Elucidating the role of non-covalent interactions in unexpectedly high and selective CO₂ uptake and catalytic conversion of porphyrin-based ionic organic polymers

Abstract

Here, we present viologen-porphyrin based ionic covalent organic polymers (H2-ICOP and Zn-ICOP) with multiple CO₂-philic sites. The specific surface areas of H2-ICOP and Zn-ICOP were found to be 9 m² g⁻¹ and 20 m² g⁻¹, respectively. CO₂ uptake analyses reveal that H2-ICOP exhibits very high CO₂ capture uptake (62.9 mg g⁻¹), which is one of the highest values among previously reported ICOPs. The results indicate very efficient non-covalent interactions between H2-ICOP and CO₂. The possible non-covalent interactions of hydrogen (O_CO2⋯H–N), tetrel (C_CO2⋯N, C_CO2⋯Cl⁻), pnicogen (O_CO2⋯N⁺), and spodium bonds (O_CO2⋯Zn) between CO₂ and H2-ICOP and Zn-ICOP are investigated via symmetry adapted perturbation theory (SAPT) analysis and electrostatic potential maps (MEP). The strength of non-covalent interactions in H2-ICOP and Zn-ICOP is decreasing in the following order ΔE_C⋯N > ΔE_C⋯Cl⁻ > ΔE_O⋯N⁺ and ΔE_Zn⋯O > ΔE_C⋯Cl⁻ > ΔE_C⋯N > ΔE_O⋯N⁺, respectively. The major CO₂ uptake contribution comes from C_CO2⋯N tetrel bonding (−22.02 kJ mol⁻¹) interactions for H2-ICOP, whereas O_CO2⋯Zn spodium bonding (−21.065 kJ mol⁻¹) interactions for Zn-ICOP. H2-ICOP has more CO₂-philic moieties with powerful non-covalent interactions compared to Zn-ICOP, which is in good agreement with the experimental results. Furthermore, the CO₂ catalytic conversion performances of Zn-ICOP and H2-ICOP gave good yields of 83% and 54%, respectively. Surprisingly, Zn-ICOP, despite having significantly lower CO₂ uptake capacity, displayed better catalytic activity than H2-ICOP, owing to a higher number of counter anions (Cl⁻) on its surface, which shows the crucial role of the counter anion (Cl⁻) in the mechanism of this catalytic reaction.

---

1. Introduction

The rise of the atmospheric carbon dioxide (CO₂) concentration has become a widespread concern due to its link to the greenhouse effect and threat to the environment.¹ Therefore, viable solutions for efficient capture and utilization of CO₂ (CCU) have received much attention in the past decades.^2,3 Among various CO₂ removal technologies, physical adsorbents such as zeolites, metal–organic frameworks (MOFs), covalent organic frameworks (COFs) and porous covalent organic polymers (COPs) came into focus owing to their high CO₂ capture capacities, catalytic transformation, and low energy requirements for regeneration.^4–8 Especially COPs have evolved for CCU due to their high specific surface areas, easily tunable pore structures, good chemical and thermal stabilities, tailored surface manipulation with polar groups, and variation in the synthesis.^9–13

A high selective uptake of CO₂ is known to arise from large surface areas and non-covalent interactions between CO₂ and surfaces of materials. In the past decade, significant effort has been mainly focused on increasing the surface areas of materials and the non-covalent interactions between CO₂ and surfaces of materials have not attracted enough attention.¹⁴ One should note that a large surface area is one of the parameters, but not the only one, that allows for high CO₂ capture. However, materials with a low surface area may still exhibit large CO₂ capture capacity with efficient non-covalent interactions with CO₂. Therefore, it is vitally important to identify and estimate the strength of these non-covalent interactions in terms of the contribution to the CO₂ capture capacity.^15,16 CO₂ is known to be able to form simultaneous non-covalent interactions via its electron-rich terminal oxygen atoms and its electron-deficient central carbon atom.^16,17 The electron-deficient central carbon atom acts as a Lewis acid (LA) and can form tetrel bonding interactions with anions or lone-pair-possessing atoms such as N, O, S, and F.^18–21 On the other hand, the electron-rich terminal oxygen atom acts as a Lewis base (LB) and can form hydrogen bonding with hydrogen atoms and, as recently reported, halogen bonding with halogen atom (I, Br, Cl) containing moieties.^22–29

In particular, electron-rich nitrogen-based functional groups, such as amines, triazoles, triazine cores, imines, tetrazoles, benzimidazoles, azo linkages, and viologens, continue to be a dominant polar functional group to enhance the CO₂ adsorption capacity and CO₂/N₂ selectivity thanks to their multiple non-covalent interactions with CO₂.^{20,27,30–38} Among nitrogen based adsorbents, viologen linked ionic covalent organic polymers (ICOPs) are of particular importance in CCU technologies because they not only have all the advantages of COPs but also possess charged cationic skeletons paired with counter-ions which provide additional CO₂ affinity owing to their electrostatic interactions with CO₂ molecules.^35,39–42 Consequently, the CO₂ uptake capacity of viologen linked ICOPs is found to be significantly higher compared to their neutral counterparts.^34,43,44 Moreover, unlike other porous materials, selective recognition of viologen linked ICOPs can be manipulated by simply controlling the redox state of viologen or the nature of the counter-ions for the desired application.^39,41

In recent years, several cationic viologen linked ICOPs with good surface areas have been reported for CCU.^{34,39,43–49} The general understanding of these studies is that the high CO₂ capture capacity of viologen linked ICOPs is credited mostly to tetrel bonding interactions between the counter anions and CO₂.⁵⁰ Commonly, nitrogen-based groups in CCU applications are preferred for their Lewis basic character to form non-covalent interactions with the electron-deficient central carbon (Lewis acid) of CO₂. However, in the case of viologen, the quaternary nitrogen (N⁺) atom lacks lone electron pairs, and hence acts as a Lewis acid; consequently, the formation of efficient interactions between cation (N⁺) nitrogen and the electron-rich terminal oxygen atoms of CO₂, commonly referred to as pnicogen bonds, is highly possible.^19,51 Yet, the pnicogen (O_CO2⋯N⁺) interaction is often disregarded, and its contribution to CO₂ capture has not been reported.

However, despite their attractive features for CCU applications, only a limited number of viologen linked ICOPs have been reported in the literature, mainly due to their low physicochemical stability. This drawback can be overcome by the incorporation of viologens into polymeric systems with suitable rigid and symmetric macrocycles such as porphyrin.⁵² The incorporation of porphyrin brings three advantages. First, unstable cationic species get stabilized by extended delocalized polymeric structures. Second, porphyrin is another excellent N containing molecular building block, which can increase the CO₂ uptake and selectivity through the CO₂ interaction with pyrroles (–C=N, –NH). Third, the metalation of the porphyrin core with Lewis acid metals such as zinc(II) can promote additional non-covalent interactions with the electron-rich terminal oxygen atoms of CO₂, known as spodium bonding.⁵³ In addition to their multiple CO₂-philic functional sides, both viologen and porphyrin are also known to exhibit effective CO₂ catalytic conversion, which makes them very promising bifunctional materials in CCU applications.^{46–48,54–56}

Herein, based on the above considerations, we report viologen-linked porphyrin ICOPs (H2-ICOP and Zn-ICOP) synthesized through the Zincke reaction of H2 and Zn(II) tetrakis(4-aminophenyl) porphyrin with viologen Zincke salt, as presented in Scheme 1.

Scheme 1 Schematic representation of the synthesis of the viologen-linked porphyrin ICOPs via a Zincke reaction under solvothermal conditions.

The resultant ICOPs were characterized with various experimental analyses including FTIR spectroscopy, solid-state ¹³C NMR spectroscopy, energy-dispersive X-ray spectroscopy (EDX), diffuse reflectance spectroscopy (DRS), thermogravimetric analysis (TGA), powder X-ray diffraction (PXRD), field emission scanning electron microscopy (SEM), transmission electron microscopy (TEM) and dynamic light scattering (DLS). Furthermore, their CO₂ capture capacity, CO₂/N₂ selectivity, and CO₂ catalytic conversion performances were explored. Brunauer–Emmett–Teller (BET) surface area analyses reveal that both H2-ICOP (9 m² g⁻¹) and Zn-ICOP (20 m² g⁻¹) possess very low surface areas. Yet, despite their nearly non-porous surfaces, especially H2-ICOP exhibits a very large CO₂ capture capacity (62.9 mg g⁻¹) and CO₂/N₂ selectivity, which indicates powerful non-covalent interactions with CO₂. Detailed theoretical analyses have been conducted to decode insights into these unexpected large CO₂ capture capacities and CO₂/N₂ selectivity. The strengths of possible non-covalent interactions, such as hydrogen, tetrel, pnicogen, and spodium bonding, between CO₂ molecules and the H2-ICOP and Zn-ICOP surfaces were calculated, in terms of the contribution to the CO₂ capture capacity. Additionally, the CO₂ catalytic conversion performances of Zn-ICOP and H2-ICOP gave good yields of 83% and 54% in mild conditions, respectively.

2. Results and discussion

2.1. Synthesis and characterization

The synthesis of the ionic covalent organic polymers (H2-ICOP and Zn-ICOP) was carried out under solvothermal conditions by the Zincke reaction of 1,1′-bis(2,4-dinitrophenyl)-[4,4′-bipyridine]-1,1′-diium dichloride (Vio) with either 5,10,15,20-tetrakis(4-aminophenyl)porphyrin (H2-Porph) or 5,10,15,20-tetrakis(4-aminophenyl)porphyrinato zinc(II) (Zn-Porph) in a mixture of 1,4-dioxane : water (4 : 1, v/v) or DMF : water (4 : 1, v/v) (Scheme 1, for details see the ESI†). A gel-like material was obtained in the DMF : water mixture, whereas a powder was obtained in the 1,4-dioxane : water mixture (Fig. S1, ESI†). Washing the solids with the appropriate solvents is found to be crucial to remove unreacted starting materials and any smaller oligomers that may have formed. Therefore, the solids were washed with ethanol and DMSO several times until a colourless solution was obtained. The purification procedure was completed by washing with water and acetone. Complementary methods were used for the detailed structural characterization of the ICOPs. The formation of the ICOPs is confirmed by Fourier-transform infrared (FTIR) spectroscopy and solid-state ¹³C NMR. In FTIR (Fig. 1), the decline of the characteristic –NH₂ peak at 3345 cm⁻¹ and –NO₂ peak at 1343 cm⁻¹ and the appearance of the characteristic pyridinium C=N peak at 1628 cm⁻¹ for H2-ICOP and 1656 cm⁻¹ for Zn-ICOP indicate the loss of the 2,4-dinitroaniline group and the formation of viologen-linked ICOPs. Besides, the peaks belonging to the porphyrin ring at 964 and 790 cm⁻¹ for H2-ICOP and at 993 and 794 cm⁻¹ for Zn-ICOP confirm the formation of the expected materials. The same structures are obtained for H2-ICOP-dx and Zn-ICOP-dx with identical FTIR as shown in Fig. S2 (ESI†). The solid-state ¹³C NMR spectra (Fig. S3, ESI†) of the materials show broad multi-peaks between δ = 100 and 170 ppm. The peaks at 151 ppm for H2-ICOP and 152 ppm for Zn-ICOP can be attributed to the carbon signals in the bipyridinium moieties, whereas the peaks at 122 ppm for H2-ICOP and Zn-ICOP correspond to phenyl carbons of the porphyrin subunit.

Fig. 1 FTIR spectra of H2-ICOP and Zn-ICOP in comparison with the corresponding monomers (H2-Porph, Zn-Porph and Vio).

The diffuse reflectance spectroscopy (DRS) of H2-ICOP and Zn-ICOP displays the characteristic Soret and Q bands of the porphyrin at 436 nm and 535 nm for H2-ICOP, and 447 nm and 538 nm for Zn-ICOP (Fig. 2a). Moreover, characteristics of viologen radical bands were observed at 627 nm for H2-ICOP and 586 nm for Zn-ICOP. H2-ICOP and Zn-ICOP exhibit an electron paramagnetic resonance (EPR) signal (Fig. 2b). The radical character of these ICOPs is the result of charge transfer from the aromatic porphyrin core to the viologen unit and is an indication of the highly conjugated structures.⁵⁷

Fig. 2 DR spectra (a), EPR signals (b) and TGA curves (c) of H2-ICOP and Zn-ICOP.

Thermogravimetric analysis (TGA) shows that H2-ICOP and Zn-ICOP exhibit thermal stability up to 300 °C with slight weight losses (Fig. 2c). However, it is noteworthy that both materials retain about 60% of their initial mass even at 900 °C, indicating good thermal stability. H2-ICOP and Zn-ICOP have similar experimental PXRD profiles with low angle diffraction peaks in the vicinity of 2θ values of 5°, which is characteristic for expected large unit-cell parameters (Fig. S4, ESI†). However, the broadening of peaks in both profiles in combination with a large peak at ca. 22.5° indicates a semi-crystalline nature of the studied materials. To determine the structures of H2-ICOP and Zn-ICOP, the geometry optimization of constructed models having the eclipsed (AA) and staggered (AB) topology was performed using the ABINIT code using the following unit cell parameters: a = b = 35.563, c = 4.09 (7.8 for the staggered structure) Å, α = β = γ = 90°, space group P1. The comparisons of both the experimental and calculated PXRD profiles based on the relaxed models are presented in Fig. S4a for H2-ICOP and in Fig. S4b (ESI†) for Zn-ICOP. The experimental PXRD profiles of both ICOPs are in good agreement with the simulated staggered AB-stacking model (Fig. S4d, ESI†). Since the PXRD patterns of H2-ICOP-dx and Zn-ICOP-dx showed an amorphous nature rather than a semi-crystalline structure, H2-ICOP and Zn-ICOP were used for further investigations. Elemental mapping by energy dispersive spectroscopy (EDS) exhibits the elements C, N, and Cl for H2-ICOP and C, N, Cl, and Zn for Zn-ICOP (Fig. S5, ESI†), which is consistent with the compositions of our ICOPs. Dynamic light scattering (DLS) analyses showed two particle populations with an average colloidal particle size in the range between 1 and 10 μm for both ICOPs, suggesting a polydisperse nature (Fig. S6, ESI†). The stability of colloidal systems of H2-ICOP and Zn-ICOP was evaluated with zeta potential measurements in water (pH = 7.4) (Fig. S7, ESI†). Interestingly, the zeta potential of H2-ICOP shows a positive value (+25), whereas Zn-ICOP shows a negative zeta potential value (−21.6). These high absolute values of the zeta potential of H2-ICOP and Zn-ICOP indicate a stable state of their colloidal systems, which is also confirmed by observation of the Tyndall effect (Fig. S8, ESI†). Furthermore, the zeta potential data not only reflect the stability of the colloidal systems of H2-ICOP and Zn-ICOP dispersions in water but also can be used to explain the adsorption and conversion mechanisms of CO₂, which we will discuss in related sections.

Depending on the solvent mixture used, structures with different morphologies were obtained. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) studies reveal that the ICOPs obtained by using the DMF : water mixture (H2-ICOP and Zn-ICOP) exhibit a sheet-like morphology, whereas the ICOPs obtained by using the 1,4-dioxane : water mixture (H2-ICOP-dx and Zn-ICOP-dx) exhibit a spherical morphology (Fig. 3). Uniform spheres with an average diameter of 1.4 μm for H2-ICOP-dx and 0.65 μm for Zn-ICOP-dx were observed by both SEM (Fig. 3b and d and Fig. S9, ESI†) and TEM analysis (Fig. 3f and h).

Fig. 3 Microscopic characterization [SEM (top row) and TEM (bottom row)] of H2-ICOP (a and e) and Zn-ICOP (c and g) prepared in DMF : water (4 : 1, v/v) and obtained as plates; and H2-ICOP-dx (b and f) and Zn-ICOP-dx (d and h) prepared in 1,4-dioxane : water (4 : 1, v/v) and obtained as spheres.

2.2. CO₂ uptake and CO₂/N₂ selectivity performance

To investigate the permanent porosity of H2-ICOP and Zn-ICOP, krypton (Kr) adsorption–desorption isotherms were measured via the Brunauer–Emmett–Teller (BET) model, at 77 K (for details see the ESI†). As shown in Fig. 4b, both H2-ICOP and Zn-ICOP show typical type-II reversible adsorption isotherms and their porosities are found to be mesoporous in the range of 3–20 nm, whereas the nitrogen (N₂) adsorption–desorption isotherms show non-porous surfaces. The specific surface areas of H2-ICOP and Zn-ICOP were found to be 9 m² g⁻¹ and 20 m² g⁻¹, respectively (Fig. 4a). This difference in the BET surface areas can be explained by the zeta potential values of H2-ICOP (+25) and Zn-ICOP (−21.6) (Fig. S7, ESI†). The positive zeta potential value suggests that the pores are occupied by nearly all of the chloride counter ions and they do not remain on the surface of H2-ICOP, whereas the negative zeta potential value indicates that some chloride counter-ions remain on the surface of Zn-ICOP. Notably, the surface areas of both ICOPs are one of the lowest values reported to date for any ICOP. These relatively low values are likely due to the pores of the materials being occupied by chloride counterions and their semi-crystalline state and staggered morphology with an AB stacking sequence (Fig. S4d, ESI†).

Fig. 4 (a) Kr adsoption/desorption isotherms of H2-ICOP and Zn-ICOP measured at 77 K, filled and empty symbols represent adsorption and desorption, respectively. (b) Pore size distribution of H2-ICOP and Zn-ICOP calculated from Kr isotherms calculated using NLDFT. CO₂ adsorption isotherms of H2-ICOP (c) and Zn-ICOP (d) collected up to 1 bar at 273, 298, and 323 K. (e) CO₂/N₂ (IAST method) selectivity of H2-ICOP and Zn-ICOP at 273 K, 298 K, and 323 K. (f) The isosteric heat of adsorption (Q_st) plots of CO₂ for H2-ICOP and Zn-ICOP.

Owing to their multiple CO₂-philic sides, despite their low surface areas, we explored the affinity of H2-ICOP and Zn-ICOP toward CO₂ and N₂. To evaluate their performance for CO₂ capture, we carried out temperature-dependent CO₂ uptake measurements at 273, 298, and 323 K up to 1 bar (Fig. 4c and d). The CO₂ capture capacity of H2-ICOP, despite the lower BET surface area, is found to be up to 26 (42.4 mg g⁻¹, 273 K), 44 (62.9 mg g⁻¹, 298 K), and 13 cm³ g⁻¹ (17.3 mg g⁻¹, 323 K) at 1 bar, which is much higher than that of Zn-ICOP, 25 (39.3 mg g⁻¹, 273 K), 16.7 (24 mg g⁻¹, 298 K) and 10.2 cm³ g⁻¹ (13.9 mg g⁻¹, 323 K) at 1 bar, respectively. Notably, the CO₂ uptake capacities are not saturated for H2-ICOP within the pressure range, suggesting that the CO₂ uptake capacities can be further improved with increased pressure. Remarkably, the CO₂ capture capacities of H2-ICOP and Zn-ICOP are one of the highest values reported among previously reported viologen based ICOFs and ICOPs with much larger surface areas (Table 1). The CO₂ capture capacity of these viologen linked ICOFs and ICOPs is attributed mainly to tetrel bonding interactions between the counter anions and CO₂. However, as the positive zeta potential value of H2-ICOP shows that the pores are occupied by nearly all of the chloride counter ions and they do not remain on the surface, yet H2-ICOP has a very high CO₂ capture capacity, the results indicate that H2-ICOP, due to our molecular design, possesses much stronger non-covalent interactions with CO₂ than the counter anions.

Table 1 BET surface area, CO₂ uptake, selectivity, and heat adsorptions (Q_st) of the ICOPs

ICOP	BET surface area (m^2 g^−1)	CO2 adsorption (mg g^−1)			CO2/N2 selectivity (15/85) (IAST) 1 bar		Q st (kJ mol^−1)	Pressure (bar)	Ref.
		273 K	298 K	323 K	273 K	298 K
H2-ICOP	9	42.4	62.9	17.3	56.5	35	39	1	This work
Zn-ICOP	20	39.3	24	13.9	5.6	6.9	36	1	This work
PCP-Cl	755	101	61.4	—	34	—	28.5	1	39
V-PCIF-Cl	174	86.7	62.4	—	—	—	56.6	1	43
POV-V1	812	—	40.5	—	—	—	25	1.1	49
POV-V2	960	—	55.9	—	—	—	24	1.1	49
c-CTF-400	744	126	83	52	—	—	49	1	34
c-CTF-450	861	99	62	38	—	—	46	1	34
c-CTF-500	1247	133	80	47	—	—	43	1	34
SYSU-Zn@IL1	38	68	40	—	—	—	27.2	1	47

---

To gain further insights from the non-covalent interactions between the materials and CO₂ molecules, the CO₂ isosteric heat of adsorption (Q_st) of H2-ICOP and Zn-ICOP was obtained with the Clausius–Clapeyron equation from the CO₂ uptake isotherms recorded (Fig. 4f). The Q_st values for CO₂ were found to be 39 and 36 kJ mol⁻¹ for H2-ICOP and Zn-ICOP, respectively, at zero coverage. This difference of the Q_st value for the two polymers shows that the non-covalent interactions of the CO₂ molecules with the skeleton of H2-ICOP are more efficient than with Zn-ICOP, which may explain the higher CO₂ capture capacity of H2-ICOP compared to Zn-ICOP. It is also worth mentioning that the Q_st values of the polymers are moderate, not exceeding 50 kJ mol⁻¹, which implies that CO₂ is physically adsorbed by both ICOPs and the recyclability of the adsorbent at a low energy penalty.

Additionally, to assess the potential in separation of CO₂ from flue gas (>70% N₂), the adsorption selectivity of CO₂ over N₂ (15/85) of H2-ICOP and Zn-ICOP was calculated with Myers and Prausnitz's ideal adsorbed solution theory (IAST) at three different temperatures (273, 298, and 320 K) up to 1 bar (Fig. 4e and Table 1).⁵⁸ The selectivity of CO₂ from CO₂/N₂ (15/85) of H2-ICOP and Zn-ICOP is 55.7 and 5.5 at 273 K, 35.1 and 7.2 at 298 K, and 5.1 and 2.55 at 323 K, respectively. As expected, H2-ICOP exhibits much higher selectivity compared to Zn-ICOP, which also indicates powerful CO₂ affinity of H2-ICOP.

2.3. Computational analysis of non-covalent bonds

The high CO₂ capture and selectivity capacities of H2-ICOP and Zn-ICOP, despite their low surface areas, clearly show the powerful non-covalent interactions between CO₂ molecules and multiple CO₂-philic moieties of H2-ICOP and Zn-ICOP. Theoretical analyses (for details of the computational methodology see the ESI†) have been performed to get insights into the possible non-covalent interactions between CO₂ molecules and the surfaces of H2-ICOP and Zn-ICOP, and their contribution to the selective CO₂ uptake of these compounds.

To model the mentioned non-covalent interactions, the fragments H2-ICOP and Zn-ICOP were ‘cut off’ from the polymeric chains and the linkers were replaced by hydrogen atoms. H2-ICOP and Zn-ICOP contain two porphyrin cores, a chlorine counterion, and a positively charged viologen linker. The σ-hole formation on the surface of H2-ICOP, Zn-ICOP, and CO₂ has been visualized using electrostatic potential maps (MEP) and the results are displayed in Fig. 5 along with the corresponding electrostatic potential values. The calculated MEP surfaces show that a maximum of positive potential (σ-hole) involves a region with the cationic viologen subunits and Zn-ICOP porphyrin core (Zn) as well as the carbon atom C_CO2, which shows the Lewis acid character of these regions. The negative electrostatic potential is located in the vicinity of oxygen O_CO2, chlorine atoms, and nitrogen atoms in the H2-ICOP porphyrin core (Fig. 6). Thus, the positive and negative MEP distributions of H2-ICOP, Zn-ICOP, and CO₂ indicate that attractive interactions are favourable for the formation of the above mentioned non-covalent bonds.

Fig. 5 Electrostatic potentials mapped on the molecular surfaces of CO₂, H2-ICOP and Zn-ICOP. The values of the MEPs lie in the intervals −7.794, +7.794 kcal mol⁻¹ for CO₂ and −18.825, +18.825 kcal mol⁻¹ for H2-ICOP and Zn-ICOP.

Fig. 6 Scheme of possible non-covalent interactions between CO₂ and H2-ICOP and Zn-ICOP.

The MEP analyses have revealed possible non-covalent interactions, such as hydrogen, tetrel, pnicogen, and spodium bonds formed between CO₂ molecules and the surfaces of H2-ICOP and Zn-ICOP, and their contribution to the selective CO₂ uptake of these compounds (Fig. 6).

These noncovalent interactions can be uniformly defined based on the positive electrostatic region present on an atom due to the anisotropic distribution of the electron density. A positive electrostatic region present on an atom along a σ-polymer is called a σ-hole. The size and magnitude of a σ-hole are dependent on both the nature of its atom and the electron-withdrawing ability of groups attached to this atom. These σ-holes can have an attractive interaction with electron donors such as Lewis bases, anions or radicals. Depending on the σ-hole origins, these interactions are called tetrel bonding (group IV), pnicogen bonding (group V), chalcogen bonding (group VI), halogen bonding (group VII) or spodium bonding (group XII).⁵⁹

Symmetry adapted perturbation theory (SAPT) analysis, which partitions the attractive forces into electrostatic (E_elst), exchange-repulsion (E_exch), induction (E_ind), and dispersion (E_disp) terms, has been carried out to calculate the interaction energy of hydrogen (O_CO2⋯H), tetrel (C_CO2⋯N, C_CO2⋯Cl⁻), pnicogen (O_CO2⋯N⁺), and spodium bonds (O_CO2⋯Zn) in H2-ICOP and Zn-ICOP and estimate the preferable binding sites to capture CO₂ molecules on their surfaces (Table 3). For all studied non-covalent interactions, the calculated attractive components E_elst, E_ind, and E_disp provide sufficient stabilization to overcome the repulsive exchange component E_exch, and the resultant interaction energy (E_int) values are negative, except hydrogen bonding. The value of E_exch for hydrogen bonds in H2-ICOP is very large and suppresses the other attractive components, and therefore the resulting interaction energy is positive, which indicates that the formation of H-bonds between CO₂ and –NH (porp) is not favourable (Table 2 and Fig. 7d). Dispersion forces (E_disp) are found to be dominant in the attractive binding energies of pnicogen (O_CO2⋯N⁺) and C_CO2⋯N tetrel bonds and spodium bonds (O_CO2⋯Zn), while electrostatic interactions (E_elst) dominate in C_CO2⋯Cl⁻ tetrel bonds in both compounds. The percentage contribution of the dispersion term in the overall attractive forces for pnicogen bonds consists of 73% in CO₂-H2-ICOP and CO₂-Zn-ICOP, while for C_CO2⋯N tetrel bonds in these systems it is equal to 58% and 66%, respectively. For spodium bonds the dispersion term constitutes 53%. For C_CO2⋯Cl⁻ tetrel bonds in both compounds the percentage contribution of electrostatic interactions in the overall attractive forces consists of 58%.

Table 2 Summary of the SAPT results (kJ mol⁻¹) for pnicogen (O_CO2⋯N⁺), tetrel (C_CO2⋯N, C_CO2⋯Cl⁻), hydrogen (H⋯O_CO2) and spodium (Zn⋯O_CO2) bonds in CO₂-H2-ICOP and CO₂-Zn-ICOP. R is the bond length (Å)

Compound	LB⋯LA	Bond type	R	E elst	E exch	E ind	E disp	E int
a The D–H⋯A angles for both CO2 are 125.8 and 118.7°.
H2-ICOP	OCO2⋯N^+	Pnicogen	3.115	−6.1	16.8	−2.0	−22.2	−13.6
	Cl^−⋯CCO2	Tetrel	3.42	−16.3	11.0	−3.8	−7.9	−17.0
	N⋯CCO2	Tetrel	2.976	−21.7	42.3	−5.5	−37.1	−22.0
	OCO2^a⋯H	Hydrogen	1.961	−177.6	398.9	−21.2	−105.9	+94.3
Zn-ICOP	OCO2⋯N^+	Pnicogen	3.115	−6.2	16.8	−2.0	−22.3	−13.7
	Cl^−⋯CCO2	Tetrel	3.42	−16.4	11.0	−3.8	−7.9	−17.1
	N⋯CCO2	Tetrel	2.976	−14.2	42.7	−5.2	−38.4	−15.2
	OCO2⋯Zn	Spodium	2.6	−19.8	30.2	−5.8	−28.8	−21.1

---

Fig. 7 Formation of (a, e, c and g) tetrel, (b and f) pnicogen, (d) hydrogen and (h) spodium bonds in H2-ICOP and Zn-ICOP.

The strength of the non-covalent interactions in CO₂-H2-ICOP and CO₂-Zn-ICOP is decreasing in the following order ΔE_C⋯N > ΔE_C⋯Cl⁻> ΔE_O⋯N⁺ and ΔE_Zn⋯O > ΔE_C⋯Cl⁻ > ΔE_C⋯N > ΔE_O⋯N⁺, respectively (Table 2 and Fig. 7a–h). The binding energy of C_CO2⋯N tetrel bonds in CO₂-H2-ICOP is −22.02 kJ mol⁻¹ and is the strongest among the studied non-covalent interactions. Interestingly, upon Zn metalation in the porphyrin core, the binding energy of C_CO2⋯N tetrel bonds in CO₂-Zn-ICOP is decreased to −15.163 kJ mol⁻¹, probably due to an increase of the positive electrostatic potential around nitrogen atoms (Fig. 6) in the porphyrin core, which resulted in the formation of weaker tetrel bonds (C_CO2⋯N) with CO₂. In addition, further analyses were performed to calculate the binding energy of the simultaneous presence, on the same porphyrin core, of two C_CO2⋯N tetrel bonds in H2-ICOP and spodium and C_CO2⋯N tetrel bonds in Zn-ICOP. The analyses reveal that the simultaneous presence of spodium and C_CO2⋯N tetrel bonds is impossible on the same porphyrin core in Zn-ICOP because of the strong repulsion between two CO₂ molecules (Table S1 and Fig. S10, ESI†), while two tetrel bonds may form in the H2-ICOP porphyrin core because of negligible repulsion between CO₂ molecules (Fig. S11 and Table S1, ESI†). Overall, in our fragment model, H2-ICOP can potentially form two pnicogen, two C_CO2⋯Cl⁻ and four C_CO2⋯N tetrel bonds with CO₂, while Zn-ICOP can form two pnicogen, two C_CO2⋯Cl⁻ tetrel, and two spodium bonds with CO₂. The major CO₂ uptake contribution comes from the four C_CO2⋯N tetrel bonding interactions for H2-ICOP, whereas the two spodium bonding interactions for Zn-ICOP. In conclusion, H2-ICOP has more binding sites with more powerful non-covalent interactions that lead to an enhancement of the CO₂ uptake compared to Zn-ICOP, which is in good agreement with the experimental results.

2.4. The catalytic performance in CO₂ conversion into cyclic carbonates

To evaluate the catalytic activities of viologen-linked porphyrin H2-ICOP and Zn-ICOP, the cycloaddition of CO₂ to propylene oxide was chosen as a model reaction and the results are listed in Table 3. The reactions were conducted under 1.0 MPa CO₂ pressure and solvent and co-catalyst free conditions at 80 °C (for details see the ESI†).

Table 3 The cycloaddition of CO₂ with propylene oxide (PO) catalysed by various ICOPs

Catalyst	Loading	T (°C)	P CO2 (bar)	t (h)	Yield (%)	Ref.
a This work (yields are determined by ^1H NMR).
CTF-P-HSA	0.1 g	130	6.9	4	83	60
cCTF	4 wt%	90	10	12	99	34
PCP-Cl	5 wt%	100	30	12	99	39
Zn-CIF2-C2H5	0.18 mol%	120	25	5	98	61
TBB-Bpy-a	0.08 g	120	10	4	99	48
COF-salen-Zn	0.1 mol%	100	20	3	90	62
SYSU-Zn@IL1	0.16 mol%	80	10	12	62	47
H2-ICOP	0.16 mol%	80	10	12	54
Zn-ICOP	0.16 mol%	80	10	12	83

---

The compared catalysts Zn-ICOP and H2-ICOP gave yields of 83% and 54%, respectively. Surprisingly, Zn-ICOP, despite having a significantly lower CO₂ uptake capacity, displayed better catalytic activity than H2-ICOP. The high activity of Zn-ICOP can be attributed to the synergistic effect of Lewis acid Zn sites and a nucleophile counter anion (Cl⁻). The zeta potential data revealed that Zn-ICOP possesses a counter anion (Cl⁻) on its surface, whereas H2-ICOP does not, which indicates the crucial role of the counter anion (Cl⁻) in the mechanism of this catalytic reaction.^46,47 Notably, the catalytic activity of Zn-ICOP was achieved with lowered temperature and CO₂ pressure, in a solvent-free reaction system, and was comparable to that of most of the reported examples (Table 3).

The catalytic performance of Zn-ICOP was compared with that of the corresponding polymer SYSU-Zn@IL1(Br⁻) since both structures are based on Zn–porphyrin and viologen. SYSU-Zn@IL1, with a BET surface area of 38 m² g⁻¹, has a nearly identical chemical composition to Zn-ICOP. Under similar conditions, Zn-ICOP had 83% conversion of PO, whereas SYSU-Zn@IL1(Br⁻) could only afford a PO conversion of 62%. The higher activity of Zn-ICOPz(Cl⁻) is mainly attributed to the higher nucleophilicity of its counter anions (Cl⁻ > Br⁻).

Under the optimized reaction conditions, the catalytic activity of Zn-ICOP was further studied over a variety of epoxides. As shown in Table 4, various epoxides are efficiently converted into the desired products with moderate yields under mild conditions by Zn-ICOP. With the increase of the size of epoxides, a steady decrease in the yield of cyclic carbonates was observed from PO (83%) and 1,2-epoxybutane (42%) to styrene oxide (4%), likely owing to their increasing steric hindrance, except for epichlorohydrin (77%) (Table 4). The deviation of epichlorohydrin's yield from the trend, despite its larger size, is probably due to additional non-covalent interactions between the chlorine functional group and CO₂.

Table 4 The addition of CO₂ to various epoxides by the Zn-ICOP catalyst. Reaction conditions: epoxide (1 mL), catalyst (0.16 mol%), 80 °C, 1 MPa, 12 h

Epoxide
Product
Conversion (%)	83	42	77	4

---

3. Conclusion

Two viologen-porphyrin based ionic covalent organic polymers were constructed by a one-pot facile Zincke reaction. The BET surface area analyses reveal that both H2-ICOP (9 m² g⁻¹) and Zn-ICOP (20 m² g⁻¹) possess a very low surface area. Yet, despite their nearly non-porous surfaces, especially H2-ICOP exhibits a very large CO₂ capture capacity and CO₂/N₂ selectivity, which indicates efficient non-covalent interactions with CO₂. The theoretical analyses show that H2-ICOP can potentially form two pnicogen, two C_CO2⋯Cl⁻ and four C_CO2⋯N tetrel bonds with CO₂, while Zn-ICOP can form two pnicogen, two C_CO2⋯Cl⁻ tetrel, and two spodium bonds with CO₂. The strength of the non-covalent interactions in H2-ICOP and Zn-ICOP is decreasing in the following order ΔE_C⋯N > ΔE_C⋯Cl⁻ > ΔE_O⋯N⁺ and ΔE_Zn⋯O > ΔE_C⋯Cl⁻ > ΔE_C⋯N > ΔE_O⋯N⁺, respectively. The binding energy of C_CO2⋯N tetrel bonds in CO₂-H2-ICOP is −22.02 kJ mol⁻¹ and is the strongest among the studied non-covalent interactions. The major CO₂ uptake contribution comes from the four C_CO2⋯N tetrel bonding interactions for H2-ICOP, whereas the two spodium bonding interaction for Zn-ICOP. Furthermore, the CO₂ catalytic conversion performances of Zn-ICOP and H2-ICOP gave encouraging yields of 83% and 54%, respectively. Surprisingly, Zn-ICOP, even with a significantly lower CO₂ uptake capacity, displayed better catalytic activity than H2-ICOP, probably due to the crucial role of the counter anion (Cl⁻) in the mechanism of this catalytic reaction.

Overall, we have demonstrated here that ICOPs with multiple CO₂-philic sites can indeed exhibit very large CO₂ capture capacity, CO₂/N₂ selectivity, and highly efficient catalytic activity of CO₂ cycloaddition of epoxides, owing to efficient non-covalent interactions with CO₂, in spite of low surface areas. These results set a useful example for the importance of non-covalent interactions with CO₂ for the capture and conversion of CO₂ into value-added products. Therefore, we believe that these findings will prove to be of key importance to the design of bifunctional charged materials for CCU applications.

Author contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors thank the TUBITAK ULAKBIM High Performance and Grid Computing Center (TRUBA resources), Turkey, for computational facilities. Y. Zorlu acknowledges financial support from The Science Academy, Young Scientist Award (BAGEP).

References

1. V. Scott, S. Gilfillan, N. Markusson, H. Chalmers and R. S. Haszeldine, Nat. Clim. Change, 2013, 3, 105–111.
2. R. M. Cuéllar-Franca and A. Azapagic, J. CO2 Util., 2015, 9, 82–102.
3. A. Modak, P. Bhanja, S. Dutta, B. Chowdhury and A. Bhaumik, Green Chem., 2020, 22, 4002–4033.
4. C. Hepburn, E. Adlen, J. Beddington, E. A. Carter, S. Fuss, N. Mac Dowell, J. C. Minx, P. Smith and C. K. Williams, Nature, 2019, 575, 87–97.
5. G. Singh, J. Lee, A. Karakoti, R. Bahadur, J. Yi, D. Zhao, K. Albahily and A. Vinu, Chem. Soc. Rev., 2020, 49, 4360–4404.
6. M. Bui, C. S. Adjiman, A. Bardow, E. J. Anthony, A. Boston, S. Brown, P. S. Fennell, S. Fuss, A. Galindo, L. A. Hackett, J. P. Hallett, H. J. Herzog, G. Jackson, J. Kemper, S. Krevor, G. C. Maitland, M. Matuszewski, I. S. Metcalfe, C. Petit, G. Puxty, J. Reimer, D. M. Reiner, E. S. Rubin, S. A. Scott, N. Shah, B. Smit, J. P. M. Trusler, P. Webley, J. Wilcox and N. Mac, Dowell, Energy Environ. Sci., 2018, 11, 1062–1176.
7. G. Sneddon, A. Greenaway and H. H. P. Yiu, The Potential Applications of Nanoporous Materials for the Adsorption, Separation, and Catalytic Conversion of Carbon Dioxide, Adv. Energy Mater., 2014, 4, 1301873–1301892.
8. A. A. Olajire, J. CO2 Util., 2017, 17, 137–161.
9. R. Dawson, E. Stöckel, J. R. Holst, D. J. Adams and A. I. Cooper, Energy Environ. Sci., 2011, 4, 4239–4245.
10. P. Bhanja, A. Modak and A. Bhaumik, ChemCatChem, 2019, 11, 244–257.
11. W. Wang, M. Zhou and D. Yuan, J. Mater. Chem. A, 2017, 5, 1334–1347.
12. H. A. Patel, J. Byun and C. T. Yavuz, ChemSusChem, 2017, 10, 1303–1317.
13. R. Bera, M. Ansari, A. Alam and N. Das, J. CO2 Util., 2018, 28, 385–392.
14. M. Cox and R. Mokaya, Sustainable Energy Fuels, 2017, 1, 1414–1424.
15. S. K. Shukla and J. P. Mikkola, Phys. Chem. Chem. Phys., 2018, 20, 24591–24601.
16. K. M. De Lange and J. R. Lane, J. Chem. Phys., 2011, 135, 0–8.
17. G. Chang, Z. Shang, T. Yu and L. Yang, J. Mater. Chem. A, 2016, 4, 2517–2523.
18. M. Li, J. Lei, G. Feng, J. U. Grabow and Q. Gou, Spectrochim. Acta, Part A, 2020, 238, 118424.
19. A. C. Legon, Phys. Chem. Chem. Phys., 2017, 19, 14884–14896.
20. X. Tan, L. Kou, H. A. Tahini and S. C. Smith, Sci. Rep., 2015, 5, 1–8.
21. D. Nam, P. De Luna, A. Rosas-hernández, A. Thevenon, F. Li, T. Agapie, J. C. Peters, O. Shekhah, M. Eddaoudi and E. H. Sargent, Molecular enhancement of heterogeneous CO₂ reduction, Nat. Mater., 2020, 19, 266–276.
22. B. Dash, J. Mol. Model., 2018, 24, 1–8.
23. P. Kanoo, R. Matsuda, H. Sato, L. Li, N. Hosono and S. Kitagawa, Chem. – Eur. J., 2020, 26, 2148–2153.
24. S. S. Rao and S. P. Gejji, J. Phys. Chem. A, 2016, 120, 1243–1260.
25. L. M. Azofra and S. Scheiner, Tetrel, chalcogen, and CH⋅⋅⋅O hydrogen bonds in complexes pairing carbonylcontaining molecules with 1, 2, and 3 molecules of CO₂, J. Chem. Phys., 2015, 142, 034307-1–034307-9.
26. L. Lodeiro, R. Contreras and R. Ormazábal-Toledo, J. Phys. Chem. B, 2018, 122, 7907–7914.
27. X. Y. Luo, X. Fan, G. L. Shi, H. R. Li and C. M. Wang, J. Phys. Chem. B, 2016, 120, 2807–2813.
28. H. Y. Li, Y. X. Lu, X. Zhu, C. J. Peng, J. Hu, H. L. Liu and Y. Hu, Sci. China: Chem., 2012, 55, 1566–1572.
29. X. Zhu, Y. Lu, C. Peng, J. Hu, H. Liu and Y. Hu, J. Phys. Chem. B, 2011, 115, 3949–3958.
30. V. Chernikova, O. Shekhah, Y. Belmabkhout and M. Eddaoudi DOI:10.1021/acsanm.0c00909.
31. S. K. Das, P. Bhanja, S. K. Kundu, S. Mondal and A. Bhaumik, ACS Appl. Mater. Interfaces, 2018, 10, 23813–23824.
32. C. Liao, Z. Liang, B. Liu, H. Chen, X. Wang and H. Li, ACS Appl. Nano Mater., 2020, 3, 2889–2898.
33. Y. Fu, Z. Wang, S. Li, X. He, C. Pan, J. Yan and G. Yu, ACS Appl. Mater. Interfaces, 2018, 10, 36002–36009.
34. O. Buyukcakir, S. H. Je, S. N. Talapaneni, D. Kim and A. Coskun, ACS Appl. Mater. Interfaces, 2017, 9, 7209–7216.
35. Y. P. Zhao, Y. Li, C. Y. Cui, Y. Xiao, R. Li, S. H. Wang, F. K. Zheng and G. C. Guo, Inorg. Chem., 2016, 55, 7335–7340.
36. J. Wang, W. Sng, G. Yi and Y. Zhang, Chem. Commun., 2015, 51, 12076–12079.
37. A. E. Sadak, Microporous Mesoporous Mater., 2021, 311, 110727.
38. V. A. Online, M. M. Unterlass, F. Emmerling, M. Antonietti and J. Weber, From dense monomer salt crystals to CO₂ selective microporous polyimides via solid-state polymerization, Chem. Commun., 2014, 50, 430–432.
39. O. Buyukcakir, S. H. Je, D. S. Choi, S. N. Talapaneni, Y. Seo, Y. Jung, K. Polychronopoulou and A. Coskun, Chem. Commun., 2016, 52, 934–937.
40. J. Ding, C. Zheng, L. Wang, C. Lu, B. Zhang, Y. Chen, M. Li, G. Zhai and X. Zhuang, J. Mater. Chem. A, 2019, 7, 23337–23360.
41. C. Hua, B. Chan, A. Rawal, F. Tuna, D. Collison, J. M. Hook and D. M. D’Alessandro, J. Mater. Chem. C, 2016, 4, 2535–2544.
42. W. Hui, X. M. He, X. Y. Xu, Y. M. Chen, Y. Zhou, Z. M. Li, L. Zhang and D. J. Tao, J. CO2 Util., 2020, 36, 169–176.
43. G. Chen, X. Huang, Y. Zhang, M. Sun, J. Shen, R. Huang, M. Tong, Z. Long and X. Wang, Chem. Commun., 2018, 54, 12174–12177.
44. Y. Zhang, K. Zhang, L. Wu, K. Liu, R. Huang, Z. Long, M. Tong and G. Chen, RSC Adv., 2020, 10, 3606–3614.
45. T. Škorjanc, D. Shetty, M. A. Olson and A. Trabolsi, ACS Appl. Mater. Interfaces, 2019, 11, 6705–6716.
46. Y. Zhang, K. Liu, L. Wu, H. Zhong, N. Luo, Y. Zhu, M. Tong, Z. Long and G. Chen, ACS Sustainable Chem. Eng., 2019, 7, 16907–16916.
47. Y. Chen, R. Luo, Q. Xu, J. Jiang, X. Zhou and H. Ji, ACS Sustainable Chem. Eng., 2018, 6, 1074–1082.
48. Y. Leng, D. Lu, P. Jiang, C. Zhang, J. Zhao and W. Zhang, Catal. Commun., 2016, 74, 99–103.
49. C. Hua, B. Chan, A. Rawal, F. Tuna, D. Collison, J. M. Hook and D. M. D’Alessandro, J. Mater. Chem. C, 2016, 4, 2535–2544.
50. T. Lu, J. Zhang, Q. Gou and G. Feng, Phys. Chem. Chem. Phys., 2020, 22, 8467–8475.
51. S. J. Grabowski, Struct. Chem., 2019, 30, 1141–1152.
52. T. Skorjanc, D. Shetty, F. Gándara, L. Ali, J. Raya, G. Das, M. A. Olson and A. Trabolsi, Chem. Sci., 2020, 11, 845–850.
53. A. Bauzá, I. Alkorta, J. Elguero, T. J. Mooibroek and A. Frontera, Angew. Chem., 2020, 132, 17635–17640.
54. N. Manoranjan, D. H. Won, J. Kim and S. I. Woo, J. CO2 Util., 2016, 16, 486–491.
55. S. Zahedi and E. Safaei, J. CO2 Util., 2020, 42, 101308.
56. X. Jiang, F. Gou, X. Fu and H. Jing, J. CO2 Util., 2016, 16, 264–271.
57. G. Das, T. Skorjanc, S. K. Sharma, F. Gándara, M. Lusi, D. S. Shankar Rao, S. Vimala, S. Krishna Prasad, J. Raya, D. S. Han, R. Jagannathan, J. C. Olsen and A. Trabolsi, J. Am. Chem. Soc., 2017, 139, 9558–9565.
58. A. L. Myers and J. M. Prausnitz, AlChE J., 1965, 11, 121–127.
59. R. Parajuli, Curr. Sci., 2016, 110, 495–498.
60. J. Roeser, K. Kailasam and A. Thomas, ChemSusChem, 2012, 5, 1793–1799.
61. J. Liu, G. Zhao, O. Cheung, L. Jia, Z. Sun and S. Zhang, Chem. – Eur. J., 2019, 25, 9052–9059.
62. H. Li, X. Feng, P. Shao, J. Chen, C. Li, S. Jayakumar and Q. Yang, J. Mater. Chem. A, 2019, 7, 5482–5492.

---

Footnote

† Electronic supplementary information (ESI) available: Synthesis and analysis details; photographs of gel-like or powder ICOPs; FTIR of H2-ICOP-dx and Zn-ICOP-dx; CP/MAS ¹³C-NMR; PXRD with simulated patterns, EDS, DLS and zeta potentials for H2-ICOP and Zn-ICOP; SEM images of H2-ICOP-dx and Zn-ICOP-dx; formation of simultaneous tetrel bonds in H2-ICOP and Zn-ICOP; SAPT results for the interaction between two CO₂ molecules placed at the same porphyrin fragments in H2-ICOP and Zn-ICOP. See DOI: 10.1039/d1ma00217a

---