Derivatized Cyclodextrins and Their Modified Synthetic Architectures for Sensing and Catalysis

The overall theme of my graduate research is to understand forces involved in supramolecular, hydrophobically-driven interactions, primarily in cyclodextrin systems and to use those interactions in applications ranging from fluorescence-based sensing to supramolecular catalysis. This research has included a highly interdisciplinary research project exploring the effects of cation-π interactions on surfactant/lipid bilayer vesicles for delivery applications. Cyclodextrins, which are commercially available, torusshaped cyclic oligoamyloses, have been selected as the supramolecular hosts in these studies because of their well-defined hydrophobic interior cavity. The hydrophilic exterior, in turn imparts substantial aqueous solubility. Moreover, the primary and secondary hydroxyl groups of the cyclodextrin provide a myriad of synthetic handles for further modification and chemical derivatization. Cyclodextrin-based catalytic systems have been envisioned for mild, environmentally friendly transformations in high-impact organic reactions. The basis of this research stems from the ability of cyclodextrins to form hydrophobic complexes with small molecules, thereby lowering the entropic barrier for the formation of a transition state in selected organic reactions. Moreover, the hydrophobic complexes of cyclodextrin with small molecules have also been shown to be more reactive from the perspective of many organic transformations. The first manuscript “Cyclodextrin-promoted Diels Alder reactions of a polycyclic aromatic hydrocarbon under mild reaction conditions” describes Diels Alder reactions of a model polycyclic aromatic hydrocarbon (PAH), 9-anthracenemethanol with Nsubstituted maleimides under mild reaction conditions (aqueous solvent, 40C) in the presence of commercially available cyclodextrins. In this system, hydrophobic complexation of the N-substituent in turn modifies the electronics of the alkene double bond, resulting in its enhanced reactivity. We found that cyclodextrin complexation of the N-substituent on the maleimide (driven by hydrophobic interactions) was the key factor in determining the rate of the reaction and the overall conversion to product. Optimal results were found using N-cyclohexylmaleimide with a methyl-β-cyclodextrin host, with 94% conversion obtained in 24 hours. A proposed model of the complexation with methyl-β-cyclodextrin has been proposed, with cyclodextrin encapsulation perturbing the electronics of the dienophile double bond and enhancing its reactivity. Results of these experiments were published in Tetrahedron Letters in 2015, and this publication has already been cited multiple times. The second manuscript “An Environmentally Friendly Procedure for the Aqueous Oxidation of Benzyl Alcohols to Aldehydes with Dibromodimethylhydantoin (DBDMH) and Cyclodextrin Scope and Mechanistic Insights” discusses the cyclodextrin-promoted oxidation of benzyl alcohols to benzaldehydes using an inexpensive, commercially available reagent, 1,3-dibromo-5,5-dimethylhydantoin (DBDMH). This newly developed reaction has two notable advantages compared to previously reported benzylic oxidation reactions: (a) more environmentally friendly (i.e. “greener”) methodology through the use of an aqueous solvent system and mild reaction conditions; and (b) high specificity for benzyl alcohol substrates with limited side reactivity, including over-oxidation and aromatic bromination, observed. This reaction proceeds with moderate to good yields for a broad scope of benzyl alcohol substrates, with the cyclodextrin additive accomplishing two main objectives: (a) enhancement of the desired reactivity as a result of the activation of the benzylic protons via interactions with the cyclodextrin rim; and (b) limitation of the undesired aromatic bromination side products as a result of steric shielding of the aromatic ring in the hydrophobic cyclodextrin pocket. Catalyst reusability up to three consecutive runs has been observed without substantial loss of product yield and selectivity, which further enhances the atom economy of this method. Results of these experiments were published in Synthetic Communications in 2016. Non-covalent energy transfer has been used as a highly sensitive investigative tool in a wide variety of supramolecular systems. Owing to its exquisite sensitivity and dependence on a host of factors, this strategy has also been employed to study dynamic conformations of biomolecules such as nucleic acids and peptides. Our group has developed highly efficient energy transfer systems using γ-cyclodextrin as a supramolecular host for promoting non-covalent energy transfer from small molecule aromatic toxicants to high quantum yield fluorophores. Although γ-cyclodextrin has a cavity size that is well-known to be able to accommodate two small molecule guests simultaneously, limitations of γ-cyclodextrin include its limited specificity and illdefined host: guest stoichiometry, as a result of its larger cavity size. There is neither control of the orientation of the guest molecule inside the cavity, nor selective binding of a single analyte in the presence of other competitive guest molecules, which often leads to sub-optimal detection sensitivity and anomalous false positive signals. In the third manuscript “Array based detection of isomeric and analogous analytes employing synthetically modified fluorophore attached β-cyclodextrin derivatives.” the scope of the cyclodextrin host has been expanded beyond that of γ-cyclodextrin, which permits us to tailor and tune the hydrophobic domain of the hosts optimally to the size of specific guest molecules. This expansion in turn offers improvements in selectivity and sensitivity for the detection for a given analyte. The chapter highlights the synthesis of a series of fluorophore-appended β-cyclodextrins with specific degree of functionalization and high levels of regioselectivity. These are powerful architectures in our group’s ongoing attempts at developing highly selective sensors for the efficient detection of persistent organic pollutants (POPs) at concentrations close to their environmental concentrations and literature-reported limits of concerns. By covalently linking a fluorophore directly to the cyclodextrin host, we obtained higher levels of system control in the cyclodextrin-promoted binding leading to unique fluorophore responses in the detection of several isomeric and analogous toxicants, including DDT pesticide analogues and polychlorinated biphenyl (PCBs) congeners. Advantages of using β-cyclodextrin include the smaller cavity size, which directly enables greater selectivity in binding as well as more efficient host-guest interactions, the lower cost of β-cyclodextrin compared to the γ-cyclodextrin isomer, and more straightforward methods for synthetically modifying the cyclodextrin host structure. We have demonstrated the ability of three architecturally distinct combinations of perbenzylatedβ-cyclodextrin/fluorophore sensor molecules to distinguish three isomeric and two analogous classes of analytes with 100% accuracy using linear discriminant analysis (LDA) of the fluorescence response signals. Each analyte-sensor binding event results in the modulation of the associated fluorophore, generating a unique chemical signature for each isomer across all the three sensors in an array based sensing strategy. Results of this work have recently been accepted for publication in New Journal of Chemistry. Additionally, the synthesis of a series of cyclodextrin-incorporated higher order architectures has also been described in the fourth manuscript “Synthetic β-cyclodextrin dimers for squaraine binding: Effect of host architecture on photophysical properties, aggregate formation and chemical reactivity.” These architectures have been designed to exhibit higher binding affinity towards larger hydrophobic analytes like stilbene, tamoxifen and biphenyls based on hydrophobic binding of the guest from two or more distinct ends of the molecule. Two cyclodextrins were tethered by aromatic/alkyl amide linkages, and binding properties of these novel receptors were investigated for high quantum yield fluorophores (squaraine dyes in this case). A comparison of the binding constants of the different hosts was drawn to reveal the contribution from a flexible aromatic linker in the binding of hydrophobic guests. Investigation of the supramolecular interactions of hosts with a series of N-alkyl-N-methylanilino squaraines of progressively increasing alkyl chain length has produced a few notable results, including: (A) the ability of the dimers to suppress the squaraine Hand Jaggregate formations in solution very effectively (a phenomenon reported by Chen et. al. previously with β-cyclodextrin); (B) the ability of the dimers to protect the squaraine core from aqueous hydrolysis, thereby prolonging its halflife (a phenomenon previously reported by Smith and co-workers for squaraine pseudo-rotaxanes with synthetic tetralactam macrocycles) and finally (C) a particular dimer being able to act as an enzyme mimic for the aqueous hydrolysis of a squaraine dye with high selectivity and turn-over numbers (TON). The results of this work are being prepared for submission to the journal Chemical Science. To develop sensitive and selective sensors, efforts have even been extended to synthetic macrocycles for the efficient binding of PAHs and other analytes. The fifth manuscript “A highly versatile fluorenone-based macrocycle for the sensitive detection of polycyclic aromatic hydrocarbons and fluoride anions” entails the synthesis and application of a fluorenone integrated triazolophane for the efficient binding of PAHs and fluoride anions. UV-vis and H-NMR spectroscopy results showed that the macrocycle has high sensitivity for selected PAHs and binds fluoride anions in a 1:2 stoichiometry. The bilateral symmetry of the macrocycle creates two binding pockets for the relatively small fluoride anion. This conclusion is well supported by the binding curve fitting of H-NMR titrations and Job’s plot analysis of the chemical shift of the triazole proton. A high association constant value of 10 M is observed for the binding of fluoride anion in DMSO. Results of these experiments have been published in a cofirst author publication in RSC Advances in 2016. The final chapter of the thesis describes the application of basic supramolecular science in an industrial setting. In cosmetic industries, hydrated surfactant vesicles are used to deliver encapsulated perfume ingredients and counter skin dryness. However, addition of small concentrations of perfume-raw-materials (PRMs) has a drastic effect on vesicular suspensions, perturbing their microstructures and altering their rheological properties. In the sixth manuscript entitled, “Impact of Nearly Water-Insoluble Additives on the Properties of Vesicular Suspensions” two model perfume-raw-material (PRM) compounds, linalyl acetate (LA) and eugenol, have been identified to have very different impacts on a multilamellar vesicular suspension made of diethylester dimethylammonium chloride (DEEDMAC) surfactant. While the former has negligible effect, the latter triggers a change from multilamellar to unilamellar vesicles, resulting in a sharp rise in the suspension viscosity. Employing time-resolved cryogenic transmission electron microscopy, microstructural changes related to viscosity variations were observed. In addition, H-NMR spectroscopy was used to examine the interactions between the additives and DEEDMAC, revealing the underlying mechanisms behind the structural transformations. To provide additional insights, changes induced upon addition of non-allyl substituted structural analogs of eugenol with increasing aromaticity, cyclohexanol, phenol, catechol and guaiacol, to DEEDMAC suspensions were investigated. These molecules are therefore characterized as ‘intermediate’ between LA and eugenol, in terms of transitioning from the non-aromatic character of LA to the highly aromatic character of eugenol. By examining NMR results from all the additives, strong interaction of the π electrons in aromatic rings with the cationic DEEDMAC head groups was determined to play a significant role in vesicular exfoliation phenomena. Such interactions are strong in eugenol but not present in LA. Results of these experiments were published in a co-first author publication in Industrial & Engineering Chemistry Research in 2016.

Non-covalent energy transfer has been used as a highly sensitive investigative tool in a wide variety of supramolecular systems. Owing to its exquisite sensitivity and dependence on a host of factors, this strategy has also been employed to study dynamic conformations of biomolecules such as nucleic acids and peptides. Our group has developed highly efficient energy transfer systems using γ-cyclodextrin as a supramolecular host for promoting non-covalent energy transfer from small molecule aromatic toxicants to high quantum yield fluorophores. Although γ-cyclodextrin has a cavity size that is well-known to be able to accommodate two small molecule guests simultaneously, limitations of γ-cyclodextrin include its limited specificity and illdefined host: guest stoichiometry, as a result of its larger cavity size. There is neither control of the orientation of the guest molecule inside the cavity, nor selective binding of a single analyte in the presence of other competitive guest molecules, which often leads to sub-optimal detection sensitivity and anomalous false positive signals. In the third manuscript "Array based detection of isomeric and analogous analytes employing synthetically modified fluorophore attached β-cyclodextrin derivatives." the scope of the cyclodextrin host has been expanded beyond that of γ-cyclodextrin, which permits us to tailor and tune the hydrophobic domain of the hosts optimally to the size of specific guest molecules. This expansion in turn offers improvements in selectivity and sensitivity for the detection for a given analyte. The chapter highlights the synthesis of a series of fluorophore-appended β-cyclodextrins with specific degree of functionalization and high levels of regioselectivity. These are powerful architectures in our group's ongoing attempts at developing highly selective sensors for the efficient detection of persistent organic pollutants (POPs) at concentrations close to their environmental concentrations and literature-reported limits of concerns. By covalently linking a fluorophore directly to the cyclodextrin host, we obtained higher levels of system control in the cyclodextrin-promoted binding leading to unique fluorophore responses in the detection of several isomeric and analogous toxicants, including DDT pesticide analogues and polychlorinated biphenyl (PCBs) congeners. Advantages of using β-cyclodextrin include the smaller cavity size, which directly enables greater selectivity in binding as well as more efficient host-guest interactions, the lower cost of β-cyclodextrin compared to the γ-cyclodextrin isomer, and more straightforward methods for synthetically modifying the cyclodextrin host structure. We have demonstrated the ability of three architecturally distinct combinations of perbenzylatedβ-cyclodextrin/fluorophore sensor molecules to distinguish three isomeric and two analogous classes of analytes with 100% accuracy using linear discriminant analysis (LDA) of the fluorescence response signals. Each analyte-sensor binding event results in the modulation of the associated fluorophore, generating a unique chemical signature for each isomer across all the three sensors in an array based sensing strategy. Results of this work have recently been accepted for publication in New Journal of Chemistry.
Much of the toxicity of PAHs is related to their highly planar structures, which enable the PAHs to intercalate in DNA and form covalent, carcinogenic adducts. 13 Converting the PAHs to non-planar products using chemical transformations disrupts this facile intercalation and limits their ability to form carcinogenic adducts. Reported herein is the ability of cyclodextrins to promote such transformations for one PAH, 9-anthracenemethanol (compound 1), via its Diels-Alder reactions with N-substituted maleimides. Mechanistic investigations demonstrate that the rate enhancements achieved in the presence of cyclodextrin rely on cyclodextrin-induced activation of the maleimide double bond via binding of the hydrophobic substituents to promote the reaction and achieve substantial rate accelerations.
The conversion of compound 1 to its corresponding Diels-Alder adduct 3 was The starting material 1 H-NMR peak used in this equation corresponds to 3 aromatic protons of the 9-anthracenemethanol and the product peak used for this equation corresponds to 1 proton at the bridgehead of the Diels-Alder adduct 3 ( Figure 1). The integration of the NMR peaks were the relative areas under the curve measured against a calibrated internal standard corresponding to the residual CHCl3 peak at 7.26 ppm. Analysis of the conversion efficiencies with different dienophiles reveals that the dienophiles that bound most strongly in the methyl-β-cyclodextrin cavity (as indicated by greatest changes in the 1 H-NMR chemical shifts) were also the most reactive ( Figure   3). This binding strength in turn depends largely on the hydrophobicity of the Nsubstituent of the maleimide, with compound 2a demonstrating the greatest binding affinities and fastest reaction rate.
The proposed mechanism by which the cyclodextrin derivatives promote the Diels-Alder reaction of compounds 1 and 2 likely involves the binding of hydrophobic Nsubstituted maleimides 2 in the hydrophobic cyclodextrin cavity, with additional stabilization provided by hydrogen bonding between the cyclodextrin hydroxyl groups and the carbonyl groups of the maleimide ( Figure 4).
This additional binding withdraws electron density from the π-bond, activating the alkene for the resultant cycloaddition reaction. This effect was maximal for the binding of 2a in methyl-β-cyclodextrin, due to the highly hydrophobic nature of the cyclohexyl substituent 15 and the optimal size match between the cyclohexyl and the methyl-βcyclodextrin cavity. 16 A similar phenomenon has been reported by Ritter and coworkers, wherein cyclodextrin binding of N-substituted maleimides led to enhanced reactivity in free radical polymerization reactions. 17 However, the mechanism by which such binding led to activation of the alkene bond in the N-substituted maleimides was not explicitly discussed. Interestingly, methyl-β-cyclodextrin was significantly more efficient than βcyclodextrin at promoting this Diels-Alder reaction, despite the fact that methyl-βcyclodextrin and βcyclodextrin have similar cavity dimensions. This trend is likely a result of the fact that methyl-β-cyclodextrin is both more flexible and has a more nonpolar cavity than βcyclodextrin, a fact that has been reported in the literature but has been rarely exploited in organic reactions. 18 A closer look at the reaction conversions ( Figure 3) reveals that as the N-substituent decreases in bulk, the conversions obtained with γ-cyclodextrin approach those observed with methyl-β-cyclodextrin. For example, the difference between the conversions achieved with methyl-β-cyclodextrin compared to γ-cyclodextrin was 39% for substrate 2a; this difference drops to 25% for substrate 2c and to 4% in favor of γcyclodextrin for substrate 2d. The less bulky substrates can form ternary complexes in γ-cyclodextrin, with both the diene and dienophile binding simultaneously in the cavity interior. γ-Cyclodextrin is known to form ternary complexes, 19 and such ternary complexes have already been used in γcyclodextrin mediated dimerization reactions. 20 This ternary complexation binding mode is distinct from the binding mode proposed in Figure 4, which is expected to be the dominant mechanism for methyl-β-cyclodextrin binding of bulky N-substituents.
Interestingly    In summary, these experiments demonstrate the ability of methyl-β-cyclodextrin to catalyze the conversion of a PAH to non-planar hydrophobic adducts under mild reaction conditions. This rate enhancement is primarily due to the superior hydrophobic binding of methyl-β-cyclodextrin to hydrophobic substituents on the N-substituted maleimides, which in turn enhances the alkene reactivity. The resulting adducts 3 are both less planar and more hydrophobic than the starting PAH, which will help to mitigate toxicity by reducing the degree of PAH intercalation in the DNA as well as the mobility of the PAH adduct in highly polar biological environments. Current efforts are focused on expanding the scope of this Diels-Alder reaction to include other hydrophobically-substituted dienophiles and other aromatic dienes, as well as investigations of other cyclodextrin-promoted organic transformations. Results of these and other investigations will be reported in due course. The starting material 1 H-NMR peak used in this equation corresponds to 3 aromatic protons of the 9-anthracenemethanol and the product peak used for this equation corresponds to 1 proton at the bridgehead of the Diels-Alder adduct 3 ( Figure S1).

Cyclodextrin-promoted Diels
a The negative changes in the chemical shift signifies an upfield movement of the peaks on complexation.    Table S7. Average percentage conversion data for Diels Alder reaction of 9anthracenemethanol with compound 2b. no CD  8  30  32  40  37  33  24  44  57  61  46  33  48  40  51  64  53  35   Table S8. Average percentage conversion data for Diels Alder reaction of 9anthracenemethanol with compound 2c.  commercial availability, 24 air-and moisture-stability, 25 and general high reactivity. 26 Initial reports of DBDMH-promoted oxidation of benzyl alcohols only demonstrated efficacy in the conversion of secondary alcohols to ketones 27 likely because of the higher sensitivity of primary alcohols to over-oxidation and other side reactions, or require the use of organic solvents such as methanol and dichloromethane. 28 The use of cyclodextrin in combination with DBDMH for accomplishing the oxidation of primary alcohols to aldehydes has not been reported to date, despite the fact that this combination is expected to demonstrate numerous operational advantages including all advantages of using cyclodextrin (aqueous solvent system, mild reaction conditions) and DBDMH (air-and moisture-insensitivity, commercial availability, limited human toxicity) to accomplish a synthetically useful transformation. Reported herein is the cyclodextrin-promoted DBDMH oxidation of a variety of benzyl alcohols 1 to benzaldehydes 2 (Equation 1), which proceeds under fully aqueous conditions, with limited generation of byproducts, and in moderate to high reaction yields. Detailed mechanistic investigations provide substantial insight that will guide further reaction development and applications.

Equation 1. Oxidation reaction of primary benzyl alcohols to benzaldehydes DISCUSSION
A variety of benzyl alcohol substrates 1 ( Figure 1) were converted into their respective aldehydes 2, using mild heating in aqueous media to achieve moderate to high conversions. The amount of cyclodextrin varied from 0.33 equivalents relative to the benzyl alcohol up to 1.5 equivalents, with the amount used independently optimized for each substrate. Table 1 summarizes the results of these experiments, and includes results obtained using the highest performing cyclodextrin host and the optimal amount of cyclodextrin, which depended strongly on the substrate structure. For example, the reaction with the smallest substrate 1a was accelerated most strongly with the smallest cyclodextrin host α-cyclodextrin. As the size of the para substituent on the benzyl alcohol increased, the optimal cyclodextrin host size increased as well, with substrate 1d (p-chloro) optimally catalyzed by β-cyclodextrin, substrate 1g (p-bromo) by methylβ-cyclodextrin, and substrate 1j (p-iodo) by γ-cyclodextrin. Moving the same substituent from the para position to the ortho or meta position required a slightly larger cyclodextrin host to accommodate this geometry and achieve optimal reactivity (compare for example 1b vs. 1c; 1f vs. 1g; 1h vs. 1i).
3 α-CD (0.33) 100 8 a Percent conversions were calculated based on the 1 H-NMR analysis of the reaction mixture. b Percentage in parentheses indicates the amount of aromatic brominated side product formed. c the reaction was run at elevated temperature (80 o C). No product formation was observed at 60 o C.
In general, electron-deficient and electron-neutral substrates displayed markedly higher conversions compared to the electron-rich substrates (see for example 1d and 1g compared to 1b), which is a consequence both of their higher reactivity as well as the lower amounts of side products resulting from bromination of the aromatic ring. These reaction conditions were not effective in oxidizing aliphatic primary alcohols to aldehydes, which confirms the importance of aromatic ring-cyclodextrin interactions in the reaction mechanism, nor did other N-halogenated reagents such as 1,3-dichloro-5,5dimethylhydantoin (DCDMH) or 1,3-diiodo-5,5-dimethylhydantoin (DIDMH) effect the desired transformation. In all cases, the conversions obtained in the absence of cyclodextrin were markedly lower than those obtained in the presence of cyclodextrin, highlighting the crucial role for this supramolecular scaffold in promoting the desired reactivity.
A plausible reaction mechanism is shown in Scheme 1, and involves the formation of a The roles of cyclodextrin in the proposed mechanism shown in Scheme 1 are two-fold: (1) Acceleration of the C-H bond cleavage in the rate-determining first step; and (2) shielding of the aromatic ring from undesired aromatic bromination. Both of these roles are enabled through binding of the aromatic guests in the cyclodextrin cavity in a geometry such as that shown in Figure 2. This complexation activates the benzyl protons through interactions between the cyclodextrin rim and the benzyl position, leading to markedly faster C-H bond cleavage, and protects the aromatic core of the substrate from electrophilic aromatic bromination through hydrophobic interactions between the aromatic ring and the cyclodextrin cavity. 30

Figure 2. Illustration of the complexation of benzyl alcohols in cyclodextrin.
In addition to interactions between the cyclodextrin host and alcohol substrates 1, cyclodextrin can also interact directly with compound 3. Literature precedent indicates that cyclodextrins interact with hydantoins, and such interaction is used for efficient chromatographic separation of hydantoin enantiomers. 31,32 In this system, 1 H-NMR analysis of DBDMH-cyclodextrin mixtures indicates that complexation of DBDMH in cyclodextrin leads to marked decreases in the rate of N-Br bond dissociation ( Table 2). Aliphatic alcohols cyclohexanol and 4-methylcyclohexylmethanol (4-MCHM) 33 were investigated as substrates, and no reaction was observed. The fact that these aliphatic alcohols were not competent substrates under these reaction conditions provides further evidence for the crucial role of the aromatic ring in ensuring favorable host-guest complexation, and argues against literature reports that the role of cyclodextrin in organic reactions is merely as a phase-transfer catalyst. 34,35 Moreover, the fact that aliphatic alcohols are inert to oxidation allows for the recycling and re-use of the aliphatic alcohol rich cyclodextrin host without concerns about interfering reactivity (vide infra).
Investigation of the effect of electron density of the aromatic ring on its reactivity reveals a strong effect on the substrate reactivity, with strongly electron donating substrates such as p-methoxy benzyl alcohol 1m yielded none of the desired aldehyde (see ESI for spectral characterization of compound 6). This is likely because the highly electron-rich aromatic ring facilitates the formation of the dibromo intermediate 5' (Scheme 1). Other somewhat less electron rich substrates, such as 1b, proceeded to give the product in high yields in the presence of the cyclodextrin host, whereas in the absence of the host aromatic bromination products were observed (Table 1). Electrondeficient substrates such as 1h and 1i were inert in the absence of the cyclodextrin, but underwent efficient reaction in the presence of the cyclodextrin. In summary, electron deficient substrates required activation by the cyclodextrin (catalytic activation), whereas electron rich substrates need protection from side reactions (chemoselective influence).

Equation 3
. Alternate reaction pathway for highly electron rich substrates.
Moreover, decreasing the amount of N-halo reagent led to drastic decreases in the conversion of compound 1m to acid bromide 6, in comparison to the less drastic decreases observed for the conversion of 1a to 2a and 1h to 2h (Table 3). This result supports the mechanism proposed in Scheme 1, wherein acid bromide 6 is formed via dibromo intermediate 5' and requires multiple equivalents of the N-halo reagent for the reaction to proceed.  In summary, a novel aqueous oxidation procedure for the conversion of primary benzyl alcohols to benzaldehydes is demonstrated, using DBDMH as an environmentally friendly oxidant and cyclodextrins as supramolecular additives that promote the highly efficient reaction and limit the formation of undesired side products. Importantly, the cyclodextrin hosts are unaltered throughout the course of the reaction, and can be recovered and reused (conversion of 1a to 2a: first run: 75%; second run: 74%; third run: 62%). This was accomplished simply by extracting the aldehyde products into an organic solvent and then re-using the cyclodextrin-containing aqueous layer. The results indicate that the aliphatic hydroxyl groups of the cyclodextrin are relatively stable to oxidation under these reaction conditions, in accord with our previously reported results on inert aliphatic alcohol substrates. This procedure has significant potential in environmentally friendly reaction optimization and complex product synthesis. Efforts in these areas are currently underway in our laboratory, and results of these and other investigations will be reported in due course.

Alcohols to Aldehydes with Dibromodimethylhydantoin (DBDMH) and
Cyclodextrin -Scope and Mechanistic Insights

MATERIALS AND METHODS
Proton NMR spectra were recorded using a Bruker 300 MHz instrument, with the singlet peak of HDO at 4.79 ppm as reference. All reagents, substrates, and solvents were     The selective detection and accurate quantification of structurally similar analytes is a major challenge for scientists, as structurally similar analytes often have widely disparate toxicities. 1 The most common strategy is to use mass spectrometry methods, such as liquid chromatography-mass spectrometry (LC-MS) 2 or gas chromatographymass spectrometry (GC-MS). 3 However, there are significant drawbacks associated with this approach, including the costs and time necessary to conduct such analyses, 4 which limits the ability to conduct high throughput assays. 5 An alternate strategy is to use array-based sensing systems, which have recently gained in popularity. 6 This approach relies on the development of a chemical signature for each analyte based on analyte-specific interactions with a sensor series. Array-based sensing systems can be combined with supramolecular sensors, which rely on differential noncovalent interactions of analytes with supramolecular hosts, including cyclodextrins, 7 fluorescent polymers, 8 molecularly imprinted polymers, 9 and metal-organic frameworks (MOFs). 10 Although supramolecular array-based systems overcome many challenges associated with mass-spectrometry based detection methods, the analyte scope explored in most of these reports have been limited to aromatic small molecules. 11 In a real-world contaminated environment, the nature of the various pollutants is highly complex, 12 and includes mixtures of aromatic and non-aromatic compounds. 13 This kind of situation requires the development of a sensing system which is rapid, simple, and efficient in classifying a broad range of persistent organic pollutants (POPs). 14 Our group has previously reported the use of β-cyclodextrin and γ-cyclodextrin in arraybased detection systems for the sensing of a wide variety of environmental toxicants and POPs. 15 The sensing strategy is based on cyclodextrin promoted analyte-tofluorophore energy transfer as well as cyclodextrin-promoted, analyte-induced fluorescence modulation. In the fluorescence modulation systems, the fluorophore was added to the cyclodextrin solution prior to analyte addition, which can result in fluorophore-cyclodextrin binding that reduces the cyclodextrin's ability to bind the target analyte. As such, introduction of the analyte to the fluorophore-cyclodextrin solution requires the analyte-cyclodextrin association constants to be higher than those of the fluorophore-cyclodextrin ( Figure 1A), or it requires the formation of higher order association complexes between the analyte, cyclodextrin and fluorophore ( Figure 1B).
Such higher order association complexation is probable only for γ-cyclodextrin. 16 Herein, we report the development of an array-based detection system using fluorophore-functionalized perbenzylated β-cyclodextrin sensors, which enables binary complex formation between the functionalized cyclodextrin and the target analyte ( Figure 1C). Each sensor is selective, meaning the array is able to distinguish three classes of isomeric analytes and two classes of structurally similar analytes, with 100% classification accuracy. High sensitivity is demonstrated as well, with limits of detection approaching or surpassing literature-reported levels of concern. Finally, preliminary efforts at using this system for the accurate identification of binary analyte mixtures are also reported.

Materials and Methods
All the reagents were obtained from Sigma Aldrich or Fisher Scientific and used without further purification, unless otherwise noted. β-cyclodextrin was dried in the oven prior to use. Reagent grade solvents (99.9% purity) were used for the synthetic reactions.

Fluorescence Modulation Experiments
Fluorescence emission spectra were obtained using a Shimadzu RF-5301PC spectrophotofluorimeter with 3 nm excitation and 3 nm emission slit widths. 0.5 mL of S1, S2, or S3 solutions (5 μM in DMSO) and 2 mL of deionized water were combined in a quartz cuvette. The solution was excited at 320 nm, and the fluorescence emission spectra were recorded.
The fluorescence emission spectra were integrated vs. wavenumber on the X-axis, and

Array Generation Experiments
Array analysis was performed using SYSTAT 13 statistical computing software with the following settings:

Limit of Detection Experiments
The limit of detection (LOD) is defined as the lowest concentration of analyte at which a signal can be detected. To determine this value, the following steps were performed for each cyclodextrin-analyte combination. In a quartz cuvette, 0.5 mL of S1, S2, or S3 solutions (5 μM in DMSO) and 2 mL of deionized water were combined. The solution was excited at 320 nm, and the fluorescence emission spectra were recorded starting at 330 nm. Six repeat measurements were taken.
All of the fluorescence emission spectra were integrated vs. wavenumber on the X-axis, and calibration curves were generated. The curves plotted the analyte concentration in μM on the X-axis, and the fluorescence modulation ratio on the Y-axis. The curve was fitted to a straight line and the equation of the line was determined.
The limit of detection is defined according to Equation 2: Where SDblank is the standard deviation of the blank sample and m is the slope of the calibration curve.

RESULTS AND DISCUSSION
We employed a series of three cyclodextrin-based supramolecular sensors ( Figure  The synthesis of supramolecular hosts S2 and S3 is shown in Scheme 1. Perbenzylated β-cyclodextrin was obtained from the reaction of β-cyclodextrin with excess benzyl chloride. 17 Regioselective debenzylation of the primary rim was affected by treating the perbenzylated β-cyclodextrin with DIBAL-H. 18 This was followed by esterification 19 with the acid derivative of fluorophore 4, yielding mono-and di-functionalized sensors S2 and S3. Compounds S2 and S3 were fully characterized by 1 H-NMR, 13 C-NMR, MALDI-TOF mass spectrometry, UV-visible and fluorescence spectroscopy.    (Table 1). These binding constants are orders of magnitude higher than the highest literature-reported binding constants for analyte 5 in β-cyclodextrin (Ka = 50-215 M -1 ). 21 Higher association constants for analyte-sensor binding are known to lead to improved sensor performance, 22 a phenomenon that is also borne out in this system (vide infra).
Similarly, in this case, strong binding of analytes 5-8 in hosts S1-S3 induced marked changes in the resulting fluorescence emission due to proximity-induced interactions between the analyte and the fluorophore. These changes were quantified according to  The sensor S1 shows a fluorescence modulation value close to 1.00 for all the tested analytes, indicating minimal to no effect on the fluorescence emission of the fluorophore with the introduction of the analyte. In contrast to this, fluorescence modulation values measured for sensors S2 and S3 are significantly different from that of S1, and display widespread variability between different classes of analytes as well as within each analyte class (Table 2). These results clearly demonstrate the effect of the sensor architecture, and in particular the effects of covalent fluorophore attachment and the number of fluorophore units. The covalent attachment ensures close proximity between the cyclodextrin-bound analyte and the fluorophore moiety(ies), causing various degree of fluorescence modulation to occur. An example of analyte-induced fluorescence modulation for analyte 8 is shown in Figure 5.  The fluorescence signals of sensors S1-S3 in the presence of analytes 5-8 were subjected to linear discriminant analysis, and enabled 100% selectivity between the different aromatic alcohol isomers ( Figure 6). This selectivity is particularly noteworthy as such isomers are challenging to separate using other analytical techniques. 23 The binding of other structural isomers and analogues in supramolecular hosts S1-S3 also led to analyte-specific changes in the fluorescence emission (Table 3)  Analytes 9-12 represent a class of aliphatic alcohols consisting of cyclohexylmethanol (11) and its isomers. These compounds are widely used as alkene precursors, 24 and a structurally similar analogue was part of a recent chemical spill. 25 While all the analytes are structural isomers, analytes 10 and 12 are also stereoisomers. Distinct fluorescence modulation values are noted for sensor S3 in combination with stereoisomers 10 and 12, highlighting the power of the cyclodextrin-based sensor in differentiating even small structural changes. Overall, the use of sensors S1-S3 in combination with these analytes enabled 100% differentiation using linear discriminant analysis (Figure 7).   Analytes 17-21 represent aliphatic n-hexane (compound 17), its commonly occurring structural isomers (compounds 18-20, generated in 10-30% yield from industrial production of hexane) 30 and its cyclopentane analogue (compound 21). The fact that hexanes co-occur as isomeric mixtures complicates a variety of applications that require accurate characterization. 31 Using this supramolecular sensing strategy, 100% accurate classification between these analytes is achieved (Figure 9). Analytes 22-26 represent polychlorinated biphenyls (PCBs), a class of POPs that cause neurotoxicity 32 and endocrine disruption. 33 As a result of these effects, the use of PCBs has been banned in many countries; however, their environmental persistence means that significant amounts of PCBs are still found in the environment. 34 100% accurate classification has been achieved for these analytes (Figure 10), which is particularly crucial because these analytes have widely disparate toxicities. The ability of this detection method to generate well-separated signals was further investigated by generating an array with all analytes from all classes. In this case, the array exhibited well-separated clusters based on compound class, as well as excellent separation within each class. Overall, 100% accurate identification was obtained (see ESI for more details).
The limits of detection for each sensor S1, S2 and S3 for each class of analytes were calculated, to determine their ability to sense analytes at environmental levels of concern and at levels that induce toxicity. In every case, the calculated limits of detection were at or below the literature reported limits of concern (Table 4), highlighting the sensitivity of this method. Table 4. Calculated limits of detection and comparisons to known levels of concern.
Analytes Sensors LOD calculated (µM) Limit of concern (µM) 5 S2 7.1 ± 0.9 a 6 S1 5.5 ± 0.2 21.27 35  0.17 ± 0.01 1.00 38 a Limits of concern have not been established for these compounds Practical applications of this system require the capability to identify analyte mixtures, because environmental contamination scenarios almost always involve such mixtures.
To that end, preliminary work focused on identification of 1:1 binary mixtures of aromatic alcohol analytes 5-8. Using the supramolecular sensors combined with linear discriminant analytical techniques, 83% accurate identification of the 1:1 binary mixtures was obtained ( Figure 11). Interestingly, the mixture of analytes 5 + 7 is grouped near the mixtures of analytes 6 + 8 and 5 + 8, which reduces the overall classification accuracy slightly. This kind of co-clustering of analyte groups has been observed previously, and can be attributed to similar sensor responses originating from competing interactions between each component of the mixture. Other than those combinations, the mixtures demonstrated excellent signal separation and accurate identification. Current work in our group is focused on improving classification accuracy of analyte mixtures, expanding such techniques to multiple analyte classes, and moving from binary mixtures to ternary and even quaternary mixtures of analytes.

DETAILED SYNTHETIC PROCEDURES
Overall Synthetic Scheme:
The solution was allowed to stir for one hour at room temperature, after which time benzyl chloride (18.5 mL, 65 mmol, 36 eq.) was added over the course of one hour. The reaction mixture was stirred for 18 hours at room temperature, followed by the addition of methanol (20 mL

Reaction 2: Synthesis of Mono-debenzylated β-cyclodextrin:
To a stirred solution of perbenzylated β-cyclodextrin (600 mg, 0. Fluorescence emission spectra were obtained using a Shimadzu RF-5301PC spectrophotofluorimeter with 3 nm excitation and 3 nm emission slit widths. In a quartz cuvette, 0.5 mL of S1, S2, or S3 solutions (5 μM in DMSO) and 2 mL of DI water were combined. Then, the solution was excited at 320 nm, and the fluorescence emission spectra were recorded. Repeat measurements were recorded for four separate trials.
The fluorescence emission spectra were integrated vs. wavenumber on the X-axis, and All of the fluorescence emission spectra were integrated vs. wavenumber on the X-axis, and calibration curves were generated. The curves plotted the analyte concentration in μM on the X-axis, and the fluorescence modulation ratio on the Y-axis. The curve was fitted to a straight line and the equation of the line was determined.
The limit of detection is defined according to Equation S2: Where SDblank is the standard deviation of the blank sample and m is the slope of the calibration curve. In cases where the slope of the trendline was negative, the absolute value of the slope was used to calculate the LOD. In all cases, the LOD was calculated in μM.

DETAILED PROCEDURES FOR THE HPLC ANALYSIS OF S2 AND S3
The HPLC analysis of the cyclodextrin-fluorophore covalent hosts was performed on a Cyclohexylmethanol -S1 Cyclohexylmethanol -S2

PCB209 -S3
2-Methylpentane -S1 One group of guests that is known to bind well in cyclodextrins is squaraine fluorophores, 15 which contain a common cyclobutene-dione core. 16 The highly unique electronic structure of the squaraine fluorophore leads to anomalously high extinction coefficients, 17 narrow Stokes shifts, and high quantum yields, 18 with absorption and emission maxima often in the near-infrared spectral region. As a result of these properties, especially the near-infrared absorption and emission that limit interference from other analytes, 19    All computational modelling was done using commercially available Spartan software, version 16. To obtain the molecular models, the structures were first energy-minimized using multiple runs of molecular dynamics simulations. Next, these structures were submitted to MMF94 molecular mechanics methods, and the minimized structure from this was further optimized and minimized using a PM3-level semi-empirical force field in a gaseous medium. The energy obtained from these calculations were used to calculate the stabilization energy of the complex using the equation below 23 : where ∆ is the stabilization energy of the host-guest complex, is the energy of formation of the host-guest complex, is the energy of formation of the β-cyclodextrin dimer hosts, and is the energy of formation of the squaraine guest.

Synthetic routes
Three novel, covalently linked β-cyclodextrin dimers (  included in the other two structures ( Figure 2). All of the cyclodextrins and squaraines were fully characterized via spectroscopic methods (see ESI for more details).

Complexation-driven spectroscopic changes
Squaraines can exist in their monomeric form under certain conditions, but are particularly prone to aggregation (as either H-aggregates or J-aggregates) due to their planar, conjugated structures ( Figure 4). 24 Cyclodextrin complexation of the squaraines affects the equilibrium between the monomeric and aggregate states, with squaraines in β-cyclodextrin complexes stabilized in their monomeric states, and squaraines in γ-cyclodextrin complexes stabilized as dimers. 25 In our system, the monomeric squaraine species (shown in Figure 5

Complexation-induced effects on squaraine hydrolysis
The  We plotted the ratio of T in the presence of cyclodextrin hosts 1-3 (Tdimer) to T in the absence of any host (Tcontrol) (Figure 9), noting that a ratio value of 1 would represent no effect of the host on rates or extents of hydrolysis. Notably, squaraines in the presence of host 1 demonstrated the lowest effects of complexation on hydrolysis behaviours (as indicated by ratios closest to 1). For most squaraines, the presence of host 2 led to moderate protection from hydrolysis, indicated by ratio values slightly higher than 1, and host 3 conferred substantial protection to the squaraine guests from hydrolysis.

Fluorescence titration on cyclodextrin dimer hosts
The guest-induced fluorescence changes of the dimers 1-3 ([dimer] = 5 x 10 -7 M) was studied in presence of increasing concentrations of squaraines 4-11. 26 Importantly, the observed behaviours are intimately dependent on the specific interactions between each squaraine guest and cyclodextrin host, which makes general trends challenging to elucidate.  Among the straight chain alkyl-substituted squaraines, compounds 5 and 6 have optimal sizes and hydrophobicities to bind in hosts 2 and 3, respectively.
Squaraines with longer alkyl chain substituents (7-9) exhibited association constants that were one and two orders of magnitude lower with dimers 2 and 3, respectively. This trend is likely due to less optimal steric matching with the host cavity for the larger squaraines.
A comparison of squaraine guests 10 and 11 revealed that compound 11 exhibited higher association constants because of the tert-butylphenyl substituent, which has been reported to bind strongly in β-cyclodextrin (Ka = 1.6 x 10 4 M -1 ). 27 Despite structural similarities between hosts 2 and 3, the association constants for compound 2 are low for most of the squaraines compared to those observed for compound 3 (Figure 11). This differential behaviour may be a result of the greater hydrophobicity of the anthracene unit in compound 3, which in turn contributes to increased cooperativity of the β-cyclodextrin units in forming a stable 1:1 hostguest complex.

Computational Modeling
The stabilities of the 1:1 cyclodextrin dimer-squaraine complexes were calculated using PM3 calculations (with a semi-empirical force field) for host 3 with guests 6, 10 and 11.
Further comparisons were drawn between complexes formed by squaraine 6 with all of the hosts to determine the oxoanion-amide distances in host-guest complexes ( Table 2).
The calculated negative stabilization energies indicate that the squaraine guests thread inside the host cavity (1-3) to form a stable host-guest association complex. Complexes 3+11 and 3+10 were found to be much more stable than their n-hexyl counterpart 6. The distances of the two oxoanion-amide pairs (d1 and d2) were compared to determine the precise position of the electrophilic squaraine core. Two possible modes of interaction are possible: (a) where the two amide groups of the linker interact closely with one oxoanion of the squaraine core, resulting in a significant difference between the two measured distances (d 2 -d 1 ) (I, Figure 13); or (b) where the two amide groups of the linkers interact equally with both oxoanions of the squaraine core, resulting in roughly equivalent distances (II, Figure 13).

Deconvolutions
The UV spectra were subjected to a piecewise linear background subtraction method.
The selection of spectral positions to run the background were identified by a custom threshold approach. After the background spectral subtraction, the spectral signal was fitted using "NonlinearModeFit" command (method set to "automatic") with three gaussian functions, · ( − ) 2 2 2 , where A, μ, σ and x have their usual meaningsamplitude, mean, standard deviation, and wavelength respectively.

Linear Fits
All linear fits were done with "NonlinearModeFit" command (method set to concentration at time zero, rate constant and the independent variable respectively. The value of c is found from the corresponding exponential fit.

Titration Curve Fits
All fluorescence titration data fits were done with Solver.xlam using "GRG-Nonlinear" method in Excel 2017 using the 1:1 and 1:2 supramolecular titration equations.

METHODS FOR COMPUTATIONAL EXPERIMENTS
All computational modelling was done using commercially available Spartan software, version 16. To obtain the molecular models, the structures were first energy-minimized using multiple runs of molecular dynamics simulations. Next, these structures were submitted to MMF94 molecular mechanics methods, and the minimized structure from this was further optimized and minimized using a PM3-level semi-empirical force field in a gaseous medium. The energy obtained from these calculations were used to calculate the stabilization energy of the complex using the equation below 1 : where ∆ is the stabilization energy of the host-guest complex, is the energy of formation of the host-guest complex, is the energy of formation of the β-cyclodextrin dimer hosts, and is the energy of formation of the squaraine guest.

Synthesis of β-cyclodextrin dimer host 2 Compound S7
Compound S6 (1.10 g, 7.04 mmol, 1.00 eq) was dissolved in dry dichloromethane (40.0 mL) and degassed with nitrogen. Under an active nitrogen stream, N-bromosuccinimide (3.75 g, 21.0 mmol, 2.98 eq) and benzoyl peroxide (172.5 mg, 0.712 mmol, 0.10 eq) were added and the suspension was degassed to give a yellow suspension. The reaction mixture was heated under nitrogen at 55 °C for 6 hours. The reaction mixture was cooled to room temperature and then washed with 2 M HCl (2 x 15 mL), 2 M NaOH (2 x 20 mL), brine, and dried with MgSO4. The solvent was evaporated to yield an off-white powder as the crude product, which was purified by column chromatography with 1:9 (vol/vol) dichloromethane/hexanes to yield the desired product in 90% yield (2.0 g, 6.34 mmol). 1

Compound S9
An aqueous NaOH (12.5 wt%, 1.0 mL) solution was added to the solution of S8 (0.1 g, 0.24 mmol) in methanol (5.0 mL). The resulting mixture was heated to 90 o C with stirring for 5 hours, after which time the mixture was treated with ice, water and hydrochloric acid to adjust the pH to approximately 2. The resulting mixture is extracted with EtOAc, dried over anhydrous Na2SO4 and under reduced pressure to yield a white solid S9 (0.08 g, 0.23 mmol, 95% yield). No further purification was needed.

Synthesis of β-cyclodextrin dimer host 3
Compound S13 To a stirred solution of anthracene S12 (1.78 g, 10.0 mmol, 1.00 eq), dry ZnCl2 (1.64 g, 12.0 mmol, 1.20 mmol), paraformaldehyde (1.50 g, 50 mmol, 5.00 eq) in dioxane (20.0 mL) was slowly added 37wt % fuming concentrated aqueous hydrochloric acid (40.0 mL) at room temperature. After stirring slowly at gentle reflux for 3 hours, the heating was stopped and the mixture was allowed to stand for 16 hours. The fine granular yellow solid that formed was separated by filtration, and washed with water and dioxane to give a crude product. The crude product was recrystallized from toluene to give compound S13 as a yellowish solid (1.

Compound S14
A mixture of excess dimethylmalonate (10.6 mL, 92.0 mmol, 97.8 eq) and sodium hydride (2.26 g, 94.3 mmol, 100 eq) in dry THF (40.0 mL) was refluxed with stirring for 3 hours. Then, a solution of compound S13 (0.26 g, 0.94 mmol, 1.00 eq) in dry THF (20.0 mL) was added to the reaction mixture over a 15 minutes time period. The mixture was kept at reflux temperature with stirring for an additional 20 hours. The reaction was stopped by the addition of ice (50 g), water (50 mL) and hydrochloric acid (to adjust the pH ~3). The solution obtained was extracted with chloroform (3× 100 mL) and dried over anhydrous sodium sulfate. The solvent and the excess diethylmalonate were evaporated to dryness, further purification being carried out by column chromatography with 2:8 (vol/vol) EtOAc/Hexane to yield S14 (0.4 g, 92% yield) as a pale-yellow solid. Rf = 0.6 (hexane/EtOAc, 6:4). 1

Compound S16
Compound S15 (0.70 g, 1.70 mmol, 1.00 eq) was dissolved in 5.0 mL of diphenyl ether and refluxed at the solution's boiling point of 259 o C for 48 hours. Carbon dioxide gas was evolved from the reaction mixture during the reflux. Precipitation was observed on cooling to room temperature. Hot sodium hydroxide (15% aqueous solution), was added into the stirred solution until all of the precipitate was dissolved. The pH was adjusted to 5.5-6.0 with dilute HCl (10% aqueous) and the precipitate was filtered and recrystallized from an ethanol-water (1:1) mixture. A dark brown colored solid S16 (0.50 g, 95% yield) 1
[9] (μΜ)                                                                Anions are important targets for binding and detection due to their ubiquitous nature and public health relevance. 11 Fluoride, for example, is of interest due to the importance of fluoridated water in promoting dental health; 12 excessive amounts of fluoride, by contrast, can lead to fluorosis. 13 Other key anions include those with negative health effects including phosphate, 14 nitrate, 15 thiocyanate 16 and cyanide. 17 A third class of anions is those that are explosive such as azide. 18 Polycyclic aromatic hydrocarbons (PAHs) are another class of important detection targets, with negative health and environmental effects, 19 and are formed from the incomplete combustion of petroleum. 20 Their environmental stability means that they bioaccumulate and biomagnify, 21 which is of concern due to their known and suspected teratogenicity, 22 mutagenicity 23 and carcinogenicity. 24 Work in the Levine group has focused on the detection of toxicants using cyclodextrinpromoted energy transfer 25 and cyclodextrin-promoted fluorescence modulation, 26 as well as on the use of synthetic macrocycles for the enhanced binding and detection of PAHs. 27 One shortcoming is that the previously synthesized macrocycles lacked easily detectable photophysically active components, which in turn meant that an external fluorophore was required to obtain a response signal. Incorporating a UV-active moiety, such as fluorenone, directly into the backbone of the macrocycle would enable the direct use of optical detection methods, and incorporation of a triazole functionality will enable the detection of a broader variety of analytes. Reported herein is the high yielding synthesis of precisely such a macrocycle, compound 1, containing a photophysically active fluorenone unit and two triazole moieties, and its versatility in binding and detecting both PAHs and anions with extremely high sensitivities.
Macrocycle 1 was synthesized from compounds 2 and 3 via a copper catalyzed azidealkyne cycloaddition (Figure 1). This reaction proceeded under high dilution conditions 28 in toluene to obtain a 71% isolated yield. The low solubility of the macrocycle in toluene caused it to crash out of the reaction mixture, and was crucial in enabling high yields. The formation of the macrocycle was confirmed by NMR spectroscopy and mass spectrometry (see ESI).   In contrast to the limited changes in the absorbance spectra, the fluorescence emission of each of the analytes decreased with the addition of the macrocycle (Table 1)  Of note, these decreases were not accompanied by significant shifts in the emission maxima, in contrast to a report of analogous system in which such a red shift is observed. 32 In that case, the red-shift is probably a result of excited state energy transfer between the anthracene host and guanine guest.  In the case of naphthalene (analyte 4), the excitation wavelength of 265 nm is a wavelength at which compounds 1, 8, and 9 have noticeable absorption cross-sections (see ESI). Although significant wavelength-dependent fluorescence decreases were observed, these observed changes are indicative of an inner filter mechanism, where the macrocycle absorbs energy and filters some of that energy from reaching the analyte. 35 The limits of detection of analytes 4-7 using this method were calculated following literature-reported procedures ( Table 2). 36 For analyte 4, the calculated detection limit is a result of the inner filter effect-induced fluorescence changes. 35 The nanomolar detection limits obtained for the analytes are close to or below the literature-reported levels of concern for three out of the four analytes (compounds 4, 5, and 7), 37 which highlights the high sensitivity of this fluorescence method for PAH binding and concomitant detection. The limits of detection for the analytes in the absence of the macrocycle were higher, which highlights the role of the macrocycle in enhancing fluorescence sensitivities. In addition to binding PAHs in the cavity interior, macrocycle 1 (10 mM in DMSO) was also investigated for its ability to bind anions. Among all anions studied (fluoride, cyanide, azide, and thiocyanate), only fluoride exhibited a noticeable spectroscopic change ( Figure 6) with increases in the molar absorptivity of the macrocycle's λmax bands at 264 and 305 nm. The response for fluoride is likely due to its ability to act as a hydrogen bond acceptor, as a result of its small size, high electronegativity, and high charge density. 40 The fluoride binding was confirmed by nonlinear curve fitting of the 1 H-NMR titration data to host-guest binding models (Figure 7 and Table 3). An excellent non-linear fit was obtained for a 1:2 binding stoichiometry between macrocycle 1 and two fluoride anions, and this stoichiometry was confirmed with a Job plot analysis that showed a      (Table 4).

INTRODUCTION
Vesicles are made up of single or multiple bilayers consisting of surfactants or lipids 1,2 and are promising delivery vehicles for drugs, enzymes, and other active ingredients. 3−6 In cosmetics, vesicles not only deliver encapsulated ingredients like perfume but also counter skin dryness as the surfactants are hydrated. 7,8 Esterquats such as diethylester dimethylammonium chloride (DEEDMAC) are doublechained cationic surfactants that are used as the major ingredients in fabric softeners. 9−11 During the latter stage of a laundry cycle, DEEDMAC adsorbs onto negatively charged fabrics, producing a thin lubricating layer that reduces friction between fabric filaments.
The inability to form hydrogen bonds also aids in the reduction of static charge. Because of ester linkages, DEEDMAC is readily degradable by hydrolysis in a post-washing cycle. 12 The phase behavior of double-tailed cationic surfactants such as didodecyl dimethylammonium bromide 13,14 (DDAB), dioctadecyl dimethylammonium bromide (DODAB), and dioctadecyl dimethylammonium chloride (DODAC) in water has been studied extensively. 15,16,17,18 These surfactants self-assemble to form unilamellar and multilamellar vesicles in the concentration range of 0.15−30 wt % and above the main phase transition temperature of the bilayer. These vesicles often exist in a kinetically stabilized state. 19−21 Fragrance is an integral part of many consumer products. 22 Perfume raw materials (PRMs) are added to these products to generate a pleasant odor over extended periods of time. PRM molecules, typically oils, usually have extremely low water solubility.
They are therefore distinct from cosurfactants and hydrotopes. However and deuterium oxide (99.9 atom % D) were obtained from Acros Organics. All materials were used as received.

Viscosity
LA and eugenol were added to the DEEDMAC suspension to a final concentration of 2 wt % PRM. The samples were vigorously hand shaken in a vial and left undisturbed at room temperature before being examined at various time points. A TA Instruments AR2000 EX stress-controlled rheometer with a 40 mm diameter and 0.5° steel cone was used for measuring the steady shear viscosity as a function of shear rate.

Cryogenic Transmission Electron Microscopy (cryo-TEM)
The DEEDMAC suspension by itself as well as a suspension diluted by a factor of 10 using a 1200 ppm of a CaCl2 solution in water were prepared. This salt concentration ensured that the dilution of the DEEDMAC was isotonic. All additives were mixed with the diluted DEEDMAC suspension, and the samples were hand shaken in a vial. A few microliters of the undiluted or diluted DEEDMAC or the mixed DEEDMAC/additive suspensions, equilibrated in a controlled environment vitrification system (CEVS) at 25 °C and 95%− 100% humidity, were deposited on a holey carbon grid. The high humidity suppresses water evaporation from the sample prior to vitrification. The grid was blotted and then plunged into a liquid ethane reservoir cooled by liquid nitrogen. Rapid heat transfer away from the grid leads to sample vitrification. The samples were vitrified at designated time points after mixing was ceased. The grid containing the sample was transferred to a cooled tip of a Gatan 626DH cryo-transfer stage. The stage was then inserted into a JEOL JEM 2100 transmission electron microscope. The sample was maintained at −175 °C; a low electron dosage (∼20 e−/Å2) and a slight underfocus (1−6 μm) were used for imaging.

Nuclear Magnetic Resonance (NMR)
1 H-NMR spectra were recorded at room temperature on a Bruker-Avance 300 MHz spectrometer with the singlet peak of HDO at 4.79 ppm as reference. All the D2O used in the NMR samples contain 1200 ppm of CaCl2. Also, 2 wt % eugenol in the CaCl2containing D2O, as well as the DEEDMAC suspension added to this mixture, were probed. Despite the low solubility of cyclohexanol, guaiacol, and eugenol in water, these compounds disperse homogeneously after vigorous shaking and provide a clean 1 H-NMR signal. It was not possible to get a homogeneous dispersion of LA in D2O, and thus a good NMR signal, as the sample phase separated within several minutes. For probing the LA-containing samples, an 80 vol % solution in acetone was first prepared.
This solution was diluted with the CaCl2-containing D2O to a 2 wt % concentration of LA. The DEEDMAC suspension was then added, and peak intensities as well as the broadening of specific peaks relevant to LA were monitored.   The layers peel off and form multilamellar vesicles when the suspension is agitated.

RESULTS AND DISCUSSION
From an application perspective, this is important, as multilamellar structures are able to "store" surfactant in the inner leaflets and thus provide greater supply of DEEDMAC surfactant/volume of suspension than their unilamellar counterparts. In this work, we did not undertake a detailed study of the phase behavior of DEEDMAC in a saltcontaining aqueous medium but focused only on the role of additives on this suspension.
The DEEDMAC suspension supplied by P&G contains CaCl2. When this suspension was diluted by a 1200 ppm calcium chloride solution in a 1:1 volume ratio, the multilamellar structures, shown in Figure 2(b), were preserved, and the samples look identical to those in Figure 2(a). Osmotic pressure changes caused by an increase in salt concentration outside dioctadecyldimethylammonium bromide (DODAB) vesicles has been known to deflate them into cup-like shapes where the poles of the vesicles approached each other, until they fuse into bilamellar twinned vesicles. 26 Figure 2 (c) shows that dilution with water without salt changes the microstructure to unilamellar vesicles because osmotic stresses drive water into multilamellar vesicles and cause them to rupture. This phenomenon has also been observed by others. 8 The absence of this structural change upon addition of a 1200 ppm of CaCl2 solution indicated isotonic conditions matching this salt concentration for our sample. Thus, all additives were formulated in water containing 1200 ppm of CaCl2. Any observed changes could then be directly attributed to the presence of additives. This issue is also important for our NMR experiments.   (c) additives. However, when eugenol was added to the DEEDMAC vesicle suspension, the microstructures changed over time. A few seconds after mixing, multilamellar vesicles were observed, as shown in Figure 4(a). After 6 hrs, we observe undulating bilayers, indicated by the arrows in Figure 4 (b) and (c). We speculate that the insertion of the eugenol into the bilayer lowers the phase transition temperature, 27 and thus the bending modulus, and promotes these undulations. Some of the external lamellae got exfoliated and broke off from the vesicles 12 h after mixing, resulting in a reduction in vesicle size. Bilayer fragments were also observed. The newly uncovered lamellae get exposed to eugenol and follow a similar path, ultimately resulting in predominantly unilamellar vesicles in the suspension 24 hrs after mixing. Some tubules were also formed in this process. These are shown in Figure 4(d).    Bilayer fragments are not very visible when seen normal to their surfaces because of insufficient contrast with the background, in Figure 4(d), but they become edge on and provide enough contrast to become visible in Figure 4(e). Tubules appear as lines, and their relative distance does not change upon tilting. 28 The exfoliation of the multilamellar vesicles over time into predominantly unilamellar vesicles, bilayer fragments, and tubules results in a more volume filling arrangement that causes a rise in the low shear viscosity.
We probed the interaction of LA and eugenol with the DEEDMAC vesicles using 1 H-NMR. Addition of 2 or 4 wt % DEEDMAC vesicle suspension to a 2 wt % eugenol suspension in D2O containing 1200 ppm of CaCl2 showed peak broadening at chemical shifts of 3.75 ppm (methoxide protons), 5.90 ppm (vinyl proton), and 6.75 ppm (aromatic protons), marked by the arrows in Figure 5. This intense peak broadening for eugenol indicates strong association with the DEEDMAC vesicle bilayer, which can be ascribed to cation−π electron interaction 29 of the electron-rich aromatic group of eugenol with the positively charged headgroup of DEEDMAC. The linear alkene portion of eugenol also promotes its insertion into the vesicles, positioning the eugenol for reduced mobility in the bilayer. Figure 6 shows an increase in the intensity of the peak associated with the protons in the methyl groups of DEEDMAC over 24 hrs, 30 because of the increased mobility of DEEDMAC caused by bilayer exfoliation and breakage.     between the DEEDMAC and the additives. The microstructures did not change when the substituents on the aromatic rings were modified, but they did respond to a change from an aliphatic to an aromatic additive. Thus, any other contributions, such as hydrophobic interactions and steric effects, which can dominate interactions between some surfactants and additives, appear to be less important for the additives used in our experiments. We note that association of a PRM with a bilayer has consequences on the olfactory effects of these materials through its impact on release kinetics. This latter issue has not been studied in this paper.