Selective alcohol adsorption in a uniformly ordered array of lipophilic mesopores by a giant macrocycle

Abstract

The crystal structure of giant macrocycle 1 was found to have a mesoporous structure. The lipophilic pores of 1 uniformly formed in an ordered array, and contributed to the selective adsorption of 1-BuOH among the isomers of butanol.

---

The sorption of porous materials is a fundamental property, providing a wide range of applications, e.g., gas storage, separation, and catalysis. Recently, considerable research on porous materials has focused particular attention on porous coordination polymers, because their topology of structures and pore sizes can be rationally designed by choosing the coordination ligands and metal ions to provide a certain functionality.¹ Another way to form porous materials is the use of a macrocyclic molecule. One of the advantages of using a macrocycle for sorption is that a macrocycle itself can act as a host molecule for capturing a guest molecule. Another advantage is its light weight and accessibility to obtain quantitative information on host–guest interactions in solution. A rather limited number of macrocyclic molecules, such as calixarene,² cucurbiturils,³ cyclotriphospazene,⁴ and imine-linked cages,^5–8 have been explored as porous materials for gas storage and separation. A growing number of porous macrocyclic molecules in this area have been developed to have a high surface area⁵ and selective CO₂ adsorption.^5–7

Although gas storage and separation are crucial characteristics for the improvement of a porous macrocyclic molecule, combining quantitative information of guest recognition in solution with guest adsorption in the solid state appears to be possible from the viewpoint of host–guest chemistry. Also, with the use of a large macrocyclic molecule, it raises an intriguing point that the cage of a macrocycle actually acts as a pore for adsorption in the solid state. We previously reported the synthesis of giant macrocyclic molecule 1.⁹ The most revealing property of 1 is its swelling with the guest molecule in solution, that is, increasing the volume of the molecule by complexation with the guest. In this paper, we report that molecule 1 has an affinity for adsorbing alcohol molecules in the solid state. The pore size of the porous host 1 was characterized by N₂ sorption isotherms, and the pore structure of 1 was revealed using theoretical calculations. Furthermore, interactions of porous host 1 with MeOH and EtOH molecules were found to be almost the same as estimated by the isosteric heats. The selective adsorption of 1 for 1-BuOH among several isomeric alcohols was also discussed.

Macrocyclic molecule 1 was synthesized by the reaction of dynamic self-assembled covalent bond formation between tris(2-aminoethyl)amine and trans-[bis(4-formylphenylethynyl)bis(triethyl-phosphine)]platinum(II).⁹ Molecule 1 has a giant tetrahedral structure consisting of imino groups at the vertices and acetylide groups at the edges (Fig. 1a). The average length of the edges is ca. 23 Å. Since molecule 1 has a single bond at the platinum acetylide moiety, the PEt₃ moiety can freely rotate around the acetylide edges. This movement of the phosphine ligands, going into and out of the cavity, enables accommodation of solvent molecules in the cavity of 1 when in solution. We estimated the host–guest complexation of 1 with various solvents by measuring the diffusion coefficient (D) using dynamic light scattering and pulsed gradient spin-echo NMR methods. The obtained value of D for 1 varied in different solvents such as dichloromethane, tetrahydrofuran, p-dioxane, benzene, and other aromatic solvents. We found that the D value for 1 has a tendency to decrease with increasing volume of solvent. For example, the D value for 1 in dichloromethane was 4.44 × 10⁻¹⁰ m² s⁻¹, while that for 1 in butylbenzene was 1.44 × 10⁻¹⁰ m² s⁻¹. The calculated hydrodynamic radius obtained from the D value increases with increasing volume of solvent. Thus, macrocycle 1 showed swelling with the solvent molecules by host–guest complexation in solution.

Fig. 1 (a) Chemical structure of macrocycle 1. (b) The X-ray crystal structure of 1·MeOH. Hydrogen atoms except MeOH omitted for clarity. Color coding: C: gray; H: white; N: blue; O: red; P: green; Pt: orange.

A crystal of macrocycle 1 was obtained by recrystallization from a mixed solvent of CH₂Cl₂ and MeOH. The crystal structure of 1·MeOH was previously reported⁹ as shown in Fig. 1b. The crystal of 1 was found to have a void, and a MeOH molecule was included close to the ligand of the PEt₃ moiety. However, the obtained crystal gradually collapsed when the crystal was removed from the solvent. The weight of the crystals decreased with increasing time of exposure to air, and finally reached a constant weight. The powder X-ray diffraction pattern of the crystal of 1 in the absence of MeOH was almost the same as the simulated powder pattern for the crystal of 1·MeOH (Fig. 2). This observation indicated that the molecular structure of 1 was sustained even if a MeOH molecule was removed from the crystals.

Fig. 2 (a) Powder XRD pattern of 1. (b) Simulated powder pattern of the crystals of 1·MeOH calculated from the crystallographic data.

Our observations of the crystal of 1·MeOH suggest that the crystal of 1 gains a void by the release of a MeOH molecule. Therefore, we decided to utilize the void of 1 as the pores for guest adsorption in the solid state. First, we estimated the surface area and the pore size of 1. The N₂ sorption isotherm of 1 is shown in Fig. 3a. The saturated adsorbed amount of N₂ was 15.3 cm³ g⁻¹ at p/p₀ ≈ 0.92. Although the amount of adsorbed N₂ was moderate, the obtained isotherm contains a hysteresis loop during the adsorption and desorption cycling. We found that the observed isotherm is classified as Type IV according to the classification of adsorption isotherms by the IUPAC.^10,11 This Type IV isotherm is considered to result from a mesoporous structure in which the pore widths are in the range of 2–50 nm. Furthermore, this Type IV isotherm is associated with capillary condensation taking place in the mesopores. The initial stage of the isotherm is attributed to a monolayer–multilayer adsorption. For 1, the N₂ molecule appears to be adsorbed in a monolayer–multilayer fashion around p/p₀ ≈ 0.2. Based on this adsorption isotherm, a Brunauer–Emmett–Teller (BET) plot¹² was obtained (Fig. S1†). A linear relationship was obtained from 0.05 to 0.30 in relative pressure, and the BET surface area was found to be 11.8 m² g⁻¹. We next determined the pore size of 1 using the Cranston–Inkley method.¹³ For the process of adsorption, the pore diameter was in the range of 1.6 nm to 7.98 nm (Fig. 3b). The peak top of the pore diameter appeared at 1.85 nm and 3.09 nm, namely, the crystal of 1 has a mesoporous structure. The cumulative surface area was 13.2 m² g⁻¹, in agreement with the BET analysis result.

Fig. 3 (a) N₂ sorption isotherms of 1 at 77 K. Filled circles (●) indicate adsorption, and filled triangles (▲) correspond to desorption. (b) Pore size distribution of 1 obtained from N₂ adsorption.

The shape of the hysteresis loop has often been linked to a specific pore structure.¹⁰ The observed hysteresis loop of the N₂ isotherm resembles a Type H1 class, implying that the pore structures of 1 are compacts of approximately uniform spheres in a regular array. This result agrees with the crystallinity of 1 as proved by the powder X-ray diffraction pattern (Fig. 2). To gain information about the pore structure of 1, we calculated the voids of the crystal of 1·MeOH (Fig. 4). The crystal of 1·MeOH contains a void volume of 1094 ų, which corresponds to 9.6% of the unit cell. Based on the assumption of the persistent cell parameter from 1·MeOH to 1, the voids of the crystal of 1 could be estimated to have a volume of 1161 ų, about 10.2% of the unit cell. The voids of 1·MeOH have an approximately uniform structure in an ordered array (Fig. 4).

Fig. 4 Viewed along the crystallographic a axis and showing the crystal packing of 1·MeOH. The yellow surfaces correspond to the crystal voids. Color coding: C: gray; N: blue; O: red; P: green; Pt: orange.

The crystal of 1 has a mesoporous structure and includes guest molecules in the presence of MeOH vapor. Therefore, we investigated the adsorption property of 1 for several kinds of alcohols. Fig. 5a shows the adsorption isotherms of 1 for MeOH and EtOH. At a temperature of 288 K, the saturated adsorbed amount of the alcohol was 59.0 cm³ g⁻¹ and 31.5 cm³ g⁻¹ for MeOH and EtOH at p/p₀ ≈ 0.9, respectively. This explains why the small volume of MeOH is more accessible and accumulates in the pores of the crystal of 1. The shapes of the isotherm are different from each other. The adsorption isotherm for MeOH is classified as a Type IV, while for EtOH, it corresponds to a Type I classification.

Fig. 5 (a) MeOH and EtOH sorption isotherms of 1 at 288 K and 308 K (●: MeOH, 288 K; ○: MeOH, 308 K; ▲: EtOH, 288 K; △: EtOH, 308 K). (b) Isosteric heats of MeOH (●) and EtOH (□) adsorption of 1.

To clarify the difference in the isotherms between MeOH and EtOH, we estimated the interaction of the guest adsorbate with 1 based on the isosteric heat. At the adsorption equilibrium, the isosteric heat corresponds to the enthalpy of the adsorption. By using the isotherms at two distinct temperatures, the isosteric heat (Q_st) can be obtained by applying the Clausius–Clapeyron equation.¹⁴ The applied isotherms for the isosteric heat were collected at temperatures of 288 K and 308 K, and the results are shown in Fig. 5b. The horizontal axis indicates the adsorbed amount, and the vertical axis refers to the isosteric heat of the adsorption. For the adsorption of MeOH, Q_st reached a maximum at 55.4 kJ mol⁻¹, then gradually decreased to 38.2 kJ mol⁻¹ followed by a plateau region. For the EtOH adsorption, Q_st was at a maximum at 56.7 kJ mol⁻¹, then gradually decreased to ≈34 kJ mol⁻¹. Noteworthy is the fact that the maxima of Q_st for MeOH and EtOH are almost the same. This indicated that the interactions of the adsorbate of MeOH and EtOH with 1 are almost equal though the volumes of the adsorbate molecules are different. Because of the limited application of the Clausius–Clapeyron equation in the high relative pressure region, Q_st around the saturated adsorbed area is hardly explained. However, based on the obtained isotherms of MeOH and EtOH, capillary condensation is present as a consequence of the multilayer adsorption. One explanation is as follows: for the adsorption of MeOH, the adsorption equilibrium is almost shifted to the liquid phase, while the equilibrium between the liquid and the vapor phases shifts to the liquid phase for the EtOH adsorption. This theory is supported by the presence of the plateau region for MeOH and the tail for EtOH after reaching the maximum Q_st.

Fig. 6 shows the adsorption isotherms of 1 for several kinds of alcohols at a temperature of 298 K. The alcohols used for the adsorption were 1-PrOH, 1-BuOH, 2-BuOH, and iso-BuOH. The obtained isotherms for these alcohols were classified as Type IV. The total adsorbed amount of the alcohols at p/p₀ ≈ 0.9 were 32.0 cm³ g⁻¹, 28.1 cm³ g⁻¹, 28.5 cm³ g⁻¹, and 26.0 cm³ g⁻¹ for 1-PrOH, 1-BuOH, 2-BuOH, and iso-BuOH, respectively. Thus, with increasing volume of the adsorbate, the total adsorbed amount decreased. This tendency was the same as observed in the isotherms for MeOH and EtOH. However, in the low relative pressure region, this is not true.

Fig. 6 Alcohol isotherms of 1 at 298 K (●: 1-PrOH; △: 1-BuOH; ○: 2-BuOH; ■: iso-BuOH).

The adsorbed amount of the alcohols for 1 at p/p₀ ≈ 0.9 and 0.3 is summarized in Table 1. To our surprise, at p/p₀ ≈ 0.3, the adsorbed amount of 1-BuOH is high at 24.0 cm³ g⁻¹, and close to the amount of MeOH (25.9 cm³ g⁻¹). On the other hand, for the other alcohols including the isomers of butanol, the adsorbed amount is much lower than that of 1-BuOH. The order of the adsorbed amount of alcohols at p/p₀ ≈ 0.3 is as follows: MeOH ≈ 1-BuOH > EtOH > 1-PrOH > 2-BuOH > iso-BuOH. Except for 1-BuOH, the small volume of the adsorbate has a tendency to accumulate in the pores of 1. For the adsorption of 1-BuOH, however, porous macrocycle 1 showed a high selectivity. In the series of the isomers of butanol, the adsorbed amount ratio between 1-BuOH and isomers 2-BuOH and iso-BuOH is 2.2 and 4.3 respectively. This high selectivity of 1-BuOH for 1 may come from the lipophilic nature of the pores of 1. Since porous macrocycle 1 contains triethylphosphine ligands, the pores of 1 would be surrounded by a lipophilic environment. Based on this assumption, the manner of the packing of the adsorbate molecule during condensation may also play an important role in the selectivity of the adsorbate. From the adsorption isotherm of 1-BuOH, the multilayer adsorption progresses around p/p₀ ≈ 0.15, and then a rapid adsorption occurs. This implies that the condensation of 1-BuOH in a multilayered fashion is more accessible in the lipophilic pores of 1. For the adsorption of the other isomers, the condensation in the pores requires a high relative pressure. This suggests that the adsorbate molecule is hardly packed during condensation because of the presence of the side chain in the isomers of butanol.

Table 1 Adsorbed amount of alcohols (cm³ g⁻¹) for 1 at the selected pressure (p/p₀). The data was collected at 298 K.

p/p0	Adsorbed amount [cm^3 g^−1]
	MeOH	EtOH	1-PrOH	2-BuOH	2-BuOH	iso-BuOH
0.3	25.9	19.1	15.4	24.0	11.2	5.6
0.9	57.3	29.8	32.0	28.1	28.5	26.0

---

In summary, we have shown that molecule 1 acts as a porous macrocycle for the adsorption of several kinds of alcohols. Macrocycle 1 has a mesopororous structure with 1.85 nm and 3.09 nm pore diameters. The pore structure of 1 is uniformly formed in an ordered array due to the crystallinity of 1. In the adsorption of alcohols for 1, the total adsorbed amount decreased, with increasing volume of the adsorbate. However, at a specific pressure, porous macrocycle 1 showed a selectivity for the adsorption of 1-BuOH as high as 4.2 among the isomers of butanol. This adsorption property could originate from the lipophilic surroundings of the pores of 1 and the constrained packing of the branched alcohols during condensation. We hope that our findings of the lipophilic pores in an ordered array for selective adsorption by a giant macrocycle acts as a model for the future design of a porous compound.

Acknowledgements

We wish to thank Dr Takehiro Ohta and Dr Satoaki Onistuka (IMCE) for their many helpful suggestions when using the adsorption apparatus. This work was partially supported by a Grant-in-Aid for the Global COE Program, “Science for Future Molecular Systems”, the Hayashi Memorial Foundation for Female Natural Scientists (A.H.), and a Grant-in-Aid for Scientific Research on Innovative Areas (no. 21106015) from the Ministry of Education, Culture, Sports, Science and Technology, Japan.

Notes and references

1. For review, see: (a) M. Eddaoudi, D. B. Moler, H. Li, B. Chen, T. M. Reineke, M. O'Keeffe and O. M. Yaghi, Acc. Chem. Res., 2001, 34, 319–330; (b) J. L. C. Rowsell and O. M. Yaghi, Microporous Mesoporous Mater., 2004, 73, 3–14; (c) J. L. C. Rowsell and O. M. Yaghi, Angew. Chem., Int. Ed., 2005, 44, 4670–4679; (d) S. Kitagawa and R. Matsuda, Coord. Chem. Rev., 2007, 251, 2490–2509; (e) J. J. Perry IV, J. A. Perman and M. J. Zaworotko, Chem. Soc. Rev., 2009, 38, 1400–1417; (f) D. J. Tranchemontagne, J. L. Mendoza-Cortés, M. O'Keeffe and O. M. Yaghi, Chem. Soc. Rev., 2009, 38, 1257–1283; (g) S. Horike, S. Shimomura and S. Kitagawa, Nat. Chem., 2009, 1, 695–704; (h) J.-Y. Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen and J. T. Hupp, Chem. Soc. Rev., 2009, 38, 1450–1459; (i) M. O'Keeffe and O. M. Yaghi, Chem. Rev., 2012, 112, 675–702.
2. (a) J. L. Atwood, L. J. Barbour and A. Jerga, Science, 2002, 296, 2367–2369; (b) J. L. Atwood, L. J. Barbour and A. Jerga, Angew. Chem., Int. Ed., 2004, 43, 2948–2950; (c) J. Tian, P. K. Thallapally, S. J. Dalgarno, P. B. McGrail and J. L. Atwood, Angew. Chem., Int. Ed., 2009, 48, 5492–5495.
3. (a) Y. Miyahara, K. Abe and T. Inazu, Angew. Chem., Int. Ed., 2002, 41, 3020–3023; (b) S. Lim, H. Kim, N. Selvapalam, K.-J. Kim, S. J. Cho, G. Seo and K. Kim, Angew. Chem., Int. Ed., 2008, 47, 3352–3355.
4. P. Sozzani, S. Bracco, A. Comotti, L. Ferretti and R. Simonutti, Angew. Chem., Int. Ed., 2005, 4, 1816–1820.
5. M. Mastalerz, M. W. Schneider, I. M. Oppel and O. Presly, Angew. Chem., Int. Ed., 2011, 50, 1046–1051.
6. Y. Jin, B. A. Voss, R. D. Noble and W. Zhang, Angew. Chem., Int. Ed., 2010, 49, 6348–6351.
7. Y. Jin, B. A. Voss, A. Jin, H. Long, R. D. Noble and W. Zhang, J. Am. Chem. Soc., 2011, 133, 6650–6658.
8. T. Hasell, S. Y. Chong, K. E. Jelfs, D. J. Adams and A. I. Cooper, J. Am. Chem. Soc., 2012, 134, 588–598.
9. A. Harano, M. Tanaka, T. Nakagaki, M. Annaka, K. Ideta, K. Goto and T. Shinmyozu, Org. Biomol. Chem., 2011, 9, 3611–3614.
10. K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J. Rouquérol and T. Siemieniewska, Pure Appl. Chem., 1985, 57, 603–619.
11. M. Kruk and M. Jaroniec, Chem. Mater., 2001, 13, 3169–3183.
12. S. Brunauer, P. H. Emmett and E. Teller, J. Am. Chem. Soc., 1938, 60, 309–319.
13. R. W. Cranston and F. A. Inkley, Adv. Catal., 1957, 9, 143–154.
14. H. Pan, J. A. Ritter and P. B. Balbuena, Langmuir, 1998, 14, 6323–6327.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details, characterizations, and BET plot. See DOI: 10.1039/c3ra45716h

---