Magnetite-Accelerated Corrosion of SA508 Tubesheet Material and Its Effect on Steam Generator Tube Denting

Abstract

The objective of this work is to investigate the magnetite-accelerated corrosion phenomenon of SA508 used as tubesheet material in simulated secondary side environments of pressurized water reactors through immersion and electrochemical tests. The presence of sulfate ions induced the fast growth of a corrosion product layer on SA508, and this phenomenon was accelerated when the SA508 was coupled to magnetite. From the perspective of electrochemical behavior, it was found that SA508 behaves as an anodic member in the coupling system with magnetite, resulting in an increased corrosion rate.

1. Introduction

Heat-transfer tubes made of Alloy 600 material are susceptible to a variety of corrosion-induced problems on the secondary side of steam generators (SGs) in pressurized water reactors (PWRs), with some of them being stress corrosion cracking (SCC), intergranular corrosion, pitting, and wastage [1,2,3,4]. To overcome these corrosion problems, Alloy 690 with a higher Cr content (30 wt (%)) was developed as a replacement for Alloy 600. There have been no complications arising from the corrosion of Alloy 690 in service since its first application in SGs in 1987 [5].

Alloy 690 has a high corrosion resistance in pressurized water reactor (PWR) secondary environments, but there still remains a potential problem that can damage the SG tubes. It is noteworthy that there are some dented tubes even in SGs with Alloy 690 [6,7]. While not directly linked to the corrosion of the SG tube itself, denting is another type of damage that affects the integrity of SG tubes. The basic mechanism is that SG tubes are distorted from the outer to the inner surface by the volume expansion of the corrosion products caused by continuous corrosion of the top of the tubesheet (TTS) or tube support plate (TSP) materials in the crevices around the tubes [2]. Apart from the corrosion degradation of the tube itself, denting may lead to SCC in the tube. This degradation mode is affected by the concentration of impurities such as chloride and sulfate ions, as well as oxidants in the crevices, leading to the fast corrosion of the tubesheet or TSP [6,7,8,9].

However, it should also be noted that the crevices between SG tubes and TTS or TSP are covered with deposits. The deposits originate from corrosion products released from the surface of the carbon steel used for the feed water piping material by flow-accelerated corrosion under the alkalized reducing condition in the PWR secondary system. Therefore, the deposit consists primarily of magnetite [10]. This magnetite deposit has numerous open pores in nature [11,12]. Therefore, most surfaces of the tubesheet or TSP materials within the heated crevice are covered with magnetite, except for the open pores, and thus only the local surface surrounded by the open pores is exposed to corrosive crevice water [13]. This means that the corrosion of the tubesheet or TSP occurs under the condition that is in electrical contact with the magnetite deposit. Magnetite behaves as a conductive material because of its high electrical conductivity [14,15]. Therefore, a galvanic cell between magnetite and the tubesheet or TSP materials will be formed. This may also accelerate the corrosion of the tubesheet or TSP. Therefore, the goal of this work is to investigate the corrosion and electrochemical properties of SA508 coupled to magnetite under PWR secondary system environments. For this purpose, a specially designed magnetite-deposited corrosion coupon is prepared, and the effect of magnetite on the electrochemical behavior is investigated.

2. Experimental

2.1. Test Materials and Solutions

SA508 specimens were machined into two sizes from low-alloy steel used as tubesheet material in the SG of PWRs: A size of 20 mm × 10 mm × 1 mm for immersion corrosion tests and a size of 10 mm × 5 mm × 1 mm for potentiodynamic polarization tests. The machined specimens were ground with silicon carbide papers of up to #600, #800, and #1000 grit, and were ultrasonically cleaned in acetone. The chemical composition of SA508 is given in

Table 1. Chemical composition of SA508 (wt. %).

C	Si	Mn	P	S	Ni	Cr	Mo	Fe
0.199	0.051	1.52	0.005	0.006	0.987	0.232	0.582	Bal.

.

The corrosion tests were conducted in two different solutions. The reference solution was deionized water (Milli-Q, Merck KGaA, Darmstadt, Germany) adjusted to a pH of 9.5 at room temperature using ethanolamine, which is used as a pH control agent of secondary water in PWRs. When necessary, 0.05 M Na₂SO₄ was added to the reference test solution to simulate the crevice environment between the SG tube and TTS.

2.2. Magnetite-Deposited Corrosion Coupon

To investigate the corrosion behavior caused by the coupling effect of SA508 and magnetite, we needed to prepare the corrosion coupon to be electrically connected to magnetite. To achieve this, a thick magnetite film was grown on the entire surface of the prepared corrosion coupon using the electrodeposition (ED) method, except for a circular area with a diameter of about 5 mm in the center (the left side of Figure 1). The ED of the magnetite film was carried out in a solution containing 2.0 M NaOH, 0.043 M Fe₂(SO₄)₃, and 0.1 M triethanolamine by using a three-electrode system under the following conditions: ED potential = −1.05 V_SCE; ED time = 60 min; and ED temperature = 80 °C. The details of the ED process are given in our previous reports [16,17,18,19,20].

The right side of Figure 1 shows the final product of the magnetite-deposited corrosion coupon after the electrodeposition process. In this corrosion coupon, the undeposited area at the center of the specimen simulates the local tubesheet surface within the crevice covered with magnetite. The characteristics of the magnetite layer electrodeposited on the SA508 specimen were investigated using a scanning electron microscope (SEM) (QUANTA 3D FEG FIB-SEM, FEI, Hillsboro, CA, USA), an energy dispersive X-ray spectroscopy (EDS), and an X-ray diffractometer (D/Max-2500, Rigaku, Tokyo, Japan).

2.3. Immersion Corrosion Tests

The immersion corrosion tests for each condition were separately conducted in two static titanium autoclaves with a capacity of 1 L, as shown in Figure 2. Two magnetite-free and magnetite-deposited specimens were located on a specimen holder made of polytetrafluoroethylene (PTFE) in each autoclave. A test solution of 700 mL was injected into the autoclave and deaerated with high-purity nitrogen gas (99.999%) at a flow rate of 100 mL/min in the closed autoclave at room temperature for 3 h. The immersion tests were conducted at 280 °C for 100 h. After the immersion tests were completed, the surfaces and cross sections of the corroded specimens were investigated using SEM.

2.4. Potentiodynamic Polarizaion Tests

The SA508 specimen for the potentiodynamic polarization test was machined into a rectangular shape (10 mm × 5 mm × 1 mm) and was spot-welded to an Fe wire that had been covered with a heat-shrinkable polytetrafluoroethylene (PTFE) tube. This specimen was used as an SA508 electrode. A magnetite electrode was prepared by the ED of magnetite film on the entire surface of the prepared SA508 electrode under the same conditions used in the preparation of the magnetite-deposited corrosion coupon.

The potentiodynamic polarization test was performed in the same solution used for the immersion corrosion test at 80 °C using a three-electrode system. A saturated calomel electrode (SCE) and platinum wire were used as the reference and the counterelectrode, respectively. The test solution was injected into the corrosion cell and was continually deaerated with high-purity nitrogen gas (99.999%) at a flow rate of 100 mL/min in the closed corrosion cell until the test was completed. After the open-circuit potential (OCP) was stabilized at 80 °C, polarization scans were performed in the positive or negative direction from the OCP at a scan rate of 30 mV/min. In addition, the reproducibility of the polarization curve was confirmed by repeating the tests at least three times using newly prepared electrodes and test solutions.

3. Results and Discussion

3.1. Electrodeposited Magnetite Layer on SA508 Substrate

Figure 3 shows the X-ray diffraction (XRD) patterns, the surface SEM image, and the EDS point analysis of the magnetite layer electrodeposited on the SA508 substrate. As shown in Figure 3a, the XRD patterns of the electrodeposited layer corresponded to pure crystalline magnetite (JCPDS card no. 01-080-6406). Polyhedral-shaped particles were homogeneously formed on the SA508 substrate after the electrodeposition process, as shown in Figure 3b. This layer was mainly composed of O and Fe, and the Fe/O atomic ratio was about 0.77, as shown in Figure 3c. Consequently, the deposited layer was confirmed to be magnetite from these results.

Figure 4 shows the SEM image and EDS line profile for a cross-section of the electrodeposited magnetite layer. As shown in Figure 4a, defects such as cracks or pores were not observed in the electrodeposited magnetite layer with a thickness of about 7 μm, indicating that the layer was tightly connected to the SA508 substrate. In addition, the uniformity of the chemical composition of the layer was confirmed from the EDS line profile, as shown in Figure 4b.

3.2. Immersion Corrosion Behavior

The corrosion product layer formed on the SA508 exposed to the reference solution at 280 °C for 100 h is presented in Figure 5. In the reference solution, polyhedral-shaped corrosion products were formed on the magnetite-free specimen, and the thickness of this corrosion product layer was about 0.2 μm. However, corrosion products were more densely formed on the magnetite-deposited specimen compared to the magnetite-free specimen. The thickness of the corrosion product layer was twice that formed on the magnetite-free specimen. This result indicated that the corrosion rate of SA 508 increased when the alloy was coupled to magnetite.

The addition of 0.05 M sulfate ions produced a significant change in the formation of the corrosion product on both the magnetite-free and magnetite-deposited corrosion coupons, as shown in Figure 6. Fine oxide particles were densely formed on the magnetite-free corrosion coupon. The thickness of the corrosion product layer was 1.7 times that formed on the magnetite-free specimen exposed to the reference solution. These results indicated that the addition of 0.05 M sulfate ions accelerated the corrosion of SA 508.

For the magnetite-deposited specimen exposed to the reference solution with 0.05 M sulfate ions, the corrosion product layer formed on the SA508 was very thick and porous. Compared to the magnetite-free specimen exposed to the reference solution, the thickness of the corrosion product layer increased by about 6.7 times. This result demonstrated that the combined effects of sulfate ions and coupling with magnetite significantly accelerated the corrosion of SA 508, leading to the fast growth of the corrosion product layer.

3.3. Potentiodynamic Polarization Curves

Figure 7 shows the polarization curves of SA 508 and magnetite in the reference solution with and without 0.05 M sulfate ions. In the reference solution, the corrosion potential of magnetite was approximately 250 mV higher than that of SA508. This implies that SA508 behaves as an anodic member of the galvanic couple with magnetite if these two materials are in electrical contact. In this couple with equal areas of materials, the electrochemical potential of SA508 is expected to shift in the positive direction. In addition, the anodic current of SA508 increases by galvanic coupling with magnetite. To evaluate the effect of the increased magnetite surface area in the couple, the cathodic curve of magnetite with an area of 20 cm² is also presented in Figure 7, and was simply estimated from that with an area of 1 cm². The corrosion potential of the coupled SA508 is polarized more in the positive direction as the magnetite surface area increases, which resulted in a significantly increased anodic current.

In the reference solution with 0.05 M sulfate ions, the polarization curves of SA508 and magnetite shifted in the direction of higher current density, as shown in Figure 7b. SA508 still behaved as the anode in the coupling system with magnetite. Therefore, if SA508 is in electrical contact with magnetite, then the galvanic corrosion rate of SA508 is expected to increase more than that in the reference solution. Consequently, it is considered that these changes in the electrochemical behavior also occur on the exposed SA508 surface in the magnetite-deposited specimen shown in Figure 5 and Figure 6 during the immersion corrosion tests.

In order to quantitatively measure the effects of sulfate ions and magnetite on the corrosion rate of SA508, the corrosion current density of SA508 for each test condition was calculated via the Tafel-extrapolation method and the mixed potential theory from the polarization curves in Figure 7, as shown in Figure 8a. The electrochemical kinetic parameters are summarized in Figure 8b in relation to the growth rate of the corrosion product layer estimated from Figure 5 and Figure 6. The corrosion rate of SA508 increased by about 1.4 times with the addition of 0.05 M sulfate ions. This markedly increased by about 22.1 times when the effects of sulfate ions and the magnetite (magnetite/SA508 area ratio = 20) were combined. Consequently, the corrosion rate of SA508 can be drastically accelerated in a region where these two factors are combined. This tendency is in good agreement with the results of the growth rate of the corrosion product layer, as shown in Figure 8b. Therefore, it is determined that these enhanced electrochemical kinetics by the combined effects of sulfate ions and magnetite caused the fast growth rate of the corrosion product layer shown in Figure 6.

This work confirms that galvanic coupling with magnetite accelerated the corrosion of SA 508, resulting in the fast growth of the corrosion product layer. This magnetite-accelerated corrosion phenomenon of SA508 can be discussed as a factor that affects SG tube denting in PWR secondary-side environments. As the operation time passes, magnetite released from carbon steel piping transported to the secondary side of SGs and deposited on the TTS and heated crevices, forming a galvanic cell. Furthermore, aggressive chemical impurities were concentrated in the heated crevices covered with magnetite deposits by local boiling.

The electrochemical corrosion behavior of SA508 used for tubesheet material in this crevice is schematically presented, as shown in Figure 9, based on the results of the potentiodynamic polarization tests in Figure 7. The electrochemical behavior of SA508 and magnetite demonstrates that the corrosion current density of SA 508 increases more by not only the impurity effect but also a galvanic corrosion mechanism with magnetite. Consequently, the change in the electrochemical corrosion behavior of SA508 by coupling with magnetite can induce a fast volume expansion of the corrosion products of the SA508 tubesheet adjacent to the tube, resulting in accelerated tube denting. Therefore, this magnetite-accelerated corrosion mechanism of SA508 should be considered as a factor that affects SG tube denting.

4. Conclusions

The corrosion behavior of SA508 coupled with magnetite was investigated in simulated PWR secondary side environments. The main conclusions are drawn as follows:

• Magnetite in contact with SA508 induced the fast corrosion of SA508. This accelerated corrosion of SA508 by magnetite was more severe in a test solution with sulfate ions.
• Regarding the electrochemical behavior, SA508 and magnetite acted as anodic and cathodic members, respectively, in the coupling system.
• In this coupling system, the corrosion current density of SA508 was increased by a galvanic corrosion mechanism with magnetite.
• Based on the experimental results, galvanic corrosion between magnetite and SA508 is proposed as a new acceleration contributor to the denting of SG tubes in the PWR secondary environment.