mercosur Congress on Chemical Engineering
th
4 Mercosur Congress on Process Systems Engineering

controlled by diffusion (in the case of H2O2 by escape of a titanium complex from the surface). The good
agreement between experimental and simulated data by using non-linear least-square fit routines is demonstrated
in Figures 6-8.

100000 80 10000000 80

medida medida
70 70
fiteo 1000000 fiteo
10000 medida 60 medida 60
fiteo 100000 fiteo
50 50

|Z| (Ω ).cm2
|Z| (Ω ).cm2

- Fase (°)

- Fase (°)
1000 40 10000 40

30 30
1000
100 20 20

100
10 10

10 0 10 0
0,01 0,1 1 10 100 1000 10000 100000 0,01 0,1 1 10 100 1000 10000 100000
Frecuencia (Hz) Frecuencia (Hz)

(a) (b)
Fig.6. Bode plots for titanium exposed at alkaline hydrogen peroxide solution, (a) 1 h and (b) 30 days.

1000 30 1000000 90

medida 80
medida 25 fiteo
100000
fiteo 70
medida
medida fiteo 60
20
fiteo 10000
|Z| (Ω ).cm2
|Z| (Ω ).cm2

- Fase (°)
- Fase (°)

50
15
40
1000
10 30

20
100
5
10

100 0 10 0
0,01 0,1 1 10 100 1000 10000 0,001 0,01 0,1 1 10 100 1000 10000
Frecuencia (Hz) Frecuencia (Hz)

(a) (b)
Fig.7. Bode plots for titanium exposed at alkaline hydrogen peroxide solution with EDTA, (a) 1 h and (b) 30 days.

1000 35 1000000 90

medida 80
medida 30
100000 fiteo
fiteo 70
25
medida
medida 60
fiteo
fiteo 10000
|Z| (Ω ).cm2
|Z| (Ω ).cm2

20
- Fase (°)
- Fase (°)

50

40
15
1000
30
(a) 10
20
100
5
10

100 0 10 0
0,01 0,1 1 10 100 1000 10000 0,01 0,1 1 10 100 1000 10000
Frecuencia (Hz) Frecuencia (Hz)

(b)
Fig.8. Bode plots for titanium exposed at alkaline hydrogen peroxide solution with DTPA, (a) 1 h and (b) 30 days.
8
 2nd Mercosur Congress on Chemical Engineering
th
4 Mercosur Congress on Process Systems Engineering

The Figure 9 shows that the Ti suffers generalized corrosion in H 2 O2 media, which is confirmed in the

microphotographies.

(a) (b)

Fig. 9. Photomicrographs (x400) showing surface morphology of Ti, (a) 5 hs and (b) 30 days.

It is important to add that a precipitated adherent is formed on the surface of Ti. This precipitated provokes
different attack depths, Figure 10.

(a) (b)
Fig. 10. Photomicrographs (x400) showing surface morphology of Ti after 30 days of immersion
(a) H2O2 with EDTA (b) H2O2 with DTPA.

4. Conclusions

The hydrogen peroxide anion, HOO-, is considered as the agent corrosive. Ti suffers generalized corrosion in
H 2 O2 media.
The corrosion of Ti in media H 2 O2 with EDTA or DTPA is not generalized.
The impedance parameters of Ti in EDTA and DTPA media oscillate in the time and, at each time according
to medium.
The corrosion rate of Ti in media H 2 O2 with EDTA or DTPA is larger than in H 2 O2 without these

chelants.
It is recommended to minimize, specially, the use of EDTA, utilizing the most exact concentration for the

9
 2nd Mercosur Congress on Chemical Engineering
th
4 Mercosur Congress on Process Systems Engineering

objectives established
It is essential that an effective means were developed for the removal of EDTA and DTPA or that they were
replaced by more degradable compounds.

References

Andreasson, P., (1995). The corrosion of titanium in hydrogen peroxide bleaching solutions. Proceedings of the 8th.
International Symposium on Corrosion in the Pulp and Paper, Stockholm, Sweden, 119.
Bambrick, D.R. (1985). The effect of DTPA on reducing peroxide decomposition. Tappi Journal, 68 (6), 96.
Been, J., Tromans. D. (2000). Titanium Corrosion in Alkaline Hydrogen Peroxide. Corrosion, 56 (8), 809.
Been, J., Tromans, D. University of British Columbia, Departament of Metals and Materials Engineering. Vancouver, BC,
Canada.
Poppius-Levlin K. (1992). Chemistry of Bleaching. INSKO Publication 8-92 V, Finland.
Pourbaix, M. (1974). Atlas of Electrochemical Equilibria in Aqueous Solutions, National Association of Corrosion
Engineers. Houston, Texas, USA.
Rämö, J. (2003). Hydrogen Peroxide-Metals-Chelating agents; Interaction and Analytical Techniques. Doctoral Dissertation,
Academic dissertation to be presented with the assent of Faculty of Technology, Univesity of Oulu, for public desission in
Auditorium D 104 of the VTT. Kaitoväylä I, Oulu, Finland.
Reichert, D.L. (1996). Corrosion behavior in Hydrogen Peroxide Bleaching Solutions. NACE – Corrosion 96, paper Nº 467.
Ruiz, E. R., Mendez, C. M., Stampella, R. S., (2003). Corrosion behavior of titanium in alkaline peroxide bleach liquors.
Proceedings of the 5to Congreso de Corrosión de la NACE Región Latinoamérica and 8vo Congreso Iberoamericano de
Corrosión y Protección. Santiago, Chile. 102.
Schutz, R. W., Xiao, M. (1995). Practical windows and inhibitors for Grade 2 titanium use in alkaline peroxide bleach
solutions. Tappi Journal 78 (11), 79.
Stampella, R. S. et al. (1999). Efecto de inhibidores y quelantes sobre la corrosión del titanio en peróxido de hidrógeno
alcalino. Informe Final Programa Nacional de Incentivos a la Investigación. Posadas, Misiones, Argentina.
Varjonen, O.A., Hakkarainen, T.J. (1995). Corrosion of titanium in alkaline hydrogen peroxide bleaching solution. Tappi
Journal 78 (6), 161.
Wyllie II, W., Brown, B.E., Duquette, D.J. (1995), The corrosion behavior of titanium (grade 2) in alkaline peroxide bleach
liquors. Tappi Journal 78 (6), 151

10