984 　塑性と加工（日本塑性加工学会誌）第 54 巻　第 634 号（2013―11）

Adhesive Strength of Oxide Scale Formed on Low-Carbon Steel

Yasumitsu KONDO* and Hiroshi TANEI *
(Received on November 20, 2012)

The scale formed on the surface of steel during hot rolling causes surface defects. Thus, controlling scale
adhesion is important; however, there have been few reports on quantitative investigations of scale adhesion at high
temperature. In this study, a method of measuring scale adhesive strength at high temperature is investigated. The
effect of the external stress on the oxide scale is also investigated. The adhesive strength is on the order of 1 MPa. The
adhesive strength increases proportionally as axial stress applies on the scale. The scale/steel interface analysis shows
that there are many voids at the interface after the oxidation, however, the number of voids decreases and a rough
interface is formed after axial compression stress is applied on. It is found that the adhesive strength of the scale
depends on the contact area and the rough interface between the scale and the steel.

Key words: rolling, steel, scale, adhesive strength, measurement, axial compression, interface.

scale and a material was affected by the joining stress.6)

1. Introduction
The authors improved the method of Kushida et al.6) Scale
In steel production, the surface of steel is oxidized and scale merging is performed by oxidation with small compression
forms on it at high temperature; this oxide scale may cause stress rather than with high compression stress.7) Scale adhesive
surface defects. Typically, scale is removed by high pressure strength was measured with low compressive stress at high
water before hot rolling because the roll may push a thick scale temperature quantitatively7). Here, a detailed study of the effect
into the steel surface. Surface defects may be caused by the of external axial strength on the scale adhesion is conducted.
spontaneous detachment of blistered scale; blistering occurs
2. Experimental
when oxide scale swells during oxidation. Thus, the adhesion
property of scale at high temperature is an important Steel materials with the chemical components shown in Table
characteristic, and understanding and controlling scale adhesion 1 were used for the experiment. Cylindrical samples with a
properties would contribute to the quality of steel produced. diameter of 10 mm were used. An additional cylindrical
Some methods of measuring scale adhesive strength specimen made of electrolytic iron was used for the sample with
qualitatively have been reported. These methods can evaluate which the scaled sample is merged. This is why the iron has
the amount of detached scale or the detaching behavior when higher scale adhesive strength than the steel used for the
compressive1) or tensile2), 3) force applies on the sample material. experiment.
A method of measuring such force when scale has been
mechanically detached after a bolt is buried in scale has been Table 1 Chemical compositions of the samples (mass%)
reported.4) However, it is difficult to use the method to evaluate C Si Mn P S Al
adhesion quantitatively in the case of a thin oxide scale (tens of A 0.094 0.028 0.45 0.004 0.003 0.016
micrometers in thickness). A method of obtaining adhesive B 0.092 0.014 0.43 0.004 0.003 0.014
energy by in situ observation when a tensile test is conducted on
an oxidized steel sample has been reported.5) It is possible to Tensile test equipment with a furnace was used for the
apply this method at high temperature. Kushida et al.6) proposed experiments. The atmosphere in the furnace could be controlled.
a method of estimating scale exfoliation stress. After generating The experimental method used to measure scale adhesive
scale on two pillar specimens at high temperature, the scales strength is shown in Fig. 1. Two cylindrical samples were placed
formed on the two samples were joined by pressurizing at high 10 mm apart at room temperature. The electrolytic iron sample
temperature. Next, by the tensile test of the joined specimen, the was placed on the top and the sample used to measure scale
force and stroke curve was obtained. The scale exfoliation stress adhesive strength was placed at the bottom. The samples were
is estimated from the curve. The exfoliation stress between a heated up to 950 °C and oxidized for 360 s. Oxide scales formed
*
Integrated Process R & D Div. Process Research Laboratories,
on both samples. Then, the electrolytic iron sample was moved
Technical Research & Development Bureau, downward and positioned on the lower sample with constant
Nippon Steel & Sumitomo Metal Corporation
20-1 Shintomi, Futtsu, Chiba 293-8511, Japan. compressive stress in the axial direction. The scale formed on

38
 Journal of the JSTP vol. 54 no. 634（2013―11） 985

the electrolytic iron sample merged with the scale formed on the
lower sample. The axial compressive stress was varied from 6.4 Opposite side sample
200 μm
kPa to 12.7 MPa during the scale merging. After the scales Scale on the opposite side sample

merged, the atmosphere was changed from oxidizing nitrogen to

non-oxidizing nitrogen. Then, the electrolytic iron sample was Scale on the sample Resin
pulled up and the load was recorded during scale detachment
from the lower sample. The scale merging and separation Fig. 3 Optical microscopy image of a cross section of the scale
processes were conducted at the oxidation temperature (950 °C). after adhesive strength measurement.7) Surface
The maximum stress was considered to be the scale adhesive appearance after tests
strength when the scale on the lower sample was completely
detached.
0.8

Oxidation Scale merging Scale separation

Pull up 0.6
Push
down

Stress [MPa]
Opposite
Opposite side
side 0.4
sample
sample
Scale
Scales Detached
formation 0.2
adhere scale
to each
Sample other Sample
0
0 0.01 0.02 0.03 0.04
Displacement [mm]
Fig. 1 Method of measuring scale adhesive strength
Fig. 4 Load change during experiment. The stress was
recorded every 10 ms
3. Results

A photograph of typical samples after the separation for The axial compressive stress applied on the scale during scale
adhesive strength measurement is shown in Fig. 2. The scale of merging changes from 6.4 kPa to 1.27 MPa using material A, as
the lower sample is completely detached and the metal surface shown in Table 1. The measured scale adhesive strengths are
appears [Fig. 2 (b)]. Fig. 3 shows an example of the cross shown in Fig. 5. Although scale adhesive strength has a certain
section of the detached scale. The scale formed on the lower range when the axial compressive stress is small, above
sample is well merged with the scale formed on the electrolytic approximately 0.1 MPa, the adhesive strength becomes higher in
iron sample. Fig. 4 shows the stress change during the proportion to the axial compressive stress. When the
measurement when an axial compression stress of 6.4 kPa is compressive stress was 12.7 MPa, the measured stress became
applied on. The stress increases almost linearly when the very high and reached 12.8 MPa and the attached part of the
electrolytic iron sample is pulled up. The sample and tools are specimen was damaged. It is assured that under these conditions
elastically deformed in this region. The stress decreases rapidly the scale adhesive strength is more than 12.8 MPa.
after the scale has detached. In this case, the adhesive strength The above results indicate that scale adhesive strength is
for the maximum stress obtained is 0.65 MPa. After the stress proportional to compressive stress above a certain value and that
reaches its maximum, the stress decreases to zero. It is it can be measured below that value. The scale adhesive strength
considered that the fraction of detaching oxide scale increases of material A is on the order of 1 MPa.
during this stress decrease.
4. Discussion

(a) (b) The method used to measure scale adhesive strength in this
study is similar to that used by Kushida et al.6) The scale
exfoliation stress obtained by Kushida et al. ranges from 15 to
50 MPa and is much higher than the adhesive strength we
obtained, which may be attributed to the high compressive stress
applied on during scale joining. The result shown in Fig. 5
5mm indicates that measured scale adhesive strength is affected by
Fig. 2 Surface appearance after tests. compression stress during scale merging above 0.04 MPa. A
(a) Opposite (upper) side sample. (b) Lower sample sufficiently low compressive stress during scale merging is
(scale is completely detached) required to measure scale adhesive strength.

39
986 　塑性と加工（日本塑性加工学会誌）第 54 巻　第 634 号（2013―11）

100 Scale
More than
12.8 MPa
Scale adhesive force [MPa]

10
Steel
1μm
1 Fig. 6 SEM image of scale/steel interface when axial
compression of 25 kPa is applied on during scale
merging7)
0.1
0.001 0.01 0.1 1 10 100
Axial compression force
Scale
at scale merging [MPa]

Fig. 5 Effect of axial compressive stress during scale merging

on scale adhesive strength Steel 1μm
Fig. 7 SEM image of scale/steel interface when axial
We should realize that the scale adhesive strength measured in
compression of 12.7 MPa is applied on during scale
this method may differ from the adhesive strength of the scale
merging
formed on an actual steel surface. The scale on the actual steel
surface has three layers: hematite, magnetite, and wustite. In this
The scale/steel interface when oxide scale is formed is shown
method, the scale structure changes from a three-layer structure
schematically in Fig. 8(a). There are some voids at the interface
to a wustite mono layer structure during merging. The scale
and the contact area is small. The interface where compressive
structure affects growth stress in oxide scale. Growth stress may
stress is applied on the scale is shown schematically in Fig. 8(b).
be a factor for scale adhesion.
There are few voids at the interface. The contact area is larger
Using material B shown in Table 1, two experiments were
and a rough interface is also formed.
conducted to examine the effect of outer force on scale adhesive
strength. In the first experiment, the axial compressive stress
was 25 kPa. After scale merging, the samples were cooled down (a)

and the axial compressive stress was maintained at 25 kPa. After Scale
cooling, a cross section of the sample was obtained using the
focused ion beam (FIB) technique to avoid damaging the
structure of the scale/steel interface. Fig. 6 shows a scanning Voids
Steel
electron microscopy (SEM) image of the cross section of the
scale/steel interface. Voids between the scale and the steel are
observed. It is suggested that the scale makes partial contact
with the steel. The void formation at the scale/steel interface is (b) Compression
associated with the outward diffusion of iron ions in the oxide
via ion vacancies, which condense and form voids near the Scale
scale/steel interface.
In the second experiment, quite a high axial compressive
stress (12.7 MPa) was applied on. As in the first experiment, the
Steel
axial compressive stress was held at 12.7 MPa while the samples
were being cooled. A cross section of the sample was made by
the FIB technique; the SEM image obtained at the cross sections
Fig. 8 Schematic representations of scale/steel interface.
is shown in Fig. 7. There are few voids between the scale and
(a) Oxidized. (b) Axial compression is applied on
the steel. It appears that both the scale and the steel intruded on
each other by plastic deformation at the interface, resulting in a
It is considered that the improvement of scale adhesion by
rough interface. It is reported that the oxide of wustite deforms
external force is due to the increase in the contact area and
plastically above 700°C8). It is considered that the increased
roughness at the scale/steel interface. It is supposed that, during
contact area and rough interface result in pronounced scale hot rolling, scale is compressed under high pressure in roll bites
adhesion. and has very high adhesive strength. In this study, we suggest

40
 Journal of the JSTP vol. 54 no. 634（2013―11） 987

that external force applied on oxide scale causes the scale/steel

References
interface to deform plastically, thereby increasing the contact
area between the scale and the steel; both the deformation and 1) Takeda, M., Onishi, T. & Mukai, Y.: Kobe Steel Engineering
the increased contact area affect scale adhesive strength. Reports, 55-1 (2005), 31-36.
However, the mechanism of scale adhesion on steel is not yet 2) Okada, H.: J. JSTP, 44-505 (2003), 94-99.
understood. Further research is necessary. 3) Krzyzanowski, M. & Beynon, J. H.: ISIJ Int., 46-11 (2006),
1533-1547.
5. Conclusion
4) Morita, M., Nishida, M. & Tanaka, T.: Tetsu-to-Hagane, 68-5
A method of measuring scale adhesive strength by oxidizing, (1982), 114.
scale-merging and separating two samples has been proposed. 5) Chandra-Ambhorn, S., Roussel-Dherbey, F., Toscan, F.,
The application of this method revealed that the scale adhesive Wouters, Y., Galerie, A. & Dupeux, M.: Materials Science and
strength is on the order of 1 MPa and is increased by the Technology, 23-4 (2007), 497-501.
application of compressive stress. The scale/steel interface is 6) Kushida, H., Maeda, Y. & Ishikawa, T.: Tetsu-to-Hagane,
deformed by external force, thereby increasing the contact area 98-11 (2012), 593-600.
between the scale and the steel. It is suggested that the contact 7) Kondo, Y. & Tanei, H.: Proc. of the 63rd Japanese Joint
area and rough interface between an actual scale and steel Conference for the Technology of Plasticity, (2012), 457-458.
greatly affects scale adhesive strength. 8) Hidaka, Y., Anraku, T. & Otsuka, N.: Oxidation of Metals,
59-1/2 (2003), 97-113.

41