Corrosion Behavior of Carbon Steel in Piperazine Solutions for Post-Combustion CO₂ Capture

Abstract

The corrosion behavior of A106 carbon steel in 30 wt.% piperazine (PZ, a cycling amine) solutions was investigated by potentiodynamic polarization, EIS, LPR, SEM/EDS, and XRD. For comparison, the corrosion was also conducted in the benchmark solvent of monoethanolamine (MEA) solution. Similar to MEA, initial corrosion testing results in PZ solvent showed that the corrosion rate increases with increases either in CO2 loading from 0.23 to 0.43 mol CO2/mol alkalinity (C/N) or in temperature from 20 to 80 oC. Short term corrosion investigations showed an approximately two orders of magnitude lower corrosion rate for A106 in 0.43 C/N CO2 loaded 30 wt.% PZ at 80 oC than in the same conditions but with 30 wt.% MEA. The difference was due to the ability to form a dense and protective FeCO3 layer on the metal surface in PZ as opposed to in MEA solutions. Possible mechanisms are discussed.

Understanding the corrosion of CO2-loaded 2-amino-2-methyl-1-propanol solutions assisted by thermodynamic modeling

“…solutions (Xiang et al, 2014;Zheng et al, 2014a;Zheng et al, 2014b). However, no dense layer of FeCO 3 was observed in the corresponding MEA solution at 80°C after 200 h but mainly residual Fe 3 C (Zheng et al, 2015;Zheng et al, 2014a).…”

“…A corrosion cell with a standard three-electrode setup was used for electrochemical testing with a saturated calomel electrode (SCE) as the reference electrode and a graphite rod as the counter electrode. The LPR method was employed to study corrosion behavior over an extended time period, and the details of this method can be found elsewhere (Zheng et al, 2015;Zheng et al, 2014a;Zheng et al, 2014b). The corrosion current density (i corr , A cm -2 ) was calculated using standard Tafel slopes of 100 mV/decade for both the anodic ( a ) and cathodic ( c ) reactions with equation 1.…”

“…Corrosion rates versus time for A106 in 30 wt.% AMP and 30 wt.% MEA solutions with a CO 2 loading of 0.43 C/N at 80°C. The corrosion rates for A106 in MEA solution are from our previous work (Zheng et al, 2015;Zheng et al, 2014a;Zheng et al, 2016a). Figure 1 shows the corrosion rates for A106 carbon steel in a naturally aerated 30…”

“…For comparison, the corrosion rates for A106 in naturally aerated 30 wt.% MEA solution with a CO 2 loading of 0.43 mol CO2 /mol MEA at 80°C, from our previous work (Zheng et al, 2015;Zheng et al, 2014a;Zheng et al, 2016a), are also presented in Figure 1. With increases in time to approximately 100 h, the corrosion rate increased from 1.55 to 2.88 mm/y for A106 in the MEA solution with CO 2 loading and subsequently stabilized up to a testing duration of approximately 200 h. Eventually, the corrosion rate in the AMP solution with CO 2 loading (0.01 mm/y) is over two orders of magnitude lower than that in the MEA solution (~3 mm/y) even though the initial corrosion rate was slightly higher in the AMP solution.…”

“…To understand the differences in corrosion behavior of carbon steel in AMP and MEA solutions with CO 2 loading, SEM and XRD were used to characterize the corroded samples after electrochemical testing (Figures 2 and 3). An approximately 5 m-thick and dense layer of FeCO 3 (a known protective corrosion layer), with a typical morphology of cubic clusters (Zheng et al, 2014a;Zheng et al, 2014b), was formed on the A106 surface in the AMP solution at 80°C after approximately 115 h. This is consistent with the observation of FeCO 3 on the surface of 1018 CS after 144 h in a deaerated 45 wt.% AMP solution with a continuous flow of CO 2 at 80°C (Campbell et al, 2016). In addition, the formation of a dense layer of FeCO 3 is similar to that in piperazine (a secondary amine, PZ) and methyl diethanolamine (a tertiary amine, MDEA)…”