Effects of Cr Cu Ni and Ca on the corrosion behavior of low carbon steelin synthetic tapwater
Yoon-Seok Choi Jae-Joo Shim Jung-Gu Kim*
Depre of Acf Mtrials Engineering Smg Uiesity 300 Chch-Dog JGu S 440-746 Sh o
Reoeived 8 July 2004; acoepted 23 July 2004Available online 11 November 2004
Abstract
electrochemical corrosion tests (potentiodynamic polarization tests electrochemical impedance spectroscopy (EIS) and analytical techniques Aqueous corrosion behaviors of the low-alloy steels with small amounts of Cr Cu Ni and Ca in synthetic tap water were studied byshifed to nle direti and crosi rate tndd t dcese as the result of allying.elts ofEIS mesurmnt hed that themre (XPS EPMA). Potentiodynamic polarization test represented that all cunves showed active corrosion behavior and corrosion potential wasthe content of element the larger the polarization resistance. I1 can be seen that these results were caused by the formation of a coherent protective film which contained alloying elements. Actually XPS results showed that Cr Cu and Ca foemed protetive metal oxide at thesurface of rust layer under stagnant condition. The results of EPMA indicated that Cr Cu and Ca were distributed densely at inner layer of the rust layer while Ni was distributed at the whole layer under flow condition. Consequently alloying elements improved corosion resistanceof low carbon steel. It was reflested that corrosion rates of all specimens were quite low: especially; Cu- Ni and Ca-containing specimens revealed the lowest corrosion rate due to the formation of insoluble rust layer at the initial stage of corrosion.
2004 Elsevier B.V. All rights reserved.
Kerywonds:： Pipeline stel; Alloying element; Corosice characteristics;Electrochemsical measurements
1. Introduction
For the decrease of internal corrosion rate in drinking wa-ter distribution systems several methods have been tried suchare not fundamental soluble problem but temporary. Thus as lining cleaning and so on [4 5]. However these methodsit is necessary to develop a new pipeline steel of low cost and high corrosion resistance. Change of pipe material is oneway which can reduce internal corrosion rate and then cor-rosion products can be decreased at the intermal region of pipes.
systems for over five centuries [1]. Although the internal cor- Iron and steel pipes have been used in water distributionrosion rate of drinking water pipes is low intemal corrosion of drinking water distribution systems leads to two majorproblems for water utilities [2 3]. The first is the failure ofthe distribution system pipes. This failure results in water leakage and loss of hydraulic capacity caused by the buildupof corrosion products. The second problem is an unwanted change in water quality as the ground water is being trans-ported through the distribution system. The response of cor- roded stee to changes in water quality is expected to be astrong function of the physicochemical characteristics of thecorrosion scales and is different from classical electrochem- ical corrosion of steel.
amounts of alloying elements such as Cr Cu Ni Si and It is well known that weathering steel containing smallP has been widely used because of its excellent resistanceto atmospheric corrosion. This is due to the development of an adherent protective layer formed on the steel [69].In our previous work the Cr and Cu pounds promote more or less protective rust layers on weathering steel in anaqueous condition as they do with an atmospheric condi- tion [10 11]. Accordingly it can be expected that alloyingwith small amount of noble elements such as Cr Cu and Ni
Chemical positions of specimens Table 1
Specimem Composition (wt.%)c Si Mn P Cu Ni CCr steel Carbon steel 0.250 0.075 0.246 0.908 0.040 0.015 0.004 0.040 0.972 - - - - -Cr-Cu-NiCa steel CrCuNi steel 0.078 0.079 0.247 0.248 0.894 0.912 0.016 0.016 0.004 0.004 0.992 1.00 0.198 0.196 0.150 0.150 0.007 -
provides a possibility for improving corrosion resistance indistribution system.
2.2. Potentiodynamic polarization test
The potentiodynamic polarization test [12 13] was carriedand calculate total charges passing through specimen during out to evaluate the overall corrosion behavior of specimendeterioration.
The aim of this investigation was to determine the influ-ence of the alloying elements on the internal corrosion rateof water distribution systems. Previous investigations have been oriented towards coupon tests under stagnant condition.However in this investigation parallel studies of coupons and rotating electrodes have been carried out to achieve a betterunderstanding of the rule of the alloying elements under flowcondition.
lary probe connected to a reference electrode (SCE) and two The cell contained the working electrode a glass capil-graphite rod counter electrodes.
The specimen was allowed to attain a stable open-circuitpotential (OCP) before starting the polarization scan. A pe-riod of ~3h was required for OCP to bee stable within ±5mV.
2. Experimental
Potentiodynamic polarization curves were generated us-ing an EG&G Model 273 A potentiostat. After immersion of the working electrode in the synthetic tap water for 3 h thefrom the initial potential of 250mV versus open circuit potential of the electrode was swept at a rate of 0.166 mV/spotential (OCP) to the final potential of 400 mV versus sat-Electrodesystem was used fortheimitationof flow condition urated calomel electrode. An EG&G Model 636 Ring-Diskand the flow rate is 1 m/s.
2.1. Specimen and solstion preparation
onpu unnsea e u Suou q paedad aam suoadstion furmace. After adding the constituent elements the cast slab was reheated at 1200 °C for 2 h and hot rolled into 13 mmthick plates. The chemical positions of the specimens (wt.%) used for experiments are given in Table 1.
Each specimen was mounted in a cured epoxy resinfor stagnant condition while the rotating cylinder electrode (1 cm diameter 1.5 cm height) was used for flow condition.The outer surface of the cylinder was exposed in solution To prevent the initiation of crevice corrosion between the epoxyleaving an exposed area of 1 cm² on the metal surface. Be- ed s aeudsxoa uds pfore each test the working electrode surface was treated asfollows: polished with silicon carbide abrasive paper (from grade 220600) then degreased with acetone rinsed withdouble-distilled water and finally immersed in the solution.
2.3. Electrochemical impedance spectroscopy and galvanostatic test
EIS instrumentation [14 15] consisted of an EG&G PAR-STAT 2263 and electrochemical impedance software. Gal-potentiostat. vanostatic tests were generated using an EG&:G Model 273 A
EIS measurements were conducted at OCP with a 10 mVrange was covered from 10kHz to 10 mHz. The experimen- (rms) perturbation and five points per decade. The frequencyusing a suitable fitting procedure elaborated by ZWinSimp tal results were interpreted on the basis of equivalent circuitprogram.
Table 2 gives the chemical position of the synthetic taptemperature. water. The experimental temperature was maintained at room
The first EIS tests were carried out after 3 h immersion.The galvanostatic tests were performed to simulate the same90 days. The applied curent density was based on corrosion extent of corrosion as specimens immersed in solution forFaraday’s law. Then the second EIS tests were performed current density (/cor) and total charge passed obtained fromafter galvanostatic tests.
Table 2Chemical positions of synthetic tap water
Parameters Synthetic tap water concentrationCI- 65ppm40ppm 30ppm50ppm uddoc7.6
C² s0 ²- HCOMg?
2.4. Surface analyses
To investigate the relationship between the alloying ele-ments and surface position of rust the surface was ex-
Fig. 1. Polarization curves of specimxms in synthetic tap water nder stag- nant condition.
amined by XPS and EPMA after EIS measurements [16].The XPS was carried out using a mercial ESCA system (Model VG Microtech ESCA 2000). The excitation sourceThe absolute binding energies of different surface species was Mg Kα radiation (photoelectron energy = 1253.6 eV).were calibrated using the C 1s line at 285 eV.
with an acceleration voltage of 15 kV an irradiation current EPMA was performed using the JEOL JXA-8600MX of 1 × 107 A and a beam diameter of 1 μm.
3. Results and discussion
3.1. Potentiodynamic polarization Iest
specimens in synthetic tap water for stagnant and flow condi Figs. 1 and 2 show potentiodynamic polarization curves of no passivation). tions. All specimens exhibited active corrosion behavior (i.e..
Fig. 2. Polarization curves of specimens in synthetic tap water under fowconditioe (1 m/s).
smaller corrosion current density (corr) and anodic current In the case of stagnant condition new alloy steels haddensity than carbon steel. Furthermore under the flow con-dition corrosion potential was shifted to noble direction and anodic current density was greatly decreased with addingments have a suppressing effect on the anodic reaction. alloying elements. This behavior confirms that alloying ele-
Corrosion rates were determined by Tafel extrapolationmethod. The corrosion current density can be measured and can yield a corrosion rate based on Faraday's law [17]:
where 0.13 is the metric and time conversion factor and EWthe equivalent weight in grams. Fig. 3 shows the variation of corrosion rates obtained by the Tafel extrapolation method.
Fig. 3. Corosin rales of specimens im synthetic tap water obtained bypolarization test: (a) stagnant condition; (b) flow condition (I m/s).
Fig. 4. Nyquist plots for specimems before galvanostatic test: (a) stagnantcondition; (b) flow condition (I m/s).
under both stagnant and flow conditions. Among new alloy New alloy steels had lower corrosion rates than carbon steelsteels CrCuNiCa steel had the lowest corrosion rate thanothers indicating the effectiveness of this steel for use in potable water distribution system.
3.2. Electrochemical impedance spectroscopy and galvanostatic test
For a better understanding of the effect of Cr Cu Ni andCa on corrosion behavior EIS and galvanostatic tests were conditions. performed in synthetic tap water under stagnant and flow
Table 3
Specimen Stagnant condion (A/cm²) Flow condition (A/cm²)Carbon steel 0.0078 0.0260Cr steel Cr-CuNi steel 0.0052 0.0044 0.0130 0.0086CrCuNiCa steel 0.0034 0.0073
Fig. 5. Nyquist plots for specimens afer galvanostatic test: (a) stagnantcondition; (b) flow condition (1 m/s).
Fig. 4 shows Nyquist plots of the specimens before gal-vanostatic test. The impedance spectra measured from thesteels indicated a single semicircle which means only one reaction existed between specimen and electrolyte. The di-ameter of the arc can be regarded as a polarization resistance
Fig. 6. Physical models and equivalent circuit for fiting the EIS data.
nen in synthetic tap water: (a) before galvanostatic test; (b) after galvanostatic test.
Fig. 8. Sectra f allyimg elements frm te st of specimms obimed by XPS mder stagat cditio: (a) Cr2p spetr: (b) C 2p2 setra; (c) Ca2pi spectra.
