Functionalized Covalent Triazine Frameworks for Effective CO2 and

Subscriber access provided by UNIV OF LOUISIANA

Energy, Environmental, and Catalysis Applications

Functionalized Covalent Triazine Frameworks for Effective CO2 and SO2 Removal Yu Fu, Zhiqiang Wang, Sizhe Li, Xunming He, Chunyue Pan, Jun Yan, and Guipeng Yu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b13417 • Publication Date (Web): 01 Oct 2018 Downloaded from http://pubs.acs.org on October 3, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Functionalized Covalent Triazine Frameworks for Effective CO2 and SO2 Removal Yu Fu§, Zhiqiang Wang§, Sizhe Li§, Xunming He§, Chunyue Pan§, Jun Yan§,* Guipeng Yu§,* §

College of Chemistry and Chemical Engineering, Hunan Provincial Key Laboratory

of Efficient and Clean Utilization of Manganese Resources, Central South University, Changsha 410083, China.

KEYWORDS: Covalent Triazine Frameworks, microporous polymers, pore engineering, SO2 removal, CO2 capture ABSTRACT: Building novel frameworks as sorbents remains highly significant target for key environmental issues like CO2 or SO2 emissions from coal-fired power plants. Here we report the construction and tunable pore structure as well as gas adsorption properties of hierarchically porous covalent triazine-based frameworks (CTF-CSUs) functionalized by appended carboxylic acid/sodium carboxylate groups. The densely integrated functionalities on the pore walls bestow strong affinity to the as-made networks toward guest acid gases, despite of their moderate BET surface areas. With abundant microporosity and integrated carboxylic acid groups, our frameworks deliver strong affinity towards CO2 with considerably high enthalpy (up to 44.6 kJ/mol) at low loadings. Moreover, the sodium carboxylate anchored framework (termed CTF-CSU41) shows an exceptionally high uptake of SO2 up to 6.7 mmol g−1 (42.9 wt%) even under a low SO2 partial pressure of 0.15 bar (298 K), representing the highest value for a scrubbing material reported to date. Significantly, such pore engineering could pave the way to broad applications of porous organic polymers.

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 20

INTRODUCTION More recently, great research efforts on carbon dioxide (CO2) capture from fuel gases by designable crystalline porous solids mainly including metal organic frameworks (MOFs) and covalent organic frameworks (COFs), have emerged because of their great potentials for Carbon Capture and Storage (CCS) processes.1-3 MOFs and COFs are usually featured by large internal surface areas and rare crystal structures, and most importantly they allow the atomically precise integration of building blocks into periodicities. In other words their properties and functionality are capable of being probed by powder X-ray diffraction spectrums or single crystal X-ray diffraction, when compared with their amorphous counterparts.4-6 The interaction of good sorbents toward guest molecules generally requires to be strong for gas capture and at the one time for reversible release, thus allowing an efficient but economically valuable cost-effective capture process with low energy consumption.7,8 Despite of several COF examples utilizing azine9,10 and boronate11 linkages that interact with CO2 to exhibit good capacity, the majority of COFs deliver very low performance in CCS, and the design of COFs and their pore engineering for efficient gas adsorption and separation remains highly significant. The structure-property relationship in COFs has been intensively investigated,12 while the capture and storage towards sulfur dioxide (SO2) has not yet been explored. SO2 is a major industrial pollutant with high corrosivity, causing adverse effects to environment and human health. Considering that in a flue gas stream the SO2 concentration is much lower than that of CO2, it still represents a significant environmental hazard, and accordingly, effective removal of SO2 requires strong absorption due to the relatively low partial pressure (e.g., 0.2 vol% SO2) in the gas mixture.13

Normally,

the

chemisorption

through

organic

amine-containing

materials14,15 or ionic liquid16-18 provides a sufficiently good capacity for gas absorption along with a high absorption enthalpy. However, it always results in great difficulty in desorption, leading to high cost due to the energy demand for regeneration. Concerning most COF-based sorbents, although the physical sorption


Page 3 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


enables an energy-effective reverse absorption-desorption, there is always an inherent drawback, i.e. a weak binding affinity towards guest acid gases.19 The weak interaction is generally associated with the structural nature of COFs that are constructed solely from light elements like carbon, hydrogen and boron. This would ultimately lead to a poor SO2 loading under ambient conditions (i.e. low pressures). The development of new absorbents for the removal and separation of SO2 (RSS) from off-gases would be a highly topical issue. Additionally, it remains a great challenge to develop ultra-stable MOFs or COFs to resist heat or moisture which is always accompanied with a SO2 emission process in coal-fired power plants.20 As a subclass of COFs, covalent triazine-based frameworks (CTFs) represents one potential platform for a practical CCS or RSS process due to their high porosity and easy synthesis.21-24 Pioneer work on CTFs was reported by Antonietti and Thomas, et al, who built the first crystalline CTF through a simple ion-thermal polymerization of terephthalonitrile by using ZnCl2 as catalyst.21 Of great importance is their excellent physicochemical stability which facilitates the design and development of functional CTFs for practical CCS processes. Notably, significantly high CO2-adsorption capacities have been demonstrated by a CTF with fluorine edges or by rich N/O co-doped hexaazatriphenylene-based CTFs.25,26 The exploring of the channel wall function of CTFs remains crucial for specific target applications. Bearing these in mind, we develop four CTFs with pre-designed carboxylic acid or sodium carboxylate units, and also provide an excellent platform to probe the effect of the appended functionalities on the pore parameters and also on the capture performances towards guest CO2 and SO2. In addition, we also report an attractive affinity of functionalized CTFs for guest SO2 and CO2, and also a record value of SO2 adsorption under mild conditions. The introduction of functional groups to the CTF architecture, together with its effect on the capacity towards guest molecules, could provide useful references for designing and manufacturing novel high performance CO2 and SO2 scavengers.

RESULTS AND DISCUSSION



Page 4 of 20

In this study, we targeted a new class of covalent triazine-based frameworks (CTFs) consisting of different functionalities held in the channel. The synthetic routes of four CTFs are outlined in Fig. 1a. These frameworks were prepared through the ionothermal polymerization of terephthalonitrile derivatives in sealed quartz ampoules in the presence of anhydrous zinc chloride.21 Unlike the commonly used post-modified strategy,24 each appended functional group was integrated into the backbone through a pre-designable technology.27 Especially the carboxylic acid or sodium carboxylate-anchored building blocks, i.e. 2,5-dicyanobenzoic acid or sodium 2,5-dicyanobenzoate were facilely synthesized in a two-step reaction (substitution and hydrolysis) with good yields starting from cost-effective 2,5-dibromobenzoate (Scheme S1 ESI†). The structure of the corresponding precursors and intermediates were confirmed by Hydrogen Nuclear Magnetic Resonance (1HNMR, Fig. S1-6, ESI†) and carbon NMR spectroscopy (13C NMR, Fig. S7-8, ESI†). Under ionothermal conditions, the precursor must be stable enough to endure the high polymerization temperature (exceeding 400oC) to avoid unexpected pyrolysis that would damage the reversibility of triazine formation. Thermo-gravimetric (TG) traces (Fig.S9, ESI†) demonstrate that only 2,5-dicyanobenzoic acid are thermally stable up to 430 oC (T10%, temperature for 10 wt%), of which the stability is comparable to those precursors like terephthalonitrile and 1,3,5-tricyanobenzene for crystalline CTFs. Concerning the fact that the optimization of the CTF growth conditions can produce crystalline frameworks, the polymerization of our terephthalonitrile derivatives is achieved following a step-by-step ion-thermal process to avoid initial decomposition of the precursors and to realize a well-ordered CTFs. For simplification, the frameworks derived from 2,5-dicyanobenzoic acid and sodium 2,5-dicyanobenzoate are termed as CTF-CSU38 and CTF-CSU39, respectively. Another two frameworks (simplified as CTF-CSU40, CTF-CSU41) with a double content of functional units anchored within pore

walls

were

obtained

in

similar

conditions

(Scheme

S2,

ESI†).

2,5-Dicyanoterephthalic acid and sodium 2,5-dicyanoterephthalate are used as starting blocks, respectively. In these cases, all frameworks were obtained as black monolithic materials in good yields (>80%). The good yields suggest low amounts of ACS Paragon Plus Environment

Page 5 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


side-reactions during theses polymerization processes. The residual Zn2+ contents are determined to be lower than 0.5at% (Table S1, ESI†), according to the Inductively Coupled Plasma-Atomic Emission Spectroscopy (ICP-AES). For each sample, it is fully degraded in nitrohydrochloric acid before the ICP-AES analysis. This indicates that the salt catalyst was removed almost completely.

Figure 1 Synthesis routine of CTF-CSUs (a), FT-IR spectra (b) and 13C CP-MAS solid state NMR of CTF-CSU38 (c, where asterisks denote spinning sidebands)

Chemical structure of the resulting frameworks was characterized by Fourier Transform Infrared (FT-IR) spectroscopy (Fig.1b),

13

C CP-MAS solid state NMR

(Fig.1c), Raman spectroscopy and elemental analysis (Table S1,S2, ESI†). A careful comparison between the precursors and their corresponding frameworks indicates a high degree of polymerization based on cyano cyclization chemistry (Fig. S10, ESI†). For example, the disappearance of intense C≡N band at around 2235 cm-1 demonstrates the almost complete consumption of the starting precursor, while the emergence of new bands for C–N (1360 cm-1) and C=N stretch (1590 cm-1) proves the generation of triazine rings.22,23 The signals at around 1720 and 1230 cm-1 that correspond to C=O and C-O stretching, respectively, proves the successful



introduction of acetic acid or sodium carboxylate into the framework backbone. Following a similar quantitative FTIR analyses on the ratio of the C=O stretch verse C≡N stretch using KSCN as an external reference (Fig.S11-S15, ESI†),27 we found that all polymerization processes display almost complete cyano conversions (exceeding 91%) (Table S3, ESI†). For CTF-CSU39 and CTF-CSU41, around 21.3% and 36.1% of sodium carboxylate were eliminated because of accelerated cleavage of the sodium carboxylate. In the case of CTF-CSU38, the carboxylic acid was efficiently integrated into the channel walls considering that around 84.5% of the functional group retained after the polymerization. Almost same trend was demonstrated for CTF-CSU40 whose carboxylic acid content significantly decreased to 72.8% of that expected, due to the easy decarboxylation in rigorous molten salts. 13

C CP-MAS solid state NMR spectrum provides further evidence to confirm the

chemical connectivity of CTF-CSU38. The expected signal at δ=163 ppm which would be assigned to the carbon for the carboxyl unit, indicating the presence of carboxylic acid functionalities. Raman scattering spectra (Fig.S16, ESI†) give diagnosis signals for the formation of C=N and C-N stretching, well correlated with FT-IR results. For all frameworks, the tested elemental values by Elemental Analysis (EA) show certain deviation when compared to the theoretical C/H/N values (Table S1, ESI†), since both incomplete combustion of the frameworks and adsorbed moisture in channels have to be taken into account. Energy Dispersive X-ray Spectroscopy (EDS) also show rather lower oxygen content and sodium content than the theoretically calculated values (Table S2, ESI†). It would be reasonable considering the presence of a significant amount of decarboxylic reaction. This is in good agreement with the FTIR and EA results. To note that, for the two sodium carboxylate-appended samples, the contents of sodium stay close to the values determined from ICP-AES analysis and from the quantitative FTIR analyses. XPS survey spectrum on CTF-CSU39 (Fig. S17, ESI†) reveals two typical environments for C=O (533.2 eV) and for C-O-C=O (533.7 eV), which correlate well to the C=O and C-O configurations. These results suggest that the carboxylic acid and sodium carboxylate functional groups have been ACS Paragon Plus Environment

Page 6 of 20

Page 7 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


successfully incorporated into the backbone.

Figure 2 (a) Observed (red) and calculated (gray) PXRD profiles of CTF-CSU38 from an optimized structure (eclipsed conformation AA; (b) Schematic representation of the structure of CTF-CSU38 (C: Gray, N: Blue, O: Red);

Powder X-ray diffraction spectrums (PXRD, Fig.2a, Fig.S18-20, ESI†) showed that the obtained polymers were at least partially crystalline. It is clear that the intensity and width of the peaks for crystalline CTFs could be determined by the nature of precursors, monomer/ZnCl2 ratio, reaction time and in most cases by polymerization temperature.21,28 CTF-CSU38 gave a significantly good crystallinity, as proved by the good correlation with the calculated in-plane Bragg peak positions of hexagonal unit sheet with the parameters. It would be reasonable concerning the fact that its precursor can endure the polymerization temperature in molten ZnCl2 and the competitive pyrolysis is restrained as identified by spectroscopic measurements and elemental analysis. Demonstrated in Fig. 2a is one sharp peak of 2θ degrees at 6.9° corresponding to the (100) facet of a primitive hexagonal lattice. Another two weak reflections at 12.5 and 13.6 are displayed, which may be assigned to (110) and (200) facet, respectively. The successful construction of 2D CTFs in a crystalline and π−π stacked form could be confirmed by the presence of a broad reflection at 2θ = ∼25.2 degrees for the (001) facet. It can be concluded that CTF-CSU38 has one-dimensional channels, along the c-axis and that the layers stack with an interlayer distance of 3.44 Å. The PXRD patterns of CTF-CSU39 and CTF-CSU41 (Fig. S19, S21, ESI†) feature



a rather ill-defined crystal structure with broadened peak widths when compared to that of CTF-CSU38 (Fig.2a). The ill-defined crystal nature could be ascribed to the fact that the accelerated thermal decomposition of the precursors during the ionothermal processes damages the reversibility of triazine formation. The low degree of reversibility of polymerization distorts significantly the long-range order of the crystalline material. The CTF-CSU40 with designed double content of carboxyl groups exhibit an identical diffraction pattern (Fig. S20, ESI†) to that of CTF-CSU38, suggesting their similar crystal structure. The obtained networks are non-swollen or insoluble in common organic solvents, implying their high physicochemical stability. Unlike the poor stability of the initial precursors, TG analysis of the CTF-CSUs demonstrates a superior stability in same conditions (Fig. S21). Typically, these frameworks deliver high decomposition temperatures exceeding 428 °C (Temperature for 10% mass-loss) under nitrogen conditions, suggesting the excellent stability of the networks once formed (Fig.S22). Their morphologies are investigated by scanning electron microscopy (SEM, Fig. S23, ESI†). CTF-CSU38 and CTF-CSU39 display a similar lamellar structure, while CTF-CSU40 and CTF-CSU41 are aggregated particles, despite of the certain differences in chemical composition. The alternating dark and bright area is shown in high-resolution transmission electron microscopy (HR-TEM) images (Fig. 3) which insinuates a porous structure of the as-made frameworks. As measured by sorption experiments which apply nitrogen as probes, the frameworks show typical Type IV isotherms (IUPAC), and a sharp uptake in the low-pressure region (P/P0 < 0.01) implies the presence of abundant micropores (Fig. 4). Another steady rise phase ranging from 0.2 to 0.8 bar suggests certain meso-porosity which may be ascribed to the cleavage of triazine intermediates during the ion-thermal polymerization.24,25 To note out, CTF-CSU41 possesses the highest mesopore content among the four samples, which would be reasonable considering the decomposition of its precursor and intermediates with lowest stability during the ionothermal polymerization. The highest Brunauer-Emmett-Teller (BET) surface area is demonstrated by CTF-CSU38 (491 m2 g-1, Table 1), followed by 402 m2 g-1 for ACS Paragon Plus Environment

Page 8 of 20

Page 9 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


CTF-CSU39, 333 m2 g-1 for CTF-CSU41 and 326 m2 g-1 for CTF-CSU40. These values are slightly lower than that of appended group-free CTF-1 (791 m2 g-1).22

Figure 3 TEM images for (a) CTF-CSU38; (b) CTF-CSU39; (c) CTF-CSU40; (d) CTF-CSU41

All frameworks demonstrated a significant fraction of micropores with pore width less than 2 nm according to the pore size distribution (PSD) curves (Fig. 4). The PSD curves were obtained by fitting Non-local density functional theory (NLDFT). CTF-CSU38 shows a relative narrow distribution where dominant channels locate at

a)

b)

Figure 4 (a): N2 sorption isotherms at 77 K; (b): NL-DFT pore size distribution curves for CTF-CSUs



Page 10 of 20

Table 1 The properties of porosity, gas uptake and isosteric heat of adsorption of CTF-CSUs

a

Sample

SBETa (m2 g-1)

Vtotalb (cm3 g-1)

Vmicro/Vtotalc (%)

CO2 uptaked (wt %)

Qst e kJ/mol

SCO2/N2

SO2 uptakef (wt %)

CTF-CSU38 CTF-CSU39 CTF-CSU40 CTF-CSU41

491 402 326 333

0.44 0.53 0.79 0.78

23.23 8.80 10.34 22.32

9.8 8.7 6.9 10.3

39.2 41.3 38.5 44.6

72.0 48.8 56.2 35.3

28.3 37.2 35.5 42.9

Brunauer-Emmett-Teller surface area.

b

Total pore volume determined from the N2 isotherm at P/P0=0.99. d

Micropore volume determined from the N2 isotherm at P/P0= 0.1. CO2 uptake at 273K/1bar.

e

c

Obtained from

adsorption isotherms at 273 and 298 K utilizing Clausius-Clapeyron equation and virial equation. f SO2 uptake at 273K/1bar

around 0.9 nm, which is close to the pore width measured from the optimized model. However, the rest of samples give broader pore width distributions in the range of 0.5 to 12 nm. For the broad distribution, one reason should be taken into account, i.e. the easy cleavage of carboxylic acid or sodium carboxylate units during the ionothermal polymerization. The decarboxylation enables these organic groups as an additional template to tailor the pore size and the skeleton ordering29,30 and to build hierarchically porous structure. Furthermore, their PSD curves show many parallels, such as abundant micropores existing from 0.5 to 1.2 nm, implying that a similar topology was obtained despite of the incorporation of different functionalities. Another bonus is that the introduction of substituting group can further boost the control on the pore structure, the crystallinity and the stability. For the carboxylic acid-containing frameworks, it is slightly odd for the trend that di-substituted CTFs deliver higher pore volume than the mono substituted one. It is generally accepted that the introduction of the appending groups would block the access of nitrogen probe, therefore giving a decreased pore volume.30,31 This trend would be reasonable considering the fact that the di-substituted precusors possess lower stability as revealed by TG traces (Fig.S9, ESI†). The easy decarboxylation of the di-substituted precusors than the mono substituted ones in rigorous polymerization condition provides their frameworks additional templates to promote accessible pore generation and carbonization. These would hence increase the overall pore volume of the as-made frameworks.


Page 11 of 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Porous sorbent provides an alternative approach to mitigate the amount of the CO2 released in the atmosphere. The CO2 adsorption isotherms at 273 and 298 K were collected (Fig. 5a, Fig.S24, ESI†), which reveal good gas uptake capacities and a reverse release ability for all samples. This would be fundamental for identifying cost-effective CO2 sorbents. Among of them, CTF-CSU41 possesses the highest CO2 uptake of 10.3 wt% (Table 1, 1bar/273 K). The highest micropore volume of CTF-CSU41 contributes to its prominence. Besides, the rich nitrogen or oxygen in the backbone of frameworks improves the CO2 affinity due to the promoted dipole-quadrupole interaction.31,32 Note that the decoration of carboxyl groups on the pore walls effectively promotes the CO2 load capacity despite of their low BET surface areas.

Figure 5 (a) CO2 adsorption isotherms (273K) and (b) isosteric heat plots of adsorption (Qst) for CTF-CSUs (c) SO2 uptake verse sorption time at room temperature under a pressure of ~0.1 bar. (d) Regeneration performance of CTF-CSU38 towards SO2.

For an optimal sorbent, a moderate still reversible adsorption-desorption towards guest gases (CO2) requires a perfect interaction between host and guest.33 However,



for most nanoporous organic polymers (NOPs), a weak adsorption affinity towards guest molecules like CO2 was usually demonstrated. Pore surface engineering like the introduction of nitrogen-rich or oxygen-rich functionalities has been proved to be powerful to enhance the host-guest interactions.24,26,34-36 Utilizing Clausius-Clapeyron equation and Virial equation, the isosteric enthalpies (Qst) towards CO2 were estimated from the adsorption isotherms. Our frameworks possess high values of Qst at a low coverage up to 44.6 kJ mol-1 (Fig.5, Table 1). Interestingly, monocarboxylic group functionalized CTF-CSU38 show a higher enthalpy than the dicarboxylic acid functionalized CTF-CSU40. It suggests that the trend arises from the almost 3-fold ratio of micropore volume verse total pore volume rather than the difference on carboxylic content. The high CO2 Qst values (>38 kJ mol-1) are sufficiently higher than the known CTFs like CTF-037 and CTF-CSU37@post27 and NOPs like MOP-10,38 PPF-439 and TPI-1.40 The high Qst values indicate a strong interaction between the framework and guest CO2 through decorating carboxylic acid or sodium carboxylate groups within our frameworks. Giving the high affinity for CO2, the selective capture of CO2 from fuel gases was probed using the simplified ideal adsorbed solution theory (IAST) model.41 The near-linear adsorption profile for N2 at 273K is indicative of its low affinity for these frameworks. Combing the experimental single-component isotherms, ideal selectivity of CO2 over N2 was calculated at an equilibrium partial pressure of 85% N2 and 15% CO2 in the bulk phase. Among the investigated frameworks, the formic acid-appended CTF-CSU38 gave the highest ideal selectivity of CO2-over-N2 of 72.0 (Fig. S25-S29, ESI†). It is reasonable to comprehend that the high ideal selectivity originates from the presence of abundant ultramicropores ( CTF-CSU39 (37.2%)> CTF-CSU40 (35.5%)> CTF-CSU38 (28.3%). It can be seen that CTF-CSUs modified by sodium carboxylate possessed higher SO2 uptake than those modified by carboxylic acid groups. The major difference on SO2 uptake capacity may originate from the intense difference on micropore volume due to the side functionalities. As shown in Table 1, an important feature of such reversible SO2 uptake strategy involves ‘tailoring’ the appended functionalities based on a pre-designable technology that will permit an overall favorable interaction between the framework and the guest gases. Altering chemical heterogeneity of channel walls through increasing the carboxylic acid contents or the sodium carboxylate contents can significantly enhance SO2 uptake. For example, the SO2 storage capacity of CTF-CSU41 is almost 1.2-fold of that of CTF-CSU39 under same conditions. More importantly, Table 2 shows that our frameworks surpasses those of the known carbons,42 metal organic frameworks43-46 and noncovalent porous materials,47,48 and the performance is also comparable to the reported best ionic liquids.16,17 For instance, at room temperature and around 0.15 bar, our frameworks with at least 4.4 mol/kg SO2 uptake capacity outperforms the known [Zn2(L1)2(bpe)]

43

and FMOF-245 even under high SO2

pressures (1bar). At low pressure (

Functionalized Covalent Triazine Frameworks for Effective CO2 and

Recommend Documents