ARTICLE IN PRESS

Science and Technology of Advanced Materials 8 (2007) 559–565

www.elsevier.com/locate/stam

Review

Introduction to atmospheric corrosion research in China

Junhua Dong, Enhou Han, Wei Ke
Institute of Metal Research, Chinese Academy of Sciences, Shenyang 110016, China
Received 8 July 2007; received in revised form 31 July 2007; accepted 31 August 2007
Available online 9 November 2007

Abstract

In this paper, we introduce the research on atmospheric corrosion in China. We describe the climate characteristics and the
classification of atmospheric corrosivity across the whole country. We also describe the rusting evolution under simulated wet/dry cyclic
conditions.
r 2007 NIMS and Elsevier Ltd. All rights reserved.

Keywords: Atmospheric corrosion; Corrosivity; Rusting evolution

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 559
2. Field studies of atmospheric corrosion in China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 560
2.1. Climate characteristics and the National Corrosion Survey Organization of China [40–44] . . . . . . . . . . . . . . . . . . . 560
2.2. Classification of atmospheric corrosivity in China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 560
2.3. Atmospheric corrosion law for mild steel and weathering steel [11,26,53–61] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
3. Study on rusting evolution of mild steel in a simulated atmospheric environment [38–39,62] . . . . . . . . . . . . . . . . . . . . . . 562
4. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563
Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 564
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 564

1. Introduction In general, the surface aqueous electrolyte induces two

The corrosion of metals exposed to the atmosphere is
governed by the interplay of the resistance of the metal to
corrosion and the corrosivity of the aqueous electrolyte on
substances on the material surface [1]. The atmospheric
corrosion of metals is usually electrochemical, involving
the anodic dissolution of the metal, the cathodic reduction
of an oxidant, such as oxygen or g-FeOOH, and the
transport of reactants and products to and from the surface
[1–5].
Corresponding author. Tel.: +86 24 23915897; fax: +86 24 23894149.
E-mail address: jhdong@imr.ac.cn (J. Dong).
effects, resulting in the atmospheric corrosion of a metal:
(1) it connects anodic and cathodic areas to form an
electrochemical circuit with electronic conduction in the
solid state and (2) it is involved in the breakdown of
the normally protective oxide film (passive film) and the
deposition and reaction of corrosive gases on the metal.
The corrosivity of trace gases such as SO2 is therefore often
connected with the formation of an aqueous electrolyte on
the material surface. Some other corrosive trace compo-
nents, such as H+ from acidic rain, may replace oxygen as
the oxidizing agent of the metal, while others, such as
NaCl, tend to facilitate the dissolution of the metal.
However, the resistance to atmospheric corrosion is mainly
governed by the chemical activity of the metal itself and the

1468-6996/$ - see front matter r 2007 NIMS and Elsevier Ltd. All rights reserved.
doi:10.1016/j.stam.2007.08.010
 ARTICLE IN PRESS
560 J. Dong et al. / Science and Technology of Advanced Materials 8 (2007) 559–565
chemical stability of its corrosion products formed on the
surface. For instance, aluminum is easily corroded because
of its high chemical activity. However, the corrosion
product of aluminum, alumina, is stable because of its
chemical inertia. This means that aluminum is a good
engineering material with high corrosion resistance. In
addition, alloying elements may have a significant effect on
improving the corrosion resistivity of the products formed
on the metal. For example, the alloying of nickel and
copper in weathering steel [6,7] may strengthen the rust
structure and increase the resistance to atmospheric
corrosion [8,9].
Atmospheric corrosion may cause serious damage to
metallic structures and equipments. The effect is so serious
that the annual cost of atmospheric corrosion is approxi-
mately half the total annual cost of all types of corrosion of
metals [10]. Corrosion scientists and engineers of many
countries have carried out numerous investigations on
atmospheric corrosion [10–15]. In past decades, atmo-
spheric corrosion research was mainly dominated by field
studies comparing the resistance to corrosion of different
materials [16–18], the corrosivity of different environments
[19–24] and on empirical correlations between corrosion
rate and environmental parameters [25–29]. A few studies
addressed the processes occurring on the material surface
that cause corrosion [30–36].
China started to establish atmospheric corrosion test
stations at the end of the 1950s. Since then, 13 national
atmospheric corrosion test stations have been successfully
established, the sites of which are selected as being typical
examples of urban, rural, coastal and industrialized
environments, and some valuable results have been
obtained [37]. Recently, a few simulative studies addressing
the processes occurring on the material surface that cause
atmospheric corrosion have been carried out, and the
mechanism of the rusting process of mild and low-alloy
steel was proposed [38,39]. This paper will concentrate
mainly on the results of the field corrosion investigations
and simulative corrosion studies of iron and steel, with
references to other metals when necessary.
2. Field studies of atmospheric corrosion in China
2.1. Climate characteristics and the National Corrosion
Survey Organization of China [40–44]
There are seven types of climate in China (Fig. 1). The
eastern part mainly has the East Asian monsoon climate,
which is warm and humid with high rainfall. However, the
northwest part has a typical inland arid climate. The
northern part includes frigid and frigid temperate zones.
The south includes a warm temperate zone, a humid
subtropical zone and a humid tropical zone. Most of the
country is in the frigid temperate zone, warm temperate
zone, or humid subtropical zone. The industrial develop-
ment of the eastern region has resulted in serious atmo-
spheric contamination, while the tableland in the western
Fig. 1. Plot of climate zones of China and distribution of 13 NASCM test
sites, shown by saffron dots.
part is exposed to strong ultraviolet radiation and frequent
sandstorms. Close to Qianghai Lake, the sand contains a
lot of salt [45]. Because of the big differences in the climate
and concentrations of corrosive substances in the atmo-
sphere across the country, the corrosion rate of materials is
strongly dependent on the site where the materials are
located.
Since the late 1950s, under the joint leadership of the
Ministry of Science and Technology (MOST) and the
National Natural Science Foundation of China (NSFC),
China has successfully built up the National Network for
the Atmospheric Corrosion Survey of Materials (NACSM)
and 13 test stations have been built across the country
(Fig. 1). Among these sites, five stations were built in the
1950s, three in the 1970s, three in the 1990s and three after
2000. The sites cover the seven types of climate and
represent the four atmospheric environments of urban,
rural, coastal and industrialized areas. Information on
these sites is given in Table 1.
2.2. Classification of atmospheric corrosivity in China
A classification of the corrosivity of the atmosphere
based on surveys of time of wetness, temperature and
contamination (ISO9223) at seven sites in eastern China is
given in Table 2 [11,37,46–52]. It is possible to place each of
these sites into one of the five categories of corrosivity.
Table 2 shows that the atmosphere at the listed sites has
corrosivity levels of 3 or 4.
In addition to the classification based on the surveys of
wetting time, temperature and contamination (ISO9223),
the corrosivity of the atmosphere was classified into seven
categories on the basis of a number of surveys of the rate of
corrosion of a standard specimen (ISO9223), as shown in
Table 3. In the table, a category lower than C1 and a
category higher than C5 have been added, and the average
corrosion rate over a 10-year exposure test was used
instead of the 1-year test in ISO9223. To allow a
 ARTICLE IN PRESS
J. Dong et al. / Science and Technology of Advanced Materials 8 (2007) 559–565 561

Table 1
The 13 NASCM test sites

Station Year Geographical position Altitude Climate type Average T Radiation Rainfall RH
built (m) ( 1C) time (h/a) (mm/a) (%)
Longitude Latitude

Mohe 1991 1221230 531010 613.0 Rural frigid temperate 3.5 2317.9 236.7 69.2
Shenyang 1991 1231260 411460 41.6 Urban frigid temperate 9.4 2279.3 408.5 66
Beijing 1957 1161160 361060 73.4 Urban warm temperate 11.9 2559.3 586.0 57
Dunhuang 2002 941410 401090 1139.9 Rural arid tropical 10.8 3257.8 35.0 45
Kuerle* 2005 861130 411240 ? Rural frigid temperate 11 3050.9 48.9 –
Qingdao 1976 1201250 301380 18.0 Coastal warm temperate 12.3 1944.0 561.7 71
Wuhan 1958 1141040 291190 23.3 Urban humid 16.8 1621.6 1146.8 75
subtropical
Jiangjin 1974 1061170 231080 208.0 Industrial humid 17.9 1317.0 1202.9 81
subtropical
Lhasa 2002 911080 291400 3648.7 Tableland frigid 9 3053.1 580.9 46
temperate
Guangzhou 1956 1131170 1131170 6.3 Urban humid 22.7 1607.0 1562.7 77
subtropical
Qionghai 1956 1101050 191020 10 Rural humid tropical 24.3 2072.0 1794.0 86
Wanning 1974 1101050 181580 12.3 Coast humid tropical 24.2 2026.5 1515.0 86
Xishuangbanna 1958 1001400 211350 626 Rural tropical 21.6 1716 1713 83

Table 2
Corrosivity classification based on time of wetness, temperature and contamination for seven NACSM test stations (ISO9223)

Station Cl SO2 Time of wetness Corrosivity Environmental

categories characteristic
mg (100 cm2 d)1 Class mg (100 cm2 d)1 Class RH480% (h a1) Class

Beijing 0.049 S1 0.442 P2 2358 T3 3 or 4 Urban

Qingdao 0.25 S1 0.704 P2 4049 T4 4 Coastal
Wuhan 0.011 S0 0.272 P1 4871 T4 3 Urban
Jiangjin 0.006 S0 0.667 P2 5304 T4 4 Industrial
Guangzhou 0.024 S0 0.107 P1 5048 T4 3 Urban
Qionghai 0.199 S1 0.15 P1 6314 T5 3 or 4 Rural
Wanning 0.387 S1 0.06 P0 6736 T5 3 or 4 Coastal

Table 3
Atmospheric corrosivity classification based on surveys of the corrosion rate at various sites in China

Level Atmospheric Calculated Average corrosion rate (mm/a, 10 years)

corrosivity corrosivity
A3 Low-alloy steel Zn Cu Al

1 C0 Lowest o1 o0.1 o0.05 o0.005 o0.002

2 C1 Very low 1–10 0.1–0.5 0.05–0.1 0.005–0.01 0.002–0.01
3 C2 Low 10–50 0.5–5 0.1–0.5 0.01–0.1 0.01–0.025
4 C3 Middle 50–150 5–12 0.5–2 0.1–1.5 0.025–0.2
5 C4 High 150–250 12–30 2–4 1.5–3 0.2–1
6 C5 Very high 250–350 30–100 4–10 3–5 1–3
7 C6 Highest 4350 4100 410 45 43

comparison with ISO9223, we also adjusted the scale of the
corrosion rate corresponding to each category in Table 3,
which was different from the standard data listed in
ISO9223. A number of surveys to determine the corrosion
rate at various sites in China made it possible to place each
of these sites into one of the seven categories. Referring to
the corrosion rate of A3 mild steel, Table 4 shows that the
atmosphere at the listed sites generally has corrosivity
levels of 3 or 4. The category of atmospheric corrosivity is
related to not only the environment but also the material
and exposure period. Table 4 shows that the category for
mild steel is generally different from those for zinc, copper
and aluminum throughout the sites. Comparing the
corrosion rate for the 1st year with the average rate over
 ARTICLE IN PRESS
562 J. Dong et al. / Science and Technology of Advanced Materials 8 (2007) 559–565

Table 4 the time of wetness is lower than those of Wanning and

Corrosivity classification based on corrosion rate of standard specimen for Qionghai, n is less than 1 for both mild and Corten-A
seven NACSM test stations (ISO9223)
steels. Comparing the data on the mdd of chloride ions and
A3 steel Zinc Copper Aluminum SO2 listed in Table 2, it can be concluded that SO2 has a
significant effect on the values of A and n. The most typical
1 10 1 10 1 10 1 10 sites are Jiangjin and Qingdao. In Jiangjin, where the mdd
Beijing 3 3 2 3 4 3 2 2 of SO2 is 0.667 and that of Cl is very low, n is 0.45. In
Qingdao 4 4 3 4 5 3 4 4 Qingdao, where the mdd of SO2 is 0.704 and that of Cl is
Wuhan 3 3 3 3 4 3 3 3 0.25, n is 0.57. The higher value of n for Qingdao is caused
Jiangjin 4 4 4 4 5 4 3 4 by the interaction of Cl and SO2 in the corrosion process
Guangzhou 4 4 3 4 5 3 4 4
of steel. Because the time of wetness in Qingdao is shorter
Qionghai 3 4 2 2 5 3 2 4
Wanning 3 5 3 3 4 4 3 3 than that of Jiangjin, A for Qingdao is smaller. Data for
Corten-A steel exposed to the atmosphere at Qingdao and
Jiangjin shows a pattern similar to that for mild steel.

Table 5 3. Study on rusting evolution of mild steel in a simulated

Corrosion data for mild steel and weathering steel atmospheric environment [38–39,62]
Station Mild steel Corten-A steel
The field investigation of atmospheric corrosion provides
A/mm n A/mm n actual information on the corrosion of materials; thus, it is
the most important method of evaluating atmospheric
Beijing 0.032 0.45 0.028 0.63
Qingdao 0.058 0.57 0.019 0.73 corrosion. However, it usually takes about 10–20 years for
Wuhan 0.047 0.39 0.048 0.40 one evaluation period, which makes it difficult to determine
Jiangjin 0.083 0.45 0.053 0.41 the rusting evolution induced atmospheric corrosion and to
Guangzhou 0.06 0.48 0.032 0.26 evaluate new materials designed to resist atmospheric
Qionghai 0.024 1.03 0.040 0.46
corrosion. In previous studies, simulative experiments
Wanning 0.033 1.60 0.034 0.30
enabling the acceleration of corrosion have been developed.
Complex cyclic corrosion tests (CCTs) based on saltwater
spray followed by drying have been widely used to accelerate
10 years, the category increases with time for mild steel, the corrosion of coated and noncoated steels. Such tests
zinc and aluminum but decreases for copper. This is caused have been also adopted in several standards. The rusting of
by the differences in the types of corrosion and the steel in the atmosphere includes many reaction steps at
corrosion products formed on the material surface. The different corrosion stages. In the initial stage, the anodic
atmospheric corrosivity at the six sites built since the 1990s dissolution of steel and the cathodic reduction of oxygen
has not yet been classified. dominate the corrosion. Then, the dissolved ferrous ions are
hydrolyzed into a deposit of Fe(OH)2. During the drying of
2.3. Atmospheric corrosion law for mild steel and weathering the aqueous film, Fe(OH)2 is oxidized into a colloid deposit
steel [11,26,53–61] of Fe(OH)3, which can transform into an amorphous
deposit of FeOOH on the dry surface of steel. The accumu-
On the basis of the regression analysis of the atmospheric lated deposit on the steel surface blocks the transmission of
corrosion data measured at the test sites, the rusting oxygen to the steel surface. Then, the cathodic reduction is
evolution of metal satisfies the exponential equation: dominated by the reduction of FeOOH, and the reductive
product is Fe3O4. Meanwhile, the crystallization of the
D ¼ Atn , (1)
amorphous FeOOH is inhibited with the development of
where D is the corrosion depth, t is the exposure time of atmospheric corrosion, a-FeOOH, b-FeOOH, g-FeOOH
the specimen, and A and n are constants. A represents and magnetite are the main components of steel rust. The
the corrosion rate of the first year and n represents the structural evolution of the rust layer of mild steel has been
propensity of corrosion. Table 5 lists some simulative data proposed by Dong et al. [38,39]. Using an accelerated CCT
for mild steel and weathering steel collected at various test proposed by Nishimura et al. [62], the weight gain of rust on
sites in eastern China. At Qionghai and Wanning, which mild steel with increasing number of wet/dry cycles was
are located on Hainan island, n is higher than 1 for mild determined; the results are shown in Fig. 2(a). The simulated
steel but less than 0.5 for Corten-A steel. Even though result shows that the rusting process includes two stages,
Wanning and Qionghai have almost the same humidity and and the transition is at approximately 30 cycles. Fig. 2b
temperature, the mdd (mg dm2 day1) of chloride ions in shows that the rate of rust formation increases with
Wanning is higher than that in Qionghai; thus, the values increasing number of wet/dry cycles in the first stage but
of both A and n in Wanning are higher than those in decreases after the transition. It was shown by XRD analysis
Qionghai. In the other stations listed in Table 5, because that the initial rust included a large amount of amorphous
 ARTICLE IN PRESS
J. Dong et al. / Science and Technology of Advanced Materials 8 (2007) 559–565 563

Fig. 2. Result of corrosion acceleration test for mild steel with 0.3% NaCl solution: (a) weight gain and (b) corrosion rate.

Fig. 3. Micromorphology evolution of the rust layer with increasing number of wet/dry cycles.

FeOOH with exiguous a-FeOOH and b-FeOOH. After the

transiton, g-FeOOH and magnetite appear as the inter-
growth in the rust. Fig. 3 shows the micromorphology
evolution during the rusting of mild steel. It shows that the
initial rust layer has low density but it transforms into a
dense layer after the transition. A flow chart of the rusting
evolution of mild steel under wet/dry cyclic corrosion
conditions is shown in Fig. 4.

4. Summary

1. In this paper, we have briefly discussed the classification Fig. 4. Flow chart of the rusting evolution of mild steel under wet/dry
of the atmospheric corrosivity in the eastern region of cyclic-corrosion conditions.
 ARTICLE IN PRESS
564 J. Dong et al. / Science and Technology of Advanced Materials 8 (2007) 559–565
China. The NASCM has set up a database containing a
substantial amount of data from the test sites located in
the eastern region. With the rapid development of the
economy of China, the NASCM will collect atmospheric
corrosion data from the other stations located in the
north and west to extend the database, which will be
used in conjunction with future construction projects.
2. The rusting evolution of mild steel in a simulated coastal
atmospheric environment was discussed. The rusting
process includes two stages. The corrosion rate increases
with time in the initial stage, during which the rust
consists of a large amount of amorphous FeOOH with
exiguous crystallized a-FeOOH and b-FeOOH in a low-
density layer. However, the corrosion rate decreases
with time in the subsequent stage, and the rust forms a
high-density layer of g-FeOOH and magnetite.
Acknowledgement
The work is funded by NSFC Projects 5049336 and 973
plan 2004CB619101. The authors acknowledge the assis-
tance.
References
[1] U.R. Evans, Corros. Sci. 9 (1968) 813.
[2] T. Misawa, K. Asami, K. Hshimoto, Corros. Sci. 14 (1974) 279.
[3] J.T. Keiser, C.W. Brown, R.H. Heidersbach, Corros. Sci. 23 (1983)
251.
[4] A. Raman, A. Razvan, B. Kuban, Corrosion 42 (1986) 447.
[5] I. Suzuki, N. Masuko, Y. Hisamatsu, Corros. Sci. 19 (1979) 521.
[6] K. Noda, T. Nishimura, H. Masuda, T. Kodama, J. Jpn. Inst. Metals
43 (1999) 1133.
[7] A. Usami, H. Kihira, T. Kusunoki, Nippon Steel Technical Report
87, 2003, p. 17.
[8] X.H. Chen, Ph.D. Thesis, Institute of Metal Research, Chinese
Academic Science, 2007.
[9] J.H. Dong, X.H. Chen, E.H. Han, W. Ke, Iron Steel 40 (2005) 127.
[10] W. Ke, Chinese Corrosion Survey Report, First ed, Chemical
Industry Press, Beijing, ISBN 7-5025-4792-4/TQx1816, 2003, p. 13.
[11] C.N. Cao, Corrosion in Natural Environment of Materials in China,
First ed, Chemical Industry Press, Beijing, ISBN 7-5025-6031-9/
TQx2062, 2005, p. 69.
[12] H.H. Uhlig, Corrosion 6 (1950) 29.
[13] K. Barton, Protection against Atmospheric Corrosion, Wiley,
London, 1976.
[14] G.Y. Wang, Corros. Sci. Prot. Technol. 1 (1989) 41, 48.
[15] I.L. Rozenfeld, Atmospheric Corrosion of Metals, NACE, Houston,
1972.
[16] C.P. Larrabee, S.K. Coburn, The atmospheric corrosion of steels as
influenced by changes in chemical composition, in: Proceedings of the
First International Congress, London, 1962, p. 276.
[17] C.R. Shastry, J.J. Fried, H.E. Townsend, in: S.W. Dean, T.S. Lee
(Eds.), Degradation of Metals in Atmosphere, ASTM, Philadelphia,
1988, p. 5.
[18] F.I. Wei, Br. Corros. J. 26 (1991) 209.
[19] L.G. Johansson, Ninth International Congress on Metallic Corro-
sion, Toronto, vol. 1, 1984, p. 407.
[20] S. Oesch, M. Faller, Corros. Sci. 39 (1997) 1505.
[21] J. Tidblaol, C. Leygraf, J. Electrochem. Soc. 144 (1997) 3513.
[22] S. Zakipour, J. Electrochem. Soc. 142 (1995) 749.
[23] M. Morcillo, B. Chiro, E. Otero, Mater. Perform. 38 (1999) 72.
[24] I.S. Cole, D.A. Patterson, S.A. Furman, A holistic approach to
modeling atmospheric corrosion, 14th International Corrosion
Congress, Cape Town, 1999, p. 265.
[25] T.E. Gradel, Corros. Sci. 38 (1996) 2153.
[26] M. Pourbaix, The Linear Bilogarithmic Law for Atmospheric
Corrosion, Wiley, New York, 1982, p. 107.
[27] J.W. Spence, F.H. Haynie, in: R. Baboian, S.W. Dean (Eds.),
Derivation of a Damage Function for Galvanized Steel Structure:
Corrosion Kinetic and Evaluation, ASTM STP 1000, 1990, p. 208.
[28] F.H. Haynie, J.W. Spence, F.W. Lipfert, S.D. Cramer, L.G.
McDonald, in: R. Baboian, S.W. Dean (Eds.), Evaluation of An
Atmospheric Damage Function for Galvanized Steel, 13th Interna-
tional Corrosion Congress, Australia, 1996, p. 36.
[29] M. Morcillo, S. Feliu, J. Simancas, Br. Corros. J. 28 (1993) 50.
[30] T.E. Gradel, R.P. Frankenthal, J. Electrochem. Soc. 137 (1990) 2385.
[31] M. Yamashita, H. Miyuki, Y. Matsuda, Corros. Sci. 36 (1994) 283.
[32] S. Rivero, E. Quagliata, V. Diaz, Mössbauer spectroscopy study of
corrosion products developed on unalloyed steel samples exposed to
natural environments, 13th International Corrosion Congress,
Australia, 1996, p. 33.
[33] A.K. Neufeld, I.S. Cole, Corrosion 53 (1997) 788.
[34] K. Noda, H. Masuda, T. Kodama, AFM study on the rusting process
on iron under atmospheric corrosion environment, 14th Interna-
tional0 Corrosion Congress, Cape Town, South Africa, 1999, p. 94.
[35] T. Aastrup, C. Leygraf, M. Wadsak, In-situ studies of the metal/
atmosphere interface, 14th International Corrosion Congress, Cape
Town, South Africa, 1999, p. 10.
[36] I. Odnevall, C. Leygraf, in: W.W. Kirk, H.H. Lawson (Eds.),
Reaction Sequences in Atmospheric Corrosion of Zinc, ASTM STP
1239, Philadelphia, 1995, p. 215.
[37] G.Y. Wang, Q.M. Shu, Bull. Nat. Nat. Sci. Found. China 6 (1992) 40.
[38] J. Dong, J.H. Dong, E.H. Han, W. Ke, Ch.M. Liu, Corros. Sci. Prot.
Technol. 18 (2006) 414.
[39] J. Dong, J.H. Dong, E.H. Han, W. Ke, Ch.M. Liu, Corros. Sci. Prot.
Technol. 20 (2008), to be published.
[40] G.Y. Wang, H.J. Wang, L.X. Lian, Y.D. Yin, Corrosion and
Protection of Natural Environment—Atmosphere, Sea Water and
Soil, Chemical Industry Publishing House, Beijing, 1997.
[41] M.Z. Li, J. Zhang, F.Y. Qi, Study on Environmental Test for
Aeronautical Materials and Equipments, Sci. Technol. Res. Inst.
Aeron. Astron. Dep., Beijing, 1993, p. 71.
[42] M.Z. Li, J. Zhang, F.Y. Qi, J. Chin. Soc. Corros. Prot. 13 (1993) 10.
[43] Zh.J. Li, F.Y. Qi, M.Z Li, Quantitative prediction of atmospheric
corrosivity and application, C-MRS and MRS-K Joint Symp. ‘96,
China, Beijing, 1996, p. 217.
[44] J.Zh. Li, Research on data accumulation of atmospheric corrosion,
accelerated test and atmospheric corrosivity of material, Major
Program of National Natural Science Foundation of China,
‘‘Corrosion of Materials in Natural Environments’’ summary data
of ‘‘the 8th Five-Year Plan’’ research, Serial number: 59290912-02.
[45] Q.X. Li, Z.Y. Wang, W. Han, E.H. Han, Corros. Sci. 63 (2007) 640.
[46] ISO 9223, Corrosion of metals and alloys—corrosivity of atmo-
spheres—classification.
[47] ISO 9224, Corrosion of metals and alloys—corrosivity of atmo-
spheres—guiding values for the corrosivity categories.
[48] ISO 9225, corrosion of metals and alloys—corrosivity of atmo-
spheres—measurement of pollution.
[49] ISO 9226, Corrosion of metals and alloys—corrosivity of atmo-
spheres—determination of corrosion rate of standard specimens for
the evaluation of corrosivity.
[50] C.F. Liang, W.T. Hou, J. Chin. Soc. Corros. Prot. 18 (1998) 1.
[51] Zh.Q. Li, Newsletters of national network for environmental
corrosion survey of materials in China 247 (1999) 7.
[52] Zh.Q. Li, Newsletters of national network for environmental
corrosion survey of materials in China 248 (1999) 9.
[53] J. Yan, Ch.Q. Qiong, J. Chin. Soc. Corros. Protec. 6 (1986) 1.
[54] W.T. Hou, Proceedings of the 11th International Corrosion
Congress, Florence, Italy, 1990, p. 2.
 ARTICLE IN PRESS
J. Dong et al. / Science and Technology of Advanced Materials 8 (2007) 559–565 565

[55] W.T. Hou, J.D. Yu, C.F. Liang, J. Chin. Soc. Corros. Prot. 13 (1993) 291. [59] C.F. Liang, W.T. Hou, J. Chin. Soc. Corros. Prot. 17 (1997) 87.
[56] W.T. Hou, J.D. Yu, C.F. Liang, Corros. Sci. Prot. Technol. 6 (1994) [60] C.F. Liang, W.T. Hou, J. Chin. Soc. Corros. Prot. 18 (1998) 1.
137. [61] W.T. Hou, C.F. Liang, Corrosion 55 (1999) 65.
[57] W.T. Hou, C.F. Liang, J. Iron Steel Res. 6 (1994) 40. [62] T. Nishimura, H. Katayama, K. Noda, T. Kodama, Iron Steel 86
[58] C.F. Liang, W.T. Hou, Corros. Sci. Prot. Technol. 7 (1995) 182. (2000) 39.