Tests spot inconsistencies predicting corrosion-inhibitor effectiveness

Fig. 1 shows the results in terms of the corrosion rate in mm/year and the efficiency of the product, measured against the condition without any inhibitor added, as a function of concentration for bare steel conditions.

The test without inhibitor established an average corrosion rate of about 1.81 plus or minus 0.21 mm/year. With 50-ppm inhibitor, the corrosion rate decreased to 0.42 plus or minus 0.02 mm/year, establishing an efficiency of 76%, which was the best obtained during the weight-loss measurements. As more inhibitor is added, the corrosion rates tend to increase, especially at 100 ppm.

This behavior is generally exhibited because of a competing effect between the growing corrosion scale and the inhibitor monolayer trying to become attached to prevent scale formation.

This condition has been interpreted as a detergency effect where the inhibitor cleans the surface for the ongoing corrosion processes that continue at the metal-electrolyte interface.¹²

LPR measurements

The efficiency of the inhibitor was retested with a linear polarization resistance (LPR) technique, but in this case, the test used a precorroded steel surface before the inhibitor was injected into the autoclave (plus or minus 5 mv vs. E_corr, 0.1 mv/sec).

This procedure, using the same solution, allows for calculating an instantaneous corrosion rate from the linear behavior observed in a typical overvoltage vs. current density plot.¹⁰ The slope of the plot is denoted as polarization resistance (R_p), which can be traced over time and is related to corrosion rate.¹¹

In this study, the starting surface condition is bare and corrosion products are intentionally grown and stabilized for a period before the inhibitor is injected into the cell. The test also involved an autoclave vessel but with the appropriate ports for LPR measurements.

The test conditions were the same as in the weight-loss tests.

Fig. 2 summarizes the corrosion measurements obtained through the LPR technique as a function of time and inhibitor concentration. It can be readily seen that in the absence of the inhibitor, the corrosion rate increases, establishing a corrosion rate of about 35 mm/year at 6,000 min.

During this time, a corrosion film forms. Once the inhibitor is added (50 ppm), the corrosion rate starts decreasing to 31 mm/year. Further inhibitor additions decrease the corrosion rate to 25 mm/year at 100 ppm, and to 23 mm/year for both 150 and 200 ppm. The best efficiency of 65% is obtained at 200 ppm.

Fig. 3 compares the results from both techniques. The measurements clearly show a difference in corrosion rates. The LPR technique resulted in measuring up to 14 times greater corrosion rates than the weight-loss techniques.

Test results

Weight-loss measurements represent an average rate for a specific exposure period and contain very little kinetic information and probably none concerning long-term exposure or steady-state rates because corrosion is not constant with time (Fig. 2).

This technique is limited because of the relatively small solution volume in the autoclave, which defines a limit for iron concentration, bicarbonate and sulfide concentration, and pH.

This causes the steel to become passive (Fig. 3) from the formation of a corrosion layer that has a composition and protectiveness dependent on the mode of formation. The formation is affected by agitation, rate of temperature increase, pressure fluctuations, C0₂ and H₂S equilibrium, etc.¹³

This problem could be solved by use of both a larger volume to exposed surface area ratio and a shorter testing time. Some information, however, will be sacrificed in terms of the average corrosion rate obtained and its actual dependence with time.

In terms of the LPR measurements, effective Tafel slopes cannot be measured once oxides have formed. As shown in Fig. 2, the corrosion rate varies with time and hence Rp. It has been pointed out that the assumption of a stable Rp with time is incorrect because this parameter will increase with time rather than stabilize or decrease.¹³

The Tafel slope extrapolation is masked because inhibitor adsorption is controlled by the electric potential. Likewise, in the presence of a corrosion layer, it is understood that all processes occurring across the solid interface are rate controlled by a combination of Faradaic and non-Faradaic processes as well as the changing environment, both liquid and solid.

The measurement of the Tafel approximation may or may not reflect the actual condition because the inhibitor interferes with the growth or dissolution of the layer but does not affect the Faradaic processes.

For the precorrosion conditions (Fig. 2) before the inhibitor injection, LPR measurements in many instances do not provide protection at the surface by forming a compact iron carbonate scale. This is probably due to the oxide dissolution that may create anodic zones where pitting takes place, accelerating the corrosive process.¹⁴

This point is important when one evaluates corrosion inhibitors because many internal corrosion failures in the oil and gas industry are a consequence of pitting.

The dissolution of iron carbonates depends on the temperature at which the steric interaction occurs between the product and the surface to be protected. In this study, the temperature is well below the limit for forming adherent protective layers,1 a fact that reaffirms this observation and explains the relatively low efficiencies obtained.

As discussed, limitations for corrosion inhibitor evaluation include:

• Drawback of evaluating inhibitors through weight-loss measurements because measurements do not take into account all kinetic information.
• Problems that occur because of the limited solution volume.
• Conditions at which corrosion films are grown, which strongly depend on the experimental set up because of the limited solution volume.
• Establishment of a Tafel slope approximation to obtain kinetic information from LPR measurements.

Corrosion inhibitor evaluations are therefore complex and should take into account more real conditions such as flow dynamics and the replenishment of fresh solution.

But weight loss and LPR are relatively simple low cost procedures that are a good approach for a first selection procedure when compared to more-sophisticated, complicated, and time-consuming experimental setups.

In any case, there remains the need for understanding the processes that take place at the solid interphase.

References

1. Ikeda, A., Ueda, M., and Mukai, S., "Influence of Environmental Factors on Corrosion in C0₂ Source Well in Advances in C0₂," Corrosion, Vol. 2, 1985, pp. 1-22.
2. Videm, K., and Kvarekval, J., "Corrosion of Carbon Steel in C0₂ Saturated Aqueous Solutions Containing Small Amounts of H₂S," Corrosion/94.
3. Kvarekval, J., "The Influence of Small Amounts of H₂S on C0₂ Corrosion of Iron and Carbon Steel," Eurocorr'97, Trondheim, Norway.
4. Ramírez, M., "Efecto de la Relación PH₂S/PC0₂ sobre la Corrosión del Acero ASTM A516," Thesis, Universidad Simón Bolívar, Caracas, 1995.
5. Valdes, A., Case, R., Ramírez, M., and Ruiz, A., "The Effect of Small Amounts of H₂S on C0₂ Corrosion of a Carbon Steel," Paper No 98022, Corrosion/98.
6. Suárez, M., "Estudio de la Velocidad de Corrosiín en Acero ASTM A515 en Sistemas de C0₂ con Presencia de Pequeñas Concentraciones de H₂S," Thesis, Universidad Central de Venezuela, October 1997.
7. Perdomo, J.J., Morales, J.L., Viloria, A., and Lusinchi, A.J., "C0₂ and H₂S Corrosion of API 5L-B and 5L-X52 Grade Steels," Paper No. 00042, NACE International Corrosion Conference, Houston, 2000.
8. McMahon, A.J., "The Mechanism of Action of Oleic Imidazoline as an Oil Field Corrosion Inhibitor," Colloids and Surfaces, Vol. 59, 1991, p. 187.
9. "Standard Practice for Preparing, Cleaning, and Evaluating Corrosion Test Specimens," ASTM Standard Practice G 1-90, Annual Book of ASTM Standards Vol. 3, No. 2, 1992, pp. 35-41.
10. "Standard Practice for Calculation of Corrosion Rates from Electrochemical Measurements," ASTM Standard Practice G 102-89, Annual Book of ASTM Standards Vol. 3, No. 2, 1992, pp. 406-11.
11. EG&G Princeton Applied Research Corp., Application Note CORR1.
12. Gutzeit, J., "Corrosion Inhibitor can Cause Corrosion," Materials Performance, July 1993, p. 64.
13. Haussler, R.H., "Corrosion Inhibition in the Presence of Corrosion Product Layers," 6th European Symposium on Corrosion Inhibitors, Ferrara, Italy, 1985.
14. Gulbrandsen, E., Sundfaer, B., Hesjevik, S.M., Skjerve, S., Nesic, S., and Burchard, T., "Effect of Precorrosion on the Performance of Inhibitors for C0₂ Corrosion in Carbon Steel," Eurocor Conference, Norway, 1997.

The authors

Jorge J. Perdomo works for the material technology department at Pdvsa E&P, Caracas. His primary responsibilities involve questions regarding corrosion, inhibitor selection, and quality control. Perdomo has a material science degree, specializing in metallurgy, from the University of Simon Bolivar, Caracas and a doctorate in material sciences from the Case Western Reserve University, Cleveland.

Marjorie D. Ramirez is an engineer in the material technology department at Pdvsa E&P, Caracas. She works on investigating corrosion and inhibitors. Ramirez has an engineering degree in materials, specializing in metallurgy, from the University of Simon Bolivar, Caracas.

Alfredo Viloria is a technology manager at Pvdsa's research group Intevep, Caracas. He has worked on various projects concerning materials and corrosion. Viloria has a chemical engineering degree from the Central University of Venezuela, Caracas, and Phd in physical chemistry from the National Polytechnic Institute of Toulouse, France. He is a member of NACE, and Venezuela's Gas Processing Association, National Council of Scientific & Technological Investigation, and Association of Corrosion Engineers.