https://doi.org/10.1016/j.petlm.2016.11.002
“It is believed that the success of bi–solvent blends will gradually lead to the application of tri–solvent blends. This invention is aimed at further utilizing the benefits of the individual solvents in tri–solvent blends to enhance CO2 capture efficiency and suppress the individual problems accompanying each solvent.
Tri–solvent blends containing AMP–PZ–DIPA (diisopropanolamine) was investigated by Haghtalab et al. at high pressures (1–40 bar) and temperature from 40 to 70 °C [114]. Their result showed that AMP–PZ blend used as a promoter for DIPA increased its CO2 loading. Freeman et al. reported increase in CO2 absorption capacity and absorption rate with new tri–solvent blends containing PZ, N-methylpiperazine (MPZ) and N,N′-dimethylpiperazine (DMPZ) when compared to both PZ and MEA [141]. More recently, Liu et al. confirmed that the CO2 desorption time of aqueous solutions of 7 wt% MEA–3 wt.% MDEA reduced by adding 1 wt% AMP (7 wt% MEA–3 wt.% MDEA–1 wt.% AMP) [116].
Nwaoha analysed highly concentrated (6–7 kmol/m3) tri–solvent blend of AMP–PZ–MEA at 93.93 kPa CO2 partial pressure and 40 °C absorption temperature [112], [113]. The high CO2 partial pressure was chosen in order to facilitate precipitation which was triggered at high CO2 loadings and high amine concentration. It was confirmed from Fig. 8 that none of the highly concentrated tri–solvent blends formed any solid precipitate when their CO2 rich solutions (40 °C absorption temperature and 93.93 kPa CO2 partial pressure) was cooled at 20 °C for over 400 h. In addition, also shown in Fig. 8, as the concentration of the tri–solvent blend (AMP–PZ–MEA) was further increased to 2.5 kmol/m3 AMP–1.5 kmol/m3 PZ–4.5 kmol/m3 MEA, its CO2 rich solution did not precipitate when cooled at 20 °C for over 400 h. Furthermore, the precipitation possibility experienced with high concentrations of AMP–PZ [88], [105] can be eliminated by the development of tri–solvent blends [112].
Fig. 8. Non-precipitation of AMP–PZ–MEA tri–solvent blends compared to AMP–PZ bi–solvent blends.
The main idea is to show that tri–solvent blends can accommodate the use of AMP–PZ in a concentrated amine solution without precipitation, unlike in AMP–PZ bi–solvent blend where none of the amine solvent concentration can be high because of possibilities of precipitation.
Also, the highly concentrated tri–solvent blends have higher absorption capacity than those of single MEA and AMP–PZ bi–solvent blends [111], [113]. This proves that tri–solvent blends can potentially offer better a CO2 capture capability compared to conventional single and bi–solvent systems.
Nwaoha reported success with tri–solvent blends containing a bicarbonate forming amine (hindered amine, AMP) and two rate promoters (PZ and MEA), other blend configurations such as two bicarbonate forming amines (hindered and/or tertiary amines) and one rate promoter (polyamine) can be formulated as to have a better insight on the optimal configuration [112]. Based on the recommendation, another tri–solvent blend containing AMP–MDEA–DETA was experimentally analysed in which a reduction in regeneration energy of up to 50% when compared to MEA was observed [97]. Their results also revealed that AMP–MDEA–DETA tri–solvent blend can reduce amine circulation rate, minimize amine waste treatment.
This further confirms the flexibility of tri–solvent blends. Another merit of applying this tri–solvent blend is that though at a high total amine concentration, AMP and PZ can be used at low to moderate concentrations, thereby limiting chances of precipitation associated with them. Additionally, it is worth noting that the more amine solvents in the blend (bi–solvent to tri–solvent) the more difficult it is to control their concentration, but significant success in their CO2 absorption–desorption capability will likely suppress this expected challenge [111].
“