Condensed from paper No. 13, presented at NACE Corrosion/96
Reproduced with permission from NACE International.

The effect of microstructure and Cr content in steels on CO2 corrosion was investigated by using steels containing Cr content from 0 to 13 mass% melted in laboratory and Steels J55, N80 and L80(API Grade) melted in the mill. Temperatures and H2S contamination were considered as environmental factor. In CO2 environments, the temperature giving a maximum corrosion rate, Tmax. existed in carbon and Cr steels. Tmax. increased together with Cr content, and Tmax. of 0, 1, 2 and 13% Cr steels was about 80, 100, 120 and 225°C, respectively. Because of this behavior, the relationship between Cr content and corrosion rate was linear at 60°C, but the corrosion rate was highest on the steel with around 1mass% Cr at 100°C. H2S contamination for CO2 corrosion suppressed the corrosion rate and localized-corrosion in the temperature region whose corrosion rate showed a maximum value. It was clarified that this was related to the formation of Fe-sulfides from EPMA analysis and the solubility of the corrosion products. Concerning microstructure, Steel J55 with ferritic-pearlitic microstructure showed good corrosion resistance for localized-corrosion compared with Steel N80 and L80 with martensitic microstructure.

The CO2 corrosion of carbon and low alloy steels called "Sweet corrosion" has been one of the important problems in oil and gas industry since 1940 because of both a high corrosion rate and a severe localized corrosion1,2,3. This CO2 corrosion affects the materials used in production, gathering transportation and processing facilities. Low cost carbon steels are susceptible to corrosion in CO2 environments. The severity of corrosion depends particularly on temperature, CO2 partial pressure, pH and material characteristics4.

The temperature where the susceptibility to the severe corrosion is highest is around 100°C5. This corrosion behavior is related to the formation behavior of FeCO3 which is a corrosion product in CO2 environments, and is classified to three types of corrosion below 60°C, at about 100°C and over 150°C. The first is a general corrosion type, the second is a deep pitting and a ringworm corrosion type and the third is a corrosion resistant type through the formation of protective FeCO3 film. Namely, the corrosion can be understood from the FeCO3 formation behavior that the higher the temperature, the smaller the solubility of FeCO36.

For the prevention of CO2 corrosion, various Cr steels are applicable according to the environmental temperature. The result of the loop tests in a CO2 environment at 60°C obtained by Ikeda et al showed the Cr dependence on corrosion rate: that the higher the Cr content in steel, the lower the corrosion rate5. For lower Cr bearing steel, the enrichment of Cr content arises in the corrosion product. The Cr-enriched film which is produced in the environment increases the corrosion resistance.

The effect of corrosion products is also important for CO2 corrosion. It is suggested that Fe3O4 and FeS are stable corrosion products in the CO2 environment with low CO2 partial pressure and in that containing a little amount of H2S, respectively6,11. Further, Vera et al discussed the influence of flow velocity and galvanic coupling on the morphology of corrosion attack and corrosion rate in the field from a viewpoint of the formations of FeCO3, Fe3O4 and Fe3C. Namely, the tight thin FeCO3 film with a relatively high concentration of Fe3C and/or Fe3O4 was found in carbon and low Cr steels under high flow velocity and galvanic couple conditions12,13.

Therefore, in this paper, the effect of the microstructure on CO2 corrosion is discussed considering environmental temperatures, Cr content in steels, the morphology of corrosion attack and corrosion products. Because there is also CO2 environments containing a small amount of H2S in actual field conditions, the CO2 corrosion in that environment is also investigated.

Materials

Materials with the chemical compositions given in Table 1 were used in this study. Both materials melted in laboratory and in mill are involved. The materials melted in mill were manufactured by the Mannesmann-mandrel mill process, and Steel J55(min. yield strength: 55ksi[379MPa]) had the as-rolled, ferritic-pearlitic microstructure and Steels N80 and L80(min. yield strength: 80ksi [550MPa]) had the quenched and tempered, martensitic microstructure. On the other hand, the materials melted in laboratory were hot rolled and heat treated after melted, and those had the normalized, ferritic microstructure.

Corrosion Test

Immersion tests were carried out by using the autoclave with a stirrer. The autoclave vessel was deaerated by using a vacuum pump and purging nitrogen. Then, deaerated-solution was poured into the vessel, and more deaeration was carried out by the operation of N2 gas bubbling and a vacuum. H2S and CO2 gases were charged to test pressure at 25°C. The temperature was raised to the testing condition. During the bubbling, the gas charging and the testing, the stirrer was used and the flow velocity at specimen surfaces was about 1 to 2.5m/s(300 to 500rpm).

In order to analyze corrosion products, X-ray diffraction, scanning electron microscope(SEM) and electron probe microanalysis(EPMA) were used in this study.

Effect of Cr Content and Temperature on CO2 Corrosion

In CO2 environments.. The temperature(Tmax.) which gives a maximum corrosion rate increased with increasing Cr content in the steels. Namely, Tmax. of 0, 1, 2 and 13%Cr steels was about 80, 100, 120 and 225°C, respectively. It is clarified by A.Ikeda et al that this behavior relates to the formations of FeCO3 and amorphous Cr-enriched oxide which is a corrosion product in CO2 environments5. The corrosion rate of 0, 1 and 2%Cr steels showed a tendency to becomes small in order of 0, 1 and 2%Cr steels at below Tmax., but in reverse order at over Tmax.. Accordingly, the relationship between Cr content and corrosion rate was linear at 60°C and the corrosion rate monotonously decreased with increasing Cr content, but was not at 100°C and the corrosion rate showed maximum on about 0.5%Cr steel(L80). The result includes data of Steels J55 and L80 melted in mill.

In CO2 + H2S environments. The effect of temperature on CO2 corrosion of Cr bearing steels produced in laboratory was investigated in the CO2 environment containing a small amount of H2S. The corrosion in the region which gave a maximum corrosion rate in CO2 environments was suppressed by the contamination of H2S. This suppression effect was large for 0%Cr steel. The corrosion rate of the steels with Cr content below 1mass% did not increase with the lowering of Cr content. The S distribution in a cross section of the surface film produced on Steel L80 at 60°C. The concentrated-S was observed on steel surface and C in the outer of that mainly existed. These compounds were identified as FeS and FeCO3, respectively. It is thought that the formation of FeS relates to the decreasing of corrosion rate.

Effect of Microstructure and Corrosion Morphology on CO2 Corrosion

In CO2 environments. The corrosion morphology of Steels J55(0.04mass%Cr, ferrite-pearlite), N80(0.05mass%Cr, martensite) and L80(0.48mass%Cr, martensite) was observed at 60 and 100°C in CO2 environments.

At 60°C, Steels N80 and L80 with martensitic microstructure suffered localized-corrosion, but Steel J55 with ferritic-pearlitic microstructure did not. This behavior is corresponding to the fact that many defects were observed in the corrosion film formed on Steels N80 and L80, but those were not seen on Steel J55 . Steel J55 containing 0.04mass% Cr gave almost the same corrosion rate as Steel N80 containing 0.05mass% Cr, but Steel L80 containing 0.48mass%Cr gave smaller one than Steels J55 and N80. Namely, from these test results, it is thought that the localized-corrosion depends on microstructure and the corrosion rate depends on Cr content in steels. However, the corrosion rate of Steel J55 in long term test might be smaller than that of Steel N80. Steel J55(ferritic-pearlitic microstructure) did not have defects in corrosion film. which was the case for Steel N80(martensitic microstructure).

At 100°C, the tendency of the localized-corrosion in Steels N80 and L80 with martensitic microstructure is smaller than that at 60°C. The localized-corrosion and surface defects on the corrosion film were not observed in Steel J55 with ferritic-pearlitic microstructure. Therefore, Steel J55 has a good corrosion resistance for localized-corrosion. The corrosion rate of the steels at 100°C was small compared with that at 60°C, and the corrosion rate of Steel L80 with 0.48mass%Cr is slightly larger than that of Steels J55 and N80 with about 0.05mass%. This behavior can be understood from the result of the following temperature dependency on CO2 corrosion.

1. the corrosion occurs in the temperature region where the corrosion rate decreases.

2. the temperature giving a maximum corrosion rate is high in high Cr containing steels.

In CO2 + H2S environments. The corrosion morphology of Steels J55, N80 and L80 was also observed at 60 and 100°C in the CO2 environment containing a little amount of H2S. Corrosion rate in the CO2 + H2S environment was smaller than that in the CO2 environment. Then, localized-corrosion was not observed in the CO2 + H2S environment. But a few defects in Steels N80 and L80 were found in the surface corrosion film. These corrosion behaviors would relate to corrosion protection due to Fe sulfide formation in corrosion products as mentioned above.

CO2 corrosion of carbon and Cr-bearing steels was discussed considering testing temperatures, H2S contaminations, Cr content in steels and microstructure of steels. The results obtained are summarized as follows:

For corrosion rate,

1.The temperature giving a maximum corrosion rate, Tmax. increased together with Cr content in steels. Tmax. of 0, 1, 2 and 13%Cr steels was about 80, 100, 120 and 225°C, respectively. Based on this behavior, the relationship between Cr content and corrosion rate was linear at 60°C, but the corrosion rate at 100°C was highest on the steel with around 1mass% Cr.

2.The contamination of H2S suppressed CO2 corrosion in the temperature region where the corrosion rate showed a maximum value. From EPMA analysis and solubility calculation of corrosion products, it was clarified that this was relating to the formation of Fe-sulfides.

For localized-corrosion,

1.Steel J55 with ferritic-pearlitic microstructure showed good corrosion resistance for localized-corrosion compared with Steel N80 and L80 with martensitic microstructure in CO2 environments.

2.H2S also suppressed localized-corrosion in the temperature region giving a maximum corrosion rate.

The authors wish to thank Sumitomo Metal Industries Ltd. for allowing publication of this research. The assistance and discussion of co-workers in laboratories are gratefully acknowledged.