This is a repository copy of Assessment of general and localized corrosion behavior of

X65 and 13Cr steels in water-saturated supercritical CO2 environments with SO2/O2.

White Rose Research Online URL for this paper:

http://eprints.whiterose.ac.uk/119263/

Version: Accepted Version

Article:
Hua, Y, Jonnalagadda, R, Zhang, L et al. (2 more authors) (2017) Assessment of general
and localized corrosion behavior of X65 and 13Cr steels in water-saturated supercritical
CO2 environments with SO2/O2. International Journal of Greenhouse Gas Control, 64. pp.
126-136. ISSN 1750-5836

https://doi.org/10.1016/j.ijggc.2017.07.012

© 2017 Published by Elsevier Ltd. Licensed under the Creative Commons

Attribution-NonCommercial-NoDerivatives 4.0 International
http://creativecommons.org/licenses/by-nc-nd/4.0/

Reuse
Items deposited in White Rose Research Online are protected by copyright, with all rights reserved unless
indicated otherwise. They may be downloaded and/or printed for private study, or other acts as permitted by
national copyright laws. The publisher or other rights holders may allow further reproduction and re-use of
the full text version. This is indicated by the licence information on the White Rose Research Online record
for the item.

Takedown
If you consider content in White Rose Research Online to be in breach of UK law, please notify us by
emailing eprints@whiterose.ac.uk including the URL of the record and the reason for the withdrawal request.

eprints@whiterose.ac.uk
https://eprints.whiterose.ac.uk/
1 Assessment of general and localized corrosion behavior of X65 and 13Cr steels in water-

2 saturated supercritical CO 2 environments with SO 2/O2

3
4 Yong Hua*a , Raghu Jonnalagadda a , Lei Zhang b, Anne Nevillea and Richard Barkera
5
6 a Instituteof Functional Surfaces, School of Mechanical Engineering, University of Leeds,
7 Leeds, LS2 9JT, UK
8 b Corrosion and Protection Centre, University of Science and Technology Beijing, 30 Xueyuan

9 Road, Beijing, China

10
11 *Corresponding author: Yong Hua, Tel: 07923359918, Email: leo.huayong@gmail.com
12

13

14 Abstract

15 To mitigate the corrosive effect encountered in carbon steel pipelines during dense phase CO2

16 transport the general consensus is that the CO 2 stream must be sufficiently dehydrated.

17 Although such a process will undoubtedly help prevent the breakout of free water, it can

18 contribute significantly towards the handling costs, particularly in the context of offshore

19 installations. As opposed to drying the CO 2 stream to excessive levels, one alternative option
20 is the application of corrosion resistant alloys such as 13Cr. This paper performs a comparison

21 between X65 carbon steel and 13Cr in pure and impure CO 2, evaluating the influence of SO2

22 and O 2 on the general and localized corrosion rate of both materials at 80 bar and 35°C. The

23 results show that 13Cr is able to perform exceptionally well in comparison to X65 in pure CO 2

24 as well as when SO 2 and O 2 are present in the system individually, producing no localized

25 corrosion and general corrosion rates below 0.02 mm/y. However, when SO 2 and O 2 were

26 combined, the formation of sulfuric acid was permitted which proved detrimental to 13Cr,

27 producing excessive localized attack much greater than that observed on X65. Raman

28 spectrosopy, XRD and SEM/EDX are used to analyse the corrosion products

29 Key words: CO2 corrosion, carbon steel, carbon capture and storage, sulfur dioxide, oxygen

30 1. Introduction

31 Fossil fuels will continue to be the dominant source of the

32 foreseeable future, yet there has been increased concern that the combustion of such carbon-

33 based fuels produces greenhouse gases (particularly CO 2), which adversely affect the global

34 climate.[1]

35 The implementation of Carbon Capture and Storage (CCS) technology would allow the

36 continued use of fossil fuels through the abatement of carbon dioxide (CO2), preventing

37 emissions into the atmosphere. Currently, CCS is the only process available to generate a

38 E t.

39 CCS refers to a process by which CO 2 is captured from large point sources (e.g. power

40 generation and industrial applications), followed by compression and transportation to a

41 storage site (e.g. a geological reservoir or depleted oil field). For the transmission of large
42 quantities of CO 2, the most logical and cost effective solution would be the development of a

43 dedicated pipeline network manufactured from carbon steel to transport CO 2 in a liquid or

44 supercritical state. However, one limitation of carbon steel is its susceptibility to corrosion in

45 flue gas environments due to the presence of CO 2, water (H2O), oxygen (O 2), sulfur dioxide

46 (SO2), nitrogen oxides (NO x) and other constituents that can result in the formation of

47 corrosive phases.

48 A number of studies have recently been published which seek to understand the implications

49 of such impurities in liquid or supercritical CO 2 on the extent of carbon steel corrosion during

50 pipeline transportation.[2-20] Research has also focused on defining the safe limits of impurities

51 that can be tolerated within the CO2 stream by systematically varying the water content or

52 concentration of other contaminants such as SO 2, O2 and NO x amongst others.[4, 5, 7, 9, 18, 20, 33,

53 34, 35, 36]

54 From a review of the literature within this subject area it appears that in order to

55 appropriately mitigate excessive corrosion rates the general consensus is that sufficient

56 drying (i.e. water removal) of impure CO 2 upstream of the pipeline is required. Although such

57 a process can prevent the breakout of free water, it can contribute significantly towards

58 handling costs, particularly in the context of offshore installations. [1] Furthermore, some

59 studies have suggested that extensive dehydration down to 50 ppm (mole) should be applied,

60 which can require the use of molecular sieves, imposing an even greater cost [9, 32]. Limits as

61 low as 50 ppm have already been implemented in the US[21] and Norway[22] for specific

62 pipelines. However, other specifications from the DYNAMIS project[23] and for the Kinder

63 Morgan pipeline[24] are less conservative, imposing limits of 500 and 650 ppm, respectively.
64 As opposed to dehydrating the CO 2 stream, one alternative option is the application of

65 corrosion resistant alloys (CRAs) such as 13Cr. However, limited data on CRA corrosion in

66 dense phase CO 2 is available in the literature with the exception of work by Choi et al. [16] which

67 was performed with 13Cr in the presence of very high SO 2/O2 concentrations (1% and 5%,

68 respectively) that are orders of magnitude greater than those typically anticipated for

69 anthropogenic CO 2 transport according to Walspurger et al.[25].

70 The purpose of this present study is to contribute to the literature by determining the ability

71 of 13Cr relative to X65 carbon steel to mitigate corrosion in water-saturated dense phase CO2

72 when SO2 and O2 are present both individually and together. Both the general and localized

73 corrosion behavior of X65 and 13Cr are determined using white light interferometry and the

74 nature and morphology of corrosion products formed on the steel surface are reviewed using

75 a combination of scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy

76 (EDX), and X-ray diffraction (XRD) in order to clarify the role low concentrations of SO 2 and O2

77 play in the degradation process and ascertain whether selection of a CRA is a suitable

78 corrosion mitigation option.

79 2. Experimental procedure

80 2.1 Sample preparation

81 Test specimens were machined into discs of diameter 25 mm and thickness of 6 mm from

82 both API 5L X65 carbon steel and 13Cr (UNS41000) bars. The chemical composition of X65

83 and 13Cr and their microstructures are provided in Table 1 and Figure 1 respectively. Surface

84 preparation for corrosion experiments consisted of wet-grinding the entire sample with up to

85 600 grit silicon carbide abrasive paper, rinsing with distilled water, followed by acetone and

86 high purity ethanol, followed by drying gently with compressed air. Samples were then stored
87 in a desiccator until required and weighed immediately before the experiment on an

88 electronic balance to within an accuracy of 0.01 mg before suspending inside the autoclave.

89 Two samples of the same material were placed within the autoclave for each individual test.

90 Surface preparation for the microstructures consisted of wet-grinding the entire sample

91 surface up to 1200 grit silicon carbide abrasive paper, followed by polishing with the help of

92 3µm diamond suspension to attain a mirror finish, rinsing with distilled water, followed by

93 acetone, high purity ethanol and drying gently with compressed air. Etchants were varied

94 depending on the sample alloy composition. A 2% Nital was used for X65 where etching

95 consisted of swabbing the surface for 10-20 seconds with a cotton pad. 13Cr was etched with

96 waterless Kalling's reagent (5 g copper chloride + 100 ml hydrochloric acid + 100 ml ethanol)

97 by swabbing the surface for 10 seconds with a cotton pad.

98

99 2.2 Autoclave testing procedure

100 A schematic representation of the autoclave experimental system layout is provided in Figure

101 2. The testing procedure has been published in a previous paper. [7]

102 All tests were conducted under static conditions in water-saturated supercritical CO 2 (a water

103 content of 3437 ppm in the dense phase at 80 bar and 35°C based on the analysis performed

104 by Spycher et al.[26]). However, in order to ensure complete saturation of CO 2 under these

105 conditions, 34000 ppm of water was introduced to the bottom of the autoclave (i.e. not in

106 direct contact with the sample). The entire matrix of the experimental conditions is provided

107 in Table 2 which describes the different materials and conditions that were evaluated. The

108 molar concentrations of SO 2 (100 ppm) and O 2 (1000 ppm) were specifically chosen to reflect

109 the recommended limits proposed by de Visser et al. [23] and Alstom (which can be found in a
110 publication by Dugstad et al. [13]) to ensure safe CO 2 transport. However, it should be noted

111 that the proposed limits within these publications are based on health and safety criteria in

112 the event of a sudden release from a pipeline, not from the perspective of corrosion/pipeline

113 integrity. This study aims to evaluate whether these proposed limits are tolerable within a

114 CO2 pipeline when the stream is saturated with water. Previous research [37, 38] has indicated

115 that the addition of these particular impurities at the aforementioned concentrations do not

116 shift the critical point of CO 2 significantly. As such, all experiments performed in this study are

117 conducted with CO 2 within its supercritical state.

118 It is difficult to monitor the actual pH of the solution due to the presence of small amounts of

119 water in the system and the considerable pressure. Consequently, the OLI software [31] was

120 used to speculate about the possible pH of the aqueous phase as shown in Figure 3.

121 Considering the experimental conditions in this work (35°C and 80 bar CO 2), the solution pH

122 is expected to reduce from 3.1 in the absence of SO 2 to 2.4 when 100 ppm SO 2 is introduced.

123 This is based on the assumption that SO 2 will partially dissolve into the aqueous phase to form

124 sulfurous acid (H2SO3) as stated by Dugstad et al. [10] It is difficult to predict the effect of

125 combined addition of O 2 and SO 2 to this system. However, based on the assumption that the

126 likely effect is to promote the formation of sulfuric acid through the oxidation of sulfurous

127 acid, the pH can be expected to decrease markedly given that sulfuric acid is a much stronger

128 acid.

129 At the end of each experiment, the specimens were dried thoroughly and subsequently

130 chemically cleaned to remove all traces of corrosion products before weighing. The cleaning

131 C

132 antimony trioxide + 50 g stannous chloride + 1000 ml hydrochloric acid) in accordance with
133 ASTM Standard G1-03[27]. This was followed by rinsing the samples with distilled water and

134 then drying with compressed air.

135 The mass loss due to corrosion was determined from the weight difference before exposure

136 and after cleaning. The corrosion rates were calculated using Equation (1):

(1)

137 where Vc is the corrosion rate of the sample in mm/year, m is the weight loss in grams, is

138 the density of the sample in g/cm3, A is the exposed area in cm2 and t is the immersion time

139 in hours.

140 3. Results and Discussion

141 3.1 General corrosion of X65 and 13Cr steels exposed to water-saturated supercritical CO2

142 condition with various SO 2/O2 concentrations

143 Figure 4 provides the general corrosion rates recorded (based on gravimetric analysis) for X65

144 and 13Cr exposed to water-saturated conditions in the presence of various concentrations of

145 SO2 and O2. In the absence of SO 2 and O 2, the general corrosion rates of X65 and 13Cr are 0.1

146 and 0.003 mm/y, respectively. Addition of 1000 ppm O 2 serves to reduce the general

147 corrosion rates of both materials to 0.03 and 0.001 mm/year. The ability of O 2 to reduce the

148 general corrosion of carbon steel in CO 2 systems has been reported previously by other

149 authors at low temperatures and can be attributed to the formation of protective oxide films

150 on the steel surface[5, 28]. In terms of 13Cr, O 2 is also known to assist in the formation and

151 replenishment of the passive chromium oxide films that can be established on the steel

152 surface, which can explain the increase in general corrosion resistance with the introduction

153 of O 2.
154 The mass loss measurements in Figure 4 indicate that O 2 has no detrimental effect on the

155 general corrosion of X65 and 13Cr in a CO 2-H2O-O2 system under these particular conditions

156 and that 13Cr exhibits improved corrosion resistance compared to X65 in both environments .

157 The superiority of 13Cr relative to X65 in CO 2-H2O and CO 2-H2O-O2 systems at high pressure

158 was also reported by Choi et al. [16] for experiments performed at 80 bar and 50°C in CO 2-

159 satuarted water. However, in contrast to the results in Figure 4, Choi et al.[16] reported an

160 increase in general corrosion rates of X65 for experiments in water-saturated CO 2 with the

161 introduction of O 2. A possible explanation for the disparity in observations could be linked to

162 the temperature difference in experiments, but also the higher O 2 contents of 1.6-5.5 bar

163 evaluated by Choi et al., [16]. As O 2 content and temperature are increased, O 2 is capable of

164 influencing the kinetics of the cathodic reaction at the steel surface (see Equation (2) for the

165 reaction under acidic environments), which may explain the accelerated corrosion reported

166 by Choi and co-workers. [16]

(2)

167

168 Referring again to Figure 4, the introduction of solely 100 ppm SO 2 to the CO 2 system resulted

169 in an increase in general corrosion rates of X65 and 13Cr from 0.1 and 0.003 mm/y to 0.65

170 and 0.01 mm/y, respectively. Generally, the acceptable internal corrosion limit for such

171 pipelines is ~0.1 mm/y, although this depends on the system design life and the anticipated

172 corrosivity of the fluid being transported. For the sake of material evaluation, adopting 0.1

173 mm/y as an acceptable benchmark, the corrosion rate of X65 under these conditions is well

174 beyond this limit, whereas 13Cr is still able to offer adequate corrosion protection even in a

175 water-saturated environment.

176 Addition of both 100 ppm SO 2 and 1000 ppm O 2 (final column in Figure 4) clearly enhances

177 the corrosion rate of both materials from the pure CO 2 environment, particularly in the case

178 of 13Cr. The general corrosion rates recorded were 0.95 and 0.75 mm/y for X65 and 13Cr,

179 respectively. The collection of results within Figure 4 indicate a noticeable synergistic effect

180 between SO 2 and O2. Corrosion rates observed when SO 2 and O2 are together exceed the sum

181 of the two individual degradation rates when the both species are present individually. The

182 observed synergistic effect is evident in both materials and results in 13Cr exceeding 0.1

183 mm/y by a large margin (by 0.65 mm/year), making it an unsuitable material choice for such

184 an environment based on this reported limit.

185 3.2 Corrosion product morphology for X65

186 Figure 5 shows the SEM images of the X65 steel surface after 48 h exposure to the water-

187 saturated CO 2 phase containing different concentrations of SO 2 and O 2. In the absence of SO2

188 and O2, iron carbonate (FeCO3) is the sole crystalline product recorded on the steel surface

189 (Figure 5(a)) and this is confirmed by the XRD pattern provided in Figure 6. Addition of 1000

190 ppm O2 inhibited the formation of FeCO 3, producing a seemingly amorphous corrosion

191 product layer which produced no XRD pattern within Figure 6. XPS analysis of the film in a

192 previous publication identified that the film formed under these conditions predominantly

193 comprises of iron oxides and/or hydroxides [5]. It is assumed that the oxidation of Fe 2+ to Fe3+

194 due to the presence of O 2 inhibits the formation of FeCO 3.

195 Addition of solely 100 ppm SO 2 (SEM image in Figure 5(c)) produced a corrosion product layer

196 comprising of both FeCO 3 and iron sulfite (FeSO 3) according to the XRD pattern provided in

197 Figure 6. This was confirmed by additional Raman spectroscopy measurements performed on

198 the same sample. The introduction of both 100 ppm SO 2 and 1000 ppm O 2 resulted in the
199 formation of FeCO 3 and FeSO 3 (confirmed by XRD patterns in Figure 6), but also the co-

200 presence of hydrated FeSO4 based on localized Raman spectra (specifically peaks at 185, 480

201 and 990 cm-1[29] identified within Figure 7).

202 These observations are in alignment with Choi et al. [16] who reported that the presence of SO2

203 alone promoted the formation of FeSO 3 on X65 steel while the addition of O 2 can form FeSO4.

204 The work is also in alignment with Dugstad et al. [10] who stated that the presence of SO2

205 permits the formation of sulfurous acid (H2SO3), and the addition of O 2 enables sulfuric acid

206 (H2SO4) to be formed via the series of reactions (3) to (5). H2SO4 is a significantly stronger acid

207 then H2SO3, and is potentially more corrosive, corroborating with the observed synergistic

208 effect between O 2 and SO 2.

(3)

(4)

(5)

209 The formation of sulfite and sulfate ions enables the precipitation of FeSO 3 and FeSO 4 via

210 precipitation reactions:

(6)

(7)

211 while FeCO 3 is produced from the precipitation of iron and bicarbonate ions formed from the

212 steel dissolution and dissociation of carbonic acid within the aqueous phase:
 (8)

213 3.3 Corrosion product morphology for 13Cr

214 13Cr samples exposed to 0 ppm SO 2 with and without 1000 ppm O 2 produced general

215 corrosion rates below 0.003 mm/year, indicating that O 2 presence has little effect on the

216 general corrosion of the CRA. SEM images of the steel surface indicated no noticeable signs

217 of general or localized corrosion and are provided in Figure 8. SEM images of the 13Cr surface

218 after exposure to 100 ppm SO 2 with and without 1000 ppm O 2 are provided in Figure 9 and

219 show clear signs of material degradation as a result of introducing SO 2, the presence of SO2

220 permits the formation of H2SO3 and the addition of O 2 enables H2SO4 to be formed. Both

221 H2SO3 and H2SO4 are more corrosive in comparison to that of H 2CO3, corroborating with the

222 increased mass loss observed. No significant corrosion product was visible on the 13Cr surface

223 exposed to solely 100 ppm SO 2 (Figure 9(a)) as degradation rates under these conditions were

224 still low at 0.01 mm/y and unlikely to result in substantial corrosion product precipitation.

225 However, the combined presence of SO 2 and O 2 resulted in corrosion rates rising to 0.65

226 mm/y, producing a thin, cracked corrosion product layer (Figure 9(b)).

227 XRD and Raman analysis of the corrosion product observed in the SEM image shown in Figure

228 9(b) proved challenging and were unable to identify the nature of the corrosion product

229 present. However, cross-section EDX analysis of the films formed on X65 and 13Cr after

230 exposure to 100 ppm SO 2 and 1000 ppm O 2 after 6 h and 48 h (Figure 10) showed that the

231 thickness of the corrosion products increased for both materials and the corrosion product

232 on 13Cr was rich in Cr, S and O, and contained very little traces of Fe. Given that the combined

233 presence of O 2 and SO 2 permits the formation of H2SO4, as discussed by Dugstad et al.[10], it is

234 suggested that this acid is capable of dissolving the Cr2O3 passive film and reacting with Cr3+
235 to produce chromium sulfate (Cr2(SO4)3), although this cannot be confirmed and requires

236 further study.

237

238 3.4 Localized corrosion of X65 and 13Cr

239 One particular concern in acidic environments, particularly in CRAs is the potential for

240 localized corrosion to occur. Each material and test environment combination considered in

241 Figure 1 (and Table 2) was evaluated for localized attack using white-light interferometry. The

242 threshold value used was 1 µm and so the measured localized depth beneath that value of

243 depth will not be considered. An example of profilometry images of X65 and 13Cr sample

244 surface exposure to water-saturated CO 2 condition in the presence of 100 ppm SO 2 and 1000

245 ppm O2 are provided in Figure 11. Smaller versions of these images are also embedded within

246 Figure 12 which provides a summary of the pit/localized depth measurements extracted from

247 X65 and 13Cr samples from all tests performed in Figure 1 and are expressed as penetration

248 rates. The localized/pit depth analysis was conducted in alignment with ASTM Standard G46-

249 94, selecting the average of the 10 deepest pits on the surface to provide an accurate

250 assessment[30]. Multiple scans (at least 3) such as those shown in Figure 11 were performed

251 across the steel surface to ensure reliable data was obtained.

252 Considering Figure 12, it is evident that X65 steel undergoes localized attack in all

253 environments. The attack manifests itself as a form of micro-pitting over 48 h (typical image

254 shown in Figure 11) and is particularly prominent in the presence of 1000 ppm O 2 and 100

255 ppm SO 2/1000 ppm O 2, exceeding pitting rates of 2 mm/year. In contrast, 13Cr was only

256 susceptible to localized corrosion when both SO 2 and O 2 were present together. Under these

257 conditions, the form of corrosion observed was a combination of extensive localized attack
258 and pitting, with the surrounding material adjacent to the attack undergoing significantly less

259 attack. SEM images of the X65 and 13Cr surfaces after 6h and 48 h of exposure are provided

260 in Figure 13. These images were taken C the

261 steel surface; they indicate that the pit initiation is not limited to solely the ferrite phase of

262 the material for X65 steel.

263 3.5 100 ppm SO2 and 1000 ppm O2 evolution of corrosion products and general/localized

264 corrosion rate vs time

265 The initial localized depth measurements at 48 h in the previous section cast ambiguity over

266 the actual growth of pits/localized areas as they are purely one measurement performed at a

267 particular instance in time. To determine the nature of surface pit growth/rate of localized

268 attack, X65 and 13Cr samples were exposed to the 100 ppm SO 2/1000 ppm O 2 environment

269 for varying exposure times between 6 and 48 h. After each experiment, gravimetric

270 measurements and surface profilometry were performed on both materials to determine the

271 general and localized corrosion rates, respectively. The data collected is provided in Figure 14

272 and shows that there is little difference between the general corrosion rate of X65 and 13Cr

273 over 48 h of exposure. However, a difference of one order of magnitude is recorded between

274 X65 and 13Cr in terms of localized attack, with 13Cr showing much greater susceptibility to

275 the SO2/O2 environment than X65 steel. Both materials display decay in their rate of localized

276 attack as a function of time. The actual general and localized corrosion rates could be higher

277 than the values recorded here due to depletion of impurities in a closed autoclave over time.

278 However, the localized attack for both materials remain high for the duration of the

279 experiment.
280 The evolution of corrosion products on the surface of X65 and 13Cr is provided in Figure 15

281 and indicates the presence of a sulfur-rich, cracked film as early as 6 h into the experiment on

282 both materials. Further work is still needed to determine when exactly pit nucleation starts

283 to occur and what causes this process. Additionally, further work is required to determine

284 whether the reduction in localized corrosion rate is attributed to corrosion product formation

285 in the localized region, re-passivation of the oxide film or depletion of impurities within the

286 autoclave.

287 Conclusions
288 The research presented has focused towards studying and quantifying the extent of both
289 general and localized corrosion of X65 and 13Cr in water-saturated supercritical CO2
290 environments containing various SO2 and O 2, representative of dense phase CO 2 transport.
291 Tests were conducted at a pressure of 80 bar and a temperature of 35°C for periods of up to
292 48 h. The main conclusions which can be drawn from this study are:

293  The general and localized corrosion rate of X65 in water-saturated dense-phase
294 environments is accentuated by the presence of 100 ppm SO 2 and 100 ppm SO 2/1000
295 ppm O 2. The effect of 1000 ppm O 2 as a sole impurity served to reduce general
296 corrosion rates, but promotes more substantial pitting on the steel surface.
297  The most aggressive environment for X65 was obtained when SO 2 and O 2 were
298 combined together which produced general corrosion rates of 0.95 mm/y. SO 2 alone
299 in the water-saturated dense phase CO 2 permits the formation of sulfurous acid,
300 which resulted in the precipitation of FeSO 3 on the steel surface. However, the
301 addition of O 2 was shown to result in the formation of SO 42- (through the formation of
302 sulphuric acid), producing FeSO 4 on the steel surface.
303  13Cr was shown to be highly resistant to corrosion in the CO 2-H2O-O2 and CO 2-H2O
304 environments, producing no indication of localized attack and general corrosion rates
305 below 0.003 mm/y. The material also performed well in experiments with 100 ppm
306 SO2 present, producing no signs of localized corrosion and general corrosion rates
307 below 0.02 mm/y.
308  The combined presence of O 2 and SO 2 was detrimental to 13Cr, producing general
309 corrosion rates of 0.65 mm/y. A clear synergistic effect was observed between SO2
310 and O 2 for both X65 and 13Cr whereby the degradation rates resulting from the
311 combined presence exceeded the total of the corrosion rates produced when the
312 species were present individually.
313  Pitting rates of X65 and 13Cr were shown to be in excess of 7 and 80 mm/year at the
314 start of the experiment for X65 and 13Cr, respectively. The pitting rates were shown
315 to reduce with time by around an order of magnitude. If was not clear whether the
316 reduction in corrosion rate was attributed to corrosion product formation, pit
317 repassivation or depletion of impurities within the system.
318

319 References

320 1. G.A. Jacobson, S. Kerman, Y.-S. Choi, A. Dugstad, S. Nesic, and S. Papavinasam, "Pipeline

321 Corrosion Issues Related to Carbon Capture, Transportation, and Storage", Materials

322 Performance, (2014): p. 24-31.

323 2. Y. Hua, R. Barker, and A. Neville, "Comparison of corrosion behaviour for X-65 carbon steel in

324 supercritical CO2-saturated water and water-saturated/unsaturated supercritical CO 2", The

325 Journal Of Supercritical Fluids, 97, (2015): p. 224-237.

326 3. Y. Hua, R. Barker, and A. Neville, "Relating iron carbonate morphology to corrosion

327 characteristics for water-saturated supercritical CO2 systems", The Journal Of Supercritical

328 Fluids, 98, (2015).

329 4. Y. Hua, R. Barker, and A. Neville, "The influence of SO 2 on the tolerbale water content to avoid

330 pipeline corrosion during the transportation of supercritical CO 2", International Journal of

331 Greenhouse Gas Control, 37, (2015): p. 412-423.

332 5. Y. Hua, R. Barker, and A. Neville, "The effect of O2 content on the corrosion behaviour of X65

333 and 5Cr in water-containing supercritical CO 2 environments", Applied Surface Science, 356,

334 (2015): p. 499-511.

335 6. Y. Hua, R. Barker, and A. Neville, "Effect of temperature on the critical water content for

336 general and localised corrosion of X65 carbon steel in the transport of supercritical CO 2",

337 International Journal of Greenhouse Gas Control, 31, (2014): p. 48-60.

338 7. Y. Hua, R. Barker, and A. Neville, "Understanding the influence of SO 2 and O2 on the corrosion

339 of carbon steel in water-saturated supercritical CO2 ", Corrosion, 71, 5 (2014): p. 667-683.

340 8. M. Halseid, A. Dugstad, and B. Morland, "Corrosion and bulk phase reactions in CO 2 transport

341 pipelines with impurities: Review of recent published studies", Energy Procedia, 63, (2014): p.

342 2557-2569.

343 9. J. Brown, B. Graver, E. Gulbrandsen, A. Dugstad, and B. Morland, "Update of DNV

344 Recommended Practice RP-J202 with focus on CO 2 corrosion with impurities", Energy

345 Procedia, 63, (2014): p. 2432-2441.

346 10. A. Dugstad, M. Halseid, and B. Morland, "Effect of SO 2 and NO2 on corrosion and solid

347 formation in dense phase CO 2 pipelines", Energy Procedia, 37, (2013): p. 2877-2887.

348 11. A. Dugstad, M. Halseid, B. Morland, and A.O. Sivertsen, "Corrosion in dense phase CO 2 the

349 impact of depressurisation and accumulation of impurities", Energy Procedia, 37, 0 (2013): p.

350 3057-3067.

351 12. A. Dugstad, B. Morland, and S. Clausen, "Corrosion of transport pipelines for CO2 - Effect of

352 water ingress", Energy Procedia, 4, (2011): p. 3063-3070.

353 13. A. Dugstad, S. Clausen, and B. Morland. "Transport of dense phase CO 2 in C-steel pipelines -

354 when is corrosion an issue?", CORROSION 2011, paper no. 70, (Houston, TX:NACE, 2011).

355 14. Y.-S. Choi, S. Hassani, T.N. Vu, and S. Nesic. "Effect of H 2S on the corroison behavior of pipeline

356 steels in supercritical and liquid CO 2 environments", CORROSION 2015, paper no. 5927,

357 (Dallas, TX: NACE, 2015).

358 15. S. Sim, I.S. Cole, Y.S. Choi, and N. Birbilis, "A review of the protection strategies against internal

359 corrosion for the safe transport of supercritical CO 2 via steel pipelines for CCS purposes",

360 International Journal of Greenhouse Gas Control, 29, (2014): p. 185-199.

361 16. Y.-S C S N D Y E CO 2

362 transmission pipeline steel in supercritical CO 2 water environments", Environmental Science

363 & Technology, 44, 23 (2010): p. 9233-9238.

364 17. Y. Xiang, Z. Wang, Z. Li, and W. Ni, "Effect of temperature on corrosion behaviour of X70 steel

365 in high pressure CO2/SO2/O2/H2O environments", Corrosion Engineering, Science and

366 Technology, 48, 2 (2013): p. 121-129.

367 18. Y. Xiang, Z. Wang, X. Yang, Z. Li, and W. Ni, "The upper limit of moisture content for

368 supercritical CO2 pipeline transport", The Journal of Supercritical Fluids, 67, (2012): p. 14-21.

369 19. Y. Xiang, Z. Wang, Z. Li, and W.D. Ni, "Effect of Exposure Time on the Corrosion Rates of X70

370 Steel in Supercritical CO 2/SO2/O2/H2 O Environments", Corrosion, 69, 3 (2012): p. 251-258.

371 20. Y. Xiang, Z. Wang, C. Xu, C. Zhou, Z. Li, and W. Ni, "Impact of SO 2 concentration on the

372 corrosion rate of X70 steel and iron in water-saturated supercritical CO2 mixed with SO2", The

373 Journal of Supercritical Fluids, 58, 2 (2011): p. 286-294.

374 21. F.W. Schremp and G.R. Roberson, "Effect of supercritical carbon dioxide (CO 2) on construction

375 materials", SPE, 15, 3 (1975): p. 227-233.

376 22. A. Oosterkamp and J. Ramsen, "State-of-the-art overview of CO2 pipeline transport with

377 relevance to offshore pipelines", Polytech Report No: POL-O-2007-138-A, (2008).

378 23. E. de Visser, C. Hendriks, M. Barrio, M.J. Mølnvik, G. de Koeijer, S. Liljemark, and Y. Le Gallo,

379 "Dynamis CO2 quality recommendations", International Journal of Greenhouse Gas Control, 2,

380 4 (2008): p. 478-484.

381 24. Kinder Morgan, "CO2 Transportation pipelines", [Online], Available at:

382 http://www.kindermorgan.com/business/co2/transport.cfm. Accessed on 11th March 2015.

383 25. S. Walspurger and H.A.J.v. Dijk, "EDGAR CO 2 purity: type and quantities of impurities related

384 to CO2 point source and capture technology: a literature study", ECN-E-12-054, (2012).
385 26. N. Spycher, K. Pruess, and J. Ennis-King, "CO2 -H2O mixtures in the geological sequestration of

386 CO2. I. Assessment and calculation of mutual solubilities from 12 to 100°C and up to 600 bar",

387 Geochimica et Cosmochimica Acta, 67, 16 (2003): p. 3015-3031.

388 27. ASTM, Standard G1-03, Standard practice for preparing, cleaning, and evaluating corrosion

389 test specimens. ASTM International: West Conshohocken, PA, 2003.

390 28. G. Schmitt and R. Forster. "Unexpected Effect of Small Oxygen Concentrations in Sales Gas on

391 Element Currents between Pipeline Steel and Magnetite from Black Powder", CORROSION

392 2015: NACE International, 2015).

393 29. C.H. Chio, S.K. Sharma, and D.W. Muenow, "The hydrates and deuterates of ferrous sulfate

394 (FeSO4): a Raman spectroscopic study", Journal of Raman Spectroscopy, 38, 1 (2007): p. 87-

395 99.

396 30. ASTM, Standard G46-94, Standard guide for examination and evaluation of pitting corrosion.

397 ASTM International: West Conshohocken, PA, 2003.

398 31. OLI is a commercial software package from Corr Science, for more information see:

399 http://www.corrscience.com/products/oli/.

400 32. R. Barker, Y. Hua, and A. Neville, Internal corrosion of carbon steel pipelines for dense -phase

401 CO2 transport in carbon capture and storage (CCS) a review. International Materials Reviews,

402 2017. 62(1): p. 1-31.

403 33. Xu, M., Q. Zhang, X. Yang, Z. Wang, J. Liu, and Z. Li, Impact of surface roughness and humidity

404 on X70 steel corrosion in supercritical CO 2 mixture with SO 2, H 2 O, and O 2. The Journal of

405 Supercritical Fluids, 2016. 107: p. 286-297.

406 34. Sun, C., J. Sun, Y. Wang, X. Lin, X. Li, X. Cheng, and H. Liu, Synergistic effect of O2, H2S and SO2

407 impurities on the corrosion behavior of X65 steel in water-saturated supercritical CO2 system.

408 Corrosion Science, 2016.

409 35. Sun, J., C. Sun, G. Zhang, X. Li, W. Zhao, T. Jiang, H. Liu, X. Cheng, and Y. Wang, Effect of O2

410 and H2S impurities on the corrosion behavior of X65 steel in water-saturated supercritical CO2

411 system. Corrosion Science, 2016. 107: p. 31-40.

412 36. Sun, C., Y. Wang, J. Sun, X. Lin, X. Li, H. Liu, and X. Cheng, Effect of impurity on the corrosion

413 behavior of X65 steel in water-saturated supercritical CO2 system. The Journal of Supercritical

414 Fluids, 2016. 116: p. 70-82.

415 37. Farelas. F, Choi. Y.-S N . S, Corrosion Behavior of API 5L X65 Carbon Steel

416 under Supercritical and Liquid Carbon Dioxide Phases in the Presence of Water and

417 Sulfur Dioxide. Corrosion, 2012. 69(3): p. 243-250.

418 38. Farelas. F, Choi Y.-S., and Nesic. S, "Effects of CO 2 phase change, SO 2 content and flow

419 on the corrosion of CO 2 transmission pipeline steel", in CORROSION 2012. 2012: Salt

420 Lake City, UT:NACE.

421
422 Table 1: Elemental composition (wt.%) of API 5L X65 carbon steel and UNS41000 13Cr
X65 13Cr
C 0.12 0.15
Si 0.18 0.36
Mn 1.27 0.56
P 0.008 0.006
S 0.002 0.008
Cr 0.11 13.5
Mo 0.17 -
Fe Balance
423
424 Table 2: Test matrix for corrosion experiments in water-saturated dense phase CO2

Pressure SO2 O2 Immersion

Temperature Materials H2O (ppm)
(bar) (ppm) (ppm) time hours
(°C)
0 0
Above solubility
0 1000
limit of 3437ppm
35 80 X65/13Cr through addition of 100 0 48
34000ppm water
100 1000

Pressure SO2 O2 Immersion

Temperature Materials H2O (ppm)
(bar) (ppm) (ppm) time hours
(°C)
Above solubility
limit of 3437ppm 6/14/24/48
35 80 X65/13Cr 100 1000
through addition of
34000ppm water
425
426

(a) (b)
427 Figure 1: Microstructure of a) API 5L X65 depicting a ferritic-pearlitic structure and b) UNS41000
428 13Cr depicting a martensitic structure

429

430 Figure 2: Schematic of the autoclave setup for impure, dense phase CO2 experiments

431

432 Figure 3: Predicted pH of CO2 -saturated water in the presence of SO2 at 35oC and 80 bar
 1

General corrosion rate (mm/year) 0.1

0.01

0.001
SO SO SO SO
O O O O
X65 13Cr
433
434 Figure 4: General corrosion rates of X65 and 13Cr after exposure to water-saturated dense
435 phase CO2 at 80 bar and 35 °C for 48 h in the presence of various concentrations of SO2
436 and O2
437
438
439
440
441

(a) (b)
 (c) (d)

442 Figure 5: SEM images of X65 samples exposed to water-saturated supercritical CO 2

443 conditions at 80 bar and 35°C containing (a) 0 ppm SO2 and 0 ppm O2; (b) 0 ppm SO2 and
444 1000 ppm O2; (c) 100 ppm SO2 and 0 ppm O2 and (d) 100 ppm SO2 and 1000 ppm O2
445

446

447 Figure 6: XRD patterns of X65 samples exposed to water-saturated supercritical CO 2

448 containing various SO 2 and O2 at 35°C and 80 bar after 48 h
449

450 Figure 7: Raman spectra of X65 sample exposed to water-saturated supercritical CO 2

451 containing 100 ppm SO2 and 1000 ppm O2 at 35°C and 80 bar after 48 h
452

(a) (b)
453 Figure 8: SEM images of 13Cr samples exposed to water-saturated supercritical CO 2
454 condition with, (a) 0 ppm SO 2 and 0 ppm O2 and (b) 0 ppm SO2 and 1000 ppm O2 after 48 h
455
456
 (a) (b)
457 Figure 9: SEM images of 13Cr samples exposed to water-saturated supercritical CO 2
458 condition with (a) 13Cr - 100 ppm SO2 and 0 ppm O2, (b) 13Cr - 100 ppm SO2 and 1000 ppm
459 O2 after 48 h
 (a) X65 6hrs Steel Fe Cr S O

Resin

100 µm

(b) X65 48hrs

Fe Cr S O

Steel
Resin

100 µm

(c) 13Cr 6hrs Fe Cr S O

Steel
Resin

100 µm

(d) 13Cr 48hrs

Fe Cr S O

Resin Steel

100 µm

Figure 10: SEM cross-section images of (a) X65 6 hours, (b) X65 48 hours, (c) 13Cr 6 hours and (d) 13Cr 48 hours samples exposed to water-saturated
supercritical CO 2 at 80 bar and 35°C with 100 ppm SO2 and 1000 ppm O2.
 (a)

(b)
Figure 11: Examples of profilometry images (after removal of corrosion products) from (a)
X65 and (b) 13Cr after exposure to water-saturated supercritical CO 2 in the presence of
100 ppm SO2 and 1000 ppm O2 at 35°C and 80 bar for 48 h
 Localized corrosion rate (mm/year)

10

No No localized No
localized corrosion localized
corrosion observed corrosion
0.1
SO SO SO SO
O O O O
X65 13Cr

Figure 12: Localized corrosion rates of X65 and 13Cr in water-saturated supercritical CO 2
environments containing varying concentrations of SO 2 and O2 at 35°C and 80 bar for 48 h

Localised attack
Localised

(a) (b)

Pitting Pitting

(c) (d)
Figure 13: SEM images of samples (after removal of the corrosion products) exposed to
water-saturated supercritical CO 2 condition with 100 ppm SO 2 and 1000 ppm O2 (a) X65
6 h, (b) X65 - 48 h, (c) 13Cr 6 h and (d) 13Cr 48 h
 0.06 100

0.05

General corrosion rate

Total mass loss (g)

0.04 10

(mm/year)
0.03

0.02 1

0.01

0 0.1
5 10 15 20 25 30 35 40 45 50 0 10 20 30 40 50
Time (h) Time hours
X65 13%Cr X65 13Cr

(a) (b)

1000 100

Localized corrosion rate

Localised depth (µm)

100 10
(mm/year)

10 1

1 0.1
5 10 15 20 25 30 35 40 45 50 0 10 20 30 40 50
Time (h) Time (h)
X65 13%Cr X65 13Cr

(c) (d)

Figure 14: (a and b) Total mass loss and general corrosion rates and (c and d) localized
depth and localized corrosion rates of X65 and 13Cr in water-saturated supercritical CO 2 in
the presence of 100 ppm SO 2 and 1000 ppm O2 at 80 bar and 35 °C for exposure times of 6,
14, 24 and 48 h
 (a) (b) (c) (d)

(e) (f) (g) (h)

Figure 15: SEM images of samples exposed to water-saturated supercritical CO 2 with 100 ppm SO2 and 1000 ppm O2 (a) X65 6 h, (b) X65
14 h, (c) X65 24 h, (d) X65 48 h, (e) 13Cr 6 h, (f) 13Cr 14 h, (g) 13Cr 24 h, and (h) 13Cr 48 h

30