See discussions, stats, and author profiles for this publication at: ​https://www.researchgate.

net/publication/299340836

Corrosion in cold rolled steel sheets for drum manufacturing

Conference Paper ​· November 2004
CITATION
READS ​1

346
2 authors​, including:
Marcelo Gines ​Tenaris
36 ​PUBLICATIONS ​815 ​CITATIONS
SEE PROFILE
All content following this page was uploaded by ​Marcelo Gines ​on 22 March 2016.
The user has requested enhancement of the downloaded file.

CORROSION IN COLD ROLLED STEEL SHEETS FOR DRUM

MANUFACTURING

M.J.L. Ginés (1) ​

(1) ​
G.J. Benítez ​ A.
Diego (2) ​

ABSTRACT

Cold rolled steel products are normally protected against corrosion by the use of rustproofing oils or aqueous nitrite
based lubricant. However, drum manufacturers that paint directly on the steel surface without any preparation require a
dry (oil-free) steel surface. Consequently, this kind of material could easily suffer the effects of atmospheric corrosion
during storage and transit if no countermeasures are taken. Corrosion rate depends on several factors such as
temperature, relative humidity, pollutants in the environment and/or residues that remain on the steel surface
(hygroscopic salts or iron fines).

At Siderar, steel for drums manufacturing can be produced by different routes, which may involve an electrolytic
cleaning process in the ECL (​E​lectrolytic ​C​leaning ​L​ine) and box annealing (EBNER or HNX). Although these
processes could help increase surface cleanliness, they might also have effects on steel corrosiveness.

In this work, the effect of the different process routes and the use of VCI (​V​olatile ​C​orrosion ​I​nhibitor) paper on steel
corrosion performance during storage is assessed through laboratory experiments. Steel panels were analyzed by visual
inspection and photographed as a function of storage time. Corrosion products were characterized by SEM/EDS
(​S​canning ​E​lectron ​M​icroscopy/​E​nergy ​D​ispersive ​S​pectroscopy) and FTIR (​F​ourier ​T​ransform ​I​nfra​R​ed)
spectroscopy.

The obtained results allowed to select a process route to lower the occurrence of atmospheric corrosion problems on
cold rolled steel sheets.

Key Words: ​corrosion, drums manufacturing, surface cleanliness, filiform corrosion.

1. INTRODUCTION

Cold rolled steel products are normally protected against corrosion by the use of rustproofing oils or aqueous nitrite
based lubricant [1, 2]. However, drum manufacturers that paint directly on the steel surface without any preparation
require a dry (oil-free) steel surface. Consequently, this kind of material could easily suffer the effects of atmospheric
corrosion during storage and transit if no countermeasures are taken [3]. Corrosion rate depends on several factors such
as temperature , relative humidity, pollutants in the environment and/or residues that remain on the steel surface
(hygroscopic salts or iron fines) [1, 4].

At Siderar, steel for drums manufacturing can be produced by different routes, which may involve an electrolytic
cleaning process in the ECL (​E​lectrolytic ​C​leaning ​L​ine) and box annealing (HNX or EBNER). Although these
processes could help increase surface cleanliness, they might also have effects on steel corrosiveness.

In this work, the effect of the different process routes and the use of VCI (​V​olatile ​C​orrosion ​I​nhibitor) paper on steel
corrosion performance during storage is assessed through laboratory experiments.

(1) Centro de Investigación Industrial (CINI-FUDETEC), CC 801, 2900 San Nicolás, Pcia. de Buenos Aires,
Argentina. E-mail: yapgin@siderar.com – www.fudetec.com (2) SIDERAR-CSGS-Gerencia de Laminación, CC
801, 2900 San Nicolás, Pcia. de Buenos Aires, Argentina. E- mail: apadan@siderar.com ​2​nd ​IAS Conference on Uses of

Steel, 2004, San Nicolas, Argentina ​227

2. EXPERIMENTAL
Cold-rolled steel coupons were sampled from coils that were produced through different process routes (Fig.1).
These routes were selected in order to assess the effects of both electrocleaning stage (ECL) and annealing
atmosphere (HNX/EBNER boxes) on corrosion performance.
Cold Cold rolling rolling ​Cold Cold rolling rolling #1 #1 Cold Cold rolling rolling #2 #2 ​Cleaning Cleaning ​ECL ECL
Annealing Annealing ​HNX HNX EBNER EBNER HNX HNX EBNER EBNER EBNER EBNER ​Tempering Tempering ​TM2
TM2 TM2 TM2 TM1 TM1 TM1 TM1 TM3 TM3 ​Final Final stage
stage T​ RL TRL TRL TRL LAI LAI LAI LAI LIF LIF ​Sample Sample ID
ID ​1 1 2 2 3 3 4 4 5 5 ​ECL: ECL: electrolytic electrolytic cleaning cleaning line, line, TM1~TM3: TM1~TM3: dry dry temper temper mills mills
#1-3, #1-3, TRL: TRL: edge edge trimming trimming facility, facility, LAI: LAI: inspection inspection and and roll-flattening roll-flattening line, line, LIF:
LIF: final final inspection inspection line line
Figure 1. Process routes for the coils used as substrates in this study.
The characterization of the different substrates is shown in Table 1. Chloride and sulfate surface contents were
determined by titration after aqueous extraction. Surface silicon content was determined by XRF (​X​-​R​ay
F​luorescence spectroscopy). Surface carbon was determined by means of a ​LECO R212 Surface carbon detector,​
which performs carbon oxidation at controlled temperature in pure oxygen and detects carbon dioxide evolution.
Evolved gas is attributed to the presence of carbon on steel surface. The temperature program consist of two
consecutive isothermal steps, the first at 400°C and the second at 600°C. Determined surface carbon values are
referred to as phase 1 carbon (CF1), and phase 2 carbon (CF2), respectively. Reported values (Table 1) show low
surface carbon contents of samples #1 and #2, which have undergone ECL processing.

Property Sample
​ ​ #2 #3 #4 #5 Thickness
ID #1 ​ [ mm ] 0.21 0.26 0.70 0.50 0.80 Days in stock a​ ​19 5 12 21 17
Surface Chloride [ mg/m​2​] 0 0 2.1 0.41 1.1 Surface Sulfate [ mg/m​2 ​] - - << 3.0 - - Surface Silicon [ mg/m​2 ​] 3.57
9.22 2.26 - 2.13 Surface carbon b​ [​ mg/m​2​] 1.62/6.42 0.96/0.92 4.15/15.49 30.14/16.55 11.74/23.99 ​a Computed
​ from the
​
end of annealing process on; b​ Result correspond to CF1/CF2 carbon content (see text).
Table 1. Characterization of steel samples used in this study.
Probes (10x10 cm​2​) cut from sampled materials, were stacked and tied up with plastic cable ties. Half of the stacks
were wrapped with VCI paper. Both wrapped and bare packages were placed in a coil storage sector in Siderar-San
Nicolas Plant. Packages were withdrawn at the following pre-established storage times: 7, 15, 30, 45, 60, 75, 90,
228
2​nd ​IAS Conference on Uses of Steel, 2004, San Nicolas, Argentina
105, 120, 150, 180, 210, and 240 days. Storage conditions were monitored by means of a ​Testo 635 ​digital hygrometer
(Table 2). ​Condition Min. Max. Average ​a T
​ emperature [ °C ] 16.5 34.0 24.0 ​± ​3.4 Relative Humidity [ % ] 24.5 78.3

52.6 ​± ​13.3 Dew point [ °C ] 0.8 24.4 13.3 ​± ​5.8 ​a ​Average of 85 measurements.

Table 2. Environmental conditions in storage site during the experiment.

After each extraction, steel panels were analyzed by visual inspection and photographed. When present, corrosion
products were characterized by SEM/EDS (​S​canning ​E​lectron ​M​icroscopy/​E​nergy ​D​ispersive ​S​pectroscopy) and FTIR
(​F​ourier ​T​ransform ​I​nfra​R​ed spectroscopy).

3. RESULTS AND DISCUSSION

3. 1. Corrosion resistance

According to some authors, the occurrence of steel coil corrosion during in-plant storage is due to water vapor diffusion
into interstitial spaces within the coils and later absorption of the water by contaminating inorganic salts present on the
steel surface [1, 4, 5]. Volatile corrosion inhibitors (VCI) are also transported by diffusion, providing protection to those
areas they can reach [5, 6]. The protection mechanism implies migration of VCI to those inner areas and cavities that are
accessible, after which it is dissolved in the moisture layer, protecting the metal surface [5, 6].

After withdrawal from storage, visual inspection of packages showed clearly a better surface condition of VCI- wrapped
packages when compared to bare ones. For a given storage time, bare packages present corrosion in the external probes
and edges of inner probes, while these areas of the VCI-packed probes are uncorroded (Figs. 2 and 3). For bare coupons,
corrosion is observed first on its edges, and later on the overall surface as small pits. These results agree with the
described mechanism for VCI protective action and suggest that at industrial scale, the protection would be effective
mainly on the outer turns and edges of coils.

(a) (b)
 Figure 2. External probes appearance after 75 days of storage. a) VCI wrapped; b) bare.

2​nd ​IAS Conference on Uses of Steel, 2004, San Nicolas, Argentina ​229
Inspection of the inner probes for both bare and VCI packages showed corrosion pits in the overall surface. Again, the
corrosion degree of bare-packed is higher to that of VCI-packed probes for the same storage time. However, it was
observed that the time elapsed until the first corrosion spots is observed is the same for both types of package (Table 3).
Results also show that process routes have incidence on the onset time for corrosion process.

(a) ​(b)

Figure 3. Internal probes appearance after 75 days of storage. a) VCI wrapped; b) bare.

​
Process route Onset time [days] Sample ID Cold rolling Annealing ECL Bare

VCI
#3 #1 HNX no - - #5 #2 EBNER no 150 - #4 #1 EBNER no 90 90 #1 #1
HNX yes 60 60 #2 #1 EBNER yes 15 15
 Table 3. Onset time for corrosion process on stored materials.

After 240 days of storage, a ranking based on overall surface corrosion performance could be established for the tested
materials (Table 4). The results clearly show that corrosion performance depends on process routes.

Packing Corrosion performance ​3 > 5 > 4 >> 1 > 2 ​Bare ​No ECL+HNX No ECL+EBNER No ECL+EBNER ECL+HNX

ECL+EBNER 3
​ ≈ 5 > 4 >> 1 > 2 ​VCI ​No ECL+HNX No ECL+EBNER No ECL+EBNER ECL+HNX ECL+EBNER

Table 4. Corrosion performance (from high to low) of studied materials (see Fig.1).

In order to illustrate the difference in corrosion performance, in Figures 4 and 5 the appearance of both types of samples
after 180 days of storage is shown. For bare samples the material with the higher performance (#3) shows only corrosion
attack on one edge, while the worst material (#2) developed corrosion spots in the overall surface (Fig. 4). For
VCI-wrapped samples, corrosion spots developed in the overall surface but no attack on edges was observed (Fig. 5).

2​nd ​IAS Conference on Uses of Steel, 2004, San Nicolas, Argentina ​230

(a) (b)
 Figure 5. Appearance of VCI-packed materials (inner samples) after 180 days of storage. a) substrate #5 (No
ECL+EBNER); b) substrate #2 (ECL+EBNER).

The results obtained show that substrates without ECL processing are more resistant to corrosion than electrocleaned
ones, and that HNX annealing leads to higher atmospheric corrosion resistance than EBNER annealing. These facts are
observed for both bare and VCI-wrapped samples. Results suggest that VCI does not delay to a great extent the onset of
corrosion process on the overall surface.

According to the presented results, corrosion susceptibility is strongly related to surface cleanliness. Given that
corrosion impact is higher for electrocleaned material, and EBNER annealing leads to more corrosion problems than
HNX annealing, it can be said that the higher the surface cleanliness, the lower the corrosion resistance. This fact must
be taken into account for proper selection of process routes according to customers’ requirements.

3. 2. Characterization of corrosion products

SEM characterization showed that the morphologies of corrosion products in electrocleaned and non-electrocleaned
material are different (Fig. 6). Even though corrosion spots look similar to the naked eye for both type of samples, a
closer look reveals the presence of filiform corrosion on electrocleaned material, and round-type corrosion spots on
non-electrocleaned samples. EDS analysis showed the presence of Cl, S, and Ca in some cases. No relationship could be
established between the presence of these pollutants and the production processes.

2​nd ​IAS Conference on Uses of Steel, 2004, San Nicolas, Argentina ​231
(a) (b)
 Figure 4. Appearance of bare-packed materials (inner samples) after 180 days of storage a)
substrate #3 (No ECL+HNX); b) substrate #2 (ECL+EBNER).

Filiform corrosion is characterized by the formation of a network of threadlike filaments of corrosion products as a
result of exposure to a humid atmosphere [7]. Although this phenomenon is most readily observed under lacquer or
paint, it can also occur on a bare metal surface. The damage caused by this type of corrosion is very slight, the major
objection being its detrimental effect on the appearance of the metal [7]. According to literature, a relative humidity
higher than 55% at room temperature is required for filiform corrosion to occur, but some studies indicate that at lower
values it could also take place [8]. Evans, reported the occurrence of this type of corrosion on steel contaminated with
calcium sulfate after 12 hours at 99 % RH [9].

(a) (b)

Figure 6. Morphology of corrosion products by SEM. a) round-type corrosion spot on non-electrocleaned material
(100X); b) filiform-type corrosion spot on electrocleaned material (200X).

The mechanism for filiform corrosion can be described as follows: after an electrolyte (water) permeates or become
occluded under a coating, corrosion process begins, favored by the presence of both organic and inorganic anions. Iron
salts and hydroxyl ions are formed, and a differential aeration cell is set within the formed filament nucleus. In its centre
(anode) there is a deficiency of oxygen, while in the periphery (cathode) hydroxyl ions are accumulated. Hydroxyl ions
diffuse and react with ferrous ions consuming oxygen, and producing iron hydroxides that precipitate onto the steel
surface. These oxides form a barrier, passivating the nearby zone and avoiding filaments crossing each other. According
to Koehler [10] anions would migrate to the head of the filament in order to compensate electrical charge due to ferrous
ions formation. The process continues while water diffuses, until ions concentration diminishes, osmotic pressure falls
down and the filament collapses, leaving an oxide trace. A scheme of this process is observed in Fig. 7.

The occurrence of this filament-type morphology on electrocleaned substrates could be explained by the presence of a
silicate layer on the steel surface. Although the main role of ECL processing is to remove surface pollution, it also
leaves a thin silicate layer on the steel surface, which is aimed at preventing coil interturn sticking and graphitization
during batch annealing [11-15]. According to unpublished Auger analysis performed by the authors on electrocleaned
samples, the thickness of this layer is ~40Å. It is suggested that this thin film would play the role of a coating in the
mechanism described above, leading to filiform corrosion.

Corrosion products were characterized by means of FTIR microscopy. Figure 8 shows the spectrum corresponding to a
corrosion spot on a substrate stored for 75 days in the bare-condition. According to the observed bands (Table 5), the
corrosion products contain different types of iron oxi-hydroxides, such as akaganeite (​β​-FeO(OH)), goethite
(​γ​-FeO(OH)), and crystallite (​δ-​ FeO(OH)) [16, 17]. The presence of a broad band corresponding to hydroxyl ions
suggests that these compounds have a high degree of hydration.

2​nd ​IAS Conference on Uses of Steel, 2004, San Nicolas, Argentina ​232
Iron oxides and oxi-hydroxides
Figure 8. FTIR spectrum on a corrosion spot. Sample #4 after 75 days storage in bare condition.
​ ​] Vibration
ν ​[cm-1
3448 O-H stretch 1635 H-O-H bend 1466 O-H bend 1354 O-H bend 1149 ​γ​-FeO(OH) 1064 ​δ​-FeO(OH) 1018
γ​-FeO(OH) 786 ​δ​-FeO(OH) 690 ​β​-FeO(OH)
Table 5. FTIR band assignation of corrosion products.
2​nd ​IAS Conference on Uses of Steel, 2004, San Nicolas, Argentina ​233
Figure 7. Mechanism of filiform corrosion [7].
1,0 1,0
e e ccnnaabbrroossbb​
AA​4000 3500 3000 2500 2000 1500 1000 500 ​Wavenumber Wavenumber [ [ cmcm​-1 -1 ​] ]
Iron oxides and 0
​ ,8 0,8
oxi-hydroxides
0,6 0,6
0,4 0,4
0,2 0,2
0,0 0,0
4000 3500 3000 2500 2000 1500 1000 500
4. CONCLUSION

The laboratory study performed on bare and VCI-wrapped steel samples with different process routes and a dry
(oil-free) surface condition allows us to conclude that:

• ​Steel strip submitted to ECL processing and EBNER annealing is prone to oxidation, probably due to its high surface
cleanliness. The highest corrosion resistance of all the studied substrates was observed for non- electrocleaned material
with HNX annealing.

• ​Electrocleaned material develops corrosion spots with a filiform morphology after a certain storage time. On the other
hand, the morphology of corrosion spots for non-electrocleaned substrates is circular or round-type.

• ​For all the analyzed process routes, corrosion spots developed are composed by the same iron oxides and oxi-
hydroxides.

• ​Laboratory analysis showed the beneficial role of VCI paper for protecting steel. Its use would give protection to the
edges and outer turns of steel coils, whereas it seems to have a lower ability to protect the inner surfaces (within coil
turns) from rusting.

• ​No relationship was observed between polluents and process route.

• ​The results of this study led to modifications in the process routes of steel for drum manufacturing at Siderar.

5. REFERENCES

[1] R. Legault, ​“The protection of flat-rolled steel products from corrosion in storage and transit”,​ Corrosion 83, Paper
303, NACE, April 18-22, 1983. [2] B.E. Andrew, ​“Corrosion inhibition of steel: Part 2 - Long-term protection by the
application of rustproofing oils”,​ Proceedings of the International Conference on Corrosion Inhibition, May 16-20,
1983, Dallas, Texas (NACE-7), Ed. R.H. Hausler, 179-184, 1983. [3] B.E. Andrew, ​“Corrosion inhibition of steel: Part
1 - Short-term protection by nitrite-based solutions”,​ Proceedings of the International Conference on Corrosion
Inhibition, May 16-20, 1983, Dallas, Texas (NACE-7), Ed. R.H. Hausler, 172-178, 1983. [4] L.E. Helwig, ​“Rusting of
steel surfaces contaminated with acid pickle salt residues”​, Metal Finishing, Vol. 78, No. 7, 41-46, 1980. [5] B.A.
Miksic, A.S. Sobkin, and R.W. Kramer, ​“VCI technology for flat rolled steel production”,​ presented at The Association
of Iron and Steel Engineers 1998 Annual Convention and Mini-Expo Lubrication Section, 1998. [6] C. Fiaud, ​“Theory
and practice of vapour phase inhibitors”,​ in ​A Working Party Report on Corrosion Inhibitors, E ​ uropean Federation of
Corrosion Publications No. 11, The Institute of Materials, 1-11, 1994. [7] L.L. Shreir, ​“Corrosion. Volume 1:
Metal/environment reactions”,​ 2​nd ​Edition, Newnes-Butterworths, 1976. [8] H. Leidheiser Jr. en “​Corrosion
Mechanisms”​, p. 183, Florian Mansfeld ed., Dekker, New York, 1987. [9] U. R. Evans, “​The Corrosion and Oxidation
of Metals”​, Edward Arnold, 1963. [10] E.L. Koehler, “​The influence of contaminants on the failure of protective
organic coatings”, ​Corrosion - NACE, Vol. 33, No. 6, 209-217, 1977. [11] L.J. Brown, ​“Silicates as cleaners in the
production of tinplate”​, Plating 2, 225-228, 1966. [12] C. Edmond, J. Lempereur, L. Renard, and V. Leroy, ​“Particular
aspects of alkaline electro-degreasing of black-plate”​, Metallurgical Report CRM n°56, 53-64, 1980. [13] J. Kresse and
H.G. Germscheid, ​“Silicate containing cleaners for steel strip tinning”​, Plating and Surface Finishing, 122-123, 1981.
[14] P.-E. Augustsson, I. Olefjord, and Y. Olefjord, ​“Electrocleaning of cold rolled steel sheets”​, Surface Engineering 2
(4), 277-281, 1986. [15] G. Benítez, M. Ginés, E. Merli, and W. Egli, ​“Surface cleanliness and silica film deposition in
electrodegreased cold-rolled steel sheets”,​ Proceedings of the 14​th ​Rolling Conference, 323-332, 2002. [16] T. Misawa,
T. Kyuno, W. Suëtaka and S. Shimodaira, “​The mechanism of atmospheric rusting and the effect of Cu and P on the rust
formation of low alloy steels”​, Corrosion Science, Vol. 11, 35-48, 1971. [17] A. Raman, B. Kuban and A. Razvan, ​“The
application of infrared spectroscopy to the study of atmospheric rust systems -I. Standard spectra and illustrative
applications to identify rust phases in natural atmospheric corrosion products”​, Corrosion Science, Vol.32, No.12,
1295-1306, 1991.
 2​nd ​IAS Conference on Uses of Steel, 2004, San Nicolas, Argentina ​234

View publication stats View

publication stats