Sulfated prism[6]arene as a water-soluble host for molecular recognition of dye pollutants in aqueous environments

Abstract

The contamination of urban groundwater by synthetic organic compounds represents a growing environmental concern; therefore, the development of supramolecular systems capable of selectively recognizing pollutants in aqueous environments is increasingly important, both for detection and remediation purposes. Herein, we report the synthesis of a water-soluble prism[6]arene extensively functionalized with sulfate groups at both rims. This host exhibits remarkable binding affinities toward organic dyes, specifically rhodamine B and methylene blue. The binding properties with these guests were investigated via UV–vis, fluorescence, NMR, DLS, DOSY, and ITC. Notably, in the presence of rhodamine B, the sulfated prism[6]arene forms both 1:1 endo-cavity and 1:2 capsular complexes, as evidenced by multi-technique characterization. Furthermore, in silico calculations suggest that two macrocycle units assemble into a water-bridged supramolecular capsule that encapsulates one molecule of rhodamine B. In contrast, methylene blue exclusively forms 1:1 endo-cavity complexes. These findings establish sulfated prism[6]arene as a promising water-soluble macrocycle paving the way for applications in environmental remediation.

Urban groundwater contamination by synthetic organic dyes poses a significant environmental challenge due to their toxicological properties and prevalence as micropollutants. Here, the authors develop water-soluble prism[6]arene hosts functionalized with sulfate groups (PrS[6]^SO₃⁻) which demonstrate exceptional potential for selective recognition and sensing of organic dye pollutants.

Introduction

Water is the most abundant solvent on Earth, and it is a precious resource essential for supporting the development of living systems¹. It also plays a crucial role in human activities, including industrial processes, which contribute to a progressive decline in water quality worldwide¹. In particular, the contamination of urban groundwater with an increasingly diverse range of organic pollutants, especially synthetic organic dyes², has become an escalating environmental concern³. These dyes are extensively used in industries such as textiles, tannery, cosmetics, and food, leading to their widespread presence in water and soil. They have been detected in water samples, sediments, and aquatic organisms, classifying them as micropollutants⁴. Given their toxicological properties, monitoring their occurrence in water matrices is crucial⁵. Consequently, the development of more efficient and selective systems for the supramolecular recognition and sensing of organic dye pollutants in water is increasingly critical⁶. Within this framework, designing novel supramolecular hosts^6,7 that can operate in aqueous environments poses a significant challenge in supramolecular chemistry, particularly with the objective of creating water-soluble systems for the recognition of organic dye pollutants in water^6,7.

This strategy aims to produce prototypes⁸ that can be readily integrated into more complex and functional architectures⁶ for water contaminant sensing and remediation. Macrocycles, including calixarenes, pillararenes, cyclodextrins, and cucurbiturils, have historically played a pivotal role in this field⁹, particularly as molecular recognition platforms in aqueous environments^10–12. Among these, the family of cyclophanes has demonstrated exceptional potential due to their ability to form highly selective host–guest complexes^13,14. The main advantage of cyclophanes lies in their chemical versatility, which enables access to a broad range of hosts with diverse functionalities.

Prismarenes^15–20 represent a novel class of deep-cavity macrocycles composed of naphthalene units connected by methylene bridges. These macrocycles are characterized by their deep hydrophobic cavities, which can accommodate organic cations. Prismarenes exhibit unique emission and absorption properties and are characterized by their planar chirality.

Currently, limited information is available regarding molecular recognition in water using prismarenes^8,18–20. Most of the existing knowledge primarily pertains to prism[5]arene⁸, while significantly much less is known about prism[6]arene²⁰. For example, our group¹⁸ and others¹⁹ have reported synthesizing water-soluble prism[5]arene derivatives functionalized with carboxylate groups on their rims, which effectively complex neutral organic pollutants in water⁸. The primary driving force for the complexation of neutral organic guests in water with carboxylato-functionalized prismarenes arises from the expulsion of frustrated water molecules from their deep cavities^8,20.

Prism[6]arene possesses a wider cavity compared to its five-membered counterpart²⁰, resulting in different recognition properties. Among the anionic moieties that can confer water solubility to organic macrocycles, sulfates exhibit superior charge density and pH stability compared to carboxylates^21,22. This increased charge density of sulfate groups is due to their compact molecular size and symmetrical geometry, which enhances hydration and promotes stronger intermolecular interactions with cationic species^21,22. Consequently, sulfate functionalities are advantageous in the design of water-soluble host molecules with unique supramolecular properties.

In this study, we report the synthesis of a novel water-soluble prismarene, designated sulfated-prism[6]arene (PrS[6]^SO3−, Fig. 1). Our primary objective is to investigate this molecule as a receptor for the recognition of emerging organic pollutants in aqueous media (Fig. 1). Rhodamine B (RhodB) and Methylene Blue (MethB) were selected due to their extensive use in the textile industry, leading to their presence in water sources. These compounds adversely affect aquatic ecosystems due to their potential to accumulate in fishes and aquatic plants, compromising their development and survival. According to the classification provided by companies to ECHA in REACH registrations, both RhodB (https://echa.europa.eu/it/substance-information/-/substanceinfo/100.001.259?_disssubsinfo_WAR_disssubsinfoportlet_substanceId=100.001.259) and MethB (https://echa.europa.eu/it/substance-information/-/substanceinfo/100.000.469?_disssubsinfo_WAR_disssubsinfoportlet_substanceId=100.000.469) are harmful to aquatic life with long-lasting effects.

Fig. 1. Chemical structures of PrS[6]^SO3− and dyes.

Chemical structures of sulfated-prism[6]arene PrS[6]^SO3− and dyes investigated in this work.

Results and discussion

Synthesis of sulfated-prism[6]arene

Scheme 1 illustrates the synthesis of PrS[6]^SO3−. The hydroxylated precursor prism[6]arene PrS[6]^OH (1) was synthesized following an established protocol²³. PrS[6]^OH was reacted with pyridinium·SO₃⁻ in pyridine at 90 °C^21,22, resulting in the formation of PrS[6]^SO3− with a yield of 62% after the addition of Na₂CO₃ and subsequent dialysis in water. The new compound was thoroughly characterized using ¹H and ¹³C NMR spectroscopy (SI, Figs. S1 and S2), 2D NMR (SI, Figs. S3 and S4), and high-resolution ESI mass spectrometry (SI, Fig. S5). Prism[6]arenes are known to exist in 8 distinct conformations¹⁶, each exhibiting different planar chirality due to rotational dynamics around the naphthalene units. The ¹H-NMR spectrum (SI, Fig. S1) of PrS[6]^SO3− in D₂O at ambient temperature is characterized by sharp singlets.

Scheme 1. Synthesis of PrS[6]^SO3−.

Schematic representation of the synthesis of sulfated-prism[6]arene PrS[6]^SO3−.

This observation indicates that the naphthalene units of PrS[6]^SO3− undergo rapid rotation on the 600 MHz NMR timescale. Due to the presence of anionic groups on the rims, we wondered whether the conformational behavior of PrS[6]^SO3− might vary at acidic or basic pH. The ¹H-NMR spectra of PrS[6]^SO3− recorded at pH 4.5 and pH 12 (Fig. S12) showed no differences compared to the spectrum of the macrocycle at pH 7. This result indicates that the conformational behavior of PrS[6]^SO3− remains consistent regardless of pH variations. The high-resolution FT-ICR ESI mass spectrum confirmed the molecular mass of PrS[6]^SO3−, with an observed molecular ion peak at m/z 728.8723, matching the calculated value of 728.8707 for [M + 9Na]³⁻ (see SI, Fig. S5).

Initially, the interaction between PrS[6]^SO3− and the dyes RhodB and MethBwas examined using ¹H-NMR spectroscopy (Figs. S6 and S9). The ¹H-NMR spectra of mixtures of PrS[6]^SO3−/RhodB and PrS[6]^SO3−/MethB at stoichiometric ratios of 1:0.5 and 1:1 exhibited significant line broadening and shifts in the signals of both the host and guest, compared to the resonances observed for the free species (Figs. S6 and S9). These results suggested the formation of supramolecular endo-cavity complexes (vide infra) between PrS[6]^SO3− and the dyes. ¹H-NMR spectra in D₂O of the mixtures PrS[6]^SO3−/RhodB and PrS[6]^SO3−/MethB, in the presence of interfering ions such as Na⁺, K⁺, Ca²⁺, Mg²⁺, Cl⁻, HCO₃⁻, and SO₄²⁻ (more details in the Supporting Information), did not show significant changes, suggesting that the complexes remain stable in the presence of the most common ions found in natural water (Figs. S15 and S16).

Based on these findings, the complexation equilibrium of PrS[6]^SO3− in water with RhodB and MethB was rigorously analyzed using calorimetry, UV-Vis titration, dynamic light scattering (DLS), and Diffusion-Ordered Spectroscopy (DOSY) NMR spectroscopy. Finally, the structures of these complexes were studied using molecular dynamics (MD) simulations based on the results obtained.

Calorimetric investigations

Calorimetric titrations were conducted to characterize the thermodynamics of the complexation process of PrS[6]^SO3− with both RhodB and MethB in aqueous solution (Fig. 2). These dyes exhibit pronounced aggregation properties at very low concentration, leading to the formation of dimers and, in some cases, even to higher order aggregates^24–27. Consequently, ad hoc calorimetric experiments were carried out (Figs. S17–S20) to enable the determination of the dimer/aggregate species formed as well as their aggregation constants and enthalpy changes (Table 1, entries a and b for RhodB and MethB, respectively), which refer to the dimerization reaction: 2 Dye ⇄ (Dye)₂. The formation of the (Dye·Dye) species occurs with comparable binding affinities for both RhodB and MethB, and these values are comparable/slightly lower than those reported in water (Table 1). The small discrepancies can be attributed to the different experimental conditions, as the aggregation process is strongly affected by factors including ionic strength, pH, temperature, and the presence of organic solvents. The determination of the enthalpy and entropy contributions to ΔG⁰ provided insights into the forces governing the aggregation process. The enthalpy gain, which drives the dimerization for both the investigated dyes, can be ascribed to favorable attractive forces, including π‧‧‧π, CH‧‧‧π, and cation‧‧‧π interactions. An additional contribution to the dye aggregation comes from desolvation, as indicated by the positive entropic term. ITC titrations, such as those shown in Figs. 2a, b and S19 for RhodB as well as in Figs. 2c, d and S20 for MethB, were conducted for the two dyes^24–27.

Fig. 2. ITC characterization of host–guest complexation between PrS[6]^SO3− and RhodB/MethB.

a ITC titration of RhodB (1.40 mM) into PrS[6]^SO3− (0.10 mM) at 25 °C and pH 5.5 (acetate buffer, 50 mM); b observed vs. calculated net heat data. c ITC titration of MethB (1.39 mM) into PrS[6]^SO3− (0.11 mM) at 25 °C and pH 5.5 (acetate buffer, 50 mM); d observed vs. calculated net heat data. e Thermodynamic parameters for the host–guest complex formation at 25 °C and pH 5.5.

Table 1.

Thermodynamic parameters and binding constants. LogK and thermodynamic parameters ^a for host (H)–guest (G) complex formation and dye dimerization at 25 °C and pH 5.5 (acetate buffer, 50 mM)

Dye	Equilibrium	LogK	ΔH0 (kJ mol−1)	ΔS0 (J K−1 mol−1)
RhodB	H + G ⇄ HG	5.5 (1)	−52.68 (1)	−71 (3)
	HG + H ⇄ H2G	3.8 (3)	0.8 (3)	76 (7)
	HG + G ⇄ HG2	4.1 (2)	−36.49 (2)	−45 (5)
(a)	2RhodB ⇄ (RhodB)2	3.2 (1)	−15.7 (5)	7.7 (2)
MethB	H + G ⇄ HG	5.0 (3)	−58.09 (7)	−100 (4)
(b)	2MethB ⇄ (MethB)2	3.13 (3)	−50.21 (8)	108.5 (8)

^aσ in parentheses.

The involved species, pertinent binding constant, and thermodynamic parameter values are listed in Table 1 and displayed in Fig. 2e. Calorimetric experiments pointed out the ability of PrS[6]^SO3− to form 1:1 host–guest complexes with both dyes and species with higher order stoichiometry in the case of RhodB. The binding affinity of PrS[6]^SO3− towards the two dyes is comparable and attributable to interactions occurring between the charged rim and the cavity of the host with the aromatic moieties of the dyes. The formation of the HG complex is enthalpically favored and driven for both guests: π‧‧‧π and CH‧‧‧π interactions with the prismarene cavity, along with electrostatic interactions between the positive nitrogen atom of the dyes and the negatively charged host, drive the formation of the host–guest complex. The unfavorable entropic contribution results from the loss of degrees of freedom of the system due to HG complex formation^24,25.

The logK value for the complexation of RhodB by PrS[6]^SO3− is about 1.7 log units (Table 1) higher than that reported for the inclusion in β- and γ-cyclodextrins²⁸ as well as in sulfonate calix[6]arene²⁹ in buffered water at pH 6.4‒6.6. A significant increase in the binding affinity (more than two orders of magnitude) is also observed for MethB when compared to the inclusion of the same guest in cyclodextrin derivatives at pH 7.2³⁰. The higher binding affinity observed for PrS[6]^SO3− compared to more traditional macrocycles can be attributed to the differences in the host structure and functionalization. In particular, the deep aromatic cavity of PrS[6]^SO3− enables π‧‧‧π interactions with the dyes, which are not accessible in cyclodextrins and may be less effective in calixarene derivatives, which have a cone-like structure. In addition, the presence of multiple sulfonate groups promotes favorable electrostatic interactions with the cationic guests, while the overall size and shape complementarity of the cavity provides a more suitable environment for dye inclusion.

PrS[6]^SO3− also forms a 2:1 complex, which could be attributed to a capsular assembly with RhodB (H₂G species), which is instead not observed for MethB. The presence of an additional aromatic moiety in RhodB limits the extent of guest encapsulation within the host cavity. This maintains an optimal separation between the two aromatic cavities of the prismarenes, which in turn mitigates the electrostatic repulsion between the macrocycle rims and promotes the formation of the capsular assembly. A similar stoichiometry has been proposed for the inclusion complexes formed by perylenediimide with curcubit[7] and [8]uril³¹. The large entropic drive for the formation of the H₂G species stems from the extensive desolvation of the HG complex and the additional prismarene cavity that assembles into the capsular structure. The desolvation process, however, requires a high enthalpy cost, making the ΔH⁰ contribution close to zero.

PrS[6]^SO3− was further observed to form a HG₂ complex with RhodB. Several hypotheses may account for this stoichiometry, which has been reported for other supramolecular complexes³². Previous studies on a structurally similar prism[6]arene bearing carboxylate groups at both rims showed that the host cavity is unable to encapsulate more than one guest molecule, even when these have a smaller size^18,20. Therefore, here it is plausible to assume that a second rhodamine molecule may bind to the (already complexed) dye residue protruding from the prismarene cavity. Alternatively, a second guest molecule might interact with the external aromatic walls of the RhodB@PrS[6]^SO3− 1:1 complex.

Finally, a further hypothesis could envisage the complexation of a rhodamine dimer, as reported for the inclusion of the pyronine B dimer (a structural analog of rhodamine) into the γ-cyclodextrin cavity³³ or for the formation of host–guest complexes with other dyes having aggregation properties³⁴. For any of the above scenarios, the exothermic nature of the process observed by calorimetry is fully consistent with the establishment of favorable π‧‧‧π interactions between aromatic moieties, which are responsible for the enthalpic driving force. The negative entropy change arises from the structural freezing upon formation of such multi-component complex species. There is evidence of an HG₂-type complex also for MethB, with a stepwise logK value of 3.5 (2) and a highly exothermic ΔH₀ term (> −200 kJ mol⁻¹). Although these values are consistent with the hypotheses listed for RhodB, the notably large enthalpy change suggests the formation of a higher-order host-dye aggregate, likely stabilized by extensive π‧‧‧π interactions. However, accurately determining the stoichiometry of this complex is very challenging, since multiple dye moieties may be part of the assembly. Similar findings have been reported for methylene blue and sulfonate calix[8]arenes that may form large complexes involving more than two dye molecules³⁵.

Dynamic light scattering (DLS) and DOSY investigations

DLS analyses were conducted by titrating PrS[6]^SO3− with both dyes (Figs. 3 and 4). A fresh solution of PrS[6]^SO3− with a concentration of 4.0 × 10⁻⁴ M was prepared in a sodium acetate buffer (0.05 M) at pH 5.5. The dyes were added in a range of 0.1 to 5.0 equivalents (see SI for experimental details). An increase in the particle diameter from approximately 1.6 nm to 4.3 nm upon the addition of up to 0.3 equivalents of RhodB corroborated the formation of complexes with a stoichiometry greater than 1:1 (Fig. 3b–d). Specifically, the diameters obtained from DLS measurements further corroborate the hypothesis of the formation of a supramolecular capsule composed of two PrS[6]^SO3− molecules encapsulating a single RhodB molecule, i.e., RhodB@(PrS[6]^SO3−)₂ (vide infra).

Fig. 3. UV–Vis titration and DLS investigation of the RhodB@(PrS[6]^SO3− complex.

a UV-Vis titrations of RhodB (4.7 × 10⁻⁶M) with PrS[6]^SO3− (2.2 × 10⁻⁴ M) in water buffered at pH 5.5 (acetate buffer 50 mM); b diameter distributions of PrS[6]^SO3− (4.0 × 10⁻⁴ M); c after addition of 0.3 equiv of RhodB and d 1 equiv of RhodB.

Fig. 4. UV–Vis titration and DLS investigation of the MethB@PrS[6]^SO3− complex.

a UV-Vis titrations of MethB (5.0 × 10⁻⁶ M) with PrS[6]^SO3− (3.3 × 10⁻⁴ M) in water buffered at pH 5.5 (acetate buffer 50 mM); b diameter distributions of PrS[6]^SO3− (4.0 × 10⁻⁴ M); c after addition of 0.3 equiv of MethB and d 1 equiv of MethB.

Furthermore, the increase in the dispersion coefficient at higher dye concentrations may indicate the formation of a variety of complexes with different stoichiometries, as well as the presence of free dye molecules or RhodB aggregates. DLS studies (Fig. 4b–d) suggest the formation of a 1:1 complex between MethB and PrS[6]^SO3−. On the other hand, the addition of methylene blue to the prismarene solution did not cause any change in particle diameter, ruling out the formation of complexes with a stoichiometry greater than 1:1 with this dye (Fig. 4b–d), in agreement with the results described above.

DOSY NMR^36–40 is well-known for its application in host–guest chemistry to investigate supramolecular complexes^36–42.

DOSY NMR experiments were conducted in acetate-buffered D₂O (pH 5.5) at 298 K to investigate the complexation processes occurring in solution between PrS[6]^SO3− and the dyes (see Supporting Information, Figs. S38–S42). A self-diffusion coefficient for free PrS[6]^SO3−, measured at D = 2.514 × 10⁻¹¹ m² s⁻¹ from the mono-exponential Stejskal–Tanner fit of a representative resonance, was utilized as a reference for the isolated macrocycle in solution (Figs. S38 and S39). In the dye-containing mixtures, the pronounced line broadening observed in the ¹H-NMR spectra—resulting from the complexation/decomplexation equilibria on the NMR timescale—impedes a strictly quantitative analysis of the diffusion data. Under these circumstances, the attenuation profiles of the resonances are less well defined, making the diffusion coefficients extracted for the supramolecular species best interpreted qualitatively. The pseudo-2D DOSY spectra of the mixtures of PrS[6]^SO3− with MethB and RhodB were analyzed comparatively with respect to PrS[6]^SO3− (Figs. S40–S42). Within this context, the DOSY behavior of the MethB system is consistent with the formation of a relatively compact host–guest complex, supporting a predominant 1:1 stoichiometry. In contrast, the observed diffusion trend for RhodB indicates the formation of a larger supramolecular assembly, consistent with the 2:1 stoichiometry identified by ITC.

UV-Vis spectrophotometric titrations

To get more insights into the host–guest interaction between the two dyes (Fig. 1) and PrS[6]^SO3−, UV-Vis (Figs. 3, 4 and S21–S24) measurements were carried out in water buffered at pH 5.5 (acetate buffer, 50 mM). As shown in Fig. 3a, in the case of RhodB, the addition of increasing amounts of the host to a solution of the guest caused a bathochromic shift (Δλ = 11 nm) of the characteristic RhodB band at 554 nm along with the formation of an isosbestic point at 560 nm, suggesting the presence of at least two species in solution. A bathochromic shift of the characteristic band of MethB from 666 nm to 675 nm was also observed upon increasing the amount of PrS[6]^SO3− in solution (Fig. 4a). However, an isosbestic point was not clearly detected. From the binding isotherms of the two titrations, reported in Figs. S22 and S24 for RhodB and MethB, respectively, it is possible to corroborate the formation of a 1:1 host–guest complex in the case of MethB, while the formation of the 2:1 host–guest complex could be inferred for RhodB. The association constants determined by UV-Vis measurements (Table S1) agree with those determined by ITC.

Notably, no changes in the fluorescence emission of PrS[6]^SO3− were observed in the presence of the two guests (see Supporting Information, Fig. S25 for the emission spectra of PrS[6]^SO3− in the presence of RhodB). Interestingly, when the fluorescence experiment was repeated, monitoring the changes in the fluorescence emission of a pH 5.5 buffered solution of RhodB upon the addition of increasing amounts of a pH 5.5 buffered solution of PrS[6]^SO3− (λ_ex = 550 nm), a quenching of the emission band was observed, accompanied by a bathochromic shift of 12 nm (from 584 nm to 596 nm) with the addition of up to 2 equivalents of PrS[6]^SO3−. This observation is consistent with the formation of a 2:1 capsule complex (Figs. S26 and S27a, b). Upon adding an excess of PrS[6]^SO3−, an enhancement of the emission intensity was noted, which was attributed to the formation of dye aggregates^43,44. In the case of MethB, a quenching of the fluorescence emission of the dye was observed in the presence of one equivalent of PrS[6]^SO3−, consistent with the formation of a 1:1 complex (Fig. S28a, b).

In silico calculations

Based on these findings, the supramolecular capsular structure formed between two molecules of PrS[6]^SO3− and RhodB was extensively investigated using MD simulations in explicit water (SI). These simulations were carried out with the YASARA software⁴⁵, which provided for detailed insights into the supramolecular interactions and stability of the assembly in an aqueous environment. As previously reported, prismarenes are planar chiral macrocycles, and prism[6]arenes are known to exist in eight distinct conformations¹⁶, each exhibiting different planar chirality due to the rotational dynamics around the naphthalene units.

The most stable conformation, which predominates in both solution and solid states, features all-pS planar chirality (or enantiomer, the all-pR). Based on this consideration, the minimum-energy complexes were explored starting with the all-pS conformer. MD simulations highlighted the formation of a molecular capsule composed of two PrS[6]^SO3− encapsulating a RhodB guest molecule (Fig. 5a). The RhodB@(PrS[6]^SO3−)₂ supramolecular assembly shows that the diethylammonium groups of the guest are accommodated within the cavities of each PrS[6]^SO3− macrocycle, while the benzoate moiety is intercalated between the two macrocycles PrS[6]^SO3−. This structure aligns with the finding that the ¹H-NMR spectra of the mixtures of PrS[6]^SO3− / RhodB, as shown in Fig. S6, exhibit an upfield shift in the NCH₂CH₃ signals of RhodB (Δδ = 0.48 and 0.42 for the NCH₂CH₃ signals, respectively), indicating host–guest interactions attributed to the encapsulation of RhodB within the aromatic cavity of the prismarene. We also performed a non-covalent interaction (NCI, Figs. S33 and S34) analysis based on promolecular density⁴⁶ to discern the nature of the interactions between the guest and the hosts. This analysis confirmed that cation···π interactions and electrostatic contacts primarily dominate within the cavity. One nitrogen atom forms strong cation···π interactions with the cavity, while the other nitrogen atom experiences C–H···π interactions.

Fig. 5. MD modeling of RhodB@(PrS[6]^SO3−)₂ and MethB@PrS[6]^SO3− complexes.

a Minimum-energy structure of the RhodB@(PrS[6]^SO3−)₂ complex. Water molecules participating in the water-mediated hydrogen-bonding network between the two macrocycles are highlighted in blue, while those bridging the guest and the macrocycle are depicted in white and red. b Minimum-energy structure of the MethB@PrS[6]^SO3− complex. Water molecules bridging the guest and the macrocycle are shown in white and red.

The average C–H‧‧‧centroid distances were determined to be approximately 3.0 Å. In contrast, the 1:1 RhodB@PrS[6]^SO3− complex exhibits multiple cation···π interactions between the C=N⁺(CH₃)₂ group and the electron-rich cavity of the macrocycle (see SI). The NCI analysis reveals numerous π···π interactions between the naphthalene units and both the aromatic component residing within the cavity and the benzoate moiety located outside, adjacent to the macrocyclic rim.

Calorimetry studies indicated that a significant entropic gain drives the formation of the H₂G capsular complex, which arises from extensive desolvation^8,20. As previously reported by us, the release of high-energy water molecules from the prismarene cavity serves as a crucial driving force during the complexation process in water with prismarene macrocycles^8,20. After the release of frustrated water molecules from the prismarene cavities following complexation with the guest, these water molecules rearrange to form a network of hydrogen bonding interactions (depicted in blue in Fig. 5a), which involves water-mediated hydrogen bonding with the benzoate group of RhodB. This phenomenon has already been proposed during the complexation of neutral aniline guests within the cavity of water-soluble carboxylate prism[6]arene²⁰. While the carboxylate group of the guest does not interact directly with the cavity, it can associate with the sulfate groups through an extensive water-mediated hydrogen-bonding network that forms outside the macrocycle, as illustrated in Fig. 5a.

¹H-NMR studies (D₂O, 600 MHz, 298 K) were conducted on both complexes to demonstrate the stabilizing role of water-mediated hydrogen-bonding interactions. As reported in the literature⁴⁷, DMSO is a competitive hydrogen-bonding solvent that weakens and ultimately disrupts supramolecular complexes. When a D₂O solution of the complexes MethB@PrS[6]^SO3− and RhodB@PrS[6]^SO3− was treated with DMSO-d₆, the signals associated with the guests exhibited a progressive shift toward the chemical shift values of the free guests, indicating complex dissociation (see Figs. S13 and S14 in the Supporting Information). These results highlight the importance of water-mediated hydrogen-bonding interactions in stabilizing the complexes.

Finally, MD simulations were conducted on the MethB@PrS[6]^SO3− complex (Fig. 5b) to investigate the host–guest interactions in aqueous solution. The minimum-energy structure (Fig. 5b), obtained after a 10 ns simulation, reveals the guest positioned within the macrocyclic cavity, where it engages in cation···π interactions (with average cation···centroid distances of 4.0 Å) between the macrocyclic framework and the positively charged sulfur atom of the guest. Additionally, four water molecules form a network of hydrogen bonding interactions (Fig. 5c), with the heterocyclic nitrogen atom of the methylene blue guest located at the center of the cavity. This nitrogen forms water-mediated hydrogen bonds with three adjacent SO³⁻ groups positioned on the same rim of the prism[6]arene ring (Fig. 5b, left). The structure of the MethB@PrS[6]^SO3− complex obtained through calculations (Fig. 5c) reveals the N(CH₃)₂ groups protruding from the cavity and adjacent to the anionic sulfate groups (Fig. 5b, right). This observation is consistent with the calorimetric data, which indicate that a significant contribution to the stabilization of the MethB@PrS[6]^SO3− complex arises from the electrostatic interaction between the positively charged nitrogen atoms of the dye and the anionic sulfate groups on the rims of PrS[6]^SO3−. Furthermore, the localization of the N(CH₃)₂ groups of MethB outside the aromatic cavity of the macrocycle aligns with the ¹H-NMR spectra of the MethB/PrS[6]^SO3− mixtures in Fig. S7. In these spectra, the N(CH₃)₂ groups of MethB do not exhibit a substantial shift after complex formation, which is consistent with the localization of the N(CH₃)₂ groups outside the cavity of PrS[6]^SO3−.

NCI analyses (Figs. S23 and S24) indicate that the complex is further stabilized by numerous π···π interactions between the aromatic cavity of the host and the electron-poor aromatic ring of the guest.

Conclusion

In this work, we present a novel water-soluble prism[6]arene functionalized with sulfate groups (PrS[6]^SO3−) at its rims, demonstrating its effectiveness as a supramolecular host for the molecular recognition of organic dyes in aqueous media. The incorporation of sulfate groups significantly enhances solubility in water, while the deep, hydrophobic aromatic cavity effectively accommodates organic guests such as rhodamine B and methylene blue.

Calorimetric and spectroscopic investigations reveal that both dyes form stable 1:1 complexes with PrS[6]^SO3−, primarily driven by favorable enthalpic contributions. Notably, rhodamine B exhibits distinctly different recognition behavior, assembling with two PrS[6]^SO3− units to create a higher-order supramolecular assembly, with association constants of log K = 5.5 for the 1:1 complex and log K = 3.8 for the 2:1 host–guest complex. MD simulations indicate that, in all cases, complex stabilization is facilitated by secondary host/guest interactions and water-mediated hydrogen bonding interactions.

These findings broaden the scope of prism[n]arene-based receptors in aqueous supramolecular chemistry and open new avenues for designing water-soluble macrocyclic systems capable of forming adaptive, higher-order assemblies for sensing and molecular recognition of organic pollutants.

Methods

General experimental details

All chemical reagents were used as obtained by TCI, Fluorochem, and Merck, with no additional purification. Reaction temperatures were measured externally and were monitored by Merck TLC silica gel plates (0.25 mm) and visualized by UV light at 254 nm, or by spraying with H₂SO₄-Ce(SO₄)₂. NMR spectra were recorded on a Bruker Avance-600 [600 (¹H) and 150 MHz (¹³C)], Avance-400 [400 (¹H) and 100 MHz (¹³C)], and Avance-250 [250 (¹H) and 62.5 MHz (¹³C)] spectrometers. Chemical shifts are reported relative to the residual solvent peak. Standard pulse programs, provided by the manufacturer, were used for 2D COSY (cosygpqf) and 2D HSQC (hsqcedetgpsisp2.2) experiments. Structural assignments were made with additional information from gCOSY and gHSQC experiments. ESI-FTICR mass spectra were recorded on a Bruker Solaris XR Fourier transform ion cyclotron resonance mass spectrometer equipped with a 7 T refrigerated actively shielded superconducting magnet. The mass spectra were externally calibrated using sodium trifluoroacetate (NaTFA) as the standard, and a linear calibration was applied. The samples were analyzed in negative ESI mode using a 10 mg mL⁻¹ aqueous solution of the analyte prepared in deionized water.

Synthesis of sulfated-Prism[6]arene—PrS[6]^SO3Na

To a solution of per-hydroxylated prism[6]arene (0.38 g, 0.37 mmol, prepared according to the literature²³) in dry pyridine (36 mL, 10.0 mM), pyridinium·SO³⁻ (2.11 g, 13.3 mmol) was added under a nitrogen atmosphere. The mixture was stirred at 90 °C for 36 h, then cooled to room temperature. A basic work-up was performed by adding solid Na₂CO₃ until the pH reached 8.5. After purification by dialysis, PrS[6]^SO3Na was obtained in 62% yield (0.50 g).

¹H-NMR (D₂O, 600 MHz, 298 K) δ 8.02 (broad, 12 H, Ar-H), 7.49 (broad, 12 H, Ar-H), 4.97 (broad, 12 H, Ar-CH₂-Ar).

{¹H}¹³C NMR (62.5 MHz, D₂O, 298 K): δ (ppm) 146.6, 131.6, 129.1, 125.1, 121.6, 24.2.

HRMS (FT-ICR ESI) m/z [M + 9Na]³⁻ calcd for: C₆₆H₃₆O₄₈S₁₂Na₉: 728.8707; found: 728.8723.

Calorimetric investigation

Calorimetric titrations were carried out at 25 °C and pH 5.5 in buffered aqueous solution using a Nano-ITC calorimeter (TA Instruments, USA) with a 988 μL active cell and a 250 μL titration syringe. The calorimeter was chemically calibrated using the recommended HCl/TRIS reaction, according to the reported protocol⁴⁸. All experiments were performed as described elsewhere⁴⁹. All dyes and PrS[6]^SO3− solutions were prepared in 50 mM acetate buffer (pH 5.5).

Calorimetric titrations were initially conducted to examine the self-aggregation process of the dyes at pH 5.5; to this aim, buffered aqueous solutions of rhodamine (1.40 mM) or methylene blue (1.39 mM) were incrementally added to the reaction cell containing buffer solution only. The concentration of each dye was chosen to ensure that the threshold concentration for the aggregation is reached throughout the experiment.

The measurements for the study of the interaction of PrS[6]^SO3− with rhodamine B or methylene blue were performed either by titrating dye solutions (0.70–1.40 mM) into PrS[6]^SO3− solutions (0.05–0.80 mM) or solutions of PrS[6]^SO3− (0.18–0.65 mM) into dye solutions (0.05–0.10 mM).

In the first set of experiments, the dyes remained mostly aggregated throughout the titrations. To study how these dyes form complexes in their monomeric state, additional measurements were performed using RhodB or MethB as the titrand in the cell at a lower concentration to minimize the presence of dimers. In both cases, the titrations were specifically designed to promote the formation of the complex species of interest during the experiment and to collect enough data points for an accurate fit of the resulting calorimetric curves. Heats of dilution were obtained in “blank” experiments by adding the proper titrant solution to a buffer-only solution. Heat values were handled with HypCal, a software that simultaneously refines multiple data collected for the different sets of titrations through a non-linear least square analysis. The stability constant and enthalpy change values obtained for the dye dimerization/aggregation process were included in the model as fixed parameters.

Dynamic light scattering (DLS)

DLS analyses were performed using a Malvern Zetasizer (Nano ZS, UK) equipped with a 4 mW Helium-Neon laser with a wavelength of 632.8 nm. The scattering angle was set at 170 °C and the temperature at 25 °C. Hydrodynamic diameter distributions were obtained volumetrically using the instrument software package. Each curve represents the average of 3 measurements (16 runs each). Before analysis, all solutions were filtered to remove any remaining dust using 0.22 μm syringe filters.

DOSY data analysis and estimation of hydrodynamic radii

Self-diffusion coefficients (D) were determined from DOSY NMR experiments conducted in acetate-buffered D₂O (pH 5.5) at 298 K. The buffer solution was prepared by dissolving 700 mg of sodium acetate in 25 mL of D₂O, with the pH adjusted to 5.5 using DCl, resulting in an acetate concentration of approximately 0.34 M.

Hydrodynamic radii (R_h) were estimated from the experimental diffusion coefficients using the Stokes–Einstein equation. All calculations were performed assuming a temperature of 298 K and a dynamic viscosity of η = 1.8 mPa s (1.8 × 10⁻³ Pa s). The viscosity of the acetate buffer was approximated rather than directly measured, accounting for the significantly increased ionic strength of the medium compared to pure D₂O, which has a viscosity of approximately 1.1 mPa s at 298 K. Literature data for aqueous acetate solutions demonstrate a noticeable increase in viscosity with concentration; therefore, an intermediate value of 1.8 mPa s was used as a reasonable estimate for a 0.34 M sodium acetate solution in D₂O. This value provides realistic hydrodynamic radii for the prismarene host and its complexes, ensuring internal consistency across all DOSY-derived size comparisons. The calculated diffusion coefficients and corresponding hydrodynamic radii are summarized in the SI.

UV-Vis and fluorescence titration experiments

UV-Vis titration experiments were performed in quartz cuvettes (1 cm path length) on a Cary 60 spectrometer (for UV-Vis measurements) and on a Cary Eclipse spectrofluorimeter (for the fluorescence measurements) at 298 K. 2.5 ml of freshly prepared 4.7 × 10⁻⁶ or 5.0 × 10⁻⁶ M solution of the dye (Rhodamine B or Methylene Blue, respectively) under study was placed in a 10 mm cuvette. Small aliquots of 2.2 × 10⁻⁴ M or 3.3 × 10⁻⁴ M solution containing PrS[6]^SO3− were added, maintaining a constant dye concentration throughout the experiment. Titration solutions were prepared in 0.05 M sodium acetate buffer at pH 5.5. (410 mg CH₃COONa was dissolved in 100 ml MilliQ water, then glacial acetic acid was added drops to reach pH 5.5). The obtained UV-Vis titration data were analyzed using the BindFit package (v 0.5), available online at http://supramolecular.org.

In silico calculations

The Dye@PrS[6]^SO3− complexes reported in this work were studied using an MD approach, and all simulations were performed with Yasara Software⁴⁵ (Ver. 25.9.17) on an AMD Ryzen 7940HX processor.

The guest@host complexes were prepared for the MD simulations starting from the all-pS conformer after modification of the X-ray structure reported by our group¹⁶. Each complex was placed in a cubic box with periodic boundaries and water molecules at pH 5.5 and Na⁺ as a counterion. The systems were simulated in NVT ensemble at 298 K in a box of water and AMBER 14 as force field for 10 ns after a steepest descent minimization and then by a simulated annealing minimization until convergence (<0.01 kcal/mol Å).

NCI analyses were performed using the Multiwfn program⁵⁰, based on the minimum-energy structures obtained from MD simulations.

The evaluation of the interactions with water molecules was carried out on the simulation box of the minimum-energy structure, which was manually cleaned of all water molecules not involved in the host–guest interaction.

The resulting structure was then visualized using Chemcraft (version 1.8), generating the NCI plots from the promolecular electron density⁴⁶.

Author contributions

P.D.S., C.C., and C.S. conceived and supervised the research; R.D.R., A.P., and N.D.F. executed the experiments and analyzed the experimental data; C.S. and G.D. G.S. performed calorimetric investigations and rationalized the experimental results; G.P. and M.Z-H. carried out the UV-Vis experiments, and C.A. performed DLS investigations. C.T. and P.D.S. performed in silico calculations. C.G. and V.I. wrote the manuscript, and M.D.R. performed manuscript edits and revisions.

Peer review

Peer review information

Communications Chemistry thanks the anonymous reviewers for their contribution to the peer review of this work.

Data availability

The authors declare that all data supporting the findings of this study are available within the paper and its Supplementary Information files.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s42004-026-02090-8.