Effect of temperature on corrosion behaviour of X70 steel in high pressure CO 2 /SO 2 /O 2 /H 2 O environments

Abstract

The corrosion study of the pipeline steel is extremely important for the safety of CO₂ pipeline. The present work studied the effect of temperature on the corrosion behaviour of X70 steel in high pressure CO₂/SO₂/O₂/H₂O mixtures. Weight loss method was utilised to measure the corrosion rate. Environmental scanning electron microscopy and focused ion beam technology were employed to observe the morphology of product scales. X-ray diffraction was used to analyse the composition of product scales. The porosity of product scales was measured, and the exhaust gas was detected by gas chromatography. Results showed that the corrosion rate increased with the increasing temperature and then started to decline with temperature with the peak corrosion rate at 348 K. Results also showed that there was hydrogen that existed in the exhausted gas, indicating that the hydrogen evolution reaction was contained in the corrosion cathodic process.

Keywords

Carbon capture and storage Temperature Corrosion Scale Weight loss

Introduction

In recent years, climate change has caused extensive attention among the government, industry, academia and public almost all over the world. Carbon capture, transport and storage has been taken as one of the effective ways to address this issue,1 and therefore, it became a worldwide hot research topic.

Pipeline transport is one of the main ways to transport the captured CO₂ to the sequestration sites under supercritical condition. There are many CO₂ transportation pipelines that existed for enhanced oil recovery in the world with a length longer than 6000 km.2 Therefore, there is industrial experience for CO₂ pipeline transportation. However, the experience is limited only for enhanced oil recovery application, in which there are almost no corrosive gases such as SO₂ that existed in the CO₂ mixtures. However, when the CO₂ is captured from exhaust gas of power plant rather than natural CO₂ well, a certain amount of impurities are present, especially when the CO₂ will be coinjected with other impurities such as SO₂ captured by oxyfuel technology. There are still many problems requiring in depth study.

A number of scholars have conducted a preliminary study on the corrosion of pipeline steel in supercritical CO₂ containing impurities like SO₂,3 ^– 8 but the job is far from the end, including the impact of temperature on corrosion behaviour. The possible factors that can affect the corrosion behaviour of pipeline steel include temperature, pressure, SO₂ concentration, O₂ concentration, relative humidity and flowrate. Among these factors, temperature is a very important factor since it greatly affects the rate of physical and chemical process as well as the growth of corrosion product scale, and it has been regarded as one of the most important parameters in corrosion rate determination.9 ^– 16 Therefore, so far, no detailed study has been published regarding the temperature effect on the corrosion behaviour of pipeline steel under supercritical CO₂ mixtures with SO₂, O₂ and water vapour as impurities. To carry out an in depth and systematic research is of necessity.

Researchers have made a thorough study and analysis on the temperature effect on carbon steel corrosion in aqueous solution containing CO₂ and other impurities.17 ^– 20 Although in the present study, CO₂ is the solvent instead of solute and the mixtures are under supercritical or liquid condition instead of aqueous condition, the research approaches should still be applicable, such as the weight loss method and the scale analysis methods. Taking into account the practical CO₂ pipeline transport temperature range, the temperature range of the present work was set to be under 100°C. Under such temperature range and environmental gaseous mixtures, there should have some similarities with carbon steel corrosion in the humid atmosphere containing SO₂.

In this work, an environmental scanning electron microscope was utilised to observe the surface and cross-section morphology of the product scales, and it was also applied to observe morphology of the specimen after the removal of corrosion product scales to check the occurrence of pitting. Focused ion beam (FIB) technology was employed to study the inner structure of the product scales. The chemical composition of the product scales was analysed by X-ray diffraction (XRD) and energy dispersive X-ray spectroscopy (EDAX). A mercury intrusion porosimeter (MIP) was used to measure the porosity of product scale, and gas chromatography (GC) was employed to detect the exhaust gas content to understand the corrosion mechanism.

Experimental

The composition of X70 steel used in this study was Fe–0·054C–1·53Mn–0·266Si–0·011P–0·05Ni–0·0052S–0·013Cr–0·014V–0·010Cu–0·019Al_s–0·021Al_t–0·0015Ca–0·209Mo–0·039Nb–0·015Ti–0·017Sn–0·0085N–0·012As. It is noted that Mn has the largest amount among the trace elements. The sample used for weight loss method had a size of ∼40×20×2·7 mm. The sample used for measuring the porosity of the product scale had a size of ∼20×15×2·7 mm. The sample size for the morphology observation and composition analysis was ∼10×10×1 mm. The samples were polished using silicon carbide papers up to 1200 grit, and then washed with acetone and anhydrous ethanol. Before the samples were weighed by an electronic balance, they were placed in vacuum desiccators to remove water.5 In this study, a CORTEST autoclave provided the high pressure reaction environment, which had a volume of 2·2 L. The test rig schematic diagram is shown in Fig. 1.

Figure 1

Test rig schematic diagram

The corroded samples were exposed to the Clarke solution to remove the corrosion products.21 The corrosion rate was determined by weight loss method.22 The traditional electrochemical methods are not suitable for this environment.15

The experimental conditions are shown in Table 1. A certain amount of water will be injected into the autoclave by pipette gun to generate a water vapour saturated CO₂ state. The solubility of water in the high pressure CO₂ was referred to open literatures.23 ^– 26 The oxygen in the autoclave was coming from the initial air in the autoclave. Using the autoclave volume data, the oxygen quantity can be calculated accurately, as shown in Table 1.

Table 1

Test conditions

Test number	Temperature/K	Pressure/MPa	Time/h	H₂O/g	SO₂/mol.-%	Speed of rotation/rev min⁻¹	Reynolds number	O₂/mol
1	298	10·0	120	5·0	2·0	120	5·32×10⁵	0·02
2	323	10·0	120	5·0	2·0	120	6·51×10⁵	0·02
3	348	10·0	120	5·0	2·0	120	5·06×10⁵	0·02
4	366	10·0	120	5·0	2·0	120	4·38×10⁵	0·02

The rotation speed was chosen to be 120 rev min⁻¹ to make a flowing environment, and the rotator was driven by a motor outside the autoclave through electromagnetic force. The rotation speed can be controlled and set through the autoclave control panel. Using the equation proposed by Wei and Ji,27 the average flow velocity of the CO₂ mixtures in the autoclave can be approximately calculated as (1) where is the average flow velocity of the CO₂ mixtures (m s⁻¹), RPM is the rotation speed (rev min⁻¹), d _J is the paddle diameter (m) and d _h is the inner diameter of the autoclave (m). In the present study, the paddle diameter was 0·058 m and the inner diameter of the autoclave was 0·127 m. Based on the calculated average flow velocity, the Reynolds numbers can be calculated, which are also shown in Table 1. The CO₂ density and viscosity were referred to NIST REFPROP software.

In addition, the instruments and their applications in the present study are listed in Table 2.

Table 2

Instrument names and models

Number	Instrument	Model	Application
1	Environmental scanning electron microscopeEDAX	FEI Quanta 200	Morphology observation and elemental analysis
2	Focused ion beam (FIB)	LYRA3 TESCAN	Inner structure observation
3	X-ray diffraction (XRD)	D8 ADVANCE	Phase analysis
4	Mercury intrusion porosimeter (MIP)	Autopore IV 9510	Measurement of porosity
5	Gas chromatography (GC)	AutoSystem XL	Exhaust gas detection

The porosity of corrosion product scale with uncorroded substrate was measured by MIP. Using the volume data before and after the entrance of mercury into the pores, the porosity can be determined. The real porosity of the product scale can be calculated as follows.

The volume of the sample before corrosion is V ₁ (m³), the surface area of the sample before corrosion is A ₁ (m²) and the immersion time is t (h). The total volume of corroded sample including the product scale is V _T (m³), the measured porosity is ϵ _F and the corrosion rate is CR (mm/year) calculated by the weight loss method. When the corrosion product scale was removed, the sample volume V ₂ (m³) will be (2) The scale volume V _P (m³) will be (3) The real porosity of scale ϵ _R should be (4) The data from porosity measurement can also be used to calculate the packing density and true density of corrosion product scale.

Results and discussion

Corrosion rate

The result of corrosion rate using weight loss method is shown in Fig. 2. As seen, when the temperature increased, the corrosion rate first increased and then decreased with the maximum corrosion rate at ∼348 K. As described above, the supercritical CO₂ is mixed with saturated water vapour. Under this environment, there should be a thin water film on the sample surface possibly due to adsorption condensation and chemical coagulation effect.7,8 The corrosion of steel occurring in a thin layer of electrolyte is quite different from that occurring in a bulk solution. The change trend of corrosion rate with temperature can be explained as follows.

Figure 2

Corrosion rate of X70 steel in supercritical CO₂ mixtures changes with temperature

As temperature increases, the mass transfer rates of the reactants diffusing across the water film increase. At the same time, the chemical reaction rates will also increase. In addition, the corrosion product substrate that formed at high temperature generally has different protection effect for further corrosion.20,28 All these effects may result in such increase first and then decrease profile as shown in Fig. 2. In the following sections, more details of the physical–chemical processes involved in the steel corrosion process in supercritical CO₂ mixtures were explained and analysed.

Product scale analysis

Morphology analysis

Figure 3 shows the surface morphology of X70 steel specimen after corrosion under different temperatures. The morphology for different temperatures was quite different. When the temperature was at 298 K, there were massive block shaped products formed, which were covered with some irregular thin product layer, and scale had tiny cracks. For a temperature of 323 K, there was two types of product scale: the dense inner scale and the loose outer scale, and the outer scale was easy to fall off. When the temperature was at 348 K, the stripped shaped products appeared. For a temperature of 366 K, the sphere shaped corrosion products accumulated on the surface together with some small irregular corrosion products on their top, and the scale seemed to be more dense and compact.

Figure 3

a 298 K; b 323 K; c 348 K; d 366 K

Figure 4 presents the cross-section morphology of corroded sample at different temperatures. Epoxy sealed corroded samples were polished to form the cross-section. The cross-section can be divided into three layers: the top layer in deep colour is the epoxy, the middle layer is the corrosion product scale with many pores or cracks and the bottom layer is the steel base. It was seen that when the temperatures were at 323 and 348 K, the product scales were relatively thicker. It was also found that the scales that formed under such two temperatures were easier to fall off and the epoxy samples could not be well polished to be a plane. Especially for scale formed at 348 K, the underneath layer had a poor connection to the substrate metal. The sphere shaped product that formed at 366 K had dense cross-section morphology, and it was closely connected to the base metal, which may explain the slowdown of the corrosion rate at 366 K as temperature increases.

Figure 4

a 298 K; b 323 K; c 348 K; d 366 K

To further verify the existence of pitting, the surface of the corroded sample without product scales was checked. Figure 5 shows the surface morphology of corroded samples after the removal of product scales. It was seen that as the temperature was low (298 and 323 K), there was no clear pitting shown on the surface, which was relatively flat. As the temperatures increased (348 and 366 K), the surface was not flat yet. However, there was no sufficient evidence to show the occurrence of pitting corrosion. The harmfulness of pitting is far greater than the uniform corrosion. Therefore, we should try to avoid pitting phenomenon when CO₂ is transported by pipeline in order to ensure the safety of pipeline transportation. Longer corrosion cycle experiments will be carried out to check the occurrence of pitting in the future work.

Figure 5

a 298 K; b 323 K; c 348 K; d 366 K

Figure 6 is a morphology diagram using the FIB technology to cut the sphere shaped product formed at 366 K. The spherical corrosion product was ∼50 μm in diameter. From the figure after cutting, it can be seen that the inner part of the sphere was solid, only with some micropores, but these micropores were not connected. This means that these micropores are not effective pores, which cannot provide the necessary channels for reactants and products in and out. No obvious cracks exist within the sphere. This indicates that the cracks on cross-section morphology in Fig. 4 are generated after processing.

Figure 6

a before cutting; b after cutting

Chemical composition analysis

Figure 7 shows the XRD spectra of corroded samples with product scale. It can be seen that when the temperature was at 298 K, the corrosion product was mainly composed of FeSO₃.3H₂O and FeSO₄.7H₂O. For a temperature of 323 K, the main component of the corrosion product was FeSO₄.4H₂O. When the temperature was at 348 K, the corrosion product was mainly composed of FeSO₄.H₂O, with a small amount of FeSO₄.4H₂O. For a temperature of 366 K, only FeSO₄.H₂O was detected in the corrosion product. The results suggest that when the temperature is higher, the number of crystal water of the ferrous sulphate will be reduced, as well as the possibility to form ferrous sulphite. The ferrous sulphite may be decomposed or oxidised to ferrous sulphate by oxygen at higher temperature where the sulphite oxidation process will be faster. In addition, there was no FeCO₃ detected, showing that SO₂ plays a critical role in the electrochemical reaction process.5

Figure 7

a 298 K; b 323 K; c 348 K; d 366 K

Figure 8 is an EDAX line scanning result of sample cross-section with scale. The scanning curve represents the element of C, O, Si, S, Mn, and Fe from top to bottom, respectively. From the EDAX line, the boundary of product scale and the substrate metal can be clearly distinguished. At the two sides of this boundary, the elemental content of Fe, S and O had a mutation. From the direction of the base metal to the scale, the elemental content of Mn also had a small degree of reduction, which means that Mn was consumed in the corrosion reaction, and MnSO₄ might be generated.29 In addition, it can be found that the elemental content of C was higher at the higher temperature in the product scale. Because product scales that formed at temperatures of 323 and 348 K were easy to fall off when preparing for the cross-section sample and it was difficult to find a flat surface for EDAX scanning, the EDAX line scanning was not carried out at these two temperatures for the cross-section samples.

Figure 8

a 298 K; b 366 K

Physical properties measurement

X70 steel samples in the autoclave were hanging vertically. When finishing the experiment, it was found that the thickness of the sample at the lower part was thicker than the higher part, as shown in Fig. 9. To reduce measurement error, the average values of multiple measurements by vernier calliper were taken. Results indicate that the corrosion product on the sample surface has mobility and is easy to gather at the bottom of the sample. The main reasons for this phenomenon should be as follows: the major corrosion product ferrous sulphate has a strong hygroscopicity, and its solubility in water is greater,30 so the ferrous sulphate dissolved in water is easily flowing from the top to the bottom with the water film because of gravity, which appears that the thickness of the sample at the three parts will have a different thickness after corrosion. If the corrosion products were mainly ferrous sulphite, this phenomenon may not appear, because ferrous sulphite has a low solubility in water, though the solubility will increase when SO₂ is contained in the solution.31

Figure 9

Thickness of corroded sample with scale at different parts

Figure 10 shows the average porosity and total attached volume of product scale. The result shows that the higher the temperature is, the lower porosity of the corrosion products would have. The trend of attached total volume of scale with the temperature was similar with the trend of corrosion rate with the temperature. In addition, the packing density and true density of scale can be calculated by the experimental data for the porosity, and the results are shown in Table 3. The true density of FeSO₄.7H₂O given by literature30 is 1895 kg m⁻³, which shows that the present results have a high credibility.

Figure 10

Average porosity and total attached volume of product scale

Table 3

Packing and true density of scales

Test number	Temperature/K	Mass of attached scale/g	Packing density/kg m⁻³	True density/kg m⁻³
1	298	0·15724	1568·63	2095·88
2	323	0·26523	1481·50	1786·98
3	348	0·34623	1761·04	1951·19
4	366	0·19669	2018·25	2235·92

Exhaust gas detection

In order to understand the corrosion reaction mechanism, there is a necessity to examine whether the hydrogen is contained in the exhaust gas. For this purpose, gas bag was used to collect the exhaust gas at the end of experiment for GC detection. For the first time, 1 L of exhaust gas was collected directly for GC detection, and the results showed no significant peak of hydrogen. There are two possible reasons: the exhaust gas did not contain hydrogen, and the exhaust gas contained a very small amount of hydrogen and it was under the GC detection limit. In order to prove whether it is the latter reason, the exhaust gas was cleaned by NaOH solution to absorb large amount of acid in the exhaust gas before collection in order to increase the hydrogen concentration. By this approach, the significant peak of hydrogen was detected by GC, which means that the hydrogen was generated by the corrosion reaction. When hydrogen exists, hydrogen embrittlement may occur.32,33 This issue needs further research to confirm.

Additional discussion

From the above experimental results, it can be seen that the temperature has a great impact on the corrosion rate, scale morphology, scale composition and scale physical properties. From the results of scale porosity, it can be found that the scale porosity at 348 and 366 K has little difference. However, the corrosion rates under these two temperatures have big difference. The reason for this might be that out of the effect of the dense scale on the corrosion rate, the impact of temperature on the solubility of SO₂ and O₂ in the sample surface solution film is also an important reason, because the temperature will reduce the solubility of SO₂ and O₂ in the water film. These two factors together have led to great reduction to the corrosion rate when increasing the temperature.

It should also be realised that because of the high pressure characteristics of this experiment, many kinds of corrosion measurement methods cannot be employed in the experiment for in situ measurements. Because FeSO₄ has a great solubility in water, when the sample was picked out for testing, the water on the sample surface would evaporate, resulting in precipitation of dissolved FeSO₄, which will affect the morphology and porosity and other physical properties of the corrosion product scale in the corrosive environment. The use of electrochemical techniques in this environment has some difficulty, because the water film on the sample surface might not be continuous and the conductivity may also be poor.15 Therefore, scholars generally use weight loss method to measure the corrosion rate of carbon steel in this kind of environment.3 ^– 8

The corrosion mechanism of X70 steel in this kind of environment has been discussed in detail by Choi et al.3 and Xiang et al.5 The anodic reaction should be (5) The cathodic reaction should contain34 (6) (7) (8) The contribution of this study is confirming the existence of hydrogen evolution reaction. This is the greatest difference of the corrosion mechanism of carbon steel corrosion in this environment and in the humid atmosphere containing SO₂.

It was noted that the X70 steel sample had a 1·55% weight fraction of Mn element. Mn²⁺ will be generated by corrosion, which is a kind of catalyst for oxidisation of sulphite ion into sulphate ion.35 The mechanism of sulphite catalytic oxidation by Mn²⁺ ions can be explained by Backstrom chain reaction theory.36 The existence of Mn²⁺ ions makes ferrous sulphite change into ferrous sulphate in a shorter period of time, and the reaction rate constant will be larger at the higher temperature, so oxidation rate of sulphite will be higher, which is the reason why ferrous sulphite was detected at low temperature and it was not detected at high temperature. The existence of sulphate will induce the pitting corrosion.37 Sulphate increases the critical concentration of metal salt in the pit, expressed as a fraction of the saturation concentration, which is required to sustain pit dissolution.37 However, this effect was not sufficiently reflected, which might be due to the short corrosion cycle in this study.

Conclusions

In this paper, weight loss method was used to measure the corrosion rate of X70 steel in high pressure CO₂/SO₂/H₂O/O₂ system for different temperatures, and a series of analytical instruments were employed to analyse morphology, composition and porosity of corrosion product scale. Through this study, the following conclusions can be concluded.

As the temperature increased, the corrosion rate increased until it reaches a peak value, and then declined. This is due to not only the high protectiveness of corrosion product scale at high temperature but also the reduction of the solubility of SO₂ and O₂ in the thin water film when increasing the temperature.

When the temperature reached 366 K, there was an accumulation of dense spherical corrosion products, which had a strong protection to the substrate metal. The product scales formed at temperatures of 323 and 348 K have a poor protection to the base metal.

The higher the temperature is, the less the number of water in crystallisation of ferrous sulphate will have, and the smaller the porosity of corrosion product scale will be.

The existence of hydrogen in the exhaust gas proves the existence of the hydrogen evolution reaction process.

Footnotes

Acknowledgements

This work received financial support from the National Key Technologies R&D Program of China (grant no. 2007BAC03A03).

References

Forbes

Verma

Curry

Friedmann

Wade

: ‘Guidelines for CO₂ capture, transport and storage’; 2008

Washington, DC

World Resources Institute.

Dooley

Dahowski

Davidson

: ‘Comparing existing pipeline networks with the potential scale of future U.S. CO₂ pipeline networks’, Energy Proc. 2009 1 (1) 1595–1602.

Choi

Nesic

Young

: ‘Effect of impurities on the corrosion behavior of CO₂ transmission pipeline steel in supercritical CO₂–water environments’, Environ. Sci. Technol. 2010 44 (23) 9233–9238.

Dugstad

Morland

Clausen

: ‘Corrosion of transport pipelines for CO₂ – effect of water ingress’, Energy Proc. 2011 4 3063–3070.

Xiang

Wang

Zhou

: ‘Impact of SO₂ concentration on the corrosion rate of X70 steel and iron in water-saturated supercritical CO₂ mixed with SO₂’, J. Supercrit. Fluids 2011 58 (2) 286–294.

Xiang

Wang

Yang

: ‘Corrosion behavior of X70 steel in the supercritical CO₂ mixed with SO₂ and saturated water’ Proc. 21st Int. Offshore and Polar Engineering Conf., Maui, HI, USA, June 2011, ISOPE,350–354.

Xiang

Wang

: ‘The impact of corrosion product scales on the corrosion rates of X70 steel in supercritical CO₂ with SO₂ and H₂O impurities’ Proc. World Cong. on ‘Engineering and technology (CET)’ Shanghai, China October 2011

IEEE

498–504.

Xiang

Wang

Yang

: ‘The upper limit of moisture content for supercritical CO₂ pipeline transport’, J. Supercrit. Fluids 2012 67 14–21.

Nesic

Lee

KLJ

: ‘A mechanistic model for carbon dioxide corrosion of mild steel in the presence of protective iron carbonate films – Part 3: film growth model’, Corrosion 2003 59 (7) 616–628.

10.

Nesic

Nordsveen

Nyborg

Stangeland

: ‘A mechanistic model for carbon dioxide corrosion of mild steel in the presence of protective iron carbonate films – Part 2: a numerical experiment’, Corrosion 2003 59 (6) 489–497.

11.

Nesic

Postlethwaite

Olsen

: ‘An electrochemical model for prediction of corrosion of mild steel in aqueous carbon dioxide solutions’, Corrosion 1996 52 (4) 280–294.

12.

Nesic

Wang

Cai

Xiao

: ‘Integrated CO₂ corrosion – multiphase flow model’ Proc. SPE Int. Symp. on ‘Oilfield corrosion’, Aberdeen, UK, May 2004, Society of Petroleum Engineers, Paper 87555.

13.

Nordsveen

Nesic

Nyborg

Stangeland

: ‘A mechanistic model for carbon dioxide corrosion of mild steel in the presence of protective iron carbonate films – Part 1: theory and verification’, Corrosion 2003 59 (5) 443–456.

14.

Song

: ‘A comprehensive model for predicting CO₂ corrosion rate in oil and gas production and transportation systems’, Electrochim. Acta 2010 55 (3) 689–700.

15.

Vitse

Nesic

Gunaltun

de Torreben

Duchet-Suchaux

: ‘Mechanistic model for the prediction of top-of-the-line corrosion risk’, Corrosion 2003 59 (12) 1075–1084.

16.

de Waard

Milliams

: ‘Carbonic acid corrosion of steel’, Corrosion 1975 31 (5) 177–182.

17.

Yin

Feng

Zhao

Bai

Lin

: ‘Effect of temperature on CO₂ corrosion of carbon steel’, Surf. Interface Anal. 2009 41 (6) 517–523.

18.

Kladkaew

Idem

Tontiwachwuthikul

Saiwan

: ‘Corrosion behavior of carbon steel in the monoethanolamine–H₂O–CO₂–O₂–SO₂ system’, Ind. Eng. Chem. Res. 2009 48 (19) 8913–8919.

19.

Nainar

Veawab

: ‘Corrosion in CO₂ capture process using blended monoethanolamine and piperazine’, Ind. Eng. Chem. Res. 2009 48 (20) 9299–9306.

20.

Cui

Zhu

Yang

: ‘Study on corrosion properties of pipelines in simulated produced water saturated with supercritical CO₂’, Appl. Surf. Sci. 2006 252 (6) 2368–2374.

21.

‘Standard practice for preparing, cleaning, and evaluating corrosion test specimens’

G1-03

ASTM

West Conshohocken, PA, USA

2003.

22.

‘Standard practice for laboratory immersion corrosion testing of metals’

G31

ASTM

West Conshohocken, PA, USA

1994.

23.

Spycher

Pruess

Ennis-King

: ‘CO₂–H₂O mixtures in the geological sequestration of CO₂. I. Assessment and calculation of mutual solubilities from 12 to 100 degrees C and up to 600 bar’, Geochim. Cosmochim. Acta 2003 67 (16) 3015–3031.

24.

Austegard

Solbraa

de Koeijer

Molnvik

: ‘Thermodynamic models for calculating mutual solubilities in H₂O–CO₂–CH₄ mixtures’, Chem. Eng. Res. Des. 2006 84 (A9) 781–794.

25.

Munkejord

Jakobsen

Austegard

Molnvik

: ‘Thermo- and fluid-dynamical modeling of two-phase multi-component carbon dioxide mixtures’, Greenhouse Gas Control Technol. 2009 1 (1) 1649–1656.

26.

Munkejord

Jakobsen

Austegard

Molnvik

: ‘Thermo- and fluid-dynamical modelling of two-phase multi-component carbon dioxide mixtures’, Int. J. Greenhouse Gas Control 2010 4 (4) 589–596.

27.

Wei

: ‘Velocity profile and gas bubble size distribution in an agitated vessel – a cold model study on the particle size and its distribution in gas–liquid dispersion polymerization processes’, Chem. React. Eng. Technol. 1986 2 (2) 14–24.

28.

Nesic

: ‘Key issues related to modelling of internal corrosion of oil and gas pipelines – a review’, Corros. Sci. 2007 49 (12) 4308–4338.

29.

Agudelo

Marco

Gancedo Jr Perez-Alcazar

: ‘Fe–Mn–Al–C alloys: a study of their corrosion behaviour in SO₂ environments’, Hyperfine Interact. 2002 139 (1-4) 141–152.

30.

Lide Dr : ‘CRC handbook of chemistry and physics: a ready-reference book of chemical and physical data’; 2008

Boca Raton, FL

CRC Press.

31.

Kertes

: ‘Solubility data series’ in ‘Sulfites, selenites and tellurites’, (ed. Masson

M R

, ed); 1987

Oxford

Pergamon

252–255.

32.

Huang

Zheng

: ‘Hydrogen permeation and corrosion behaviour of high strength steel 35CrMo under cyclic wet–dry conditions’, Corros. Eng. Sci. Technol. 2008 43 (3) 241–247.

33.

Zheng

Huang

Huo

: ‘Effect of H₂S on stress corrosion cracking and hydrogen permeation behaviour of X56 grade steel in atmospheric environment’, Corros. Eng. Sci. Technol. 2009 44 (2) 96–100.

34.

Allam

Arlow

Saricimen

: ‘Initial stages of atmospheric corrosion of steel in the Arabian Gulf’, Corros. Sci. 1991 32 (4) 417–432.

35.

Zhao

Zhuo

Tong

Zhang

Chen

: ‘Mass transfer and kinetics study on the sulfite forced oxidation with manganese ion catalyst’, Korean J. Chem. Eng. 2007 24 (3) 471–476.

36.

Bäckström

HLJ

: ‘The chain-reaction theory of negative ctalysis’, J. Am. Chem. Soc. 1927 49 (6) 1460–1472.

37.

Moayed

Newman

: ‘Deterioration in critical pitting temperature of 904L stainless steel by addition of sulfate ions’, Corros. Sci. 2006 48 3513–3530.