Materials Science and Engineering A 432 (2006) 47–51

Corrosion and its inhibition in SA213-T22 TIG weldments used in

power plants under neutral and alkaline environments
S. Natarajan ∗ , S.P. Kumaresh Babu
Department of Metallurgical & Materials Engineering, National Institute of Technology, Tiruchirappalli 620 015, Tamilnadu, India
Received 18 February 2006; received in revised form 21 April 2006; accepted 12 June 2006

Abstract
Failures of industrial boilers have been reported to be mainly due to water side corrosion and scaling. Boiler corrosion is due to highly alkaline
or acidic conditions of the boiler water. The present work has been aimed at investigating the corrosion behaviour of T22 (2.25 Cr–1 Mo) boiler
steel TIG weldments in neutral and alkaline media. The parent metal and weld root regions are chosen as regions of exposure for the study made at
room temperature and at 100 ◦ C. Electrochemical polarization technique such as Tafel line extrapolation (Tafel) was used to measure the corrosion
current. Corrosion inhibition using thiourea at 100 ppm concentration was studied in these experiments. Micro-structural observation and surface
characterization using SEM and XRD studies have been made on samples exposed at 100 ◦ C. The results show that parent metal experiences higher
corrosion rate than weld metal and thiourea as inhibitor is found unsuitable in NaOH medium.
© 2006 Elsevier B.V. All rights reserved.

Keywords: Boiler corrosion; T22 TIG weldments; Neutral and alkaline media; Thiourea Inhibitor; SEM and XRD

1. Introduction ing results in a number of consequences in material corrosion

behaviour and several metallurgical phase changes occur in dif-
Fossil boilers experience corrosion problems in both water- ferent zones of a weldment. Because the occurrence of corrosion
side and fireside of the tubing as well as elsewhere in the plants. is due to electrochemical potential gradient developed in the
Failures of industrial boilers have been reported mainly due to adjacent site of a weld metal, it is proposed to study the effects
waterside corrosion and scaling. Corrosion causes difficulty in of welding on the corrosion behaviour. The fossil fuel fired boil-
two respects. The first is the deterioration of the material itself ers [5–7] and power generating equipment experience corrosion
and second is the deposition of the corrosion products in high problems in such components as steam generators, waterwalls
heat release areas of the boiler. Corrosion beneath certain types surrounding the furnace and in front and rear portions of the
of boiler deposits can so weaken the metal that tube failure may superheater and reheater. These components are often made
occur. In steam condensate systems, replacement of lines and of carbon and low alloy steel. The water used for steam rais-
equipment due to corrosion can be costly. ing in any boiler installation often contain gaseous impurities
On load boiler corrosion results from the local formation of and dissolved solids. These can cause scaling and corrosion
a concentrated alkaline or acidic solutions at the metal surface. in the boiler plant. Apart from these, some of the inorganic
Two types of accelerated on load corrosion can be caused by the salts hydrolyze to produce acidity causing corrosion of boiler
accumulation of boiler water impurities. The first type caustic tubes.
gouging is caused by boiler water treatment additives and second This paper describes an experimental work carried out at RT
is caused by local concentration of chloride impurities which are and 100 ◦ C to evaluate and compare corrosion and its inhibition
acidic or become acidic when become heated. in SA213 Gr.T22 (2.25 Cr–1 Mo) steel weldments prepared by
In all arc welding processes [1–4], the intense heat source TIG welding process in ammonium nitrate and sodium hydrox-
produced by the arc and the associated local heating and cool- ide media. Organic compounds such as hexamine and thiourea
are normally used as inhibitors for the prevention of corrosion.
In the present study thiourea is employed as inhibitor. SEM
∗ Corresponding author. Tel.: +91 431 2500133; fax: +91 431 2500133. and XRD analyses have also been made on samples exposed at
E-mail address: drs natarajan@yahoo.com (S. Natarajan). 100 ◦ C.

0921-5093/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.msea.2006.06.051
48 S. Natarajan, S.P. Kumaresh Babu / Materials Science and Engineering A 432 (2006) 47–51

2. Experimental work Table 2

Welding process parameters for TIG welding
2.1. Preparation of weldments Parameters Values

Current 80–90 A
T22 Cr–Mo steel in tube form was welded by TIG process Voltage 18 V
and it was stress relived at 730 ◦ C. The studies were made using Electrode diameter 3.15 mm
small coupons measuring 90 mm × 10 mm, cut from the plate Gas flow rate 20 L/min
comprising parent metal and weld metal regions. Shielding gas Argon

2.2. Corrosive media and Inhibitor Table 3

Corrosion current in mA/cm2 at RT for T22/TIG/NH4 NO3 /NaOH/thiourea
NH4 NO3 and NaOH solutions of 0.01, 0.1, 0.5 M concen-
Concentration PM WM
trations were prepared using analar grade chemicals. The pH of medium
of the solutions were found to be: 5.6, 6.0, 6.7, respectively for NaOH NH4 NO3 NaOH NH4 NO3
NH4 NO3 and 12.0, 13.0, 13.7, respectively for NaOH medium. 0.01 M
Thiourea at 100 ppm was used as the inhibitor throughout the WOI 1.35 9.80 0.68 8.20
study. Each test coupon was surface polished using conventional WI 11.0 1.50 1.10 0.98
methods (degreasing, polishing with emery papers of various PIE (IAC) (84.6) (IAC) (88.0)
grades, etc.). Except the desired area of regions of exposure, 0.1 M
the other regions were masked using Teflon. Studies were made WOI 1.40 9.00 2.20 5.50
both at room temperature and at 100 ◦ C. WI 9.50 3.50 6.60 2.40
PIE (IAC) (61.1) (IAC) (56.7)

2.3. Electrochemical polarization studies 0.5 M

WOI 0.28 2.80 1.10 6.00
WI 9.40 1.80 4.70 2.00
All studies were made using Model-362 scanning poten- PIE (IAC) (35.7) (IAC) (66.6)
tiostat (Princeton Applied Research Corp., USA). The studies
were made at room temperature and at 100 ◦ C with and with- Note—PIE: percentage inhibitor efficiency; the term within the parentheses: PIE
for TIG process; WOI: without inhibitor; IAC: inhibitor accelerates corrosion;
out inhibitor mentioned above. The following technique was WI: with inhibitor.
adopted to find Icorr (corrosion current) values.

2.4. Tafel line extrapolation radiation and λ = 1.5418 Å (Ni filter) with a scanning range
3–65◦ 2θ was used to investigate the weld region of the speci-
Application of potentials between ±50 and 250 mV is taken men.
for Tafel slope. A graphical plot of E versus log i was made in all
the experiment and tangents were drawn which on extrapolation 3. Results and discussion
to Ecorr intersected at a point that represented on the X-axis, the
Icorr value. The chemical composition, mechanical properties and weld-
ing conditions for the SA 213 T22 grade Cr–Mo steel are given
2.5. SEM analysis in Tables 1 and 2.

Studies on the surface morphologies of T22 Cr–Mo steel sam- 3.1. Selection of electrochemical technique
ples for WM region were carried out for the 0.5 M NH4 NO3 and
NaOH media containing thiourea as inhibitor at 100 ppm level Among the many corrosion monitoring techniques available,
exposed at 100 ◦ C. Tafel method is selected due to the reason that it provides greater
electrochemical perturbation to the test electrode. The values of
2.6. XRD studies Icorr in Tafel electrochemical technique for room temperature
and 100 ◦ C are given in Tables 3 and 4. The PM shows higher
A computer controlled wide angle X-ray diffractometer sys- corrosion rate than WM region both at room temperature and
tem JEOL (Japanese make) model; JDX 8030 using Cu K␣ 100 ◦ C.

Table 1
Chemical composition in wt.% and mechanical properties in MPa for SA213 Gr.T22 steel
Element

C Mn Si P S Mo Cr T.S Y.S

Base metal 0.16 0.60 0.10 0.040 maximum 0.040 maximum 1.00 2.25 415.0 207.0
Filler wire 0.105 1.41 0.47 0.017 0.01 0.01 maximum 0.01 maximum 480.0 400.0
 S. Natarajan, S.P. Kumaresh Babu / Materials Science and Engineering A 432 (2006) 47–51 49

Table 4 to high pH levels leading to caustic corrosion because the mag-

Corrosion current in mA/cm2 at 100 ◦ C for T22/TIG/NH4 NO3 /NaOH/thiourea netite film is unstable and soluble at pH values above 12. This
Concentration PM WM happens at both parent metal and weld metal regions because
of medium of micro-level galvanic cells setup due to differences in surface
NaOH NH4 NO3 NaOH NH4 NO3
profiles. At high pH levels, the protective oxide layer becomes
0.01 M soluble which leads to enhanced corrosion rate. In the case of
WOI 0.39 10.2 2.00 1.50
WI 0.94 2.40 2.60 0.98
exposure to lower pH corrosive environments namely acids,
PIE (IAC) (76.4) (IAC) (34.6) Natarajan et al. [9–11] have found enhanced corrosion rate in
all regions.
0.1 M
WOI 1.43 8.60 1.25 5.80
WI 1.60 0.90 2.40 3.85 3.3. Influence of welding process and micro-structure
PIE (IAC) (89.9) (IAC) (33.6)
0.5 M Compared to parent metal, the weld metal region shows lower
WOI 0.92 1.40 0.50 0.72 corrosion rate which indicates that TIG is beneficial to Cr–Mo
WI 1.30 5.05 0.76 3.00 low alloy steel. The low heat input of TIG process results in less
PIE (IAC) (IAC) (IAC) (IAC) dilution of WM and minimum loss of alloying elements even-
Note—PIE: percentage inhibitor efficiency; the term within the parentheses: PIE tually leading to low corrosion rate in the WM region. Also the
for TIG process; WOI: without inhibitor; IAC: inhibitor accelerates corrosion; alloy steel under consideration will not undergo any metallur-
WI: with inhibitor. gical phase transformation at 100 ◦ C and so no micro-structural
Table 5
variations are to be expected between room temperature and
Tafel slope values at RT for T22/TIG/NH4 NO3 /NaOH/thiourea 100 ◦ C. The PM microstructure consists of polygonal ferrite and
pearlite. The WM shows a combination of widmanstatten type
Medium Molar ba , bc (mV/decade)
concentration (M)
ferrite, pearlite and bainite. The presence of Cr and Mo in the
PM WM alloy enhances hardenability and promotes bainite formation
WOI WI WOI WI even on relatively slow cooling. During the bainitic formation,
coarsening of carbide particle occurs and the number of poten-
NaOH 0.01 60, 112 90, 120 70, 110 81, 124
0.1 86, 113 90, 110 82, 121 86, 119
tial galvanic cells per unit area of the exposed surface decreases
0.5 90, 129 86, 122 89, 111 86, 122 and hence the corrosion rate decreases. The microstructures of
PM and WM are shown in Figs. 1 and 2.
NH4 NO3 0.01 90, 110 64, 112 88, 114 70, 119
0.1 86, 120 69, 118 64, 115 65, 120
0.5 85, 116 79, 120 87, 127 69, 122 3.4. Role of inhibitor

It is the weld root region which always carries the corrosive

Tafel slope values namely ba and bc have been furnished in fluid during the operation of any plant. In NH4 NO3 medium,
Tables 5 and 6, respectively for RT and 100 ◦ C test conditions. thiourea shows 88.0% efficiency at room temperature and 34.6%
efficiency at high temperature (100 ◦ C) for WM region whereas
3.2. Effect of alkali concentration for PM region it shows 84.6% efficiency at room temperature

With increase in alkali concentration, increase in Icorr value

is not much prominent. The Icorr value for both PM and WM
increases up to a particular concentration (0.1 M) and then starts
to decrease to lower values. Natarajan et al. [8] have observed the
similar trend in pressure weldments. This results from formation
of protective oxide film over the surface. NaOH can concentrate

Table 6
Tafel slope values at 100 ◦ C for T22/TIG/NH4 NO3 /NaOH/thiourea
Medium Molar ba , bc (mV/decade)
concentration (M)
PM WM

WOI WI WOI WI

NaOH 0.01 87, 120 86, 122 81, 120 78, 119
0.1 77, 111 79, 122 77, 113 83, 122
0.5 81, 120 82, 122 89, 127 86, 125
NH4 NO3 0.01 90, 115 72, 118 79, 121 78, 119
0.1 89, 117 69, 116 83, 115 88, 124
0.5 88, 116 71, 115 79, 116 81, 119 Fig. 1. Microstructure consists of coarse polygonal ferrite and pearlite T22 (2.25
Cr–1 Mo)/TIG/parent metal.
50 S. Natarajan, S.P. Kumaresh Babu / Materials Science and Engineering A 432 (2006) 47–51

Fig. 2. Microstructure consists of widmanstatten ferrite and bainite T22 (2.25

Cr–1 Mo)/TIG/weld metal. Fig. 3. SEM photo: T22/TIG/WM/0.5 M NaOH/100 ppm thiourea at 100 ◦ C.
Features: steam chimneys with small globules.
and 89.9% at high temperature (100 ◦ C) independent of the con-
centrations of the medium used.
In NaOH medium the presence of thiourea has not brought
any beneficial effect at all. Instead it has shown the negative
effect leading to acceleration of corrosion. However, at this junc-
ture, this should not be taken to mean that thiourea becomes
responsible for acceleration but striking observation is that due
to high pH levels of NaOH, the protective layer of the steel
becomes soluble leading to rapid corrosion. In a nut-shell, it can
be inferred that the high pH of the NaOH has outweighed the
expected beneficial role of thiourea. Thus, thiourea accelerates
corrosion in NaOH medium both at room temperature and at
100 ◦ C. A comparison of PIE for different media at room tem-
perature and at 100 ◦ C is shown in Table 7.

3.5. SEM analysis

Fig. 4. SEM photo: T22/TIG/WM/0.5 M NH4 NO3 /100 ppm thiourea at 100 ◦ C.
The WM region exposed at 100 ◦ C in 0.5 M NH4 NO3 and Features: honeycomb type structure with microfissures.
NaOH containing 100 ppm thiourea as inhibitor was subjected to
SEM examination and the morphological features are provided growth of film; (b) existence of high pH levels of the alkali used.
in Figs. 3 and 4. In this connection it is worthwhile to quote The authors of this technical paper have also noticed similar
relevant literature [12–16] available on surface characteristics fissuring of the film in this present investigation as depicted in
of oxide film formation on steels used in fossil fired boilers SEM photographs.
pertaining to case studies. Similar observation has been made
by Natarajan and Kumaresh Babu [17–20] in the investigation 3.6. XRD studies
carried out on laboratory scale employing 1Cr–0.5Mo power
plant low alloy steel. Magnetite film (Fe3 O4 ) formation [21] is The XRD pattern obtained on WM for both media shows
the dominant effect of corrosion. The fissuring of the oxide film peaks which correspond to formation of iron oxide. This can be
is very much prone to occur under two situations: (a) excessive seen from Figs. 5 and 6.

Table 7
Comparison of PIE at RT and 100 ◦ C for T22/TIG/NH4 NO3 /NaOH/thiourea
Concentration of medium (M) PM WM

NaOH NH4 NO3 NaOH NH4 NO3

RT 100 ◦ C RT 100 ◦ C RT 100 ◦ C RT 100 ◦ C

0.01 IAC IAC 84.6 76.4 IAC IAC 88 34.6

0.1 IAC IAC 61.1 89.5 IAC IAC 56.33 33.6
0.5 IAC IAC 35.7 IAC IAC IAC 66.6 IAC
 S. Natarajan, S.P. Kumaresh Babu / Materials Science and Engineering A 432 (2006) 47–51 51

beneficial in NH4 NO3 medium with maximum inhibitor effi-

ciency of around 90%.
• SEM examination reveals even distribution of a globular
adherent corrosion product with few steam chimneys and
micro-fissures.
• The XRD shows few peaks of appreciable sharpness indicat-
ing formation of iron oxide (Fe3 O4 ).

References

[1] R.L.O. Brien, Welding Handbook, vol. 2: Welding Processes, 8th ed.,
Fig. 5. XRD pattern of WM/T22/TIG/0.5 M NaOH/100 ppm thiourea at 100 ◦ C. AWS, Miami, FL, USA, 1995.
[2] T.G. Gooch, Process Industry Corrosion: Theory and Practice, NACE
Publication, 1982, pp. 739–755.
[3] M. Henthorne, Corrosion 30 (1974) 39.
[4] G.E. Linnert, Welding Metallurgy, Carbon and Alloy Steels, vol. 1, 4th
ed., AWS, Miami, FL, USA, 1994.
[5] G.S. Theus, P.L. Daniel, Proceedings of the Eighth International Brown
Boveri Symposium, Baden, Switzerland, 1983, pp. 185–188.
[6] NACE, Corrosion Testing of Chemical Cleaning Solvents, vol. 21: Mate-
rials Performance, 3M, NACE Publication, 1982, p. 48.
[7] A.N. Rothwell, J. Corros. Prev. Contr. 39 (1992) 113–118.
[8] S. Natarajan, F. Ginoprakash, R. Vaideeswaran, J. Steel Relat. Mater. 2
(2004) 159.
[9] S. Natarajan, V. Sivan, Bull. Electrochem. 6 (5) (1990) 501–504.
[10] P.J. Antony, S. Natarajan, R. Vaideeswaran, J. Corros. Prev. Contr. 50
(2003) 181–188.
[11] S. Natarajan, V. Sivan, Int. J. Join. Mater. 15 (2003) 17–23.
Fig. 6. XRD pattern of WM/T22/TIG/0.5 M NH4 NO3 /100 ppm thiourea at [12] W.M.M. Huijbregts, Proceedings of the Fifth ICMC, Tokyo, 1972, p.
100 ◦ C. 857.
[13] P.H. Effertz, Proceedings of the Fifth ICMC, Tokyo, Japan, 1972, p.
4. Conclusions 920.
[14] G.M.W. Mann, Br. Corros. J. 12 (1977) 6–13.
[15] P. Mayer, A.V. Manolesu, Proceedings of the Seventh ICMC, Ontaria,
• The T22 grade parent metal experiences by and large two Canada, 1980, pp. 1480–1487.
to three times higher corrosion rate when compared to weld [16] L. Kayafas, J. Corros. (1990) 443.
metal region at all concentrations in both the media at both [17] S. Natarajan, V. Sivan, J. Corros. Prev. Contr. 50 (2003) 7–19.
temperatures. [18] S.P. Kumaresh Babu, S. Natarajan, J. Steel Relat. Mater., in press.
[19] S. Natarajan, Australas. Weld. J. Second Quart. 50 (2005) 33–39.
• In general, NH4 NO3 medium shows higher corrosion rate
[20] S.P. Kumaresh Babu, S. Natarajan, Int. J. Join. Mater. 18 (1) (2006)
than NaOH at room temperature and at 100 ◦ C. 19–26.
• Thiourea as inhibitor is found unsuitable in NaOH medium [21] Metals Hand Book, vol. 13: Corrosion, 9th ed., ASM International, Ohio,
both at room temperature and at 100 ◦ C whereas it is found USA, 2001, pp. 985–995.