In a previous study, we reported the use of in situ 1 H-and 13 C-NMR to elucidate mechanistic pathways for the reaction of carbon dioxide with a broad range of amines (pKa ~4.5-15.5), including alkanolamines of commercial interest, in water. In the aqueous systems of that study, water most importantly functions as a Brønsted acid/Lewis base and as the amine is consumed and pH decreases hydrolyzes the initially formed carbamate species (1:2 CO 2 :amine stoichiometry), into the alkyl ammonium bicarbonate with a more beneficial 1:1 CO 2 :amine stoichiometry. This study has been extended herein to amines, amidines and guanidines dissolved in non-aqueous solvent systems such as dimethylsulfoxide, sulfolane, toluene, 1-methyl-2-pyrrolidinone and the ionic liquid 1-ethyl-3-methyl-imidazolium acetate. The use of non-aqueous organic solvents shuts off some CO 2 reaction pathways available in aqueous solution. However, more importantly, it opens up new possibilities and reaction pathways for amine based carbon capture. Two important aqueous-system pathways are eliminated: the direct hydration of CO 2 with tertiary amines or guanidines to form bicarbonates, and the hydrolysis of carbamates at lower pH to form bicarbonates. In non-aqueous solution, the initial step for the reaction of primary and secondary amines with CO 2 is the same as in aqueous solution -nucleophilic attack by the amine nitrogen on CO 2 . However, additional mechanistic pathways are enabled in non-aqueous solvents, particularly the stabilization of carbamic acid(s) (rather than carbamates) products in certain organic solvents. The formation of carbamates requires no water and is favored by higher amine concentrations and basicities (higher amine pKa). In contrast, carbamic acid/zwitterion formation is favored by lower amine concentrations, higher CO 2 partial pressures, lower amine pKa, and selection of more polar organic solvents that promote hydrogen bonding. The new amine-CO 2 reaction pathways enabled here by the use of non-aqueous solvents introduce stabilizing interactions between the non-aqueous solvent and the amine-CO 2 reaction products, facilitating higher capacity and selectivity for carbon capture than in water solutions. The effects of temperature, amine basicity, solvent electronic structures, and concentration on amine-CO 2 reaction products (carbamic acid/zwitterion/carbamate and equilibria between neutral and ion-paired forms) are discussed in detail herein.

Layer-stacking structures are very common in two-dimensional covalent organic frameworks (2D COFs). While their structures are normally determined under solvent-free conditions, the structures of solvated 2D COFs are largely unexplored. We report herein the in situ determination of solvated 2D COF structures, which exhibit an obvious difference as compared to that of the same COF under dried state. Powder X-ray diffraction (PXRD) data analyses, computational modeling, and Pawley refinement indicate that the solvated 2D COFs experience considerable interlayer shifting, resulting in new structures similar to the staggered AB stacking, namely, quasi-AB-stacking structures, instead of the AA-stacking structures that are usually observed in the dried COFs. We attribute this interlayer shifting to the interactions between COFs and solvent molecules, which may weaken the attraction strength between adjacent COF layers. Density functional theory (DFT) calculations confirm that the quasi-AB stacking is energetically preferred over the AA stacking in solvated COFs. All four highly crystalline 2D COFs examined in the present study exhibit considerable interlayer shifting upon solvation, implying the universality of the solvent-induced interlayer stacking rearrangement in 2D COFs. These findings prompt re-examination of the 2D COF structures in solvated state and suggest new opportunities for the applications of COF materials under wet conditions.

Pyridine-modified COF-10 exhibits enhanced stability in humid air relative to un-modified COF-10. Solid state NMR and computational studies were used to probe the nature of pyridine interactions with the framework. We propose two models for pyridine-framework interactions with different stabilities.

No abstract

Covalent organic frameworks (COFs) have found wide applications due to their crystalline structures. However, it is still challenging to quantify crystalline phases in a COF sample. This is because COFs, especially 2D ones, are usually obtained as mixtures of polycrystalline powders. Therefore, the understanding of the aggregated structures of 2D COFs is of significant importance for their efficient utilization. Here we report the study of the aggregated structures of 2D COFs using 13C solid-state nuclear magnetic resonance (13C SSNMR). We find that 13C SSNMR can distinguish different aggregated structures in a 2D COF because COF layer stacking creates confined spaces that enable intimate interactions between atoms/groups from adjacent layers. Subsequently, the chemical environments of these atoms/groups are changed compared with those of the nonstacking structures. Such a change in the chemical environment is significant enough to be captured by 13C SSNMR. After analyzing four 2D COFs, we find it particularly useful for 13C SSNMR to quantitatively distinguish the AA stacking structure from other aggregated structures. Additionally, 13C SSNMR data suggest the existence of offset stacking structures in 2D COFs. These offset stacking structures are not long-range-ordered and are eluded from X-ray-based detections, and thus they have not been reported before. In addition to the dried state, the aggregated structures of solvated 2D COFs are also studied by 13C SSNMR, showing that 2D COFs have different aggregated structures in dried versus solvated states. These results represent the first quantitative study on the aggregated structures of 2D COFs, deepen our understanding of the structures of 2D COFs, and further their applications.

extensive listing of intrinsic proton binding tendencies at N(l)and N(7) of purines. The N(l)/N(7) intrinsic binding ratio for protons is 2 log units greater for the 6-oxopurines than for adenosine.The N(l) to N(7) dienPd2"1" binding ratios in the last column of Table VI have ramifications for the ligand distribution curves.The relatively pronounced favoring of N(l) for IMP and inosine in Figures 9 and 10 results in BM, clearly predominating over M7B" in basic solutions. In contrast the negative logarithm for GMP in Table VI means that the mole fraction of M7B" exceeds that for BM, and no crossover occurs; the N(7) metalated complex predominates at all pH values. Basic solutions of GMP and dienPd2"1" contain a mixture of four species, all present in significant amounts.The intrinsic N(l)/N(7) binding ratios of Table VI show quantitatively that the tendency to favor N(l) over N( 7) is H+ > CH3Hg+ > Pd(II). Indeed for dienPd2"1", the binding constant at N( 7) is often comparable to or even exceeds that at N(l). For inosine the crossover pH, where the M7BH, and BM, mole fractions are equal, occurs at pH 6.1 with dienPd2"1", 1.8 log units higher than with CH3Hg+, which is due to the lesser N(l)/N(7) binding ratio for dienPd2"1".For AMP the values of log (Ki/K•,) and log ( \/ ') are -0.1 and 0.4, respectively, the primes referring to phosphate protonated species. Compared to the monoanionic phosphate, the dianionic phosphate group favors dienPd2+ binding at N(7) over N(l). Thus there are two crossovers between N(7) and N(l) metalated species for AMP in Figure 11. The first crossover appears at pH 2.2 as M7BH,HP gives way to BM,HP when dienPd2"1" successfully competes with H+ for N(l). The maximum role fraction of , occurs near pH 5. Phosphate deprotonation near pH 6 yields a crossover close to pH 7, above which M7B predominates marginally over BM, according to the -0.1 entry in the last column of Table VI.We now compare the order of nucleotide binding strengths of dienPd2"1" with the proton and CH3Hg+, the only other metal for which a series of reliable values exists. As previously, we designate the nucleoside by its capitalized first letter and the nitrogen binding site by the usual numbering scheme.2 The order of decreasing proton binding strengths is then T3 > U3 > G1 >11 » C3 > Al > G7 > 17 > A7. The order of CH3Hg+ is similar, with a promotion of G7 and 17 to greater than Al. These two sites undergo further significant promotion in the series for decreasing dienPd2"1" stability constants T3 > U3 > II > G7 > G1 > 17 » C3 > Al > A7 with Al and A7 now of comparable magnitude.Because of the proton at N(l) in purines and N(3) in pyrimdines the stability order, as opposed to stability constants, is pH dependent. At pH 7 the stability order for dienPd2"1" becomes G7 > 17 > II > G1 > U3 > T3 > C3 > Al > A7. This order agrees with that observed in neutral solutions of nucleotide mixtures.7 Thus N(7) of GMP and IMP offers the strongest effective nucleic base binding site for dienPd2"1" in neutral solutions.

The kinetics of the irreversible phase conversion of covalent organic frameworks-1 (COF-1) has been investigated using time-resolved, in situ environmental X-ray diffraction (EXRD) and modeled with the Avrami–Erofe’ev model. Tightly fitting mesitylene solvent is found to be present in both the AB staggered and AA eclipsed polymorphs, which plays a key role in the phase change. Solid-state NMR (SSNMR) showed the presence of discrete dipolar coupling between residual mesitylene solvent and the framework in both polymorphs, indicative of a host–guest adsorptive interaction. Binding energy calculations indicate two different adsorbed mesitylene configurations in the AB and AA phases, both with short distances to the framework pore walls to generate the observed dipolar coupling. The mechanism of phase change has been illustrated using molecular dynamics simulations and was found to be a displacive transition from AB staggered to AA eclipsed COF-1 structures, which was made possible due to low in-plane shear modulus of 2D COFs. Our findings highlight the polymorphic nature of COF-1 material mediated by the interactions with guest molecules and the irreversibility for polymorph formation and conversion.

Covalent organic frameworks (COFs) have been considered promising adsorbent materials for postcombustion CO2 capture due to their high porosity, tunable functionalities, and excellent framework stability. Nevertheless, few research studies have systematically investigated the structure–performance relationships and the effect of moisture on CO2 capture performance of COFs. In this study, a series of Schiff-base COFs with different functionalities, pore sizes, and framework dimensions are prepared and evaluated for potential applications in postcombustion CO2 capture. Gas sorption isotherms and ideal CO2/N2 sorption selectivity calculations reveal the following: (1) COFs undergoing enol-to-keto transformations outperform other studied COFs with imine functionalities and similar pore sizes. (2) CO2 uptake capacity of a COF is not necessarily a function of its pore aperture and specific surface area. TpPa-1 with keto-enamine moieties exhibits an impressive CO2 uptake of 0.6 mmol g–1 and a CO2/N2 sorption selectivity of 114. Dynamic breakthrough experiments of wet CO2/N2 mixed gas (17% relative humidity) indicate that both keto-COFs studied, NUS-2 and TpPa-1, retain about 70% of their dry CO2 adsorption capacities, which can be attributed to the moderately hydrophobic pore environment of the COFs. Considering the noticeable cost of flue gas desiccation, our study suggests that COFs with moderate hydrophobicity would be promising adsorbent candidates for practical postcombustion CO2 capture.