Crevice corrosion of copper for radioactive waste packaging material in simulated groundwater

Abstract

The investigation described here was conducted to clarify the corrosion behaviour of high level radioactive waste containers made of copper. The influences of oxygen, chloride ion and sulphate ion on copper crevice corrosion were studied in solutions simulating groundwater characteristic of northwest China. The results showed that oxygen, chloride ion and sulphate ion promote crevice corrosion. Chloride ion was found to play a significant role in the crevice corrosion mechanism in copper, but sulphate ion had no effect on the mechanism.

Keywords

High level radioactive waste Copper Weight loss Crevice corrosion

Introduction

As a source of clean and efficient energy, nuclear power is widely used in many countries. However, the utilisation of nuclear energy produces large amounts of high level radioactive waste (HLW). For example, up to the year 2010, the total burden of HLW produced by nuclear power stations in China was ∼1000 tons. Therefore, the question of how to deal safely with nuclear waste is both critical and urgent. After many years of investigation and discussion, the scientific consensus appears to be that the most reasonable approach is to incorporate nuclear waste into glass, which is then sealed into a metal container that is subsequently buried in 400-500 m deep rock strata.^1,2 The waste container is thus the first barrier preventing release of HLW into the environment, and therefore, its structural integrity must be maintained. At present, different packaging materials are selected for geological disposal in different countries, depending on the local geological conditions. Copper containers have been adopted as either the reference material for containers or an optional container material in Finland, Canada, Switzerland, Sweden and Japan, and studies have shown that copper can serve as an excellent HLW container material, with an estimated lifetime exceeding thousands of years.^3,4

Copper is thermodynamically stable, with no tendency to corrode in water under reducing conditions⁵ or in non-oxidising acids free of dissolved oxygen. However, in water containing complexing species such as Cl⁻ and HS⁻, copper may no longer be immune to corrosion.^6,7 In fact, the environment will change constantly during long term geological disposal, particularly in relation to temperature, oxygen content and the ion concentration of the groundwater. Oxygen, which is present in the backfill material during the early stage of disposal, will be gradually consumed. As copper is among the candidates under consideration as a container material for use in China, it is necessary to study its corrosion behaviour under different conditions in a particular geological disposal environment in China.

Localised corrosion, especially crevice corrosion and pitting corrosion, is the main form of damage that may shorten the lifetime of containers in an environment containing Cl⁻ ions. Crevices will occur when HLW containers are in contact with supporting structures or with undersurface deposits, corrosion products, etc.^8,9 King et al. ¹⁰ believe that general pitting of copper was absent, but some crevice corrosion was noted in deaerated saline groundwater solution at 150°C. However, Kursten et al. ⁶ did not believe that crevice corrosion was a threat to the integrity of copper canisters. They had two reasons for this: first, hydrolysis of Cu⁺, especially when complexed by Cl⁻, and local acidification in crevices were unlikely to occur; second, the formation of Cu²⁺requires the presence of O₂,which is unlikely to occur in occluded regions, such as crevices, where O₂ access is restricted.

In the past few decades, China has carried out site selection for an HLW repository, and several candidate sites are undergoing investigation.^11,12 This paper describes a study aimed at determining whether or not crevice corrosion of copper occurs in geological disposal environments in China. The effects of oxygen, Cl⁻ and on susceptibility to crevice corrosion were studied.

Experimental

The material used in this study was pure Cu (>99.9%). It was machined into 50 × 25 × 3 mm samples. The surface of each sample was ground with SiC paper progressively up to 150 grit. After drying in hot air, the samples were weighed and then stored in a desiccator. The crevice corrosion experiments were conducted according to the procedure described in Ref. 12, using the specimen design shown in Fig. 1. To ensure reproducibility, at least three groups of identical samples were studied for each test condition.

Specimen for crevice corrosion immersion test

The solution simulated the groundwater composition in the northwest part of China and was made using analytical reagents and deionised water. Its chemical composition (mg L^{− 1}) was 1027.0 Na⁺, 16.1 K⁺, 1.9 F⁻, 138.0 , 30.2 , 206.0 Ca²⁺, 51.2 Mg²⁺, 1155.0 Cl⁻, 0.057 Br⁻ and 1074.0 . The pH value of the solution was adjusted to 7.2 using H₂SO₄. To study the effect of oxygen, the solution was divided into portions with high O₂ concentration (open air conditions) and low O₂ concentration (the oxygen was removed by purging with nitrogen gas for 1 h, and the solution was then sealed in a conical flask by glue). The high O₂ solution is henceforth described as aerated and the low O₂ solution as deaerated. The temperature was controlled at 90°C using a thermostat water bath with immersion durations of 1 and 6 months. Moreover, immersion tests were also carried out in Cl⁻ free or free simulated groundwater (with the concentrations of the other ions remaining constant) for 1 month at 90°C to investigate the influences of Cl⁻ and on the crevice corrosion process.

After the immersion tests, the corrosion products were examined using Rigaku D/MAX-RB X-ray diffraction. The surface morphology of the corrosion product film was examined using a Zeiss Evo 18 scanning electron microscope (SEM). The crevice corrosion depth was analysed using a Dektak 150 surface profiler. Corrosion products were removed according to ASTM G1-03 (2011).¹³ The weight loss of the specimen was measured to calculate the overall corrosion rate (mm/year) by applying the following equation 1 where Δm is the weight change after the corrosion test (g), S is the surface area of the specimen (cm²), ρ is the density of the test copper (g cm^{− 3}) and t is the immersion time (h). Three specimens were used to calculate the average corrosion rate, and all results were expressed as mean ± standard deviation.

Results and discussion

Effect of oxygen

Figure 2 shows the corrosion rates, as measured by weight loss, of copper exposed to the simulated groundwater at 90°C. With increasing immersion time, the corrosion rate decreased, which indicated that the corrosion product film on the copper had a protective effect on the matrix. After 1 month of immersion, the corrosion rates of copper exposed to the aerated groundwater were higher than those of copper exposed to the deaerated groundwater by a factor of ∼2.5. However, the difference in corrosion rates clearly decreased after 6 months. These results show that oxygen promoted the corrosion of copper, which is similar to the results of Iva et al. ¹⁴

Corrosion rate of copper in aerated and deaerated simulated groundwater at 90°C

The substrate morphologies of copper following removal of corrosion products are shown in Fig. 3. Crevice corrosion occurred in both the aerated and deaerated simulated groundwater at 90°C, as indicated by the arrows in Fig. 3. Figure 4 shows that the crevice depth of copper after immersion in the aerated groundwater for 6 months was more than twice as high as that after immersion for 1 month, whereas the crevice depth changed only a little with time after 1 month in the deaerated groundwater. This result indicates that the presence of oxygen clearly accelerated the crevice corrosion process.

a aerated, 1 month; b aerated, 6 months; c deaerated, 1 month; d deaerated, 6 monthsSubstratemorphologies of copper after immersion in aerated and deaerated simulated groundwater at 90°C for 1 and 6 months

Crevice depths of copper in aerated and deaerated simulated groundwater at 90°C

For most metals, crevice corrosion occurs inside the crevice. However, a reverse crevice corrosion is often observed on copper or copper alloys, with slight corrosion inside the crevice but severe corrosion at its edge. For example, Fig. 5 shows the SEM morphology and the surface profile curve of the copper around the crevice edge (followingremoval of corrosion products)after immersion in the aerated simulated groundwater at 90°C for 6 months. It can be clearly seen that severe corrosion occurred only at the edge of the crevice, as indicated in Fig. 5a by the arrows.

a scanning electron microscopy morphology and b profile curve of copper around crevice edge following removal of corrosion products after immersion in aerated simulated groundwater at 90°C for 6 months

The surface morphologies of copper inside the crevice after immersion in the simulated groundwater at 90°C for 1 month and for 6 months are shown in Fig. 6. The corrosion products were granular and crystalline. Energy dispersive spectroscopic analysis showed that the corrosion products contained the elements Cu and O. The longer the corrosion time, the greater the grain size of the granular corrosion product. The granular corrosion products formed in the aerated solution were larger than those formed in the deaerated solution.

a aerated, 1 month; b aerated, 6 months; c deaerated, 1 month; d deaerated, 6 monthsSurface morphologies of copper inside crevice after immersion in simulated groundwater at 90°C for 1 and 6 months

From the experimental results on crevice depth and granular corrosion products shown in Figs. 4 and 6 respectively, it can be shown that the crevice depth was correlated with the amount and the size of corrosion products inside the crevice. The crevice corrosion of copper is believed¹⁵ to be caused by the formation of an ionic concentration cell between the inside and the outside of the crevice. The metal ions formed by the initial corrosion reactions accumulate within the crevice, where solution flow or diffusion is restricted. Once a difference in metal ion concentration is established, accelerated dissolution (that is, Cu → Cu⁺ + e⁻, Cu⁺ → Cu²⁺ + e⁻) is anticipated at anodes at the crevice edge, whereas reduction of metal ions (that is, Cu²⁺ + e⁻ → Cu⁺ or Cu²⁺ + e⁻ → Cu) occurs at the cathode within the crevice. The corrosion of copper in the aerated simulated groundwater was faster than that in the deaerated groundwater, which reveals the ease with which the copper ion concentration cell became established when the copper was exposed to the aerated groundwater. Therefore, the crevice corrosion depth of copper immersed in the aerated groundwater was higher than that in the deaerated groundwater, and the granular corrosion products formed in the aerated groundwater were larger than those formed in the deaerated groundwater. Similarly, the longer the immersion time, and the more Cu²⁺ that accumulated inside the crevice, the greater the copper ion concentration gradient between the inside of the crevice and the crevice edge, which increased the crevice corrosion depth and accelerated the growth of granular corrosion products.

Effects of Cl⁻ and SO₄ ²⁻

Figure 7 shows the surface morphologies of copper after 1 month of immersion in different aerated groundwater simulants at 90°C. It can be seen that the corrosion scale inside the crevice was red brown when copper was exposed to the complete groundwater simulant and to the -free solution, whereas it was black when exposed to the Cl⁻ -free solution. The corrosion scales in the box marked in Fig. 7 were analysed by X-ray diffraction, and the results are shown in Fig. 8. The compositions of the corrosion scales formed in the complete groundwater simulant and in the free solution were different from those formed in the Cl⁻ free solution. Cu₂O was detected in the corrosion scale with a red brown colour. A very small amount of CuO was detected in the black corrosion scale. These results indicate that the red brown corrosion scale was dominated by Cu₂O, whereas the black corrosion scale contained CuO. Therefore, it can be concluded that the presence of did not change the composition of the corrosion scale, but the presence of Cl⁻ did.

a groundwater; b Cl⁻ free; c free

X-ray diffraction analysis of corrosion products on copper after immersion in different aerated groundwater simulants at 90°C for 1 month

The corrosion rates in different aerated groundwater simulant solutions at 90°C are shown in Fig. 9. Compared with the rate in the complete groundwater simulant, those in the -free and the Cl⁻ -free solutions were lower. Both Cl⁻ and are aggressive ions, with Cl⁻ being more aggressive than , which explains why the corrosion rate in the Cl⁻ -free condition was the lowest.

Corrosion rate of copper after immersion in different aerated groundwater simulants at 90°C for 1 month

Figure 10 shows the substrate morphologies after corrosion product removal; the arrows represent the edge of the crevice. The specimen suffered more serious damage in the -free solution than that in the Cl⁻ -free solution. The crevice depths as measured by the surface profiler are shown in Fig. 11. Crevice corrosion did not occur in the Cl⁻ -free solution, which is a similar result to that reported by Fujii et al.,¹⁶ who concluded that pitting did not occur in copper in water free of residual Cl⁻ even under air saturated conditions. The crevice corrosion of copper in the complete groundwater simulant was slightly more severe than that in the -free solution. The average crevice depth of the former was 0.067 mm, compared with 0.048 mm in the latter.

a Cl⁻ -free; b -free

Crevice depth of copper after immersion in different aerated groundwater simulants at 90°C for 1 month

The surface morphologies of copper inside the crevice after 1 month of immersion in the different aerated groundwater simulants at 90°C are shown in Fig. 12. It can be seen that granular corrosion products did not form in the Cl⁻ -free solution. The number and size of granular corrosion products in the free solution were similar to those in the complete groundwater simulant (as shown in Fig. 6a).

a Cl⁻ free; b free

Initially, uniform corrosion of copper occurs inside and outside the crevice when copper is exposed to the solution, during which the following anodic and cathodic reactions occur:

Anodic 2 3 Cathodic 4 If the solution contains Cl⁻, the main initial corrosion product of copper is cuprous chloride¹⁷ 5 6 Then, hydrolysis reactions of Cu⁺, Cu²⁺ and CuCl occur^18,19 7 8 9 10 When the oxygen in the shielded crevice area has been consumed after some incubation period, the oxygen reduction reaction will stop, but the dissolution of copper will continue, leading to a supersaturated positive charge in the solution, and thus migration of mobile negative ions (e.g. Cl⁻) into the crevice area is required to maintain charge balance. The concentration of Cl⁻ is increased inside the crevice, which promotes the corrosion of copper inside the crevice, and therefore, the concentration of CuCl greatly increases inside the crevice. CuCl is hydrolysed to Cu₂O and H⁺, leading to a decrease in pH inside the crevice. Cu₂O is soluble in acid solution and thus liable to a disproportionation reaction (that is, Cu₂O+2H⁺ → Cu+Cu²⁺ + H₂O),¹⁸ causing the accumulation of Cu²⁺ inside the crevice, in contrast to the outside, and eventually, a metal ion concentration cell is formed between the inside and outside, leading to crevice corrosion.

If the solution does not contain Cl⁻, then sulphate and carbonate ions may also migrate into the crevice area, and a competitive adsorption relationship will become established between them. Carbonate ions have an inhibitoryeffect on the corrosion of copper, since they have been found to counteract the aggressive effect of the sulphate ions,²⁰ with the result that there is no rise in copper ion concentration inside the crevice, no copper ion concentration cell forms and no crevice corrosion phenomenon is observed, as can be seen in Fig. 12a. At this point, Cu²⁺ hydrolysis occurs according to reactions (7) and (8), and CuO then forms as follows 11 The aggressivity of is less than that of Cl⁻, and the dissolution behaviour of Cu in Cl⁻/ mixtures follows the same mechanism as in Cl⁻ solutions.²¹ The presence or absence of did not affect the composition of the corrosion products, as shown in Fig. 10. However, in the solution containing and Cl⁻ ions, the corrosion behaviour appears to have been synergistic,²² with copper in the complete groundwater simulant suffering more damage than that in the free solution.

The above results show that copper will suffer crevice corrosion, especially in aerated conditions, but some authors^3,10,23–26 nevertheless think that copper is a promising container material for use in saline environments because Cl⁻ promotes the active dissolution of Cu, with little tendency to deep localised corrosion. Crevice corrosion may not happen or may be very slight in absolutely oxygen free conditions. However, it is not yet clear whether, once the copper has suffered crevice corrosion during the aerated stage, this corrosion will develop further under the conditions of deep geological disposal. Therefore, further investigation is required to quantify crevice corrosion kinetics.

Conclusions

The crevice corrosion behaviour of copper as a candidate container material for the geological disposal of HLW has been studied to determine the effects of oxygen, Cl⁻ and in the groundwater. The following results have been obtained. 1.

Copper will suffer crevice corrosion after immersion in groundwater in aerated conditions.

The presence of oxygen could promote crevice corrosion of copper.

The corrosion products formed inside the crevice in a Cl⁻ free solution are composed of CuO, whereas they are predominantly Cu₂O in a complete groundwater simulant and in an free solution.

The presence of Cl⁻ plays a dominant role in the crevice corrosion of copper, but has no effect on the mechanism of crevice corrosion.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (grant no. 51271024).

References

Lee

C. T.

Qin

Odziemkowski

and Shoesmith

D. W.

: ‘The influence of groundwater anions on the impedance behavior of carbon steel corroding under anoxic conditions’ Electrochim. Acta 2006 51 1558–1568. doi: 10.1016/j.electacta.2005.02.115

Féron

Crusset

and Gras

J. -M.

: ‘Corrosion issues in nuclear waste disposal’ J. Nucl. Mater., 2008 379 16–23. doi: 10.1016/j.jnucmat.2008.06.023

King

: ‘Container materials for the storage and disposal of nuclear waste’ Corrosion 2013 69 986–1011. doi: 10.5006/0894

Bennett

D. G.

and Gens

: ‘Overview of European concepts for high-level waste and spent fuel disposal with special reference waste container corrosion’ J. Nucl. Mater., 2008 379 1–8. doi: 10.1016/j.jnucmat.2008.06.001

Szakálos

Hultquist

and Wikmark

: ‘Corrosion of copper by water’ Electrochem. Solid State. Lett., 2007 10, (4), C63–C67. doi: 10.1149/1.2772085

Kursten

Smailos

Azkarate

Werme

Smart

N. R.

and Santarini

: ‘COBECOMA (State-of-the-art document on the COrrosion BEhaviour of COntainer MAterials)’ contract no. FIKW-CT-20014-20138 final report, 2001 to 2003, 5th Euratom Framework Programme 1998-2002, Luxemburg; 2004.

Boman

Berger

Andersson

Hahlin

Björefors

Gustafsson

and Ottosson

: ‘Corrosion of copper in water free from molecular oxygen’ Corros. Eng. Sci. Technol., 2014 49 431–434. doi: 10.1179/1743278214Y.0000000199

Hornus

E. C.

and Rodríguez

M. A.

: ‘Effect of temperature on the crevice corrosion resistance of Ni–Cr–Mo alloys as engineered barriers of nuclear repositories’ Prog. Mater. Sci., 2012 1 251–258.

Dunn

D. S.

and Csontos

A. A.

: ‘Corrosion of similar and dissimilar metal crevices in the engineered barrier system of a potential nuclear waste repository’ Electrochim. Acta 2007 52 7556–7569. doi: 10.1016/j.electacta.2006.12.077

10.

King

P. J.

Lamand

K. W.

and Dautovich

D. P.

: ‘The corrosion behaviour of copper under simulated nuclear waste repository conditions’ Can. Metall. Q., 1983 22 125–132. doi: 10.1179/cmq.1983.22.1.125

11.

Q. F.

Pang

X. L.

Liu

Q. L.

and Gao

K. W.

: ‘Crevice corrosion of low alloy steel and carbon steel in the simulated groundwater at 90°C’ Acta Metall. Sinica 2014 50 659–666.

12.

Q. F.

Wang

Pang

X. L.

Liu

Q. L.

and Gao

K. W.

: ‘Investigation of corrosion behaviours of high level waste container materials in simulated groundwater in China’ Corros. Eng. Sci. Technol., 2014 49 480–484. doi: 10.1179/1743278214Y.0000000191

13.

‘Standard practice for preparing, cleaning, and evaluating corrosion test specimens’. ASTM G1-03, ASTM International, West Conshohocken, PA, USA, 2011.

14.

Betova

Beverkog

Bojinov

Kinnunen

Mäkelä

Pettersson

S. O.

and Saario

: ‘Corrosion of copper in simulated nuclear waste repository conditions’ Electrochem. Solid State. Lett., 2003 6, (4), B19–B22. doi: 10.1149/1.1557033

15.

Farmer

J. C.

Van Konynenburg

R. A.

McCright

R. D.

and Bulle

D. B.

: ‘Survey of degradation modes of candidate materials for high-level radioactive-waste disposal containers’. Lawrence Livermore National Laboratory, Livermore, CA, USA, 1988.

16.

Fujii

Kodama

and Baba

: ‘The effect of water quality on pitting corrosion of copper tube in hot soft water’ Corros. Sci., 1984 24 901–912. doi: 10.1016/0010-938X(84)90127-6

17.

Kear

Barker

B. D.

and Walsh

F. C.

: ‘Electrochemical corrosion of unalloyed copper in chloride media – a critical review’ Corros. Sci., 2004 46 109–135. doi: 10.1016/S0010-938X(02)00257-3

18.

X. G.

C. W.

Dong

C. F.

and Gao

: ‘Approach to copper crevice corrosion phenomenon in 0.5 mol/L NaCl solution’ Equip. Environ. Eng., 2007 4, (3), 31–34.

19.

Palit

and Pehkonen

S. O.

: ‘Copper corrosion in distribution systems: evaluation of a homogeneous Cu₂O film and a natural corrosion scale as corrosion inhibitors’ Corros. Sci., 2000 42 1801–1822. doi: 10.1016/S0010-938X(00)00024-X

20.

Milošev

Metikoš-Huković

Drogowska

Ménard

and Brossard

: ‘Breakdown of passive film on copper in bicarbonate solutions containing sulfate ions’ J. Electrochem. Soc., 1992 139 2409–2418. doi: 10.1149/1.2221241

21.

King

Ahonen

Taxén

Vuorinen

and Werme

: ‘Copper corrosion under expected conditions in a deep geologic repository’. Swedish Nuclear Fuel and Waste Management Company report no. SKB TR-10-67, 2001, and PosivaOy report no. POSIVA 2002-01, 2002, 64.

22.

Duthil

J. P.

Mankowski

and Giusti

: ‘The synergetic effect of chloride and sulphate on pitting corrosion of copper’ Corros. Sci., 1996 38 1839–1849. doi: 10.1016/S0010-938X(96)88250-3

23.

King

and Padovani

: ‘Review of the corrosion performance of selected canister materials for disposal of UK HLW and/or spent fuel’ Corros. Eng. Sci. Technol., 2011 46 82–90. doi: 10.1179/1743278211Y.0000000005

24.

King

Ahonen

Taxén

Vuorinen

and Werme

: ‘Copper corrosion under expected conditions in a deep geologic repository’. SKB report no. TR 01-23, Swedish Nuclear Fuel and Waste Management Company report, 2001, http://www.skb.se and PosivaOy report no. POSIVA 2002-01, 2002, http://www.posiva.fi.

25.

King

and Newman

R. C.

: ‘Stress corrosion cracking of copper canisters’. SKB report no. TR 10-04, Swedish Nuclear Fuel and Waste Management Company Report, 2010. http://www.skb.se.

26.

King

Lilja

Pedersen

Pitkänen

and Vähänen

: ‘An update of the state-of-the-art report on the corrosion of copper under expected conditions in a deep geologic repository’. Swedish Nuclear Fuel Supply Company report no. SKB TR-10-67, PosivaOy report no. POSIVA 2011-01, 2011..

Crevice corrosion of copper for radioactive waste packaging material in simulated groundwater

Abstract

Keywords

Introduction

Experimental

Results and discussion

Effect of oxygen

Effects of Cl− and SO4 2−

Conclusions

Acknowledgements

References

Effects of Cl⁻ and SO₄ ²⁻