An Ideal C3-Symmetric Sulfate Complex: Molecular Recognition of Oxoanions by m-Nitrophenyl- and Pentafluorophenyl-Functionalized Hexaurea Receptors.

The anion-binding properties of two tripodal-based hexaureas appended with the m-nitrophenyl (1) and pentafluorophenyl (2) groups have been studied both experimentally and theoretically, showing strong affinities for sulfate over other inorganic oxoanions such as hydrogen sulfate, dihydrogen phosphate, bicarbonate, nitrate, and perchlorate. The structural analysis of the sulfate complex with 1 reveals that the receptor organizes all urea-binding sites toward the cavity at precise orientations around a tetrahedral sulfate anion to form an ideal C3-symmetric sulfate complex that is stabilized by 12 hydrogen-bonding interactions. The receptor and the encapsulated sulfate are located on the threefold axis passing through the bridgehead nitrogen of 1 and the sulfur atom of the anionic guest. The high-level density functional theory calculations support the crystallographic results, demonstrating that the C3-symmetric conformation of the sulfate complex is achieved due to the complementary NH···O between the receptor and sulfate.

Introduction

Molecular recognition of class="Chemical">sulfate is imclass="Chemical">portant because of its significant roles in biological and environmental class="Chemical">processes.[1−14] In nature, class="Chemical">pan class="Chemical">sulfate is found within the hydrophobic pocket of a sulfate-binding protein (SBP) derived from Salmonella typhimurium(15) and DNA helicase RepA.[16] The structure of the sulfate complex of the SBP reported by Pflugrath and Quiocho in 1985 reveals that the sulfate is encapsulated within the protein’s cleft through a total of seven hydrogen-bonding interactions, where three oxygen atoms are bound with six NH···O interactions and the forth oxygen atom is held with one OH···O bond.[15] The structural identification of the sulfate complex of DNA helicase RepA reported by Xu et al. suggests that the sulfate is encapsulated via six hydrogen-bonding interactions occupying the six ATPase active sites of RepA and one OH···O interaction with a water molecule.[16] Similar binding arrangement is also observed in a urea-based synthetic receptor complexed with hydrogen sulfate, showing six NH···O bonds and one OH···O hydrogen bond.[17] In the context of environmental viewpoints, sulfate is an inorganic contaminant in soil and water and is associated with acid rain that may cause health-related problems after a long-term exposure.[18,19] In industry, sulfate is known to interfere with the vitrification process during nuclear waste management;[20,21] thus, the separation of sulfate from the nitrate-rich waste mixtures is critical before the clean-up process.[21]

Inspired from nature’s rule, several types of neutral receptors including class="Chemical">amides,[22−24] class="Chemical">pan class="Chemical">thioamides,[25,26] ureas,[27−34] thioureas,[35−37] pyrroles,[38−43] and indoles[44−47] have been reported, which provide complementary hydrogen-bonding interactions for sulfate anions. Among them, urea-/thiourea-based receptors have been the focus of considerable interests because of the presence of two directional hydrogen-bond donors in a single unit and their potential applications in quantitative extractions[48,49] and transmembrane transport[50] of sulfate. Previous theoretical calculations by Hay et al. predicted that a sulfate anion can bind up to six urea groups to provide the optimal saturation with 12 H-bonds.[51] This prediction was later confirmed experimentally by Custelcean et al., showing a coordinatively saturated sulfate complex formed by two C3-symmetric tripodal trisurea ligands bridged with silver(I) ions.[27] Increasing the binding sites with the urea groups to a tren-based ligand, Wu et al. synthesized hexaurea ligands appended with p-nitrophenyl or ferrocenyl groups that were shown to encapsulate tetrahedral sulfates, while each oxygen atom of the guest anion is coordinated with three H-bonds, thus achieving the saturation of the coordination sphere of sulfates with six urea groups.[52,53] However, to the best of our knowledge, a sulfate complex bound to a synthetic receptor with a perfect C3 symmetry has not been reported so far.

In an effort to synthesize C3-symmetric class="Chemical">urea-class="Chemical">pan class="Chemical">based receptors with a higher order of binding sites, we recently synthesized a pentafluoro-substituted hexaurea receptor 2 that was found to encapsulate one carbonate anion; however, we were unsuccessful in obtaining the crystals of the sulfate complex with this receptor.[54] To examine the influence of attached groups on the binding strength and selectivity, we have been further interested in functionalizing the core cavity and synthesized the m-nitrophenyl-substituted hexaurea receptor 1. Herein, we report that the receptor 1 assembles the urea groups at accurate positions around a tetrahedral sulfate anion, thus forming an ideal C3-symmetric sulfate complex from the interactions of all NH-binding sites with the anion and NH···π interactions. It is shown by NMR studies that the receptors exhibit similar selectivities for sulfate over other oxoanions. Theoretical calculations based on the density functional theory (DFT) have been carried out to understand and support the binding insights on our experimental results.

Results and Discussion

Synthesis

The hexafunctional class="Chemical">urea receclass="Chemical">ptors (Chart ) were synthesized from the reaction of class="Chemical">pan class="Chemical">tris(2-aminophenyl)urea with 3 equiv of m-nitrophenyl isocyanate (for 1) or pentafluorophenyl isocyanate (for 2) in a toluene–tetrahydrofuran (THF) mixture to give about 90% yield. The crystals of a sulfate complex [1(SO4)](n-Bu4N)2 were grown from slow evaporation of a dimethyl sulfoxide (DMSO) solution of 1 in the presence of an excess n-tetrabutylammonium sulfate. Attempts to isolate X-ray quality crystals of other anion complexes with the receptors were unsuccessful.

Chart 1

Chemical Structures of the Receptors 1 and 2

Solution Binding Studies

Binding interactions between the receptors and class="Chemical">inorganic oxoanions were investigated using class="Chemical">pan class="Chemical">1H NMR experiments in DMSO-d6 at room temperature. The anions included in the titration studies were SO42–, HSO4–, HCO3–, H2PO4–, ClO4–, and NO3– in the form of their tetrabutylammonium (TBA) salts. The receptors, owing to their C3-symmetric conformations, exhibited four distinct NH resonances. To make an initial evaluation of the receptor–anion interactions, 1 equiv of each oxoanion was added separately to the receptors (Figure a,b). As shown in Figure a, the free receptor 1 displays urea–NH resonances at 9.65 (H1), 8.18 (H2), 7.98 (H3), and 6.55 ppm (H4). These peaks were significantly shifted downfield because of the addition of 1 equiv of SO42–, HSO4–, or H2PO4–. For HCO3–, a new set of NH signals appeared downfield along with the peaks of the free ligand. This observation suggests a slow exchange reaction on the NMR time scale. Similar shift changes were also observed for 2 upon the addition of 1 equiv of anions in DMSO-d6 (Figure b). On the other hand, the addition of ClO4– or NO3– ions showed a negligible change in the NH chemical shifts of the receptors, thus indicating weaker interactions of the receptor with these anions. For a quantitative assessment of the host–guest interactions, the receptors were titrated with varying amounts of anionic guests in DMSO-d6. The titration curves, as obtained from the changes in the chemical shifts or integrated intensities, were fitted to a 1:1 stoichiometry, as evaluated by nonlinear regression analysis[55] and supported by Job plot analysis. The addition of H2PO4– to the receptors also resulted in the downfield shifts of the NH signals. However, the binding constants could not be determined due to the broadening of peaks. Inspecting the observed binding data, as summarized in Table , suggests that both receptors display a similar binding trend, whereas 1, as compared to 2, shows comparatively stronger affinity for the anions. It might be due to the aromatic interactions between the peripheral substituents in 1 as supported by the X-ray structure as well as the theoretical calculations, thereby increasing the overall anion-binding ability of the host.

Figure 1

Partial 1H NMR spectra of 1 (a) and 2 (b) in the presence of 1 equiv of different anions in DMSO-d6 ([receptor]0 = 2 mM), showing changes in the NH chemical shifts in DMSO-d6 (see, Chart for the assignment of NH peaks). Overlapped NH peaks are not marked.

Table 1

Binding Constants (log K) of the Receptors with Anionsa

anions	1	2
SO42–	5.78(3), 5.85(3)b	5.55(2)
HSO4–	3.51(2)	3.35(2)c
HCO3–	3.28(2)	3.25(2)c
H2PO4–	d, 3.08(2)b	d
NO3–	<1	<1
ClO4–	<1	<1

Determined by 1H NMR titrations in DMSO-d6.

Determined by UV–vis titrations in DMSO.

Binding constant as taken from ref (54).

Binding constants could not be determined owing to the broadening of NH peaks.

class="Chemical">Partial class="Chemical">pan class="Chemical">1H NMR spectra of 1 (a) and 2 (b) in the presence of 1 equiv of different anions in DMSO-d6 ([receptor]0 = 2 mM), showing changes in the NH chemical shifts in DMSO-d6 (see, Chart for the assignment of NH peaks). Overlapped NH peaks are not marked.

Determined by pan class="Chemical">1H NMR titrations in class="Chemical">pan class="Chemical">DMSO-d6.

Determined by UV–vis titrations in pan class="Chemical">DMSO.

Both receptors exhibit remarkable selectivities for class="Chemical">sulfate anions among the other oxoanions included in this series. The strong affinity for class="Chemical">pan class="Chemical">sulfate is attributed to the structural complementarity, with the receptor’s cavity formed by six urea groups. This is also supported by the crystal structure analysis of the sulfate complex of 1 (discussed later), as well as from DFT calculations. The 1H NMR titration experiments of 1 or 2 with SO42– anions show a distinct slow exchange process in DMSO-d6. As illustrated in Figure a, the NH signals of free 1 were completely disappeared upon the addition of 1 equiv of sulfate, whereas a new set of NH resonances appeared downfield due to the formation of the sulfate complex. However, unlike the case of a p-nitrophenyl-substituted hexaurea, exhibiting a two-step binding mechanism with 1:1 and 1:2 complexes (receptor/sulfate),[52] the receptor 1 reached complete saturation with 1 equiv of SO42–, suggesting a purely 1:1 binding mechanism. This binding mode was further supported by the curve fitting isotherm (Figure b)[55] that was obtained from the relative changes in the integrated intensity of NH resonances for the complex and the free 1, as a result of the incremental addition of SO42– to the receptor. The calculated binding constants (in log K) for 1 and 2 are 5.78 and 5.55, respectively, which are higher than those observed for the sulfate complexes of trisureas substituted with pentafluorophenyl (log K = 4.72)[56] or p-cyanophenyl groups (log K = 4.7),[17] suggesting a chelate effect[2] due to the increased number of binding sites in hexaureas.

Figure 2

(a) Partial 1H NMR titration of 1 (2 mM) showing changes in the NH chemical shifts with an increasing amount of SO42– (20 mM) in DMSO-d6. Overlapped H4 peak is not marked in the complex. (b) 1H NMR titration plot of 1 (2 mM) with an increasing amount of (TBA)2SO4 in DMSO-d6. The plot was obtained from the relative change in the integration intensity ϕ (ϕ = INHc/[INHf + INHc]), where INHc is the intensity of NH1 signal in the sulfate complex, and INHf is the intensity of the corresponding NH signal in free 1.

(a) class="Chemical">Partial class="Chemical">pan class="Chemical">1H NMR titration of 1 (2 mM) showing changes in the NH chemical shifts with an increasing amount of SO42– (20 mM) in DMSO-d6. Overlapped H4 peak is not marked in the complex. (b) 1H NMR titration plot of 1 (2 mM) with an increasing amount of (TBA)2SO4 in DMSO-d6. The plot was obtained from the relative change in the integration intensity ϕ (ϕ = INHc/[INHf + INHc]), where INHc is the intensity of NH1 signal in the sulfate complex, and INHf is the intensity of the corresponding NH signal in free 1.

In addition, ultraviolet–visible (UV–vis) spectroscopy was employed to investigate the host–guest interactions in solution. Previous work demonstrated that the optical sensing of anions could be achieved through the functionalization of nitrophenyl groups to the class="Chemical">urea-/class="Chemical">pan class="Chemical">thiourea-binding sites.[29,33] The nitrophenyl-functionalized receptor 1 showed an absorption at λmax = 351 nm in DMSO, whereas no absorption was observed for the pentafluoro-substituted receptor 2 because of the absence of an effective chromophore. The addition of (TBA)2SO4 or TBAH2PO4 to the solution of 1 resulted in a change in the absorption, suggesting the interactions between the host and the anion. However, the receptor did not show any appreciable change in the absorption when TBAHSO4, TBANO3, or TBAClO4 was added (see Supporting Information), which is in agreement with the results of 1H NMR experiments. Figure shows the UV–vis titration spectra of 1, displaying a gradual bathochromic shift of the absorption band at λmax = 351 nm because of the incremental addition of TBAH2PO4, while the absorption is decreased. This spectral change is also accompanied by a visible color change from colorless to yellow (Figure S13), which could be the effect of the deprotonation of the receptor’s NH by the relatively basic phosphate anion.[32] The relative change (I/I0) in the λmax of 1 (where I0 and I represent the λmax of 1 before and after the addition of TBAH2PO4, respectively) upon the gradual addition of H2PO4– gave the best fit for a 1:1 binding mode (Figure , inset), yielding the binding constant log K = 3.08. The host showed a similar spectral change when it was titrated with sulfate anions; however, in this case, a sharp saturation was achieved after the addition of 1 equiv of sulfate (Supporting Information), which is in agreement with the 1H NMR titrations. From the UV–vis titrations, the binding constant for sulfate was found to be 5.85 (in log K), as estimated from the nonlinear regression analysis of a 1:1 binding model, which is fairly comparable to that obtained from the 1H NMR titrations.

Figure 3

UV–vis titration of 1 (1.5 × 10–4 M) with an increasing amount of TBAH2PO4 (1.5 × 10–2 M) in DMSO (inset showing the titration plot).

UV–vis titration of 1 (1.5 × 10–4 M) with an increasing amount of pan class="Chemical">TBAH2PO4 (1.5 × 10–2 M) in class="Chemical">pan class="Chemical">DMSO (inset showing the titration plot).

Crystal Structure Analysis

Single crystals of the class="Chemical">sulfate comclass="Chemical">plex of 1 were obtained by slow evaclass="Chemical">poration of a class="Chemical">pan class="Chemical">DMSO solution of the receptor in the presence of an excess tetrabutylammonium sulfate, (TBA)2SO4. The structural analysis by single-crystal X-ray diffraction reveals that the complex crystallizes as [1·SO4](TBA)2 in the trigonal space group P3. The asymmetric unit of the crystal contains three identical units (A, B, and C, Figure a), each sitting on a threefold rotation axis with an encapsulated sulfate. All three arms of each unit adopt an ideal C3-symmetric conformation in a folded umbrella to organize all urea-binding sites toward the cavity at precise positions around a tetrahedral sulfate anion, thus generating the unique C3-symmetric sulfate complex that is stabilized by a total of 12 H-bonds provided by six urea groups. Inspection of the structural and bonding features of the sulfate complex, as shown in Figures b,c for unit A, the receptor and the encapsulated sulfate are located on the threefold axis along the tertiary amine (N1A) and the sulfur atom (S1A). The distance from the terminal nitrogen to each of the three centroids of o-phenylene rings is 7.370(1) Å, whereas the three corresponding angles between the arms are the same (83.84(1)°), thereby creating an ideal C3-symmetric cavity. In the sulfate complex, each oxygen atom is held via three NH···O bonds, providing a total of 12 NH···O hydrogen bonds (dN···O = 2.834(6)–3.082(7), dH···O = 1.96–2.37 Å, and ∠N–H···O = 137.6–173.7°) that are comparable to those of the analogous complex of p-nitrophenyl-appended hexaurea (dN···O = 2.903–3.157 Å and ∠N–H···O = 142.68–172.68°).[52]

Figure 4

Crystal structure of the sulfate complex, [1·SO4](TBA)2: (a) asymmetric unit showing three identical units of A, B, and C; (b) side view of the unit A showing 12 H-bonds with six urea groups, where the symmetry-related atoms (N7A, N14A, and N17A) are leveled as unprimed, single primed, and double primed; (c) threefold axis view showing the perfect C3-symmetric sulfate complex (distances are shown in three significant figures; exact H-bonding distances with standard deviations are listed in Table ); (d) space-filling model of the sulfate complex. Non-acidic hydrogen atoms on (b,c), and TBA cations are omitted for clarity.

Crystal structure of the class="Chemical">sulfate comclass="Chemical">plex, [1·class="Chemical">pan class="Chemical">SO4](TBA)2: (a) asymmetric unit showing three identical units of A, B, and C; (b) side view of the unit A showing 12 H-bonds with six urea groups, where the symmetry-related atoms (N7A, N14A, and N17A) are leveled as unprimed, single primed, and double primed; (c) threefold axis view showing the perfect C3-symmetric sulfate complex (distances are shown in three significant figures; exact H-bonding distances with standard deviations are listed in Table ); (d) space-filling model of the sulfate complex. Non-acidic hydrogen atoms on (b,c), and TBA cations are omitted for clarity.

Table 2

H-Bonding Parameters (Å, °) for the Crystal Structure of [1·SO4]2– Motif as Shown in Figure b

	NH···O	H···O	D···O	∠DHO
arm 1	N4AH···O1A	2.08	2.960(6)	173.9
	N7AH···O2A	2.37	3.082(7)	137.6
	N14AH···O2A	1.96	2.834(6)	173.1
	N17AH···O2Aia	2.06	2.888(7)	157.1
arm 2	N4AiH···O1A	2.08	2.960(6)	173.9
	N7AiH···O2Aii	2.37	3.082(7)	137.6
	N14AiH···O2Aii	1.96	2.834(6)	173.1
	N17AiH···O2Aa	2.06	2.888(7)	157.1
arm 3	N4AiiH···O1A	2.08	2.960(6)	173.9
	N7AiiH7···O2Ai	2.37	3.082(7)	137.6
	N14AiiH···O2Ai	1.96	2.834(6)	173.1
	N17AiiH···O2Aiia	2.06	2.888(7)	157.1

Symmetry transformations used to generate equivalent atoms: −y + 1, x – y, and z.

Crystal structure of the class="Chemical">sulfate comclass="Chemical">plex of 1 showing class="Chemical">perfectly class="Chemical">parallel class="Chemical">planes formed by the three symmetry-related class="Chemical">pan class="Chemical">oxygen atoms (O2A) of sulfate (red) and by the three symmetry-related nitrogen atoms (light blue) of ureas (N7A, N14A, or N17A). The interplanar distance between the corresponding plane of sulfate and the N7A plane (formed by the three symmetry-related nitrogen atoms N7A) is 0.423 Å and is not shown for clarity.

Owing to the C3-symmetric conformation, one class="Chemical">oxygen atom (O1A) of the class="Chemical">pan class="Chemical">sulfate pointing to the tertiary amine lies on the threefold axis, forming three equidistant hydrogen bonds (NH···O = 2.960(6) Å) with the inner NHs (N4) from three different arms. The three arms of the receptor and the three oxygen atoms of the sulfate that remain on the C3 rotation axis are symmetrically equivalent; thus, one-third of the atoms of these groups is required to be specified. Each symmetry-related oxygen atom (O2B) of the sulfate is coordinated with three urea–NHs: two with o-phenylene-linked urea–NHs (N7A and N14A) from one arm and another with one m-nitrophenyl-linked urea–NH (N17A) from an adjacent arm [(N7H···O2A = 3.082(7), N14H···O2A = 2.834(6), and N17AH···O2Ai = 2.888(7) Å)] (see Figure b).

Along the class="Disease">rotation axis, three symmetry-related class="Chemical">pan class="Chemical">oxygen atoms (O2A) of the sulfate lies in an eclipsed conformation with respect to three α-carbon atoms connecting with the tertiary nitrogen (N1A) at a distance of 4.849 Å between N1A and S1A (Figure b). Because of the formation of an ideal C3-symmetric encapsulated sulfate complex, the trigonal plane of three symmetry-related oxygen atoms of the sulfate lies parallel to each of the four planes formed by the three symmetry-related nitrogen atoms of ureas (N4A, N7A, N14A, and N17A), with interplanar distances of 3.599, 1.939, 0.423, and 1.178 Å, respectively (Figure ). No dihedral angle was observed between these planes. Within the complex, each terminal m-nitrophenyl ring is folded toward the cavity and is almost perpendicularly aligned to an adjacent phenylene ring showing a dihedral angle of 89.58°, where the m-nitro group remains at the opposite side of the phenylene group to minimize the steric interactions between the aromatic rings (Figure c,d). These aromatic rings are stacked via T-shaped CH···π interactions (3.598–3.736 Å), thus further stabilizing the complex. This orientation of the two rings is in agreement with its p-nitrophenyl-substituted analogue with the sulfate complex,[52] but in contrast to its pentafluoro-substituted analogue 2 with the carbonate complex in which the corresponding two rings are parallel with respect to each other.[54] To the best of our knowledge, a perfect C3-symmetric sulfate complex with a synthetic receptor has not been reported previously.

Figure 5

Crystal structure of the sulfate complex of 1 showing perfectly parallel planes formed by the three symmetry-related oxygen atoms (O2A) of sulfate (red) and by the three symmetry-related nitrogen atoms (light blue) of ureas (N7A, N14A, or N17A). The interplanar distance between the corresponding plane of sulfate and the N7A plane (formed by the three symmetry-related nitrogen atoms N7A) is 0.423 Å and is not shown for clarity.

Computational Studies

To elucidate the geometries and to understand the energies within the multifunctional cavities of 1 and 2, high-level DFT calculations were performed on the interactions between the hosts and the pan class="Chemical">sulfate anions. All quantum mechanical calculations were carried out with the hybrid meta-exchange correlation functional M06-2X,[57−59] in conjunction with 6-31G(d,class="Chemical">p) basis using the Gaussian 09 class="Chemical">package of class="Chemical">programs.[60] Prior calculations have shown that the M06-2X functional accurately class="Chemical">predicts the binding modes of organic hosts that organize the binding sites to encaclass="Chemical">psulate the anions.[32,36,61−63] To establish a direct correlation between the theoretical and exclass="Chemical">perimental results, we have used a 1:1 binding model, as observed in the solution-binding studies as well as in the crystal structure of [1·class="Chemical">pan class="Chemical">SO4]2–. To this aim, the receptors were first optimized at the M06-2X/6-31G(d,p) level of theory.[58]

Using this optimized geometry, a pan class="Chemical">sulfate anion was incorclass="Chemical">porated at the center of each cavity and the comclass="Chemical">plex was reoclass="Chemical">ptimized at the M06-2X/6-31G(d,class="Chemical">p) level of theory. From the oclass="Chemical">ptimized structures, the binding energies of 1 and 2 for class="Chemical">pan class="Chemical">sulfate were calculated using the following equation

The binding energies for class="Chemical">sulfate, as calculated from the oclass="Chemical">ptimized geometries, are −235 and −206 kcal/mol, agreeing with the exclass="Chemical">perimental results showing the higher binding constant for class="Chemical">pan class="Chemical">sulfate with 1 than that with 2. We previously demonstrated that a tris(3-aminopropyl)amine-based tripodal urea encapsulates the sulfate with the binding energy of −173.0 kcal/mol in the gas phase.[36] The higher stability of [1·SO4]2– than that of [2·SO4]2– may be due to the geometrical complementarity between the receptor and the tetrahedral sulfate as discussed below.

The optimized structures of the free receptors and their class="Chemical">sulfate comclass="Chemical">plexes are shown in Figures and 7, resclass="Chemical">pectively. The class="Chemical">pan class="Chemical">hydrogen-bonding parameters of the DFT-optimized sulfate complexes are listed in Table . From our calculations, we found that 1 and 2 adopt different geometries, whereas both of them are organized to maintain their C3 conformations. Specially, the terminal aromatic rings of 1 are folded to form three pairs of T-shaped CH···π interactions from the interactions of m-nitrophenyl ring with an adjacent o-phenylene ring, thereby creating a tetrahedral cavity in a folded cone shape. The cavity is further stabilized by three intramolecular H-bonding interactions between the urea–NH and the carbonyl O groups (see Figure a), as also observed previously in a free p-cyanophenyl-based tripodal urea.[64] On the other hand, the free receptor 2 adopts a cone shape, exhibiting hydrogen-bonding interactions at both inner and outer cavities (see Figure b). From these calculations, it may be suggested that the geometry of 1 provides the best complementarity to fit a tetrahedral sulfate within the host’s self-generated cavity. The structures of 1 and 2 complexes with the sulfate, as shown in Figure , suggest that each receptor encapsulates SO42– inside the cavity utilizing all six urea groups, providing a total of 12 H-bonds in each case. This binding arrangement, however, leads to the expansion of the cavity, as expected, whereas the C3 symmetry is preserved in both complexes.

Figure 6

Optimized structures (a) 1 and (b) 2 showing perspective views, calculated at the M06-2X/6-31G(d,p) level of theory.

Figure 7

Optimized structures (a) [1(SO4)]2– and (b) [2(SO4)]2– showing perspective views, calculated at the M06-2X/6-31G(d,p) level of theory.

Table 3

Hydrogen Parameters (Å, °) for the Sulfate Complexes of 1 and 2a

		[1(SO4)]2–			[2(SO4)]2–
	NH···O	NH···O	H···O	∠DHO	NH···O	H···O	∠DHO
1st arm	N2H···O7	2.991	1.973	176.5	2.907	1.886	177.5
	N3H···O8	2.937	1.941	165.5	2.914	1.919	164.2
	N8H···O8	2.888	1.881	169.3	2.932	2.113	136.5
	N9H···O9	2.942	1.937	167.1	2.792	1.770	172.6
2nd arm	N4H···O7	2.991	1.973	176.5	2.895	1.874	177.1
	N5H···O10	2.937	1.941	165.5	2.937	1.939	165.4
	N10H···O10	2.888	1.881	169.3	2.891	1.934	155.8
	N11H···O8	2.942	1.937	167.1	2.870	1.843	177.8
3rd arm	N6H···O7	2.991	1.973	176.5	2.928	1.908	176.3
	N7H···O9	2.937	1.941	165.5	2.899	1.895	167.5
	N12H···O9	2.888	1.881	169.3	2.877	1.901	159.8
	N13H···O10	2.942	1.937	167.1	2.843	1.816	178.2

Calculated with DFT at M06-2X/6-31G(d,p).

Optimized structures (a) [1(pan class="Chemical">SO4)]2– and (b) [2(class="Chemical">pan class="Chemical">SO4)]2– showing perspective views, calculated at the M06-2X/6-31G(d,p) level of theory.

Notably, as in the crystal structure of [1·class="Chemical">SO4]2–, a class="Chemical">perfect C3 symmetry is again observed in the DFT-oclass="Chemical">ptimized structure of the former case (see Table ), where each arm forms four H-bonds (NH···O = 2.991, 2.937, 2.888, and 2.942 Å) with the encaclass="Chemical">psulated class="Chemical">pan class="Chemical">sulfate. The optimized complex is further stabilized by three T-shaped CH···π interactions (3.54 Å) between the terminal m-nitrophenyl rings and adjacent o-phenylene rings. By contrast, the optimized [2·SO4]2– complex, as shown in Figure b, adopts a different geometry in which the terminal aromatic groups are stacked through strong CF···π interactions (2.72, 2.85, and 3.07 Å).[65] To corroborate the experimental binding constants, we also calculated the binding energies for other oxoanions, showing the binding trend SO42– > HSO4– > HCO3– > H2PO4– > NO3– > ClO4– (see Supporting Information, Table S1). These calculated data are in agreement with the experimental titration results, showing the highest binding energies for dinegatively charged sulfates with both receptors, whereas the m-nitro analogue 1 exhibits stronger anion-binding affinity than its pentafluoro analogue 2.

Conclusions

We have designed and synthesized two tripodal-class="Chemical">based class="Chemical">pan class="Chemical">hexaureas appended with the m-nitrophenyl (1) and pentafluorophenyl (2) groups. Their binding properties have been investigated for inorganic oxoanions such as hydrogen sulfate, dihydrogen phosphate, bicarbonate, nitrate, and perchlorate, showing high binding affinities for the sulfate anion. We have isolated and structurally characterized the sulfate complex of 1 by single-crystal X-ray analysis, confirming the formation of the crystallographically perfect C3-symmetric sulfate complex of 1 in which all six urea groups excellently organize toward the center of the cavity to encapsulate a sulfate anion with 12 NH···O hydrogen bonds. Both the receptor and the anion are located on the threefold axis passing through the tertiary nitrogen of the receptor and the sulfur atom of the encapsulated sulfate. The computational studies performed by the high-level DFT calculations demonstrate that the C3-symmetric sulfate complex is achieved because of the best complementarity between the receptor and the sulfate anion. Although the complete saturation of the coordination sites of a C3-symmetric sulfate was achieved previously by six urea groups provided by two trisureas[27] or single hexaureas[52] providing the optimal 12 hydrogen bonds as predicted theoretically,[51] however, to the best of our knowledge, a perfect C3-symmetric sulfate complex with a synthetic receptor has not been reported so far. The receptor 1 represents an exceptional example that encapsulates a sulfate anion to form an ideal C3-symmetric sulfate complex.

Experimental Section

General

All reagents and solvents were purchased from Sigma-Aldrich and were used as received. The synthesized compounds were characterized using the common laboratory techniques as described before.[36]

Synthesis of the Receptors 1 and 2

The synthesis of 1 was carried out by the reaction of class="Chemical">tris(2-aminophenyl)urea[54] (0.60 g, 1.09 mmol) with 3 equiv of class="Chemical">pan class="Chemical">3-nitrophenyl isocyanate (0.54 g, 3.38 mmol) in a solution of toluene and THF (2:1, 150 mL). The reaction mixture was refluxed overnight at 100–110 °C under a nitrogen atmosphere and was cooled at room temperature. The precipitate thus formed was collected by filtration and washed with CH2Cl2. The compound was dried under vacuum to give 1 as a chalky yellow powder. Yield: 1.01 g (89%). mp: 220–222 °C, 1H NMR (500 MHz, DMSO-d6, TSP): δ 9.66 (s, 3H, ArNH), 8.52 (s, 3H, ArH), 8.18 (s, 3H, ArNH), 7.98 (s, 3H, ArNH), 7.86 (d, J = 9.0 Hz, 3H, ArH), 7.68 (d, J = 9.0 Hz, 3H, ArH), 7.56 (m, 6H, ArH), 7.43 (s, 3H, ArH), 7.03 (t, J = 6.1 Hz, 6H, ArH), 6.55 (s, 3H, NH), 3.20 (d, J = 6 Hz, 6H, NHCH2), 2.61 (t, J = 6.1 Hz, 6H, NCH2). 13C NMR (125 MHz, DMSO-d6): 156.7 (ArCO), 153.4 (NHCO), 148.5 (ArC), 141.7 (ArC), 132.3 (ArC), 131.3 (ArC), 130.4 (ArC), 124.7 (ArC), 124.5 (ArC), 116.5 (ArC), 112.4 (NHCH2), 54.4 (NHCH2CH2). Anal. Calcd for C48H48N16O12: C, 55.38; H, 4.65; N, 21.53. Found: C, 55.22; H, 4.52; N, 21.56. ESI-MS (+ve) m/z: calcd for C48H49N16O12, 1041.36 [M + H]+; found, 1041.25. The receptor 2 was synthesized following the procedures as described before.[54]

Synthesis of the Sulfate Complex of 1 ([1·SO4](TBA)2)

The class="Chemical">sulfate comclass="Chemical">plex of 1 was obtained from slow evaclass="Chemical">poration of a class="Chemical">pan class="Chemical">DMSO solution of 1 (30 mg, 0.029 mmol) in the presence of excess (∼2.0 equiv) n-tetrabutylammonium sulfate in a vial at room temperature in 5 days. Yield: 23 mg (70%). Anal. Calcd for C48H48N16O12: C, 59.24; H, 7.46; N, 15.54. Found: C, 59.28; H, 4.47; N, 15.56. ESI-MS (−ve) m/z: calcd for [C48H48N16O12SO4)/2], 568.16 [(M·SO4)/2]−; found, 568.18. The compound was further characterized by a single-crystal X-ray diffraction analysis.

NMR Binding Studies

The binding constants of the receptors with different oxoanions (class="Chemical">SO42–, class="Chemical">pan class="Chemical">HSO4–, H2PO4–, ClO4–, and NO3– in the form of their TBA salts) were obtained by the 1H NMR titrations in DMSO-d6 using a 500 MHz Bruker instrument at room temperature. The initial concentrations of the receptors and anions were 2 and 20 mM, respectively. Sodium salt of 3-(trimethylsilyl) propionic-2,2,3,3-d4 (TSP) acid in DMSO was used as an external reference in a capillary tube. Each titration was performed by 12–14 measurements. The association constants (K) were calculated using a 1:1 binding model[55] from the changes of chemical shifts of NH for the fast exchange reactions or relative changes in the intensity of the NH resonance for complexes and free receptors as described previously.[54]

UV–Vis Binding Studies

The receptor 1 showed an absorption at λmax = 351 nm in class="Chemical">DMSO, whereas no absorclass="Chemical">ption was observed for 2 because of the absence of an oclass="Chemical">ptically active chromoclass="Chemical">phore. UV–vis titration studies were class="Chemical">performed by titrating 1 with different oxoanions as their class="Chemical">pan class="Chemical">TBA salts in DMSO at room temperature. The initial concentrations of the receptor and the anions were 1.5 × 10–4 and 1.5 × 10–2 M, respectively. Each titration was performed by 15 measurements in the range of 0–35 equiv of anions, and the binding constant K was calculated by fitting the relative UV–vis absorbance or wavelength with a 1:1 binding model.[55]

X-ray Crystallography

The single-crystal structure of 1 was analyzed using a diffractometer with a Bruker pan class="Gene">APEX CCD area detector,[66] as described before.[67] Details of the crystal data and structure refinement are listed in Table . The structure was refined by a full-matrix least-squares method using the SHELXL2013 class="Chemical">program.[68]

Table 4

Crystal Data and Structure Refinement for the Sulfate Complex of 1

chemical formula	C48H48N16O12·SO4·2(C16H36N)
formula mass	1621.99
a/Å	27.389(2)
b/Å	27.389(2)
c/Å	9.7549(7)
α/deg	90
β/deg	90
γ/deg	120
unit cell volume/Å3	6337.3(10)
temperature/K	100(2)
space group	P3
no. of formula units per unit cell, Z	3
radiation type	Mo Kα
absorption coefficient, μ/mm–1	0.114
no. of reflections measured	50 226
no. of independent reflections	14 912
Rint	0.0399
final R1 values (I > 2σ(I))	0.0710
final R1 values (all data)	0.1024
goodness of fit on F2	1.007