Degradation and corrosion in water-lean solvents

Shoukat et al. [216], [217] have analyzed the thermal degradation of mixtures between amines, water and glycols, and drawn comparisons with the thermal degradation of aqueous amines. The results do not look very promising for water-lean solvents in this aspect. In Shoukat et al. [216], it has been shown that aqueous 30%wt . MEA mixtures loaded up to 0.5 mol CO_2∙mol MEA⁻¹ experience about 40% of amine loss after 50 days at 135 °C. In comparison, mixtures with triethylene glycol suffer more than 60% of amine loss, while mixtures with ethylene glycol lose 80% of their amine content. Though the results for a similar experiment carried with MDEA are a lot better, once MDEA is inherently more resistant to degradation than MEA, the substitution of water for glycol increase all degradation rates. This behavior can be observed in Fig. 16, adapted from data available in Shoukat [218]. In Shoukat et al. [217], an array of various tertiary amines mixed with glycols plus water was tested. This time, the result was more mixed, quite often being advantageous for water-lean solvents. More recently, Høisæter and Knuutila [219] observed high thermal degradation rates in MEA-based water-lean solvents when compared to aqueous amines. However, information is still very limited to form an overall picture of how degradation occurs in these mixtures.”

”

There is valuable published data regarding the degradation of the DEEA–EMEA solvent developed by Chen et al. [220], which is wasted by reacting with SO₂ present in flue gases, and of the Sulfinol-X® solvent (aqueous TMS + MDEA + PZ), whose oxidative degradation produces sulfate, sulfite and thiosulfate [221].

Many other studies do try to carry some sort of stability analysis. These, however, are mostly constrained to a few hours of absorption–desorption loops, far from the thousands of hours to be expected of a realistic solvent application. And even then they sometimes return quite unpromising results. As an example, Tao et al. [194] observed terrible losses of solvent in mixtures 30 %wt. MEA + water-free PEG200 at high temperatures. Their solvent lost 20% of mass after 5 h of N₂ purging at 40 °C, and 30% after 50 min at 80 °C. This seems to indicate some severe degradation reaction between the strong base and the ether groups of PEG200. With DEA, MDEA and DGA, the mass losses at 80 °C after 5 h of N₂ purging were kept at 19%, 8% and 26% respectively.

Results for corrosion studies are similarly incongruent. While in Shoukat et al. [216] one can clear that shifting from water to glycol reduces the corrosion caused by MEA solutions, a phenomenon understood since the 1950 s [222], this effect is not necessarily observed for tertiary amines. In a following paper, the results for corrosion in tertiary amine–glycol mixtures is even more difficult to interpret [217]. This is not to the detriment of the authors, but it might just be a very hard task to form general opinions about a wide range of different chemical compounds. It is also difficult to assess the effect of trace amounts of water in corrosion experiments with glycols, since these are hygroscopic compounds, and water concentrations after loading are usually untested for. Other results for water-lean solvents found in the literature indicate that the addition of MDEA reduces corrosion in systems operating with propylene carbonate [101], shifting from water to dimethyl formamide decreases the corrosivity of piperazine solutions [192] (and also causes the precipitation of piperazine carbamate, being a biphasic solvent). Finally, the addition of a series of organosulfur compounds in small concentrations to aqueous MEA, among them sulfolane, has been verified as successful in inhibiting corrosion mechanisms [223].”