Larger Substituents on Amide Cavitands Induce Bigger Cavities

Dec 19, 2018 - A series of quinoxaline cavitands bearing pendant amide groups with various substituent sizes (Et, iPr, tBu) were synthesized, and thei...
1 downloads 0 Views 2MB Size
Letter pubs.acs.org/OrgLett

Cite This: Org. Lett. 2019, 21, 201−205

Larger Substituents on Amide Cavitands Induce Bigger Cavities Safwan Aroua,† Andrew N. Lowell,‡,§ Ankita Ray,† Nils Trapp,† W. Bernd Schweizer,† Marc-Olivier Ebert,† and Yoko Yamakoshi*,†,‡ †

Laboratorium für Organische Chemie, ETH Zürich, Vladimir-Prelog-Weg 3, CH8093 Zürich, Switzerland Department of Chemistry, University of Pennsylvania, 231 South, 34th Street, Philadelphia, PA19104-6323, United States



Org. Lett. 2019.21:201-205. Downloaded from pubs.acs.org by TULANE UNIV on 01/09/19. For personal use only.

S Supporting Information *

ABSTRACT: A series of quinoxaline cavitands bearing pendant amide groups with various substituent sizes (Et, iPr, tBu) were synthesized, and their cavity size/structure were investigated by X-ray and NMR analyses. In the case of the Et or iPr amide cavitand, the conformation of the molecule was in the vase form, while the bulky tBu amide cavitand gave the kite conformation at room temperature. X-ray crystal structures of Et and iPr cavitands clearly showed the intramolecular H-bondings to influence the conformation and the cavity sizes dependent on the bulkiness of functional groups. The 1H NMR spectrum revealed that the Et cavitand can encapsulate an adamantane guest compound with slow exchange.

C

avitands, initially reported by Cram and co-workers,1 have been investigated (1) as molecular containers for smaller guest molecules or ions and (2) as molecular switches based on their temperature- or pH-triggered conformational changes between two distinct forms (so-called the vase and the kite, Figure 1a). By utilizing these unique features, a variety of cavitand derivatives have been synthesized by modifying the upper and lower rim parts. For instance, the Rebek group reported a number of cavitands providing the systems such as (1) capsule-type cavitand dimers via complementary intermolecular H-bondings of urea or thiourea in the upper rim,2,3 (2) water-soluble cavitands by the introduction of PEG,4,5 guanidinium,6 or sugar7 in the lower or upper rim, (3) expanded cavities to incorporate larger guest molecules by the addition of larger aromatic groups in the upper rims,8 and (4) self-folding cavitands through intramolecular H-bondings of amide groups in the smaller benzene-type upper rim.9,10 These molecules provide unique environments for guest compounds inside the cavity to allow the Menschutkin reaction,11 conformational control of n-alkane via the attractive C−H/π interaction, and acid/base pair interactions.12 In addition, the Ballester group reported a catalytic hydrogenation reaction using Rebek’s self-folding cavitand.13 Independently, the Diederich group has focused on the conformational switching properties of cavitands and has developed unique cavitandbased functional molecules such as (1) new FRET probes for temperature and pH sensing by adding FRET donor and acceptor moieties in the upper rim,14−16 (2) redox-induced switching cavitands by quinone substitution on the upper rim,17−19 and (3) halogen-bonding capsules.20−22 Separately, a different type of H-bonding capsule cavitand was reported by the Paek group.23 More recently, the group of Choi and coworkers prepared an acetamidoquinoxaline cavitand and © 2018 American Chemical Society

demonstrated a hindered ring inversion of cyclohexane in the confined cavity.24 However, there are still very few reports of the cavitands for the encapsulation of larger and bulkier organic molecules, with some exceptions.8 In the present study, we aimed to control the cavity size and conformation of the quinoxaline-based cavitands by introducing amide groups at the top of the upper rim (Figure 1). We expected that the amide groups in the upper rim contribute to the intramolecular or intermolecular H-bonds, while being affected by the bulkiness of the substitution providing a different shape of the molecule. While Rebek and co-workers reported benzene-based amide cavitands,9,10 the quinoxalinebased amide cavitands in the current study were expected to possess larger cavities suitable for the encapsulation of larger guest molecules. We chose three amide substituent groups (Et, i Pr, and tBu amide) with different bulkiness to investigate the shapes and the conformations of the cavities (vase or kite) resulting from the size of substituents in both solution and solid phase using NMR spectroscopy, X-ray crystallography, and fluorescence spectroscopy. As a result, the bulkier the amide functionality in the upper rim was, the more open the conformation was (kite). Detailed investigation using spectroscopic analyses are described below. Three new cavitands 1−3 bearing ethyl (Et), isopropyl (iPr), or tert-butyl (tBu) amide substituent groups in the upper quinoxaline rims were synthesized based on a literature protocol25 through the nucleophilic addition of quinoxalines 5−7 to the octol 4 using Cs2CO3 as a base (Figure 1b). Dichloroquinoxaline derivatives 5−7 were prepared from 2,3Received: November 15, 2018 Published: December 19, 2018 201

DOI: 10.1021/acs.orglett.8b03660 Org. Lett. 2019, 21, 201−205

Letter

Organic Letters

Figure 1. (a) Structures of three quinoxaline cavitands 1−3 with amide groups in the upper rims and side views of two possible conformations (the vase and the kite). (b) Syntheses of cavitands. Reagents and conditions: (for 1) 5 (5.0 equiv), Cs2CO3 (3.1 equiv), 20 h, 19%; (for 2) 6 (8.0 equiv), Cs2CO3 (10 equiv), 48 h, 3%; and (for 3) 7 (10 equiv), Cs2CO3 (8.0 equiv), 48 h, 10%.

dichloro-6,7-diaminoquinoxaline by amide formation (details are in the Supporting Information). Attachment of four quinoxaline flaps to octol 4 was complete after 2 days, providing each cavitand 1−3. The purifications of cavitands 1− 3 were carried out on florisil column adsorbent using an eluent of a gradient mixture of n-hexane and THF. Conformations of cavitands 1−3 in solution phase were analyzed by 1H and 13C NMR spectroscopies in THF-d8 at room temperature (Figure 2a, c). As mentioned in the initial study of Cram’s cavitands,26 the quinoxaline cavitands have two potential conformationsthe so-called vase and kite (Figure 1a)dependent on the environments such as temperature and pH. It is known that the 1H NMR chemical shift of the bridging methine (Ha in Figure 2d) can be used as a primary indicator for these conformations. In the 1H NMR spectra of 1 and 2 (Figure 2a), Ha peaks were observed at ca. 5.8 ppm indicating that both 1 and 2 were in the vase conformation. The 13C NMR spectra (Figure 2c) further confirmed the vase conformation of 1 and 2 based on the number of peaks corresponding to C4v symmetry.26 In contrast, the 1H and 13C NMR spectra of cavitand 3 suggested a kite-type conformation even at room temperature. The 1H NMR chemical shift of the indicator Ha peak in 3 was shifted to higher field (ca. 4.4 ppm) (Figure 2a) suggesting that the molecule was in the kite conformation. In addition, we speculated that the broadening of the Ha peak was due to slow interconversion between two kite conformations (Figure 2b) on the NMR time scale. Split and broadened peaks for the quinoxaline proton (Hd) and two sharp and distinct tBu peaks further supported the kite conformation of 3 with C2v symmetry (Figure 2a). Furthermore, split peaks of C1, C2, C3, and C5 in the 13C NMR spectrum of 3 suggested C2v symmetry of the molecule confirming the kite conformation

Figure 2. (a) 1H NMR spectra of cavitands 1−3 in THF-d8 at 25 °C, (b) interconversion of two kite conformations for 3, (c) 13C NMR spectra of cavitands 1−3 in THF-d8 at 25 °C, and (d) their assignments. The conformations of 1 and 2 were identified as vase, while 3 was identified as kite, mainly by the chemical shift of Ha. Consequently, splitting of the signals for Hd, C1, C2, C3, and C5 were observed in 3 due to the lower symmetry of the kite (C2v) compared to the vase conformation (C4v). The marks ** and *** are impurities.

(Figure 2c).26 Based on the observation above, it can be suggested that the bulkiness of the tBu groups at the top of the upper rim of 3 enforces a more open and kite conformation at room temperature in THF, while the cavitands 1−2 and Cram’s original unsubstituted quinoxaline cavitand exhibit a vase conformation under the same conditions. The lower shift of the NH protons of cavitand 3 indicates that the H-bonding possibly arises from dimerization of two molecules with kite conformations. In the 1H NMR spectrum of 3 in THF-d8, the peaks corresponding to the alkyl chains in the lower rim were very broad, in contrast to the sharp signals of the aromatic protons (Hb and Hc) in the central rim (Figure 2a). A similar situation was also observed in other solvents (C6D6, CDCl3, CD2Cl2, DMF-d7, DMSO-d6, and TFE-d3). This observation could be explained by very slow motion of the alkyl chain on the NMR time scale due to aggregation or micelle formation of 3. To probe this conjecture, DOSY NMR experiments of 1−3 were performed in CDCl3 (Figures S44−S46). However, as a result, the diffusion coefficient of 3 was higher (ca. 5.7−6.1 × 10−10 m2 s−1) than those of both 1 and 2 (ca. 4.4 × 10−10 m2 s−1) indicating that 3 was diffusing faster than 1 and 2 in CDCl3. 202

DOI: 10.1021/acs.orglett.8b03660 Org. Lett. 2019, 21, 201−205

Letter

Organic Letters This result suggested that the formation of a larger cluster from 3 was unlikely.

Figure 4. (a−c) X-ray crystal structures of cavitands 1 and 2 crystallized from n-hexane and THF by vapor diffusion or by evaporation method. Upper row: top views (the lower rim parts are not shown); lower row: side views. (d) H-bonding arrays in the crystal lattice of 1 prepared by the vapor diffusion. (e) H-bonding dimers in the crystal lattice of 1 prepared by evaporation. Hydrogen atoms are omitted for clarity. Distances between the top of quinoxaline flaps on the opposite side are shown with green arrow.

Figure 3. (a) UV−vis spectra of cavitands 1−3 in THF (10 μM), (b) fluorescence spectra of cavitands 1−3 in THF (100 μM, λex: 310 nm), and (c) fluorescence spectra of 3 in varied concentration in THF (0.1−100 μM, λex: 310 nm). (Same machine but different sizes of slit were used for the fluorescence measurements in experiments (b) and (c).)

different crystal structures of 1 were obtained by two different crystallization methods (solvent vapor diffusion method and solvent evaporation method); one crystal structure was obtained for 2 (details of the methods are in the SI). The crystal structure of 1 obtained by the solvent diffusion method revealed the “closed” conformation on top of the cavity through four intramolecular (two intra- and two interannular) H-bonds between the amide groups of the two quinoxaline flaps on the opposite side (blue dotted lines in Figure 4a). The distance between the top of these two quinoxaline flaps was quite short (ca. 5.2 Å), and one THF molecule was encapsulated inside the cavity. The amide groups on the remaining two quinoxaline flaps contributed to the intermolecular H-bonds with the amide groups of neighboring molecules, forming a linear H-bonding array in the crystal lattice (Figure 4d). Interestingly, by using the same solvent system, but with the solvent evaporation method, a different crystal structure of 1 with a more “open” conformation (distances between the top of quinoxaline were 8.5 and 10.2 Å) was constructed via inter- and intraannular H-bond network (Figure 4b). Two THF molecules were found in the cavity (Figure 4b) and two intermolecular H-bonds with neighboring molecules were observed forming a kind of dimer (Figure 4e). The crystal structure of 2 obtained by solvent vapor diffusion method showed a larger “open” conformation (distances between the top of quinoxaline were 8.9 and 10.5 Å) with a complete circular H-bond network, and only one THF molecule, was found to be encapsulated inside the cavity (Figure 4c). No intermolecular H-bonding interaction with any neighboring molecule was observed.

To investigate the structures of cavitands 1−3 in greater detail, fluorescence spectroscopy was measured. Prior to fluorescence measurements, the UV−vis spectra of 1−3 were recorded in THF (Figure 3a). Interestingly, the absorption of 3 was significantly higher with bathochromic shift (λmax = 360 nm) than the ones of 1 and 2 (λmax = 335 nm). Fluorescence spectra of 1−3 were measured with an excitation at 310 nm. While sufficient emissions were observed in solutions of 1 and 2 with a maximum at 394 nm, 3 displayed a significant bathochromic shift (emission maximum at 435 nm) with a broader and less intense band (Figure 4b). By combining the results of the kite conformation of 3 as suggested by NMR spectroscopy, it was speculated that fluorescent emission of the quinoxaline moieties in cavitand 3 was quenched by stacking while cavitands 1 and 2 in the vase conformation maintained a higher level of emission intensity. This result was in line with the previous report on cavitands by Rebek and co-workers on cavitands forming dimers in the kite conformation (D2d velcraplex),4 and low fluorescent emission intensity in 3 may be also due to the dimerization of two kite forms facing the aromatic surfaces each other. Concentration-dependent fluorescence quenching of cavitand 3 was observed in THF (inset of Figure 3c) to support this speculation, but further investigations are still needed. Single crystals of cavitands 1 and 2 were successfully obtained from a mixture of THF and n-hexane (Figure 4). Two 203

DOI: 10.1021/acs.orglett.8b03660 Org. Lett. 2019, 21, 201−205

Letter

Organic Letters Accession Codes

CCDC 1861109−1861111 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Yoko Yamakoshi: 0000-0001-8466-0118 Figure 5. Host−guest complexation between 1 and adamantane observed by 1H NMR (400 MHz, in TMB-d12).

Present Address

Finally, the complexation capability of 1 was investigated by H NMR spectroscopy in deuterated mesitylene (trimethylbenzene, TMB-d12), a large solvent that does not compete with the guest molecule (Figure 5). For calibration of the concentration, a noncompeting guest 1,3,5-trimethoxybenzene (a large guest) was employed. Adamantane was used as a guest molecule for the complexation. As expected, the introduction of amide units onto the upper rim lowered the exchange rate on the NMR time scale, with two sets of adamantane signals.27 The two signals for the free guest were situated between +1.7 to +1.9 ppm, while the bound guest in the cavity was shifted upfield to −1.7 to −2.1 ppm. The association constant Ka was calculated by the integration of the NMR signals and was found to be Ka = 18 M−1 (Figure S43). These experiments supported the idea that the H-bonds in the upper rims can lower the host−guest exchange rate. In conclusion, three kinds of new cavitands with different sizes of amide functional groups (1: Et, 2: iPr, 3: tBu) were synthesized and their conformations were analyzed in solution by 1H, 13C NMR spectroscopies and fluorescent spectroscopy and in solid by X-ray crystal structure analyses. Cavitands 1 and 2 with smaller amide groups showed vase conformations in non-H-bond-competing solvents due to the effects of intramolecular H-bonds, while cavitand 3 revealed an open kite conformation due to suppression of the intramolecular Hbonds by bulky tBu groups on the top of the upper rim. As a result, the cavity size and the conformation of the cavitands (vase-or-kite) could be tuned by the attachment of the amide group in the upper rim as a function of the bulk of amide groups affecting the intramolecular H-bonds. The introduction of the intramolecular H-bonds was also responsible for the slow host/guest exchange rate on the NMR time scale. Together with the larger cavity of these amide quinoxaline cavitands, the cavitands in the present study can be unique and interesting molecules for further development of the new functional materials such as molecular grippers.

Notes

§

Department of Chemistry, Virginia Tech.

The authors declare no competing financial interest.

1





ACKNOWLEDGMENTS This research was supported in part by the Swiss National Foundation (200021_140451, 200021_156097, 205321_173018), PRESTO program from Japan Science and Technology Agency, ETH Research Grant (ETH-21-15-2, ETH-25 11-1), and the Researcher’s Developing Grant from the American Heart Association (0930140N).

■ ■

DEDICATION Dedicated to Professor Dr. Amos B. Smith III, editor-in-chief. REFERENCES

(1) Moran, J. R.; Karbach, S.; Cram, D. J. J. Am. Chem. Soc. 1982, 104, 5826−5828. (2) Heinz, T.; Rudkevich, D. M.; Rebek, J. Nature 1998, 394, 764− 766. (3) Rebek, J. Angew. Chem., Int. Ed. 2005, 44, 2068−2078. (4) Haino, T.; Rudkevich, D. M.; Shivanyuk, A.; Rissanen, K.; Rebek, J., Jr. Chem. - Eur. J. 2000, 6, 3797−3805. (5) Lledo, A.; Rebek, J., Jr. Chem. Commun. 2010, 46, 8630−8632. (6) Javor, S.; Rebek, J. J. Am. Chem. Soc. 2011, 133, 17473−17478. (7) Ryan, D. A.; Rebek, J. J. Am. Chem. Soc. 2011, 133, 19653− 19655. (8) Tucci, F. C.; Rudkevich, D. M.; Rebek, J. J. Org. Chem. 1999, 64, 4555−4559. (9) Rudkevich, D. M.; Hilmersson, G.; Rebek, J. J. Am. Chem. Soc. 1997, 119, 9911−9912. (10) Rudkevich, D. M.; Hilmersson, G.; Rebek, J. J. Am. Chem. Soc. 1998, 120, 12216−12225. (11) Purse, B. W.; Gissot, A.; Rebek, J. J. Am. Chem. Soc. 2005, 127, 11222−11223. (12) Purse, B. W.; Butterfield, S. M.; Ballester, P.; Shivanyuk, A.; Rebek, J. J. Org. Chem. 2008, 73, 6480−6488. (13) Sarmentero, M. A.; Fernandez-Perez, H.; Zuidema, E.; Bo, C.; Vidal-Ferran, A.; Ballester, P. Angew. Chem., Int. Ed. 2010, 49, 7489− 7492. (14) Azov, V. A.; Schlegel, A.; Diederich, F. Angew. Chem., Int. Ed. 2005, 44, 4635−4638. (15) Pochorovski, I.; Knehans, T.; Nettels, D.; Muller, A. M.; Schweizer, W. B.; Caflisch, A.; Schuler, B.; Diederich, F. J. Am. Chem. Soc. 2014, 136, 2441−2449. (16) Pochorovski, I.; Breiten, B.; Schweizer, W. B.; Diederich, F. Chem. - Eur. J. 2010, 16, 12590−12602. (17) Pochorovski, I.; Milic, J.; Kolarski, D.; Gropp, C.; Schweizer, W. B.; Diederich, F. J. Am. Chem. Soc. 2014, 136, 3852−3858.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.8b03660. Synthesis and characterization of cavitands 1−3 and their intermediates with spectroscopic data; details of the X-ray crystallographic analyses (PDF) 204

DOI: 10.1021/acs.orglett.8b03660 Org. Lett. 2019, 21, 201−205

Letter

Organic Letters (18) Pochorovski, I.; Diederich, F. Acc. Chem. Res. 2014, 47, 2096− 2105. (19) Pochorovski, I.; Ebert, M. O.; Gisselbrecht, J. P.; Boudon, C.; Schweizer, W. B.; Diederich, F. J. Am. Chem. Soc. 2012, 134, 14702− 14705. (20) Dumele, O.; Trapp, N.; Diederich, F. Angew. Chem., Int. Ed. 2015, 54, 12339−12344. (21) Gropp, C.; Husch, T.; Trapp, N.; Reiher, M.; Diederich, F. J. Am. Chem. Soc. 2017, 139, 12190−12200. (22) Gropp, C.; Trapp, N.; Diederich, F. Angew. Chem., Int. Ed. 2016, 55, 14444−14449. (23) Park, Y. S.; Seo, S.; Kim, E. H.; Paek, K. Org. Lett. 2011, 13, 5904−5907. (24) Nguyen, Q.-T.; Oh, D.-W.; Kim, W.; Sahoo, S. K.; Choi, H.-J. Asian J. Org. Chem. 2015, 4, 729−732. (25) Azov, V. A.; Skinner, P. J.; Yamakoshi, Y.; Seiler, P.; Gramlich, V.; Diederich, F. Helv. Chim. Acta 2003, 86, 3648−3670. (26) Azov, V. A.; Jaun, B.; Diederich, F. Helv. Chim. Acta 2004, 87, 449−462. (27) Rudkevich, D. M.; Hilmersson, G.; Rebek. J. Am. Chem. Soc. 1998, 120, 12216−12225.

205

DOI: 10.1021/acs.orglett.8b03660 Org. Lett. 2019, 21, 201−205