Effect of Amine Functionalization of MOF Adsorbents for Enhanced CO2 Capture and Separation: A Molecular Simulation Study

Abstract

Different types of amine-functionalized MOF structures were analyzed in this work using molecular simulations in order to determine their potential for post-combustion carbon dioxide capture and separation. Six amine models -of different chain lengths and degree of substitution- grafted to the unsaturated metal sites of the M₂(dobdc) MOF [and its expanded version, M₂(dobpdc)] were evaluated, in terms of adsorption isotherms, selectivity, cyclic working capacity and regenerability. Good agreement between simulation results and available experimental data was obtained. Moreover, results show two potential structures with high cyclic working capacities if used for Temperature Swing Adsorption processes: mmen/Mg/DOBPDC and mda-Zn/DOBPDC. Among them, the -mmen functionalized structure has higher CO₂ uptake and better cyclability (regenerability) for the flue gas mixtures and conditions studied. Furthermore, it is shown that more amine functional groups grafted on the MOFs and/or full functionalization of the metal centers do not lead to better CO₂ separation capabilities due to steric hindrances. In addition, multiple alkyl groups bonded to the amino group yield a shift in the step-like adsorption isotherms in the larger pore structures, at a given temperature. Our calculations shed light on how functionalization can enhance gas adsorption via the cooperative chemi-physisorption mechanism of these materials, and how the materials can be tuned for desired adsorption characteristics.

Keywords: CO2 capture; MOF-74; Monte Carlo simulation; amine; chemisorption; functionalization; metal-organic frameworks.

Figure 1

Molecular representation of the MOF structures studied in this work, and amine molecules used for functionalization [Color code: C, H, O, N, Mg (metal) in gray, white, red, blue, and green, respectively].

Figure 2

Validation of simulated (filled symbols) adsorption isotherms of carbon dioxide with available experimental data (open symbols) for (A) bare materials [squares from Mason et al. (2011), at 313 K; circles from McDonald et al. (2015), at 298 K; diamonds from Caskey et al. (2008), at 296 K], and (B) functionalized materials [triangles from Lee et al. (2014), at 298 K; circles and diamonds from McDonald et al. (2015), at 313 K]. Lines are to guide the eyes. For bare Mg-MOF-74, dotted line represents pristine material and straight line is the isotherm after using a scaling factor according to Dzubak et al. (2012).

Figure 3

Three-dimensional structure representation of carbamic acid functionalization for mmen-Zn/DOBPDC (Color code: C, H, O, N, Zn in gray, white, red, blue, and purple, respectively].

Figure 4

Comparison of simulated CO₂ adsorption isotherms at 313 K. (A) MOF-74 functionalized with ethylenediamine, (B) MOF-74 functionalized with methanediamine, and (C) MOF-74 functionalized with dimethylamine and ammonia moieties (dashed lines for 16%, dotted lines for 50%, and long dash dotted lines for 100% functionalization); (D) M/DOBPDC structures functionalized with different amines.

Figure 5

(a,b) Calculated selectivity values of binary 15% CO₂ / 85% N₂ mixtures (T = 313 K and P_total = 100 kPa) for M-MOF-74 and M/DOBPDC materials studied in this work (Color code: Mg-MOF-74 based structures in green, Zn-MOF-74 in violet, Mg/DOBPDC in blue and Zn/DOBPDC in orange). Percentages correspond to the degree of functionalization.

Figure 6

Carbon dioxide working capacity and regenerability values for functionalized structures under TSA conditions (T_regen = 373 K). The sizes of the circles represent the selectivity of the given material for CO₂/N₂ separation (Color code: same as in Figure 5).