The Impact of Solvent and the Receptor Structure on Chiral Recognition Using Model Acyclic Bisamides Decorated with Glucosamine Pendant Arms.

We investigated the influence of various factors (including solvent mixtures) on chiral recognition of chiral carboxylates, using the titration method under 1H NMR control. We found that strong binding carboxylates (geometrical matching) is not enough for the satisfactory differentiation of enantiomers. Moreover, solvent mixture studies indicate a significant influence of environment on the formation of diastereomeric complexes and variations among them. Our findings offer insights into the complementarity of chiral recognition processes.

Introduction

Solving the puzzle of how nature worn class="Chemical">ks remclass="Gene">ains class="Gene">as class="Gene">an unending source of chclass="Gene">allenges for reseclass="Gene">archers.[1] One such class="Gene">an class="Disease">insufficiently understood problem lies in the subtlest types of selectivity known as chiral recognition.[2] Therefore, current research focuses on aspects such as asymmetric catalysis,[3] molecular recognition,[4] chiral separation,[5] interaction on surfaces,[6] and supramolecular assemblies of chiral molecules in solution.[7]

Chiral recognition phenomena originate from differences in the Gibbs free energy (ΔGtotal) of diastereomeric complexes formed ban class="Gene">etween chirclass="Gene">al molecules.[8] ΔGtotclass="Gene">al depends on the energy of intermoleculclass="Gene">ar interclass="Gene">actions (Einter), the energy from conformclass="Gene">ationclass="Gene">al chclass="Gene">anges (ΔEintrclass="Gene">a), class="Gene">and free solvclass="Gene">ation energy (ΔGsolv). The lclass="Gene">atter includes the free solvclass="Gene">ation energy of the complex (ΔGsolv), free receptor (ΔGsolv), class="Gene">and free guest (ΔGsolv) (Figure .).[9]

Figure 1

ΔGtotal equation.

Differences in interactions bn class="Gene">etween solvent class="Gene">and class="Gene">an class="Chemical">anionic species and receptors lead to slight differences in ΔGsolv and depend on the medium. The solvent’s parameters define this distinctiveness. Relative polarity, dielectric constant (ε), and Gutmann numbers[10] are among the criteria describing divergences in solvation (see Table ). If host–guest interactions are electrostatic and the solvent weakly solvates both, the magnitude of the binding constant is inversely correlated to the dielectric constant of the solvent.[11] Gutmann’s Donor Number (DN) defines the donicity of a solvent, meaning its behavior as a Lewis base solvent, while Gutmann’s Acceptor Number (AN) reflects a solvent’s character as a Lewis acid. Hence, the interactions between a solvent and a charged anion and the H-donor cavity lie along the ion–dipole and dipole–dipole interactions. Analysis of differences in solvent features indicates that the medium could also be a meaningful factor in chiral recognition. To our knowledge, however, there have been no in-depth studies describing the influence of the medium on chiral recognition of anions.

Table 1

Properties of Solvents Used: Gutmann Donor Number (DN) (Kcal Mol–1), Gutmann Acceptor Number (AN) (Kcal Mol–1), Relative Polarity (ET), and Dielectric Constant (ε)

entry	solvent	DN[12]	AN[12]	ET[13]	ε[13]
1	CH3CN	14.1	18.9	0.460	37.5
2	DMSO	29.8	19.3	0.444	46.68
3	CHCl3	4.0	23.1	0.259	4.89
4	H2O	54.8	18.0	1.000	80.1

Another factor is the proper design of the chiral receptor, crucial for binding the n class="Chemical">anion.[14] class="Gene">an class="Chemical">Many factors govern the chiral recognition phenomenon, making a prior prediction difficult. The appropriate arrangement of hydrogen bond donors and chiral fragments should ensure enantioselective interactions with chiral anions. Hence, effective chiral recognition of anions requires the synthesis and determination of binding affinities and enantioselectivities of prospective chiral receptors.[15]

Based on these considerations, we designed a series of chiral receptors (1a–1e) of various sizes and binding pockn class="Gene">et geomclass="Gene">an class="Gene">etries, consequently with different arrangements of hydrogen bonding donors (Figure ). To achieve this goal, we took the approach of using the covalent attachment of the chiral moiety to an anion binding backbone. Based on our experience[16] showing bisamides to be attractive building blocks for achiral receptors, we applied different simple aromatic platforms (benzene, pyridine, azulene, and pyrrole). As a chiral part, we chose a peracetylated glucosamine derivative,[17] which is a cheap and readily available source of chirality and can be easily functionalized by changing the protecting groups.

Figure 2

(A) Idea of a chiral receptor structure; (B) chiral receptors 1a–1e investigated in this study.

Results and Discussion

First, we n class="Gene">synthesized class="Gene">a series of class="Gene">an class="Chemical">amide-based receptors using per-O-acetyl-D-glucosamine hydrochloride with acid dichlorides previously prepared from the corresponding dicarboxylic acid.[18] Due to the high nucleophilicity of the 1- and 3-position of the azulene moiety, we applied HBTU (2-(1-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate) to obtain 5,7-bisamide azulene receptor 1a (Scheme ).[19]

Scheme 1

Synthesis of Model Bisamides 1a–1e

We dn class="Gene">etermined the class="Gene">an class="Chemical">anion binding properties using the 1H NMR titration method. This technique keeps track of the binding process and shows differences in the formation of diastereomeric complexes.[20] Utilizing changes in chemical shifts, we obtained all global stability constants by nonlinear curve fitting to the 1:1 and 1:2 (receptor: anion) binding model[21] using the program HypNMR2008.[22]

Hence, as to evaluate the influence of geomn class="Gene">etry on binding class="Gene">affinity class="Gene">and modes of class="Gene">an class="Chemical">anion binding by amide-based receptors, we conducted titration experiments with the series of receptors 1a–1e with achiral benzoate anion as tetrabutylammonium salt (TBA). Due to the solubility of receptors (all studied ligands are insoluble in aqueous media), we conducted titration experiments in CD3CN + 0.5% H2O. The respective binding constants toward benzoate anion (Ka), the geometrical parameters of the binding pocket, and the maximum chemical shift (Δδmax) of the protons in the binding cavity (green H in Figure ) are presented in Table .

Figure 3

Geometrical parameters presented in Table .

Table 2

Comparison of Geometrical Parameters of the Binding Pocket[16] with Binding Constants Ka [M–1]a for Complexes of Receptors 1a–1e with Benzoate in CD3CN + 0.5% H2Oa−c

receptor	1a	1b	1c	1d	1e
Ka	460	49	1400	>104c	>104c
Δδmax [ppm]	0.33	−	0.77	4.16	1.61
d [Å]b	4.0	4.6	5.0	5.4	5.6
α [o]b	117	117	123	139	145

Values determined by 1H NMR spectroscopy titration experiments at T = 303 K; estimated errors <10%; TBA salts were the sources of the anions.

Geometrical parameters determined for conformation syn–syn using X-ray for R = n-Bu.

Binding model 1:2 (host:guest), K1:2 is omitted, for more details see Supporting Information.

Geoman class="Gene">etricclass="Gene">al pclass="Gene">arclass="Gene">amclass="Gene">an class="Gene">eters presented in Table .

Values dan class="Gene">etermined by class="Gene">an class="Chemical">1H NMR spectroscopy titration experiments at T = 303 K; estimated errors <10%; TBA salts were the sources of the anions.

Geoman class="Gene">etricclass="Gene">al pclass="Gene">arclass="Gene">amclass="Gene">an class="Gene">eters determined for conformation syn–syn using X-ray for R = n-Bu.

Binding model 1:2 (host:guest), an class="Gene">K1:2 is omitted, for more dclass="Gene">an class="Gene">etails see Supporting Information.

In all cases, n class="Chemical">anion complexclass="Gene">ation cclass="Gene">aused class="Gene">a downfield shift of C–H class="Gene">and N–H protons locclass="Gene">ated in the receptor binding pockclass="Gene">an class="Gene">et. The calculated affinity constants generally increased with the size of the binding cavity. Receptors 1d and 1e, based on a five-membered aromatic core, revealed the highest binding affinity toward benzoate anion (binding constants up to 10,000 M–1) and the 1:2 (host:guest) binding model. Dipicolinic acid derivative 1b binds the anion with the lowest constant Ka = 50 M–1 (Figure ). The repulsive interaction between anion and the electron free pair located on the nitrogen atom may attenuate the formation of hydrogen bonds. Growing changes in chemical shifts of green protons demonstrating weak C–H hydrogen bonds (for 1a, 1c, and 1e) and strong N–H bonds (for 1d) are in accordance with the calculated affinity constants.

Figure 4

Dependence of logKa with the size of the binding pocket.

Dependence of logKa with the size of the binding pockan class="Gene">et.

Next, to estimate the receptors’ potential for chiral recognition, we evaluated their binding properties with respect to two pairs of enantiomeric n class="Chemical">carboxylate derivclass="Gene">atives: class="Gene">an class="Chemical">mandelic acid (Man) and N-Ac-phenylglycine (N-Ac-Phg) (see Figure ). Given the magnitude of the binding constants with benzoate, we conducted titration experiments in CD3CN + 0.5% H2O as a solvent mixture. We used chiral anions as TBA salts. We calculated the association constants (KR and KS) determined by separate titration experiments and then compared them. To evaluate the enantioselective properties, we applied thermodynamic selectivity (α), which is a ratio of the binding constants of two diastereomeric complexes (α = KR/KS).

Figure 5

Titration curves of receptor 1e with Man (R, pink; S, blue) in (A) CD3CN + 0.5% H2O and (B) CDCl3. (C) Receptor 1a with Man in CD3CN + 0.5% H2O.

Titration curves of receptor 1e with an class="Chemical">Man (R, pink; S, blue) in (class="Gene">an class="Gene">A) CD3CN + 0.5% H2O and (B) CDCl3. (C) Receptor 1a with Man in CD3CN + 0.5% H2O.

Influence of Receptor’s Geometry on Chiral Recognition

By analogy to the achiral n class="Chemical">benzoate anion, we observed thclass="Gene">at class="Gene">an class="Chemical">anions formed complexes with respect to receptors with the same stoichiometry (Table ). Receptors 1b–1d did not exhibit meaningful enantioselective properties toward the investigated chiral pairs of anions (see the Supporting Information, Figure S1). Ligand 1e recognized the model chiral pair of anion derivatives with α = 3.1 and α = ∼2, respectively, for Man and N-Ac-Phg. Receptor 1a bound the (R) enantiomer more strongly with α = 2.1, with no chiral recognition for Phg derivatives. Comparison analysis of titration curves revealed that the source of the differentiation of chiral mandelic anions for the 5,7-disubstituted azulene derivative 1a is the difference in binding of the anionic part with amide groups of the receptor (Figure c). For receptor 1e we observed perturbation in chemical shifts both for binding cavity protons and for sugar moieties (Figure a). Firstly, this may suggest that the sugar parts adopt a conformation preventing interaction with the side chain of the anions; secondly, it may indicate a close distance of chiral fragments from the binding pocket. Some explanation of origin enantioselective discrimination by 1a and 1e we could find in NOESY experiments, where we observe forming hydrogen bonding only by (R)-enantiomers of mandelate with azulene receptors (for more information see Supporting Information). The absence of significant chiral recognition for ligand 1d and its presence for 1e shows strong binding of the anion by the chiral receptor is insufficient for successfully recognizing enantiomers.

Table 3

Stability Constants Ka [M–1]a and Chiral Recognition α for Complexes of Hosts 1a–e with Chiral Anions in CD3CN + 0.5% H2O

		1a		1b		1c		1d		1e
entry	anion	Ka	α	Ka	α	Ka	α	Ka	αd	Ka	αd
1	R-Man	230	2.1	11	1.1	170	1.1	1400c	1.2	4400c	3.1
2	S-Man	110		10		160		1200c		1400c
3	L-N-Ac-Phg	130	1.1	18	1.1	350	0.9	6900c	0.9	>104b,c	b,c
4	D-N-Ac-Phg	140		20		330		6200c		5030c

Values determined by 1H NMR spectroscopy titration experiments at T = 303 K: estimated errors <10%; TBA salts were the sources of the anions.

Stability constants above the limit of the 1H NMR titration technique (Ka > 104); α was estimated.

Binding model 1:2 (host:guest), K1:2 is omitted, for more details see ESI

α given for KR1:1/KS1:1.

Values dan class="Gene">etermined by class="Gene">an class="Chemical">1H NMR spectroscopy titration experiments at T = 303 K: estimated errors <10%; TBA salts were the sources of the anions.

Stability constants above the limit of the an class="Chemical">1H NMR titrclass="Gene">ation technique (Kclass="Gene">a > 104); α wclass="Gene">as estimclass="Gene">ated.

Binding model 1:2 (host:guest), an class="Gene">K1:2 is omitted, for more dclass="Gene">an class="Gene">etails see ESI

α given for an class="Chemical">KR1:1/class="Gene">an class="Chemical">KS1:1.

Impact of the Guest on Chiral Recognition

Model chiral pairs of n class="Chemical">carboxylates (class="Gene">an class="Chemical">Man and N-Ac-Phg) exhibit different sizes, acidity, and ability to form hydrogen bonds on the substituent of the α carbon atom. All ligands revealed higher affinity constants toward N-Ac-Phg derivatives in comparison to mandalates. We observed enantioselectivity with amino acid derivatives only for receptor 1e. The results indicate that the guest’s structure and complex-forming properties influence the extent of chiral recognition.

Influence of Solvent on Chiral Recognition

Next, to estimate the influence of the solvent on chiral recognition properties, we conducted titration experiments with receptor 1e in the presence of model chiral n class="Chemical">anionic guests, class="Gene">an class="Chemical">mandelate derivatives, used as TBA salts in various solvent mixtures.

The addition of n class="Chemical">water into the solvent mixture cclass="Gene">aused class="Gene">a lowering of the binding constclass="Gene">ant of receptor 1e with the class="Gene">an class="Chemical">anions examined (Table , entry 1–6). The admixture of 5% water into CD3CN changed the affinity toward the (R) enantiomer. Experiments conducted in CD3CN + 5% H2O and DMSO-d6 + 0.5% H2O revealed similar binding affinities. Data fitting showed the considerable chiral recognition ability of receptor 1e in the CD3CN + 0.5% H2O mixture and in chloroform (α = 3.1 and 2.1, Table , entry 3 and 4, and 9 and 10, respectively). Comparison of the relative polarity of dimethylsulfoxide and acetonitrile (0.460 and 0.444, respectively, Table , entry 1 and 2) indicates the low impact of this parameter on the binding affinity. The higher value of DN points to stronger solvation of the host’s binding pocket and simultaneously weakened interaction of guest with the ligand (DNDMSO = 29.8 kcal mol–1, DNACN = 14.1 kcal mol–1). Similarly, the addition of water (DN = 54.8 kcal mol–1) into the solvent mixture causes an increase in environmental competitiveness.

Table 4

Stability Constants Ka [M–1]a and Chiral Recognition for Complexes of Host 1e with TBA Man in Various Solvent Mixtures

entry	solvent mixture	anion	Ka	αd
1	CD3CN	(R)-Man	>10 000b,c	b
2		(S)-Man	>10 000b,c
3	CD3CN + 0.5% H2O	(R)-Man	4400	3.1
4		(S)-Man	1400
5	CD3CN + 5% H2O	(R)-Man	11	0.7
6		(S)-Man	16
7	DMSO-d6 + 0.5% H2O	(R)-Man	11	1.0
8		(S)-Man	11
9	CDCl3	(R)-Man	2700	2.1
10		(S)-Man	1300

Values determined by 1H NMR spectroscopy titration experiments at T = 303 K; estimated errors <10%; anions added as TBA salts.

Stability constants above the limit of the 1H NMR titration technique (Ka > 104), α could not be determined.

Binding model 1:2 (host:guest), K1:2 is omitted, for more details see ESI.

α given for KR1:1/KS1:1.

Values dan class="Gene">etermined by class="Gene">an class="Chemical">1H NMR spectroscopy titration experiments at T = 303 K; estimated errors <10%; anions added as TBA salts.

Stability constants above the limit of the an class="Chemical">1H NMR titrclass="Gene">ation technique (Kclass="Gene">a > 104), α could not be dclass="Gene">an class="Gene">etermined.

Binding model 1:2 (host:guest), an class="Gene">K1:2 is omitted, for more dclass="Gene">an class="Gene">etails see ESI.

α given for an class="Chemical">KR1:1/class="Gene">an class="Chemical">KS1:1.

Comparative analysis of the titration curves of receptor 1e in the n class="Chemical">CD3CN + 0.5% class="Gene">an class="Chemical">H2O mixture and in chloroform revealed differences in the formation of diastereomeric complexes, depending on the experimental mixture (Figure a,b). In CD3CN + 0.5% H2O, we observed disparities in chemical shifts for both the cavity’s protons and sugar moieties in relation to the respective enantiomer (Figure a). In contrast, similar comparison analysis for chloroform revealed that recognition of stereoisomers occurs by the interaction between the side chain of anion and sugar derivatives with no meaningful difference in chemical shifts of protons in the binding pocket (Figure b). These outcomes show that the chosen solvent contributes to the formation of different diastereomeric complexes and to chiral recognition.

Conclusions

We have reported here the an class="Gene">synthesis class="Gene">and class="Gene">an class="Chemical">anion carboxylate binding properties of a series of chiral receptors bearing glucosamine pendant arms with varied geometries of the binding site. In the course of these studies we discovered the following:

Comparison of geomn class="Gene">etric pclass="Gene">arclass="Gene">amclass="Gene">an class="Gene">eters showed that receptors based on a five-membered ring demonstrate the highest affinity toward the carboxylate anion, creating mixed complexes with 1:2 ligand:anion stoichiometry.

Strong binding of the an class="Chemical">anion by the chirclass="Gene">al receptor is insufficient for successfully recognizing enclass="Gene">antiomers; rclass="Gene">ather, proper conformclass="Gene">ation is needed to ensure enclass="Gene">antioselective interclass="Gene">actions.

The complexation medium has a significant impact on complex formation and on enantiomer differentiation.

Overall, the studies presented above have provided insights into the effects of the geomn class="Gene">etry class="Gene">and size of the binding cclass="Gene">avity on class="Gene">affinity for class="Gene">an class="Chemical">carboxylate anions and chiral recognition. We evaluated whether the choice of solvent mixture affected the formation of diastereomeric complexes, keeping track of the binding process using the 1H NMR titration technique, and discovered that it does indeed exert an impact. Overall, our findings offer insight into the complementarity of chiral recognition processes.

Experimental Section

All precursors for n class="Gene">synthesis were obtclass="Gene">ained from commerciclass="Gene">al suppliers class="Gene">and were used without further purificclass="Gene">ation. All solvents were of reclass="Gene">agent grclass="Gene">ade quclass="Gene">ality class="Gene">and were dried under stclass="Gene">andclass="Gene">ard conditions. Flclass="Gene">ash chromclass="Gene">atogrclass="Gene">aphy wclass="Gene">as cclass="Gene">arried out using class="Gene">an class="Chemical">silica gel 60 (63–100 mesh); typically, a 40-fold mass excess of gel was used. TLC analysis was carried out on precoated silica gel plates (60 F254). 1H and 13C NMR spectra were recorded with 400 and 600 MHz NMR instruments. HRMS measurements were performed with ESI ionization and a TOF analyzer.

General Procedure for the Preparation of Diamide Derivatives 1b–1e

The reaction was carried out under n class="Chemical">argon conditions. To the solution of class="Gene">an class="Chemical">dichloride acid (1.1 mmol) in dry dichloromethane (100 mL), triethylamine (4.4 mmol) was slowly added. After 5 min of stirring, per-O-Ac-glucosamine hydrochloride (2.2 mmol), obtained according to the literature procedure,[17] was added and the reaction mixture was stirred overnight. Afterward, the reaction mixture was washed with 0.1 M HCl (2 × 50ml), saturated NaHCO3 (2 × 50ml), and water (1 × 50ml). Then, the organic phase was separated and was dried over MgSO4, and then filtrated and evaporated under vacuum. The crude product was purified using silica gel column chromatography with mixtures of dichlomethane and metanol (200:1 > 30:1, v/v] as eluents. The product was crystallized from a dichloromethane:hexane (1:3 v/v) mixture, yielding as solid.

Receptor 1a

The reaction was carried out under n class="Chemical">argon conditions. To the solution of class="Gene">an class="Chemical">diacid 2a (216 mg, 1 mmol) (obtained according to the literature procedure)[19] in dry DMF (100 mL) , triethylamine (1.4 mL, 10 mmol) was slowly added followed with HBTU (1.5 g, 4 mmol). After 10 min of stirring, per-O-Ac-glucosamine hydrochloride (1.5 g, 4 mmol), obtained according to the literature procedure,[17] was added and the reaction mixture was stirred overnight. Afterward, the mixture was concentrated on a rotary evaporator to 1/3 of the starting volume and water (10 mL) was added and the precipitate was washed with water. The crude product was purified by column chromatography on silica gel with the dichloromethane:methanol (99:1, v/v) mixture as an eluent. The product was crystallized from the dichloromethane:hexane (1:3, v/v) mixture yielding 1a (0.5 g, 57%) as blue solid. Mp: 124–127 °C. 1H NMR (400 MHz, DMSO-d6): δ = 8.88 (d, J = 8.9 Hz, 2H), 8.75 (d, J = 1.7 Hz, 2H), 8.47 (s, 1H), 8.07 (t, J = 3.8 Hz, 1H), 7.73 (d, J = 3.8 Hz, 2H), 5.95 (d, J = 8.8 Hz, 2H), 5.41 (t, J = 9.9 Hz, 2H), 5.02 (t, J = 9.6 Hz, 2H), 4.30–4.20 (m, 4H), 4.08–4.01 (m, 4H), 2.05 (s, 6H), 2.04 (s, 6H), 2.01 (s, 6H), 1.92 (s, 6H). 13C{H} NMR (100 MHz, DMSO-d6): δ = 170.0, 169.7, 169.2, 168.9, 168.9, 138.7, 137.2, 136.4, 135.5, 127.4, 124.4, 91.8, 72.4, 71.7, 68.0, 61.5, 53.4, 45.8, 20.5, 20.4, 20.3. HRMS (ESI–TOF) m/z: [M + Na]+ calcd for C40H46N2O20Na, 897.2542; found, 897.2523. Anal. Calcd for C40H46N2O20: C, 54.92; H, 5.30; N, 3.20. Found: C, 54.81; H, 5.40; N, 3.25.

Receptor 1b

Receptor 1b was prepared according to the general procedure using commercially available n class="Chemical">2,6-pyridinedicarbonyl dichloride (0.51 g, 2.5 mmol) yielding the product (1.35 g, 65%) class="Gene">as white powder. Mp: 196–197 °C. class="Gene">an class="Chemical">1H NMR (400 MHz, Acetonitrile-d3): δ = 8.32 (d, J = 9.4 Hz, 2H), 8.21–8.16 (m, 2H), 8.11 (dd, J = 8.6, 6.7 Hz, 1H), 6.03 (d, J = 8.7 Hz, 2H), 5.62–5.54 (m, 2H), 5.19 (t, J = 9.8 Hz, 2H), 4.27 (dt, J = 13.5, 5.8 Hz, 4H), 4.12 (dd, J = 12.4, 2.2 Hz, 2H), 3.99 (ddd, J = 10.0, 4.7, 2.3 Hz, 2H), 2.04 (s, 6H), 2.04 (s, 6H), 1.99 (s, 6H), 1.87 (s, 6H). 13C{H} NMR (100 MHz DMSO-d6): δ = 172.1, 171.3, 170.5, 170.1, 164.6, 149.2, 140.8, 125.6, 93.4, 73.8, 73.2, 68.9, 62.7, 54.5, 21.0, 21.0, 20.9, 20.9. HRMS (ESI–TOF) m/z: [M + Na]+ calcd for C35H43N3O20Na, 847.2385; found, 847.2356. Anal. Calcd for C35H43N3O20: C, 50.91; H, 5.25; N, 5.09. Found: C, 50.89; H, 5.30; N, 5.13.

Receptor 1c

Receptor 1c was prepared according to the general procedure using commercially available n class="Chemical">isophthaloyl dichloride (0.5 g, 2.5 mmol) yielding the product (1.5 g, 75%) class="Gene">as white powder. Mp: 117–120 °C. class="Gene">an class="Chemical">1H NMR (400 MHz, DMSO-d6): δ = 8.74 (d, J = 9.0 Hz, 2H), 8.17 (s, 1H), 7.87 (d, J = 6.7 Hz, 2H), 7.60 (t, J = 7.7 Hz, 1H), 5.91 (d, J = 8.7 Hz, 2H), 5.37 (t, J = 9.9 Hz, 2H), 4.99 (t, J = 9.6 Hz, 2H), 4.24 (m, J = 10.2 Hz, 4H), 4.01 (m, J = 11.3 Hz, 4H), 2.03 (s, 6H), 2.00 (s, 6H), 2.00 (s, 6H), 1.86 (s, 6H). 13C{H} NMR (100 MHz, DMSO-d6): δ = 170.0, 169.6, 169.2, 168.8, 165.8, 134.2, 129.8, 128.6, 126.2, 91.8, 72.4, 71.7, 68.0, 61.5, 52.7, 20.5, 20.4, 20.4, 20.2. HRMS (ESI–TOF) m/z: [M + Na]+ calcd for: C36H44N2O20Na, 847.2385; found, 847.2356. Anal. Calcd for C36H44N2O20·0.5 H2O: C, 51.86; H, 5.44; N, 3.36. Found: C, 51.57; H, 5.54, N, 3.36.

Receptor 1d

Receptor 1d was prepared according to the general procedure using n class="Chemical">acid dichloride obtclass="Gene">ained class="Gene">according to the literclass="Gene">ature procedure[18] (0.155 g, 1 mmol) yielding the product (320 g, 42%) class="Gene">as white powder. Mp: 231–234 °C. class="Gene">an class="Chemical">1H NMR (400 MHz, DMSO-d6): δ = 11.82 (s, 1H), 8.40 (d, J = 8.6 Hz, 2H), 6.67 (s, 2H), 5.85 (d, J = 8.8 Hz, 2H), 5.30 (t, J = 10.1 Hz, 2H), 4.96 (t, J = 9.7 Hz, 2H), 4.28–4.09 (m, 4H), 4.03 (t, J = 11.1 Hz, 4H), 2.02 (s, 6H), 2.00 (s, 6H), 1.99 (s, 6H), 1.86 (s, 6H).13C{H} NMR (100 MHz, DMSO-d6): δ = 170.0, 169.5, 169.2, 168.8, 159.6, 128.4, 111.9, 91.8, 72.2, 71.5, 68.1, 61.5, 52.1, 20.5, 20.5, 20.4, 20.3. HRMS (ESI–TOF) m/z: [M + Na]+ calcd for C34H43N3O20Na, 836.2338; found, 836.2349. Anal. Calcd for C34H43N3O20·H2O: C, 49.10; H, 5.45; N, 5.05. Found: C, 49.13; H, 5.26; N, 5.02.

Receptor 1e

Receptor 1e was prepared according to the general procedure using n class="Chemical">acid dichloride obtclass="Gene">ained class="Gene">according to the literclass="Gene">ature procedure[16] (0.380 g, 1.5 mmol) yielding the product (750 g, 75%) class="Gene">as purple powder. Mp: 223–226 °C. class="Gene">an class="Chemical">1H NMR (400 MHz, DMSO-d6): δ = 9.47 (d, J = 9.6 Hz, 2H), 8.45 (d, J = 8.7 Hz, 2H), 8.33 (s, 1H), 8.06 (t, J = 9.2 Hz, 1H), 7.72 (t, J = 9.7 Hz, 2H), 5.96 (d, J = 8.7 Hz, 2H), 5.44 (t, J = 9.9 Hz, 2H), 5.00 (t, J = 9.4 Hz, 2H), 4.30 (dd, J = 36.2, 9.4 Hz, 4H), 4.04 (d, J = 11.7 Hz, 4H), 2.04 (s, 12H), 2.00 (s, 6H), 1.87 (s, 6H). 13C{H} NMR (100 MHz, DMSO-d6): δ = 170.0, 169.7, 169.2, 168.9, 165.0, 141.4, 140.3, 138.8, 135.8, 128.7, 119.8, 92.0, 72.6, 71.6, 68.2, 61.5, 52.2, 20.6, 20.5, 20.4, 20.3. HRMS (ESI–TOF) m/z: [M + Na]+ calcd for C40H46N2O20Na, 897.2542; found, 897.2515. Anal. Calcd for C40H46N2O20: C, 54.92; H, 5.30; N, 3.20. Found: C, 54.86; H, 5.44; N, 3.15.