Journal of Applied Electrochemistry 29: 1317±1322, 1999.
1317
Ó 1999 Kluwer Academic Publishers. Printed in the Netherlands.
New passivating pastes for stainless steel without nitric acid
G. CAPOBIANCO1, G. FACCIN1, A. GAMBIRASI2, G. MORETTI2*, G. QUARTARONE2
and G. SANDONAÁ1
1
University of Padova, Department of Physical Chemistry, v. Marzolo 1 I-35131 Padova, Italy
2
University of Venezia, Department of Chemistry, D.D.S. Marta 2137 I-30123 Venezia, Italy
(*author for correspondence, bmor@unive.it)
Received 11 September 1998; accepted in revised form 15 April 1999
Key words: passivating paste, stainless steels
Abstract
Handled stainless steel (SS) structures often need accurate passivating treatments before use. Until now the most
frequently used passivating pastes have been based on nitric (¯uoridric) acid and, as a result, are responsible for
consequent environmental problems. This work addresses the SS passivation quality of two low environmental
impact pastes, based on peroximonosulfate (P2) or sodium perborate (P3). Tests were conducted on various
materials using electrochemical tests (cyclic voltammetry, anodic potentiostatic transients, open-circuit potential (E)
decay), Microscopic and Inductively Coupled Plasma (ICP) measurements. The results, when compared with those
obtained on the same materials by using a conventional nitric acid paste (P1), indicate that the proposed pastes P2
and P3 are time stable, non toxic and as ecient as the P1 paste. Moreover, the protection of the tested SS can be
further enhanced by adding small amounts of citric acid to, in particular, the P3 paste.
1. Introduction
2. Experimental details
One of the most important factors that determines the
resistance of SS structures to pitting and crevice
corrosion is the quality of the metal surface. Therefore,
even if the high chromium content of SS makes it selfpassivating upon exposure to air [1], any equipment
whose surface has been altered by handling must undergo accurate cleaning (degreasing), pickling and passivating procedures to obtain a uniform passive layer.
The procedures reported in ASTM standard A380
recommend the use of warm nitric acid (from 20 to 70%,
in the 49±71 °C temperature range for 30 min or from
10 to 15% at 66 °C for 4 h). There are, however, other
patents, some of which have developed one-step processes for pickling and passivation [2±7]. But, in recent
years, the ecological issues raised by these pickling and
passivating processes have become increasingly important.
The aim of this paper is to formulate and analyse new
passivating pastes that can provide a protective coating
for SS surfaces against aggressive agents and, at the
same time, decrease the chemical waste currently produced by nitric acid pastes [5±7]. The study was
conducted using electrochemical techniques, (voltammetry, anodic potentiostatic transients, E decay), and by
comparing the results from these tests to those obtained
with the same materials oxidised by using a conventional nitric acid paste.
The materials tested in this work include commercial
AISI 304L, AISI 316L, AISI 446 and SAF 2205 SSs of
certi®ed composition as reported in Table 1. The results
obtained on AISI 304L SS are described in detail; the
behaviour of the other examined SSs is reported by
comparison. The disc-shaped samples had an exposed
surface of 1 cm2. The specimens were polished with ®ne
grained emery paper (down to 1200), cleaned ultrasonically in acetone and then depassivated by cathodic
reduction to remove the oxide ®lm formed during
previous treatment [8]. All the galvanostatic reduction
curves (see Figure 4 and related comments) were carried
out in deaerated (N2) borate/boric acid (BB) solution
(pH 8.4).
The specimens were then subjected to an oxidation
through: (i) electrochemical anodization (in a deaerated
(N2) 0.1 M Li2SO4 solution at potential corresponding
to the passivating zone: +400 mV vs SCE), (ii) by
exposing them to the atmosphere for 24 h, or (iii) by
treating them chemically with three kinds of passivating
pastes composed as follows:
(a) Commercial HNO3 paste (P1): HNO3 60±70%,
SiO2 5±7%, surfactants 1±3%, commercial corrosion inhibitor 1±3%, H2O balance.
(b) Oxone peroximonosulfate paste (P2): H2SO4 10%,
Oxone peroximonosulfate 5%, SiO2 5%, surfactants 3%, H2O balance.
1318
Table 1. Chemical composition (wt %) of AISI 304L, AISI 316L,
AISI 446 and SAF 2205
Element
AISI 304L
AISI 316L
AISI 446
SAF 2205
C
Ni
Mn
Mo
Cr
S
P
Si
Fe
0.018
9.15
1.75
0.36
18.1
0.016
0.028
±
balance
0.017
10.52
2.0
2.10
16.8
0.03
0.045
0.87
balance
0.18
±
0.81
±
26.5
0.003
0.19
0.51
balance
0.011
5.52
1.55
2.92
22.11
0.001
±
0.32
balance
(c) Sodium perborate paste (P3): H2SO4 10%,
Na2B4O7 5%, stabilizer 1%, SiO2 5%, surfactants
3%, H2O balance.
Before the voltammetric experiments, performed in
deaerated (N2) 10)3 M H2SO4 solutions, the chemically
passivated specimens were washed in bidistilled water to
remove the paste until no further acid reaction was
detected.
The specimen, mounted on a Te¯on electrode holder
with the electrode contact made by an inner metal rod,
was then transferred into a three electrode cell. After a
cathodic polarization (30 s) at a potential near to that of
zero current, the cyclic voltammetric curve started at a
potential scan rate of 5 mV s)1.
Electrochemical measurements were carried out using
Amel (Italy) apparatus consisting of potentiostat, signal
generator, interface, current integrator and dierential
electrometer. All the potential values are referred to the
saturated calomel electrode (SCE).
In the potentiostatic current±time transients (carried
out in deaerated (N2) 10)3 M H2SO4 solutions), the
anodic voltage step was preceded by specimen stabilization (60 s) at the corrosion potential (Ecorr). The
transients were carried out through the following
procedure: after stabilization, the specimen was rapidly
brought to the chosen potential and the current decay
was then recorded over time.
The thickness of the oxide of the passive ®lms was
estimated by determining (using an inductive coupled
plasma (ICP) technique), the amount of Fe, Cr and Ni
in the solutions after the SS oxide dissolution in HNO3
10% at 60 °C following a standard procedure [9]. The
thickness of the oxide layer (around 3±4 nm) was
calculated based on the hypothesis that all the oxides,
after 2 h at 300 °C in a vacuum oven under Ar
atmosphere, are present in the anhydrous form and by
assuming a 5.2 g cm)3 average density of the passive
®lm. After preliminary measurements on depassivated
and passivated specimens, the appropriate immersion
time was determined, assuring that a minimum amount
of the oxide was always present on the surface and that
the bare metal could be etched. This has been veri®ed by
voltammetric curves on the depassivated material. Thus,
the calculated thickness of the oxide layer was presumably 10% lower than the actual value. This was veri®ed
by a X-ray photoelectron spectroscopy (XPS) determi-
nation performed on an electrochemical anodized specimen. However the calculated thickness is useful in
comparing the dierent surface treatments.
3. Results and discussion
Figure 1 shows the voltammograms in deaerated (N2)
10)3 M H2SO4 solutions carried out on AISI 304L
specimens depassivated or passivated by applying the
commercial paste P1 for 1 or 2 h [1]. It can be seen that,
as the oxidation time increases, the decrease of the
passivating current (ip) is not appreciable.
The values of ip obtained with the dierent materials
after passivation in the atmosphere or by applying
the commercial paste on the surface are reported in
Table 2.
The passivation current decreases as the chromium
content of the chemically passivated specimen increases
and is practically unaected by the presence of Mo when
comparing AISI 304L (Mo 0.36 wt %) and AISI 316L
(Mo 2.1 wt %), whereas it changes signi®cantly by
comparing AISI 446 (Mo 0 wt %) and SAF 2205 (Mo
2.9 wt %). ip reaches the same value for AISI 316L,
AISI 446 and SAF 2205 spontaneously passivated in the
Fig. 1. Voltammograms obtained with AISI 304L specimens passivated with P1 (deaerated (N2) 10)3 M H2SO4 solutions; scan rate:
5 mV s)1; the arrow shows the scan direction). Reversal scan of test
with P1 (2 h) was superimposed to the 1 h test. Key: (- -- -)
depassivated specimen; (- - - - - -) specimen passivated by P1 for 1 h;
(б) specimen passivated by P1 for 2 h.
8
Table 2. Passivation current (ip in lA cm)2) obtained with dierent
SSs passivated spontaneously or by P1 treatment
Specimens
Spontaneously
passivated
Passivated by
P1 (1 h)
Passivation current, ip/lA cm)2
in deaerated (N2) 10)3 M H2SO4
AISI 304L
AISI 316L
AISI 446
SAF 2205
41.6
20.3
20.5
20.3
5.1
5.0
4.1
2.0
1319
Fig. 2. Anodic transients obtained with depassivated and passivated
(P1 30 min) AISI 316L specimens at +400 mV vs SCE (deaerated (N2)
10)3 M H2SO4 solutions). Key: (- -- -) passivated with paste P1;
(-ámá-) passivated in the atmosphere; ( Ð Ð) depassivated.
8
atmosphere. Figure 2 shows the trend of the anodic
potentiostatic transients performed with depassivated
and passivated AISI 316L specimens [10, 11]. Figure 3
shows the transients obtained with passivated (P1
30 min) AISI 304L, AISI 316L and SAF 2205 specimens. The limiting passivation current is reached faster
as the chromium content of the specimen increases.
The galvanostatic reduction curves, obtained in BB
solution (pH 8.4) with AISI 304L specimens passivated
1 or 12 h, are shown in Figure 4. The change in slope of
the curve ranging from +150 to )350 mV are correlated
to the reduction of chromate, while those from )350 to
)900 mV to the reduction of Fe2O3 [12].
As can be seen in Figure 4, the content of Fe2O3 in the
oxide layer is low and decreases on increasing the
contact time with the paste [13±15]. This may be
ascribed to the transformation of Fe2O3 in Fe3O4,
which is not visible at this solution pH [15]. The same
behaviour was obtained when the specimens were
passivated by the other pastes.
The variation of E over time for AISI 304L SS,
measured in aerated or deaerated (N2) 10)3 M H2SO4
solutions, is shown in Figure 5.
It is apparent that, while the potential of the specimen
electrochemical anodized at +400 mV decreases slowly
from +250 mV reaching a steady value in the passivity
potential range, the potential of the chemically passivated specimens (P1 1 h) reaches a steady value in the
prepassivation zone (Epp) and then slowly shifts to more
Fig. 3. Anodic transients obtained with passivated AISI 304L (m),
AISI 316L ( ) and SAF 2205 () specimens (P1 30 min) at +400 mV
vs SCE (deaerated (N2) 10)3 M H2SO4 solutions).
8
noble values, due to the presence of oxygen in the
solution. In the deaerated (N2) solution the potential
reaches a stable steady value in the activity zone.
These results suggest that, because chromium oxide is
insoluble in the medium used, its content in the
chemically obtained passive layer should be lower than
that produced by electrochemical anodisation. On the
other hand, ICP measurements show that the Fe/Cr
ratio, which in the electrochemically anodised specimens
is about 0.5, increases to 4.3 in the oxide layer obtained
after P1 treatment (Table 3).
Similar results were obtained with AISI 316L whereas
chemically oxidised AISI 446 and SAF 2205 specimens
behave like those that were electrochemically anodized.
Furthermore, the time to reach the steady value in
deaerated (N2) solutions is linearly related to the
chemical oxidation time when the paste application
time is lower than 2 h (Figure 5).
Some authors report that in the E/t curves, before the
rapid potential decrease, the time of the slow E decrease
can be related to the oxide ®lm thickness [16, 17]. By
comparing the voltammetric results (Figure 1) with
those of the E/t curves (Figure 5) it can be seen that,
even if the value of the passivating current (ip) after 1
and 2 h of contact with oxidizing paste is practically
the same, the time of the slow E decrease (related to the
oxide ®lm thickness) results roughly doubled. The thickness of the passive ®lm obtained with P1 treatment in
1320
Table 3. ICP measurements on AISI 304L SS specimens
Type of treatment
Thickness of
oxide layer/nm
Fe/Cr
ratio
Chemically anodized
P1*
P2*
P3*
P3* + citric acid
±
3.3
4.0
2.8
4.1
0.5
4.3
3.7
3.2
3.0
* Time for P1, P2 and P3 treatments: 12 h
Fig. 4. Galvanostatic reduction curves obtained in BB solution (pH
8.4) with AISI 304L specimens passivated (P1) 1 h (curve 1) or 12 h
(curve 2).
prolonged time tests (12 h) on AISI 304L specimens is
around 3.3 nm (Table 3).
Figure 6 shows the voltammetric behaviour of AISI
304L specimens chemically passivated with the three
dierent used pastes. In the 1 h tests, the lowest
passivation current was obtained with the oxone per-
Fig. 5. AISI 304L SS open-circuit potential vs time (10)3 M H2SO4
solutions; curves 1 and 2: aerated solutions; curves 3 and 4: deaerated
(N2) solutions). After 1 h of electrochemical oxidation (curve 1); after
1 h of P1 treatment (curves 2 and 3); after 2 h of P1 treatment (curve 4).
oximonosulfate paste (P2), but the best result was
achieved with the sodium perborate paste (P3) after a
prolonged application time (12 h). Nonetheless anodic
potentiostatic transients carried out on the specimen
passivated with P1 show a faster current decay and a
value of ip lower than that obtained after P2 treatment
(Figure 7).
We have also veri®ed that the time required to obtain
a protective passivation is substantially reduced when
the specimens are pickled in an HNO3 pickling paste
after the surface emery paper treatment [8, 18].
The change in E over time in aerated 10)3 M H2SO4
solutions is reported in Figure 8. This Figure also
reveals the substantial eect obtained by adding 5%
citric acid to the P3 paste.
As far as the time stability of the dierent pastes is
concerned, Figure 9 shows the decrease in the persulphate and hydrogen peroxide percentages determined by
sodium sulphite and potassium permanganate titrations
[19]. Figure 9 depicts the degradation of the P2 paste
which reaches a 70% stable activity after 120 days of
storage, while the eciency of the P3 paste linearly
decreases with time, reaching a similar value after 150
days. After the paste is removed by rinsing with water,
the specimen surface is smooth and, thus, both pastes
seem suitable for SS passivation.
Fig. 6. Voltammograms obtained with AISI 304L specimens passivated with: (s) P1 (1 h), () P2 (1 h), (- - - -) P3 (1 h) and (б) P3 (12 h)
(deaerated (N2) 10)3 M H2SO4 solutions; scan rate 5 mV s)1).
1321
Fig. 9. Paste stability over time. Key: (ÐeÐ) P2; ( ) P3.
proposed passivating pastes cannot be applied to low
alloy steels that can undergo serious corrosive attack.
Fig. 7. Anodic potentiostatic transients obtained with AISI 304L SS
passivated by P1 (ÐsÐ) or P2 (- -- -) treatment (1 h) at +400 mV vs
SCE (deaerated (N2) 10)3 M H2SO4 solutions).
With regards to the chromium content in the oxide
layer on the metal surface, the results in Table 3 indicate
a signi®cant decrease of the Fe/Cr ratio in the presence
of citric acid. This eect could be due to a reaction of the
acid with Cr3+ to give a citric acid Cr3+ complex
avoiding its oxidation to soluble Cr6+ species.
Some tests also revealed that the surface treatments
(cleaning, degreasing etc.) recommended before the
application of the commercial paste cause the pitting
attack, particularly when carbon impurities are present
on the metal surface. This is especially prevalent in the
welding zone and was con®rmed when mild steel
specimens were used. Thus, both the commercial and
4. Conclusions
The following can now be stated:
(i) Pastes P2 and P3 provide protective steel coatings
that are equally ecient as the commercial nitric
acid paste (P1).
(ii) The proposed pastes P2 and P3 show no dissociation or degradation after lengthy storage and do
not aect the metal surface by pit formation.
(iii) In comparison to the commercial paste, the new
pastes decrease chemical waste and are nontoxic.
(iv) Surface chemical pickling prior to paste application
signi®cantly improves the eect of the paste.
(v) Carbon impurities have to be carefully eliminated
from the welding zone because carbon causes discontinuities of the passivated metal surface.
(vi) The passivating pastes cannot be applied to low
alloy steels which undergo serious corrosive
attacks.
Acknowledgements
The research was supported by Italian Ministry of
University and Scienti®c and Technological Research
(MURST) funding (1997).
References
Fig. 8. AISI 304L SS open-circuit potential against time measured in
aerated 10)3 M H2SO4 solutions, after prolonged treatment (12 h) by
P1, P2, P3 and P3 + 5% citric acid. Key: (б) P3; (- -- -) P1;
(- -h- -) P2; (--- ---) P3 with citric acid.
8
1. R. Ramasubramanian, N. Preocanin and R.D. Davidson, J. Electrochem. Soc. 134 (1985) 739.
2. J. Gallus-Olender, B. Franc, A. Lanowy, M. Szlezak, J. Francuz,
G. Ratka and B. Kapko, POL. PL 158, 060 (CI. C23F11/04) (31
July 1992), Appl. 272, 551 (17 May 1988), 3 pp.
3. I. Magyar, D. Titkos, F. Magyar, and T. Szoke (Vas Megyei
Kemenysepro es Tuzelestechnikai Vallalat, Szombathely), Hung.
Teljes HU 47, 651 (CI. C23G1/02) (28 Mar. 1989), Appl. 87/1, 685
(16 Apr. 1987), 7 pp.
1322
4. Z. Kizkova, Z. Marcanova and J. Dostal, Czech, CS 218, 389 (CI.
C23G1/18) (15 Feb. 1985), Appl. 81/2, 588 (6 Apr. 1981), 3 pp.
5. Z. Svobodny, J. Kos, J Nemcova, V. Stradova, P. Andel and
F. Kalny, (SVU Ochrani Materialu G.V. Akinova), Czech. Rep.
CZ 278, 566 (CI. C23G1/00) (16 Mar. 1994), Appl. 6, 186 (1 Nov.
1989), 5 pp.
6. W. Wiedner, R. Bielecki, R. Heuber, G. Meissner, G. Petzold
and H.W. Wolf (VEB Chemiecombinat Bitterfeld), Ger (East) DD
248, 379 (CI. 23G1/108) (5 Aug. 1987), Appl. 289, 458 (22 April
1986).
7. X. Deng and X. Xu (Damin steel works) Faming Zhuanli Shenqing
Gongkai Shuomingshu CN 86, 101, 522 (CI. C23G1/04) (27 Jan
1988), Appl. (7 July 1986), 6 pp.
8. `Handbook of Metals' Vol. 5, ASM International (1987) pp. 558±
61.
9. UNI 3666, Italy (1954).
10. A. J. Bard, `Electroanalytical chemistry' (Dekker, New York
1996).
11. F. Graham and H. Brookes, Electrochim. Acta 33 (1988) 1475.
12. K. Ogura and T. Majima, Electrochim. Acta 25 (1979) 325.
13. K. Ogura and K. Sato, Electrochim. Acta 25 (1980) 857.
14. A. Rauscher, H. Konno and M. Nagayama, Electrochim. Acta 22
(1977) 823.
15. K. Ogura and Y. Miwa, J. Electroanal. Chem. 130 (1981) 189.
16. J. Kruger, Corros. Sci. 29 (1989) 149.
17. J.A. Bardwell, B. Macdougall and M.J. Graham, J. Electrochem.
Soc. 135 (1988) 413.
18. C.P. Dillon, Mater. Performa., 33 (5) (1994) 62.
19 P. Ippoliti, E. Ippoliti, C. Zeggioululy and A. Mascioli, `Analisi
Chimica Quantitativa' (Signorelli Editore, Napoli-Italy, 1986).