Counterion Anchoring Effect on the Structure of the Solid-State

Article pubs.acs.org/JPCC

Counterion Anchoring Effect on the Structure of the Solid-State Inclusion Complexes of β‑Cyclodextrin and Sodium Perfluorooctanoate Abdalla H. Karoyo,† Paul S. Sidhu,‡ Lee D. Wilson,*,† Paul Hazendonk,*,‡ and Alex Borisov‡ †

Department of Chemistry, University of Saskatchewan, Saskatoon, SK S7N 5C9, Canada Department of Chemistry and Biochemistry, University of Lethbridge, Lethbridge, AB T1K 3M4, Canada

‡

S Supporting Information *

ABSTRACT: Characterization of the structure and dynamics for the solid inclusion complexes (ICs) between β-cyclodextrin (βCD; host) and sodium perfluorooctanoate (SPFO; guest) was carried out using 1H/19F/13C NMR spectroscopy. The 1:1 and 2:1 β-CD/SPFO solid complexes were prepared by a modified dissolution method. Evidence for the formation of β-CD/SPFO ICs was provided by 13C DP (direct polarization) and CP (cross-polarization) solid-state NMR spectroscopy with magic angle spinning (MAS) at 20 kHz. The complexation-induced shifts (CIS) of 1H/19F/13C nuclei between solution and the solid state for β-CD/SPFO complexes and the closely related complexes of β-CD/PFOA (perfluorooctanoic acid) were compared. The counterion effect for SPFO and PFOA was observed according to their variable structure and binding as inclusion compounds with β-CD. The effect of sodium versus hydronium counterions on the structure and dynamics of inclusion complexes for these systems was supported by DSC, TGA, FT-IR, and powder X-ray diffraction (PXRD). Simulations of the CF3 19F NMR with MAS at 25 kHz and selected dipolar coupling strengths were utilized in conjunction with deconvolution analyses of the experimental CF3 lineshapes to probe the dynamic properties of SPFO and its complexes with β-CD. The dynamics of the guest are influenced by the host/guest binding geometry and the stoichiometry of the complex, where free rotation of the CF3 group as well as rotations of the C−F bonds occur. 19F DP/MAS NMR results and spin−lattice (T1) and spin−spin (T2) relaxation times in the laboratory frame at variable temperatures in the solid phase indicate that the dynamics of SPFO in β-CD/SPFO complexes are unique compared to those of PFOA in β-CD/PFOA complexes, due to the role of counterion effects of the guest.

1. INTRODUCTION

differences in surface activity. PFOA and PFOS are the most common PFCs that exist predominantly as anions in aqueous environments due to their low pKa values (PFOA ∼ 2.5, PFOS < 0).9,10 SPFO and APFO (ammonium perfluorooctanoate) are salts of PFOA commonly used as surfactants. PFCs (e.g., SPFO and APFO) are also used as essential processing agents in fluorotelomer-based formulations for making consumer products (e.g., films and membranes for nonstick cookware and outerwear and cleaning agents for carpets).11 The toxicological effects of PFCs are exacerbated by their environmental

Perfluorinated compounds (PFCs) of the type CF3-(CF2)n-R′ with diverse functional groups (R′ = CF2−OH, COOH, CO− NH2, or CF2−SO3H) and variable chain lengths have received considerable attention due to their environmental and healthrelated concerns. Such materials have applications that range from fire-fighting foams to pesticides and surface-active agents.1−3 PFCs have unique physicochemical properties in comparison to their hydrocarbon analogues. For example, a comparison of the critical micelle concentrations (CMC) of perfluorooctanoic acid (PFOA; CF3-(CF2)6-COOH, ∼ 8.7−10 mM), perfluorooctanesulfonate (PFOS; 6.0−8.0 mM), sodium perfluorooctanoate (SPFO: CF3 -(CF 2) 6-COO−Na +, ∼32 mM),3−7 and sodium octanoate (∼400 mM)8 highlights © 2015 American Chemical Society

Received: July 28, 2015 Revised: August 30, 2015 Published: September 1, 2015 22225

DOI: 10.1021/acs.jpcc.5b07302 J. Phys. Chem. C 2015, 119, 22225−22243

Article

The Journal of Physical Chemistry C

Scheme 1. Molecular Structures of (a) β-CD Oligomer where n = 7, (b) β-CD Represented As a Toroidal Macrocycle, and (c) SPFOa

a

Note that the conformation of SPFO is not depicted accurately.

persistence in aquatic environments and biological samples. For example, evidence of liver damage was noted in primates that were treated with APFO.12 Cyclodextrins (CDs) have been widely used to lower the surface activity of perfluorocarbon compounds through formation of host−guest complexes.13,14 CDs are a family of macrocyclic oligomers consisting of six (α-CD), seven (β-CD), or eight (γ-CD) glucopyranose units that are linked by α-(1 → 4) glycosidic linkages (cf. Scheme 1a,b).15 We previously reported the structural and dynamic properties for the complexes formed between β-CD and PFOA (C8)16,17 in solution and the solid state, along with a similar study for a shorter-chain perfluorobutyric acid (PFBA; C4).18 It was concluded that PFOA forms 1:1 and 2:1 host−guest inclusion compounds with β-CD where variable conformation and dynamics of the guest were observed. By contrast, PFBA was reported to form chiefly 1:1 host−guest complexes due to its shorter alkyl chain. SPFO (C8; cf. Scheme 1c) is a commonly studied7,19−25 PFC carboxylate salt because it shows strong affinity with β-CD to form complexes due to its unique amphiphilic nature. Reinsborough et al.19−21 and Guo et al.7 were among the first to study CD/SPFO complexes using conductometry and highresolution NMR spectroscopy, respectively. Guo et al.7 reported a detailed and systematic binding study of the complexes formed in solution between CDs and a series of sodium PFC carboxylates comprising four to nine carbon atoms. In that study, it was reported that the association of CDs with shorter (C ≤ 5) PFC chains favors 1:1 host−guest complexes, whereas longer chains form both 2:1 and 1:1 complexes. A preferred extended (all-trans) conformation of the PFC guest systems with longer alkyl chains was reported for the 2:1 host−guest complexes in solution. Similar conclusions were drawn from previous studies of β-CD/PFOA systems in aqueous solution and the solid state.16,17 For structural considerations, Wenz26 described host−guest complexes involving CDs as consisting of single or multiple host−guest associations since such complexes can form inclusion as well as noninclusion (or facial) associations with a variety of topologies. Wilson and Verrall27,28showed that higher-order host−guest complexes (e.g., 2:1 and 1:2) were formed in aqueous solution between modified β-CD and long-chain PFCs. Facial-type complexes have also been reported of a low molecular weight halogenated ethane guest with α-CD and β-CD.29

A variety of other reports on the complexes between CDs and SPFO in solution have been documented. These include a study of the effects of SPFO monomer−micelle exchange rates in the presence of CD,22 binding constant determination using native and substituted CDs,23,24 the selective association of CD to SPFO in the presence of other hydrocarbon surfactants,25 and a study of hydrophobic effects on complex formation.30 It was recently demonstrated that complexation of SPFO by CDs is important to the biomedical field since these perfluorocarbon surfactants can be used as oxygen carriers, imaging contrast agents, delivery agents for pulmonary drugs, and genes.31−33 The fate, transport pathways, and persistence of such recalcitrant materials are poorly understood, partly due to their unique surface activity, chemical mobility, and stability when compared with their hydrocarbon counterparts. In spite of the numerous solution studies that have reported the structure of the complexes of β-CD and SPFO, similar studies in the solid state were not reported. This may be due to the challenges associated with obtaining good quality single-crystal results; however, high-resolution multinuclear solid-state NMR techniques offer a versatile alternative for the study of such host−guest inclusion complexes.34−37 In this study, we report the complex formation of SPFO with β-CD in aqueous solution and the solid state using 1H/19F/13C NMR and FT-IR spectroscopic methods, thermal analyses (DSC/TGA), and powder X-ray diffraction (PXRD). The complexes in the solid state were prepared at the 1:1 and 2:1 βCD/SPFO stoichiometric ratios using a modified dissolution method described elsewhere.17 The complexation-induced shift (CIS) values in solution and solid phases for β-CD/SPFO complexes were compared to analogous values for β-CD/ PFOA complexes to distinguish the mode of inclusion of the guests within the host and to determine the role of the sodium vs hydrogen ions on the binding of the guest. The dynamical properties of SPFO in the free and bound states were examined using a comparison of spectral simulations and deconvolution analyses of the 19F CF3 NMR resonance in conjunction with 19 F relaxation times in the solid state. The results of this study are expected to contribute a greater understanding of the structure and dynamics of both native SPFO and its complexes with β-CD, in both solution and the solid phase. Previous results on the structural and dynamical properties of the complexes of β-CD with PFOA (C8; acid) and PFBA (C4; acid) are highlighted to provide a relative comparison to the structure of SPFO (C8; salt) and its complexes with β-CD. 22226


Article


Scheme 2. Stepwise Formation of the 1:1 and 2:1 Complexes of β-CD (Toroid) and SPFO According to the 1:1 and 2:1 Host/ Guest Mole Ratiosa

a

Note that solvent is omitted for clarity.

curve and width parameters for the adiabatic 1H → 13C CP experiments were set at 50 and 10 000 Hz respectively, while those for 19F → 13C CP were set at 50 and 50 000 Hz, respectively. Optimal Hartmann−Hahn matching conditions were achieved with a contact time of 5 ms and approximate powers of 68 and 59.5 kHz for the 19F and 13C channels, respectively. The 19F T1 values (T1F) were measured by inversion recovery (180°-τ- 90°-acquire), while the T2F values were obtained with a rotor synchronous Hahn echo (90°-τ-180°-τ-acquire).38 The T1ρF relaxation times were measured with a pulse sequence that contains a spin-lock (SL) pulse out of phase by 90° with respect to the initial 90° pulse (e.g., 90°x-SLy-acquire).39 Relaxation time measurements were obtained at ambient and variable temperatures (0−70 °C). Simulation analysis of CF3 resonance as a function of dipolar coupling strength (Hz) was done using MestreNova 9.1.0. 2.5. FT-IR. Fourier Transform (FT) IR spectra were obtained using a Bio-Rad FTS-40 spectrometer with a resolution of 4 cm−1. All samples were prepared with spectroscopic grade KBr which constituted ∼80% (w/w) of the total sample. Samples were run as finely ground powders in reflectance mode. 2.6. Thermal Analyses (DSC and TGA). Differential scanning calorimetry (DSC) of the native β-CD, unbound SPFO, and the inclusion compounds was performed using a TA Q20 thermal analyzer over a temperature range of 30−370 °C. The scan rate was set at 10 °C/min, and dry nitrogen gas was used to regulate the sample temperature and purge the sample compartment. Solid samples for DSC were analyzed in hermetically sealed aluminum pans where the sample mass ranged from 3.80 to 4.00 mg. Thermogravimetric analysis (TGA) was performed using a TA Q50 over a temperature of 30−400 °C. Solid samples for TGA were heated in open pans where sample masses ∼7.0−8.0 mg were analyzed. 2.7. Powder X-ray Diffraction. PXRD spectra were collected using a PANalytical Empyrean powder X-ray diffractometer using monochromatic Cu Kα1 radiation. The applied voltage and current were set at 45 kV and 40 mA, respectively. The samples were mounted in a vertical configuration as evaporated hexane films and PXRD patterns were measured in a continuous mode over a 2θ angle range of 5−45° with a 2θ scan rate of 0.5/min.

2. EXPERIMENTAL SECTION 2.1. Materials and Chemicals. β-CD hydrate (∼10% w/w H2O) was purchased from Sigma-Aldrich Canada Ltd. (Oakville, ON), while SPFO (98%) was purchased from SynQuest Laboratories Inc. (USA). All chemicals were used as received without any further purification, and the structure of the host and guest systems is illustrated in Scheme 1. The water content of the materials was determined using thermogravimetric analysis during preparation of the sample mixtures. 2.2. Preparation of β-CD/SPFO Solid Inclusion Compounds. The solid β-CD/SPFO complexes were prepared at the 1:1 and 2:1 mol ratios (cf. Scheme 2) using a modified dissolution (evaporation) method adapted from a previous report.17 The solid products were ground into fine powder for characterization by solid-state NMR, DSC/TGA, FT-IR, and PXRD. 2.3. Solution NMR. Solution NMR experiments were performed on a three-channel Bruker Avance DRX 500 NMR spectrometer operating at 500.13 MHz for 1H and 470.53 MHz for 19F. Samples for 1H and 19F NMR spectroscopy were prepared in D2O at the 1:1 and 2:1 β-CD/SPFO mole ratios. The 1H and 19F NMR chemical shifts (δ) were measured with respect to tetramethylsilane (TMS; δ = 0 ppm) and trifluoroborane (BF3; δ = −131 ppm) as internal standards, respectively. In the case of CIS measurements, the 1H and 19F NMR signals were referenced externally to residual water (HOD; δ = 4.72 ppm) and 2,2,2-trifluoroethanol (TFE; δ = −76.9 ppm), respectively. COSY spectra for the complexes were acquired in the absolute value mode over a 7.5 kHz spectral window (ca. δ = −115 to −130 ppm). The 19F 90° pulse for 2-D COSY was set to 15 μs with a recycle delay of 3 s, and 16 scans were obtained for each of the 256 FIDs which contained 1 k data points in f2 over the 7.5 kHz spectral width. All NMR spectra were obtained in the solution state with D2O as the solvent at pH ∼ 5 at 295 K. 2.4. Solid-State NMR. All solid-state NMR spectra were obtained using a Varian INOVA NMR spectrometer operating in a triple-channel HFC mode using a 2.5 mm T3 HFXY probe operating at 125.55 MHz for 13C, 499.99 MHz for 1H, and 469.89 MHz for 19F. Solid-state 13C MAS spectra were referenced externally to adamantane (δ = 38.5 ppm) as a secondary standard with respect to TMS, while 19F MAS spectra were referenced to hexafluorobenzene (HFB; δ = −164.9 ppm). All samples were measured at the magic angle with a spinning rate of 20 kHz using 2.5 mm Vespel rotors equipped with Kel-F turbine caps, inserts, and end-caps, unless stated otherwise. All 13C NMR spectra were obtained using a 100 kHz sweep width with 8192 points in the FID and were zero-filled to 64 k data points, unless stated otherwise. The

3. RESULTS AND DISCUSSION 3.1. Thermal Analyses (DSC/TGA). DSC and TGA are complementary techniques that were used to characterize thermal properties of various polymers40 and host−guest systems.41,42 In general, DSC enables measurement of physical 22227


Article


Figure 1. DSC traces for β-CD hydrate, SPFO, and the 1:1 and 2:1 β-CD/SPFO inclusion complexes.

Figure 2. TGA traces for β-CD hydrate, SPFO, and the 1:1 and 2:1 β-CD/SPFO inclusion compounds.

phase transition temperatures and phase impurities as a function of heating rate, and TGA reveals weight loss profiles due to decomposition processes. The DSC and TGA results for β-CD, SPFO, and the 1:1 and 2:1 solid inclusion compounds prepared by the modified dissolution method are shown in Figures 1 and 2, respectively. In Figure 1, the DSC thermogram of native β-CD exhibits two endotherm events at ca. 115 and 320 °C that relate to dehydration and decomposition

transitions, respectively. The main endotherm of the unbound SPFO consists of a sharp peak at ca. 275 °C that is attributed to a melting transition. The thermograms of the 1:1 and 2:1 βCD/SPFO ICs display endotherm transitions at ca. 130 °C due to dehydration processes. The exotherm transitions over the temperature range between ca. 250−285 °C and 300−350 °C for the β-CD/SPFO ICs may be related to the decomposition of the guest and host, respectively. The absence of free guest is 22228


Article


Figure 3. FT-IR spectra of (a) β-CD, (b) SPFO, and the (c) 1:1 and (d) 2:1 β-CD/SPFO inclusion compounds.

thermal events which support the presence of more than one structural environment of the guest in this complex. The DSC results suggest that the guest molecule of the 2:1 complex may be fully encapsulated within the cavity of β-CD, whereas the guest in the 1:1 complex is partially included in variable microenvironments.26 In the case of the 1:1 complexes, there is likely more guest exposed to the aqueous environment that contributes to unfavorable guest−water interactions. By contrast, the 2:1 complex affords a more fully encapsulated guest that minimizes unfavorable guest−water interactions. The inclusion compounds decomposed at higher temperatures relative to native β-CD and pure SPFO which supports the formation of stable βCD/SPFO inclusion compounds. The greater stability of the βCD/SPFO complexes reported herein relative to the PFOA counterparts (cf. Figure 1 in ref 17) is evidenced by the increased decomposition temperature values of SPFO in agreement with the sodium counterion effect. The DSC results of the β-CD/SPFO complexes are supported by the TGA results in Figure 2. The TGA results in Figure 2 are presented as weight derivatives versus temperature. It is worthwhile to note that the difference in the temperature onset values for the DSC and TGA results may be related to the type of sample configuration (open vs sealed sample pans), as described in Section 2.6. The thermal events in Figure 2 at ca. 280 °C for SPFO and 310 °C for β-CD are associated with melting and decomposition processes, respectively. The inclusion compounds show several thermal events at ca. 60, 260, and 310 °C due to dehydration, guest volatilization, and host decomposition processes, respectively. The TGA plot for the 1:1 complex reveals two distinct peaks at ca. 250 and 270 °C which support the presence of multiple microenvironments, in agreement with the DSC results described above. The 2:1 complex is characterized by a broad range of thermal events between 240 and 290 °C which suggest a unique guest conformation that differs from the 1:1 complex. As illustrated by the DSC and TGA results, the

supported by the absence of a sharp melting transition for pure SPFO at 275 °C for both types of β-CD/SPFO preparations. This strongly suggests that a molecular-scale inclusion compound was formed with good phase purity. The appearance of exotherm events for the inclusion compounds provides further evidence for the formation of β-CD/SPFO inclusion complexes.42 In a previous study of β-CD/PFOA systems, the decomposition of the host−guest complexes was characterized by endotherm events17 that highlight some structural variability of the complexes reported herein. The observed difference between PFOA and SPFO reveals differences in the structure of the host−guest complexes due to the presence of Na+ versus H+ counterions since PFO exists as an anion at ambient pH conditions due to its relatively low pKa value (∼2.5).9 It is inferred from the DSC results that some Na+ ions may be associated with the β-CD/SPFO complex.20 The dehydration endotherms for the inclusion compounds in Figure 1 are distinct and reveal a variety of hydration environments, as compared with the β-CD hydrate. The 2:1 complex revealed a relatively sharp and more intense dehydration endotherm shifted to higher temperature (ca. 135 °C), as compared with the 1:1 complex which shows a broader signal with less intensity. The dehydration endotherms for the complexes suggest a distribution of bound and unbound water according to the availability of inclusion or extracavity microenvironments depending on the mode of inclusion and the conformation of the bound guest (cf. Scheme 2). The 2:1 complex appears to contain more cavity-bound water relative to the 1:1 complex according to the intensity and shift of the DSC signal.43 In a previous study of β-CD/PFOA complexes,17 the cavity- and interstitial-bound waters were shown as distinct peaks in the DSC thermograms (cf. Figure 1 ref 17). In contrast, such signatures are not deconvolved for the β-CD/ SPFO complexes reported herein and may indicate the coordination effect of the sodium counterion that results in a more ordered hydrate.44 The decomposition of the guest in the 1:1 complex (between 250 and 285 °C) revealed two distinct 22229


Article


Figure 4. PXRD patterns for (a) β-CD hydrate, (b) SPFO, and (c) the 1:1 and (d) 2:1 β-CD/SPFO inclusion compounds.

amount of unbound species. Red-shifts of ca. 8 and 6 cm−1 for the frequencies of the −CO bands for the 1:1 and 2:1 β-CD/ SPFO ICs with respect to the unbound guest, respectively, reveal differences in host−guest dipolar interactions, variable guest binding geometry, and conformation of the complexes, according to Figure 3. The region from 1100 to 1250 cm−1 in the FT-IR spectrum provides spectroscopic evidence of the conformational preferences of PFC chains in the solid phase. Vibrational bands (1−3) are labeled in Figure 3 and are assigned to the CF2 symmetric stretching mode (1 νs/CF2; 1250 cm−1), CC bending and CCC stretching modes (2 β/C−C, νs/C−C; 1215 cm−1), and CF2 asymmetric stretching (3 νas/CF2; 1150 cm−1).50 In particular, band 2 was shown to provide useful structural information regarding the conformational change of a PFC guest in the bound state with β-CD.46 In a previous report on β-CD/SPFO complexes in solution,7 a preferred all-trans conformation of the PFC chain was reported for the 2:1 complex. PFOA and other C8 PFC chains with different head groups have been reported to adopt helical (gauche) conformations in the pure solid compared to the zigzag (trans) conformation adopted by PFCs with chain lengths less than eight carbons (e.g., PFBA).51,52 A conformational change of the PFC chain may occur if its microenvironment is shifted via complex formation with β-CD.53,54 The attenuation of band 2 in Figure 3 for the 2:1 complex (d) relative to the 1:1 complex (c) and the unbound guest (b) may be attributed to a conformational change of SPFO between gauche and trans for this complex. According to the IR results reported herein, the native conformation of SPFO is not completely retained even in the 1:1 complex where attenuation of band 2 is observed. In previous reports16,17 of β-CD/PFOA complexes, the guest molecule was concluded to adopt a gauche conformation in the 1:1 host−guest complex, similar to the unbound guest in the solid phase. In contrast to the gauche

thermal stability of the 2:1 complex is similar or slightly greater than the 1:1 complex. The magnitude of the binding constants reported for the 1:1 β-CD/SPFO complex (K1:1) exceed those for the K2:1 values according to equilibrium studies in solution.25,28 The variable thermal stabilities of the 1:1 and 2:1 complexes reported herein may arise due to artifacts due to enthalpic contributions from 1:1, 2:1, and 1:2 β-CD/SPFO complexes. The thermal events for the TGA results are presented in the Supporting Information (Figure S1). The weight loss (%) due to the dehydration processes at ca. 60 °C was evaluated as ∼9% (β-CD), ∼ 3% (1:1 complex), and ∼6% (2:1 complex), in agreement with DSC results and the relative availability of inclusion sites. In general, the formation of the inclusion compounds tend to increase the thermal stability of the β-CD host and suggests the formation of stable β-CD/SPFO ICs.45 Differences in the thermal stability, enthalpy of dehydration, and the occurrence of multiple thermal transitions provide support for differences in the 1:1 and 2:1 β-CD/SPFO complexes due to structure. Further details on the molecular structures of the complexes are shown by the FT-IR and NMR spectroscopic results (vide infra). 3.2. FT-IR Spectroscopy. FT-IR spectroscopy is a useful technique for systems that contain functional groups such as the carbonyl and hydroxyl groups. For example, this technique was used to characterize guest conformational changes (i.e., gauche vs trans) of perfluoroalkyl chains46,47 and to study the structure of other cyclodextrin inclusion systems.48,49 The FTIR spectra of β-CD, SPFO, and the 1:1 and 2:1 host−guest complexes prepared by the modified dissolution method are shown in Figure 3. The relative intensities of the −OH (∼3400 cm−1), −CH (∼2900 cm−1), and −CO (∼1700 cm−1) vibrational bands are shown in Figure 3 for the 1:1 and 2:1 host−guest complexes. The trends in Figure 3 correlate with the relative mole ratios of the host/guest and the relative 22230


Article


Figure 5. 1H NMR spectra for (a) β-CD, (b) the 1:1 and (c) 2:1 β-CD/PFOA, and (d) the 1:1 and (e) 2:1 β-CD/SPFO inclusion compounds in D2O at 295 K.

CD and the β-CD/PFBA “cage-type” complexes reported elsewhere.18 β-CD has a strong tendency to form head-to-head dimer units with long-chain PFCs (C ≥ 8) that result from multiple H-bonding at the secondary rim of β-CD (cf. Scheme 1), where one or more guests can be accommodated.61 The broadened signals of the inclusion compounds for 2θ > 20° in Figure 4c and d indicate that the inclusion of SPFO within the host contributes to a reduction of the long-range order within the system. The PXRD pattern for the 2:1 complex (Figure 4d) has signatures at various 2θ values (14.5, 15.7, and 18.9°; denoted by asterisks in Figure 4) that are either not seen or become attenuated in the 1:1 complex (Figure 4c). The additional reflections for the 2:1 complex indicate that its structure differs uniquely from the 1:1 complex. By comparison with the results for the 2:1 complex, broader reflections for the 1:1 complex are observed at 2θ > 20° due to possible increased disorder arising from multiple host/guest topologies and the bound water according to the DSC/TGA results. Wenz26 described that the packing of cyclodextrin inclusion compounds depends on the relative dimensions of the guest and the host. In cases where the guest is partially or completely encapsulated by the host, variable guest conformations exist, in agreement with the IR results (cf. Figure 3). Variation of the guest conformation, different types of host−guest topology (i.e.,1:1 vs 2:1 complexes), and variable hydrate water contribute to structural order−disorder effects, in agreement with the DSC results.46 The structural ordering of complexes due to hydration factors was described in a review by Caira.62 3.4. 1H and 19F Solution-State NMR Spectroscopy. 3.4.1. 1H NMR in Solution. Numerous studies have been reported on the structure of host−guest inclusion systems in solution using complexation-induced 1H/19F NMR shifts (CIS).7,25,27,63 Comparative results for PFOA and its complexes

conformation adopted by the guest in the 1:1 complex, a preferred all-trans conformation of the guest was reported for the 2:1 complex. By contrast, the PFBA (C4) guest gauche conformation occurs in the 1:1 complex with β-CD.18 Differences in the conformation of the bound guests for the 1:1 β-CD/PFOA (gauche) complex vs 1:1 β-CD/SPFO (trans) complex provide further support that structural variation occurs due to the presence of H+ vs Na+ counterions, respectively. 3.3. Powder X-ray Diffraction. Powder X-ray diffraction (PXRD) is sensitive to long-range order or to the periodic structure of the host−guest framework. β-CD has diagnostic features in its PXRD patterns for all of its known crystal phases, where the 2θ at low angle indicates structural changes upon complex formation with β-CD.55 Binding of the guest can be deduced from changes in the PXRD patterns of the guest diffraction lines. In this case, no PXRD features of the unbound guest (Figure 4b) are supported by the XRD patterns for both complexes (Figure 4c,d). The results indicate that the longrange order of the unbound guest is lost upon complex formation, in agreement with inclusion of SPFO within β-CD. The XRD patterns for the host molecule (cf. Figure 4a) consist of prominent lines near 2θ values (9 and 12°), with other minor signatures at higher 2θ values that support “cagetype” crystalline structures.56−58 SPFO (b) has unique 2θ values at 7.5 and 11.3° with several minor peaks at higher 2θ. The PXRD results of the inclusion compounds in Figure 4c,d are characterized uniquely by several sharp reflections that provide further evidence for the formation of β-CD/SPFO inclusion compounds. The patterns of the 1:1 and 2:1 host/ guest complexes show major peaks at 11.5 and 17.5° which indicate the formation of head-to-head “channel-type” structures.59,60 These topologies are similar to those reported for the β-CD/PFOA complexes;17 however, they differ from native β22231


Article

The Journal of Physical Chemistry C Table 1. CIS Values (ppm) for the 19F Nuclei of β-CD/SPFO and β-CD/PFOA Complexes in Solution at 295 K 19

δfree (ppm)

F nuclei

Δδ (ppm)

δfree (ppm)

Δδ (ppm)

carbon no.

carbon ID

PFOA

1:1 CD/PFOA

2:1 CD/PFOA

SPFO

1:1 CD/SPFO

2:1 CD/SPFO

C8 C2 C4 C5 C6 C3 C7

CωF3 CαF2 Cγ F 2 Cδ F 2 Cχ F 2 CβF2 Cε F 2

80.90 −117.66 −121.94 −122.16 −122.88 −123.38 −126.19

+0.03 +0.14 +0.37 +0.36 +0.23 +0.61 +0.02

−0.09 +0.34 +0.32 +0.24 +0.50 +0.58 −0.09

−80.90 −117.65 −121.94 −122.16 −122.88 −123.38 −126.19

+0.04 +0.16 +0.41 +0.40 +0.26 +0.66 +0.04

−0.18 +0.43 +0.24 +0.11 +0.61 +0.43 −0.17

Figure 6. (A) 19F NMR expanded spectra for (a) PFOA and (b) the 1:1 and (c) 2:1 β-CD/PFOA inclusion compounds. (B) 19F NMR expanded spectra for (a) SPFO and (b) the 1:1 and (c) 2:1 β-CD/SPFO inclusion compounds in D2O at 295 K and pH 5. The resonance lines are assigned according to 2D-COSY and the accompanying structure in part a. 22232


Article

The Journal of Physical Chemistry C with β-CD are presented here and in the following sections, to assist in further understanding the structure of SPFO and its complexes with β-CD. Figure 5 shows a stack plot of the 1H NMR spectra in solution for β-CD and its inclusion complexes with PFOA (b, c) and SPFO (d, e) maintained at pH ∼ 5 and 295 K, where the residual HOD (∼4.70 ppm) signal was used as an internal standard. The assignment of the resonance lines for the 1H NMR spectra in Figure 5 agrees well with previous literature reports64,65 as illustrated by the structures in Scheme 1(a,b). The 1H/19F complexation-induced shift (CIS) values for the nuclei are reported as chemical shift difference (Δδ) as defined by eq 1. The 1H CIS results are listed in the Supporting Information (cf. Table S1), and the 19F CIS results are shown in Table 1. Note that negative Δδ values indicate that the nuclei of the bound host/guest system are shielded with respect to the free material, whereas positive Δδ values represent deshielding. The determination of the δ values for the partly overlapped signals in Figure 5 was confirmed from the known signal splitting patterns and the corresponding integrated coupling constants. Δδ = δcomplex − δfree

values are generally very small for linear PFC chains, particularly when fast rotational motion of the C−C bonds applies.69 The CεF2 group is adjacent to the CF3 group and is usually the most upfield signature in linear PFC chains and appears at ca. −126 ppm.7,68 The signal next to the CF3 resonance at ca. −117 ppm (cf. Figure 6) is a triplet and is assigned to the CαF2 owing to deshielding effect due to electron-withdrawing effects by the neighboring carboxyl group. The other CF2 groups were assigned based on the intensity of cross peaks, where four-bond couplings give the most intense peaks which implicitly presume rapid C−C bond rotations throughout the PFC chain. As this assumption may not be valid for the complexes, one has to accept the possibility of permutation between the assignments of δ and γ, as well as χ and β, when comparing the spectra of the free guest with the complexes (cf. Figure S2-b,c). The well-resolved 19F resonance lines in Figure 6a, 6b allow for more accurate quantitative analysis of the CIS values. 19F NMR CIS values in solution were used7,27,28 to elucidate the binding affinity and geometry of a series of PFC carboxylate guests in CD-based complexes since such guests can undergo significant environmental changes between the free and bound states.63,64 Therefore, changes in δ values of the host/guest nuclei can provide a measure of the degree of complex formation in host−guest systems, in accordance with the results in Table 1. In the context of host−guest chemistry, the firstprinciples described by Emsley et al.70 were applied to the study of host−guest complexes by Guo et al.7 along with Wilson and Verrall.27,28 The factors which influence the chemical shift changes of CD-based host−guest complexes in solution are not straightforward to deconvolve and can be classified into three types of interactions in the case of PFC guests: (i) the hydrophilic carboxylate headgroup must reside outside the cavity forming hydrogen bonds with the CD hydroxyl groups and water near the cavity resulting in reduced shielding of the CαF2 and CβF2 groups due to inductive effects, (ii) the displacement of polar water from the apolar cavity interior by the apolar guest upon complex formation will contribute to increased shielding because of the different polarizability of the cavity interior relative to the bulk aqueous environment, and (iii) conformational effects that lead to CIS values where a favored all-trans conformation results in greater deshielding due to increased distance between adjacent CF2 groups. Furthermore, shielding caused by conformational changes favoring an all-trans configuration results in a net shielding effect due to subsequent optimization of the host−guest van der Waals interactions. Palepu et al.19,20 observed positive shifts for most 19F resonances in an equimolar mixture of β-CD and SPFO with respect to pure SPFO. As the guest is included within the cavity of the host, it displaces water from the cavity as suggested by the DSC/TGA results. The effect of polarizability on chemical shift changes may be straightforward to interpret; however, other factors such as conformational effects of the guest and the variable polarizability of the β-CD macrocycle (annular hydroxyl region vs apolar cavity interior) provide an understanding but may also complicate the interpretation of CIS values. In the case of a PFC guest, the electron density around the 19F nuclei is less polarized in an aqueous environment since the interactions between the C−F dipoles are shielded due to the high dielectric constant of water when compared to environment in the β-CD cavity. As a result, the paramagnetic shielding contribution to 19F shielding will be correspondingly

(1)

In the case of the 1H NMR results (cf. Figure 5 and Table S1), the intracavity protons (i.e., H3, H5) experience the largest (negative) CIS values due to the pronounced changes in microenvironment of the host, which support the formation of host−guest inclusion compounds (cf. Schemes 1 and 2). In particular, the magnitude of the Δδ values for these intracavity protons is consistently larger in the 1:1 complexes as compared with the 2:1 host/guest complexes (cf. Table S1) due to increased steric effects in the former. These observations are consistent with the variable stoichiometry (1:1 vs 2:1) of the complex and configurations of the bound guest within the host−guest complexes. As the splitting patterns are preserved in the 1H spectra of the free host and complexes, conformational changes in the host must be relatively minor since the pucker units of the sugar rings are essentially unchanged.66 The preserved splitting patterns of the host, accompanied by minor CIS effects in the extra cavity protons (i.e., H2, H4, and H6; cf. Table S1), imply that the large CIS values for H3 and H5 nuclei relate to direct interactions between the host and the guest or the displacement of water from the cavity upon guest inclusion. In the case of the 1:1 complexes, the shielding effect is notably greater (cf. Figure 5b,d). The DSC/TGA, FTIR, and PXRD results (cf. Figures 1−4) jointly indicate that the 1:1 complexes contain the least amount of cavity-bound water. The greater displacement of water by the guest from the CD cavity for the 1:1 over the 2:1 complexes coincides with the larger 1H CIS values observed herein, as described by the partially hydrated 2:1 complex in Scheme 2.67 3.4.2. 19F NMR in Solution. The 19F NMR spectra of the complexes in solution formed between PFOA and SPFO with β-CD are shown in Figure 6a and 6b, respectively, where the CIS values are listed in Table 1. The assignments of the 19F NMR resonance lines for PFOA/SPFO in Figure 6 were achieved using 2D-COSY (see Supporting Information; Figure S2a−c). The results are in agreement with the results of Ribiero68 and the accompanying structure shown in Figure 6a. The 19F signal at ca. −81 ppm is assigned to the CF3 (C8) group based on structural considerations.68 Expansion of this spectral region reveals a triplet of triplets with 19F−19F scalar couplings of ∼10 and ∼2.3 Hz. These values are typically considered to be 4JFF and 5JFF couplings, respectively, as 3JFF 22233


Article


Scheme 3. Proposed Inclusion Geometry of (a-i−ii) PFOA and (b) SPFO in the 2:1 β-CD/Guest Complex and (c-i−ii) Proposed Inclusion Geometry of the 1:1 β-CD/Guest Complexa

a

Note the CD torus and SPFO/PFOA structures are not drawn to scale.

(C6), δ (C5), and γ (C4) reveal an increase for the 1:1 complexes but are relatively reduced for the 2:1 complexes. At C3 and C2, the CIS behavior varies where the chemical shifts of β and α increase significantly for the 1:1 stoichiometry and increase somewhat further for the 2:1 complex. In the case of the inclusion complexes, the chemical shifts for individual 19F nuclei are affected by one or more of the three types of interactions noted above. A major driving force for complex formation between a PFC guest and a CD host is the minimization of the contact between solvent and the apolar guest, especially in aqueous media. The 1:1 host−guest complex has several possible binding configurations (cf. Scheme 3), where the terminus exposed to the bulk water will resemble a portion of the free guest, especially for 19F nuclei that extend outside the annular hydroxyl region of β-CD. The observed small Δδ values for the CF3 (C-8) group, and by extension the CεF2 (C-7) group for the 1:1 complexes of β-CD with PFOA (+0.03 ppm) and SPFO (+0.04 ppm), suggest that the apolar end of the PFC chain is not deeply included in the cavity. The influence of the conformational change is relatively small for the CF3 group since it would reside outside or near the rim of the CD cavity, and the effects of both interactions (ii) and (iii) contribute to negligible CIS value (Δδ ≈ 0). The large deshielding CIS effects for β-carbon observed for the 1:1 complex suggest that the carboxyl group extends into the aqueous environment to allow the apolar CF3 group to minimize its contact with water. Therefore, the shielding effect of the apolar CD cavity interior with the CβF2 group is minimized. The Cβ F 2 group experiences an increased deshielding effect in the annular region which is less polarizable because of the −OH and −CH groups. The reduced deshielding for the α-carbon observed for the 1:1 complex provides further support for a host/guest topology where the CF3 resides out or near the annular region; the carboxylate headgroup extends into the aqueous environment; and the CαF2 experiences attenuation of type (i) interactions. The other CF2 groups experience the main effect of type (ii) and (iii) interactions due to conformational changes.

increased upon complexation due to enhanced interactions between C−F dipoles. Paramagnetic shielding terms tend to increase the strength of the external magnetic field, consequently resulting in an increased chemical shift for most of the fluorine nuclei of the complexes with respect to the free guest. Conformational changes of the guest that occur upon complexation contribute to changes in the 19F chemical shifts, as mentioned above. The γ-gauche effect is common in alkyl and perfluoroalkyl chains longer than three carbons and is known to affect changes in 19F chemical shifts. The γ-gauche effect is a net shielding effect that results from a carbon in a gauche position to a vicinal 1H or 19F nucleus. In the limit of fast motion, the γ-gauche effect is averaged over all rotational isomers. In the case of chains with cis−cis helical structure, the γ-gauche shielding effect occurs for only half the 19F nuclei. By comparison, for PFC chains with an all-trans conformation the γ-gauche effect occurs for all 19F nuclei. Therefore, conformational changes favoring trans configurations lead to net shielding. PFC chains generally adopt a cis−cis helical configuration when dispersed in aqueous environments,53,71 and this can be understood on the basis of minimization of the cost of Gibbs energy of hydration due to the lipophilic nature of PFC compounds, especially PFCs with longer alkyl chains. Therefore, the guest in its unbound state will experience a combination of the shielding effect due to the bulk water environment along with a partial γ-gauche shielding effect as compared with its bound form as a host−guest complex. As noted in Figure 6 and Table 1, the 19F chemical shifts for both free guests (PFOA; 6a, and SPFO; 6b) experience the same CIS pattern upon complexation with β-CD in both types of host/guest stoichiometry. Note that the CF3 (C-8) shift increases (deshielding) for the 1:1 complexes by 0.03 ppm (PFOA) and 0.04 ppm (SPFO). For 2:1 complexes, a measurable decrease (shielding) occurs, 0.09 ppm (PFOA) and 0.18 ppm (SPFO), as listed in Table 1. The identical trend persists for fluorine (CεF2) on the next carbon along the chain (C-7). Similarly this trend continues but is less pronounced further along the PFC chain. The chemical shift values of χ 22234


Article


Figure 7. 1H → 13C CP spectra for (a) β-CD and (b) the 1:1 β-CD/SPFO and (c) 1:1 β-CD/PFOA complexes. The 13C DP spectrum for (d) the 1:1 β-CD/SPFO and (e) 1:1 β-CD/PFOA complexes. All spectra were acquired at MAS (20 kHz) and 295 K.

In the case of the 2:1 complexes, the α-carbons experience additional deshielding, while increased shielding is observed for the CβF2 groups. Furthermore, the central CF2 groups experience greater shielding for the 2:1 versus the 1:1 complexes. The adoption of an all-trans conformation and rearrangement of the position of the PFC chain occurs with the formation of a 2:1 channel-type host−guest complex. The additional deshielding/shielding for the α-/β-carbons can be explained in terms of the position of the carboxylate headgroup near the annular region of the cavity. As the second CD stacks onto the 1:1 complex, the carboxylate headgroup is more deeply included within the cavity where the deshielding of the CαF2 occurs due to type (i) interactions. The same effect extends to the CβF2 group of the 2:1 complex but to a lesser extent because of the deeper inclusion within the CD cavity. The increased shielding for the central CF2 groups in the 2:1 complexes indicates a significant increase in the degree of trans configurations for these complexes as suggested by FT-IR results. 2:1 complexes yield optimal 19F−19F distances due to greater van der Waals interactions as shown by decreased chemical shifts, according to type (ii) and (iii) interactions. The shielding of CIS values for the 2:1 β-CD/SPFO complex is 2fold greater as compared to the 2:1 β-CD/PFOA complex. This indicates that the SPFO guest experiences a deeper inclusion within the CD cavity compared to the PFOA guest. It can be inferred from the above results that the sodium counterion is associated with the β-CD/SPFO complex and promotes variable binding of the PFC guest when compared with the hydronium ion for the β-CD/PFOA system. 3.5. 13C and 19F Solid-State NMR Spectroscopy. Solidstate NMR spectroscopy is a versatile technique that can be

used to study the structure and dynamic properties of host− guest systems.46,47,72−74 In particular, multinuclear NMR methods can provide unequivocal evidence for the inclusion of a guest within the host.74 Additionally, relaxation techniques in the solid state may provide molecular details about the dynamic processes of guest motions within the host.75 3.5.1. 1H → 13C CP/MAS and DP/MAS NMR. 13C MAS NMR spectroscopy with DP (direct polarization) and CP (cross-polarization) techniques were studied to provide evidence for the formation of β-CD/SPFO inclusion compounds. The 1H → 13C CP and 13C DP results for β-CD (a) and 1:1 β-CD/SPFO (b and d) and 1:1 β-CD/PFOA (c and e) complexes at 20 kHz MAS conditions are shown in Figure 7. The 13C resonance lines in Figure 7 were assigned according to previous reports58 and are illustrated in Scheme 1(a,b). The CP/MAS NMR lines for the pure host in Figure 7a cannot readily be resolved into each respective glucopyranosyl unit due to spectral overlap. The spectral lineshapes indicate the presence of several phases where at least one component is highly ordered with narrow signals and line widths of ∼50 Hz. The other phase is disordered with one or more components that show a broad line of ∼300−500 Hz. The presence of several CD phases is consistent with its water content (10% w/ w; see section 2.1), where fully hydrated samples have greater resolution, as compared with anhydrous samples with poorer resolution.62 This indicates that there is an inhomogeneous distribution of hydration states of β-CD throughout the sample, ranging anywhere from 0 to 14 water molecules. The lineshapes in the 13C NMR spectra of the inclusion complexes (cf. Figure 7b−e) display broader features relative to 22235


Article


Figure 8. 19F → 13C CP results for (a) SPFO, (b) 1:1 β-CD/SPFO, (c) PFOA, and (d) 1:1 β-CD/PFOA inclusion compounds under MAS 20 kHz at 295 K.

significantly longer T1. Therefore, the host in the PFOA and SPFO complexes provide different local dynamic environments, respectively. The DP spectra of the complexes show markedly different lineshapes for the 13C guest signals, where the PFOA signals are much sharper (∼80−120 Hz wide; Figure 7e) than the signals for SPFO (∼250 Hz wide; Figure 7d). This suggests that the SPFO complex is characterized by less extensive dynamic/ structural disorder relative to the PFOA complex, in agreement with the PXRD results. Furthermore, the differences in binding geometry of the SPFO guest relative to its PFOA counterpart are influenced by the nature of the counterion (Na+ vs H+). The variable structure for these systems according to the 1 H/19F NMR CIS results in solution may contribute to differences in guest motional dynamics. The structure of the βCD/SPFO complex differs from that of the β-CD/PFOA complex, where the dynamics of the guest in the former are “anchored” by the sodium counterion (compare Schemes 3a and b) due to its size and electrostatic interaction with the carboxylate headgroup of the guest. 3.5.2. 19F → 13C CP/MAS NMR. The 19F → 13C CP NMR spectra in Figure 8 provide unequivocal evidence for the inclusion of the guest within the host as they indicate that direct dipolar interactions occur between the host and guest. The dipolar coupling, without consideration of the scaling due to molecular motion and MAS, between 13C and 19F over a distance of one C−F bond (∼1.36 Å) is estimated at 12 and ∼1 kHz for distances greater than 3 Å. The 5 ms optimal contact time (tcp) applied for this experiment suggests that the scaling effect is very pronounced; hence, couplings over long distances would not give rise to significant CP transfer. In this case, optimal CP transfer over 3 Å would be expected near tcp = 60 ms, a point at which the majority of the signal would be lost to relaxation. The 19F → 13C CP spectra for the 1:1 complexes

the sharp features of the pure host (Figure 7a); however, the resonance lines of the inclusion complexes are generally more narrow compared to the broad contributions seen in the spectra of the pure host, excepting the presence of some sharp features for the host. This suggests that the hydration states of the complexes have reduced hydrate content on average or greater structural order of the complexed host due to disorder effects of the guest. In the DP/MAS spectra of the 1:1 complexes (d, e), the signals from both the host (ca. 60−05 ppm) and guest (ca. 110−125 ppm) molecules occur at different chemical shift regions compared with unbound SPFO and β-CD (cf. Figures 7 and 8a), respectively. The emergence of unique NMR signatures provides further support that host−guest complexes are formed. The 13C CP (b, c) and DP (d, e) spectra in Figure 7 indicate large variations in the line-shapes, -widths, and -intensities between the SPFO (b, d) and PFOA (c, e) complexes, respectively. Upon closer examination, the lineshapes in the CP MAS spectrum of the 1:1 PFOA complex (c) can be deconvolved into two comparable contributions, 100 and 350 Hz wide. By contrast, the CP spectrum of the 1:1 SPFO complex (b) has two contributions, 150 and 350 Hz wide, where the latter is dominant. The low signal-to-noise ratio in the DP spectra (d) and (e) does not allow for reliable deconvolution of the host signal. In both experiments, either differences between signal saturation (for DP) or 1H T1ρ (for CP) between the complexes can cause variable line intensities, and this may account for the observed dynamic disorder between the host structures. In particular, the relative intensity of the C6 resonance line of the host for the SPFO (b, d) complex is much greater compared to that for the PFOA complex (c, e) and is most likely due to saturation of the remaining carbon signals. This implies that while most of the carbon T1 values in the PFOA complex (∼1.0 s) are significantly shorter than in the SPFO counterparts C6 has a 22236


Article


Figure 9. 19F DP NMR spectra of (a) unbound PFOA, (b) unbound SPFO, and (c) the 1:1 and (d) 2:1 β-CD/SPFO complexes.

though it is not a carbon associated with the cavity. The deshielding of the 13C CIS values for the CαF2 group are about 4-fold greater in the SPFO (+1.62 ppm) complex with β-CD compared to the PFOA (+0.47 ppm) analogue (cf. Table S2). The effect of the Na+ counterion for the β-CD/SPFO complex would decrease the effective negative charge on the carboxylate headgroup by ion pairing that results in an increased inductive effect on the α-carbon. The discerning Na+ counterion effect on the structure of the β-CD/SPFO complexes was revealed by the DSC/TGA and the solution NMR. The trends are consistent with the solid MAS/NMR results reported herein. 3.5.3. 19F DP/MAS NMR at Ambient Temperature. The 19F DP/MAS NMR spectrum of neat SPFO, neat PFOA, and their 1:1 complexes with β-CD prepared using the modified dissolution method are presented in Figure 9. A comparison of the spectra of pure PFOA (cf. Figure 9a) with pure SPFO (b) reveals broad 19F signals for both the methyl (CF3) and methylene (CF2) regions of SPFO (b) in direct contrast to those of PFOA (a). The results indicate that there must be significant differences in the rate of C−C bond rotations along the PFC chains of the unbound guests, consistent with the large difference in their melting temperatures in Figure 1 for SPFO (∼275 °C) and PFOA (∼45 °C). In an independent study, the 19 F NMR spectra of the β-CD/PFOA complexes were 10-fold broader relative to pure PFOA (cf. Figure 4 ref 17). For the SPFO complexes, the line widths are slightly reduced upon complexation. The differences in line widths for the SPFO guest in the complexes strongly suggest that its structure and dynamic behavior are markedly different from the pure unbound compound, in accordance with differences in the complex stability due to enthalpy effects that result upon inclusion complex formation (cf. Figures 1 and 2). The CF2 resonance for the 2:1 β-CD/SPFO complex in Figure 9(d) displays relatively sharp (500 Hz) features for the terminal CεF2 (500 Hz, ca. −127.2 ppm) along with the

revealed host resonances of ca. 60 ppm (C6), 75 ppm (C2, C3, and C5), and 85 ppm (C4) (cf. Figure 8 and Scheme 1a, b). The most intense resonance at ca. 75 ppm arises from CP transfer to the intracavity 13C nuclei. Such CP transfer suggests that the PFC guest is in close spatial proximity to the β-CD cavity which is only possible if the guest is included within the host cavity. CP transfer occurs for at least two carbonyl signals for the 1:1 β-CD/SPFO complex at ca. 165 and 170 ppm (cf. Figure 8b, shown by arrows). This result further supports the presence of multiple configurations of the guest in the 1:1 complexes as indicated by the DSC/TGA and solution NMR results. The absence of similar carbonyl signal(s) in the 1:1 β-CD/PFOA complex is the result of relatively rapid guest dynamics that reduce the 19F to 13C dipolar couplings, thereby reducing the efficiency of CP transfer to the host 13C nuclei. The results are consistent with the structure of the β-CD/SPFO relative to that of the β-CD/PFOA complex as described in the solution and solid NMR results, where the presence of different counterions affect the structure and dynamics of the bound guest to varying degrees. Evidence for host−guest binding was further sought in the 13 C CIS values from the 19F → 13C CP and 13C DP spectra to supplement the NMR CIS results in solution. The 13C δ values for the respective 1:1 complexes of SPFO and PFOA with βCD were extracted by deconvolution analysis of 13C NMR lineshapes. The Δδ values (listed in brackets) are presented in Table S2. The Δδ values for the CαF2 group in the 1:1 β-CD/ SPFO complex (+1.62 ppm) and the 1:1 β-CD/PFOA complex (+0.47 ppm) indicate differences in the dynamics and binding geometry of the different guest systems, in agreement with the chemical shift difference for the C3/C5 host signals for the two complexes (∼1.12 ppm). Furthermore, the role of the H+ and Na+ counterions must be considered, which is further implicated by the Δδ for the host signals of the complexes, where C4 is increased by 0.89 ppm in the SPFO complex even 22237


Article

The Journal of Physical Chemistry C Table 2. Structural Properties of PFOA, SPFO, and PFBA and Their Complexes with β-CD77 guest molecule

PFOA

physical state solid melting point 40−50 host:guest 1:1 and 2:1 stoichiometry native gauche conformation conformation in gauche (1:1) and trans (2:1) bound state host−guest variable topology packing channel-type structure free coiling and uncoiling motional dynamics of unbound guest motional significant CF3 rotation and dynamics of axial motions in both the bound guest 1:1 and 2:1 complexes

SPFO

PFBA

solid 277−280 1:1 and 2:1

liquid n/a mainly 1:1

gauche

trans

gauche (1:1) and trans (2:1)

gauche

deeper penetration of the guest

variable

channel-type

cage-type

restricted coiling and uncoiling

n/a

reduced CF3 rotations and facilitated C−F bonds rotation in the 2:1 complex

extensive CF3 rotation and appreciable axial motions of C−F bonds

signature for several CF2 resonances (CmF2; 1800 Hz, ca. −120 ppm), while the CαF2 signal (1200 Hz, ca. −117.2 ppm) appears to be broader. The spectrum of the 1:1 β-CD/SPFO sample (c) is similar apart from small CIS values and increased line widths in the CαF2 and CεF2 signals. The differences in line widths between the 19F nuclei of the 2:1 and 1:1 β-CD/SPFO complexes are attributed to conformational and dynamic differences of the guest as illustrated by the 1H NMR results in solution and the solid NMR results (13C DP, 1H → 13C CP, 19 F → 13C CP). The PFOA chain in its unbound form can be concluded to involve free dynamics of coiling and uncoiling, whereas such dynamics are relatively restricted in the sodiumcontaining SPFO guest. By comparing the line widths of the spectrum of SPFO in its unbound form (Figure 9b) in the 1:1 complex with β-CD (c), it is noteworthy that the dynamics of SPFO in the 1:1 complex (Figure 9c) slightly resembles that of the native guest since part of the guest is exposed. The full encapsulation for the guest in the 2:1 β-CD/SPFO complex along with sodium counterion effects is expected to modulate the distinct dynamics of the guest for this complex. It can be inferred from the 19F MAS solid NMR results that the CF2 groups of the guest chain may experience more free and efficient rotations in the 2:1 relative to the 1:1 complex. The geometry of the guest in the 2:1 host−guest complex may promote the formation of a pseudo-rotaxane topology where the sodium counterion and the CF3 group may act as anchors and the central chain experiences efficient C−F bond rotations. This is consistent with the decreased line width of the central CF2 nuclei in Figure 9. The dynamics of PFOA17 and PFBA18 in the unbound and bound states with β-CD were described elsewhere and are summarized in Table 2 along with the physicochemical properties of the different guest (i.e., PFOA, SPFO, and PFBA) systems. The presence of spinning sidebands (denoted by asterisks in Figure 9) at a lower MAS speed (20 kHz in this case) complicates the direct comparison of the lineshapes for the spectra of the 1:1 and 2:1 complexes due to signal overlap. Consequently, deconvolution analyses of the CF3 signals were performed using spectra obtained at greater MAS frequency (25 kHz) to avoid signal overlap. A comparison of the deconvolved CF3 signals with simulation results provides further insight into the time scales of the dynamics of the SPFO guest. As for the CF2 groups, spectral overlap of the

NMR lines from the six CF2 groups between −110 and −130 ppm makes it very difficult to discern each signal separately. Therefore, simulation of the spectra is not practical in this case. Unlike 13C nuclei, 19F nuclei are highly abundant with a large magnetic moment, and their spectra are often complex due to strong 19F−19F homonuclear dipolar couplings. Although the isotropic resonance frequencies of the CF3 are the same, the orientations of the shielding tensors are not and are also not coincident with 19F−19F internuclear vectors. As a result, the effect of rotational resonance can lead to complex splitting patterns (cf. Figure 10) despite the fast MAS rate (25 kHz) employed. Such complex splitting patterns disappear when fast

Figure 10. Simulated CF3 resonance as a function of selected dipolar couplings. 22238


Article

The Journal of Physical Chemistry C motion effectively scales the dipolar coupling down to zero or averages shielding tensors to be essentially coincident. The presence of splitting patterns in host/guest systems indicates that local motions are on a time-scale of the dipolar coupling and/or the chemical shielding anisotropy. The former effect ranges from 10 to 15 kHz with correlation times of 66 to 100 μs, while the latter ranges from 24 to 47 kHz with correlation times of 2 to 4 μs. Hence, the degree of splitting is a measure of the homonuclear dipolar coupling. By comparing the inferred dipolar coupling from the splitting to the theoretical maximum value under static conditions, a resulting scaling factor can be used to estimate the correlation time associated with the local motion which contributes to the scaling effect. In Figure 10, the CF3 lineshapes at MAS 25 kHz were simulated using SIMPSON software as a three-spin system with the following assumptions and parameters: principle axis of the 19 F shielding tensors oriented at a 109° valence angle with respect to the C−C bond, a dihedral angle spaced at 120° intervals, an isotropic shift of −84.75 ppm, a chemical shielding anisotropy of 80 ppm, and an asymmetry parameter of 0.5. The shielding tensor parameters were determined from the sideband splitting pattern. Simulations with homonuclear dipolar coupling varied from 0.1 to 10 kHz as shown in Figure 10. Stronger coupling (e.g., 10 kHz) in Figure 10 corresponds to slow dynamics (τ > 100 μs), whereas weaker coupling (e.g., 100 Hz) corresponds to fast dynamics (τ < 10 μs). For strong dipolar couplings ranging from 3 to 10 kHz, the CF3 resonance is clearly split into several (16 to 41) lines. For weaker couplings ranging from 2 to 0.1 kHz, the signal collapses into what appears to be a single resonance that decreases in line width and eventually converges to the natural line width, set to 200 Hz, where the effective coupling is essentially zero. The lineshapes at 10, 7.5, 5, 3, 2, 1, 0.5, and 0.1 kHz have envelopes covering line widths of approximately 1500, 1250, 750, 575, 400, 325, 240, and 205 Hz, respectively. On the basis of these line widths, scaling factors of 1.00, 0.81, 0.42, 0.29, 0.15, 0.096, 0.031, and 0.004 are predicted, from which exchange rates, k, can be estimated. The exchange rates are expressed in multiples of the static homonuclear coupling Do (k = nDo; Do = 10 kHz), where n = 1.00, 1.27, 1.37, 1.81, 3.36, 16.7, and 125. Do is approximated on the basis of a C−F bond length of 1.36 Å and a valence angle of 109°. The deconvolved CF3 signals for neat SPFO and the 1:1 and 2:1 β-CD/SPFO complexes are presented in Figure 11. The CF3 signal of SPFO appears to be a single line at −83.8 ppm; however, it can be deconvolved into a set of lines, covering a range of 400 Hz. Hence the scaling factor is approximately 0.15 with an effective dipolar coupling in the range of 2 kHz and an associated k value of ∼34 000 s−1. The CF3 signal of the 1:1 complex is composed of at least three sets of lines centered at −84.0, −82.4, and −81.6 ppm, covering 950, 750, and 400 Hz, which have corresponding scaling factors 0.58, 0.42, and 0.15 with k values of ∼11 500, 14 000, and 34 000 s−1, respectively. The 2:1 complex also contains three sets of lines centered at −84, −82.5, and −81.5 ppm, covering a range of 400 to 600 Hz wide with corresponding rate constants of 34 000 to 17 200 s−1. These results indicate that the CF3 group experiences faster dynamics in the 2:1 complex relative to the 1:1 complex. While the distinct guest dynamics for the 2:1 vs 1:1 complexes may be related to the differences in the structures of the two complexes as described in the NMR results, contributions from other mixtures (1:2 and 1:1 complexes) as mentioned above in the

Figure 11. Deconvolution of the CF3 line shape for (a) SPFO and (b) the 1:1 and (c) 2:1 β-CD/SPFO complexes.

DSC/TGA results cannot be ignored and may account for the additional CF3 components observed. The deconvolution parameters for SPFO and its complexes are presented in the Supporting Information (Table S3). 3.6. Solid-State 19F Relaxation Studies at Variable Temperature. The 19F T1 and T2 relaxation times as a function of temperature (0−70 °C) for the unbound SPFO and its inclusion compounds with β-CD are depicted in Figure 12. It should be noted here that the relaxation parameters were measured for the CF3 (ca. −83 ppm) and the main methylene (CmF2) groups at ca. −125 ppm. In general, relaxation of spin1/2 nuclei (e.g., 19F) is driven by random fluctuations in local fields resulting from chemical shift anisotropy (CSA) and proximal dipole−dipole (DD) coupling interactions modulated by local motion.35,76 The nature and rate of such motion affect T1 and T2 relaxation times differently.76 Both T1 and T2 decrease monotonically with increasing correlation time; however, when the rate of motion is comparable to the Larmor frequency, T1 passes through a minimum, while T2 continues to decrease. As the correlation time increases past the T1 minimum, T1 and T2 diverge. The ratio of T1/T2 in the slow motion regime is therefore a good indicator of the correlation time of local motion (cf. Figure 13). The T2 values (ms) in Figure 12 are generally shorter than the T1 values (s) by over 4 orders of magnitude indicating that the slow motion regime applies in all cases. The T1 values for pure SPFO (Figure 12a) are consistently much larger than in the complexes, and correspondingly the T2 values (Figure 12b) 22239


Article


are consistently much smaller. Hence the local motion in pure SPFO is generally much slower (cf. Table 2). This is seen quite dramatically when comparing the T1/T2 ratios for pure SPFO which are an order of magnitude greater than in the complexes, as seen in Figure 13. This indicates that the pure SPFO must be densely packed, imposing significant steric constraints on chain motion. Furthermore, the T1/T2 ratios for the CF3 signals are about four times lower than the corresponding values for the CF2 signals indicating a higher degree of mobility for the CF3 signals, presumably due to rapid C−C bond rotation on the chain end. In general, the overall rotational dynamics of the pure SPFO guest are expected to be slower than in the complexed state. Turning to the relaxation results for the complexes in Figure 12a, the low-temperature T1 values differ significantly. In particular, those of the CF3 moieties are nearly twice as long in the 1:1 complex compared to the 2:1 complex. A similar disparity is seen for the CF2 groups, yet it is less dramatic. In all cases the T1 values increase with increasing temperature meaning that local motion is activated; however, the degree of activation is much greater for the 2:1 complex. This means that, initially, the guest is more constrained in the 2:1 complex than in the 1:1 complex at low temperature; however, at higher temperature the two complexes appear to have comparable motion. These trends are difficult to appreciate using the T2 data; hence, one must turn to the T1/T2 results in Figure 13. The T1/T2 results indicate that the CF3 group in the 1:1 complex experiences more constraint than in the 2:1 complex, which becomes less pronounced with increasing temperature. The difference in CF2 is more dramatic, where chain motion is much more impeded in the 1:1 complex than in the 2:1 complex, and even though the difference does diminish with increasing temperature, it remains significant. The ease with which the chain motion is achieved in the 2:1 host−guest complex is consistent with the possible formation of a pseudo-rotaxane topology, as described above. Furthermore, it is worthwhile to note that the T1/T2 ratios in the 2:1 complex for the CF2 and CF3 groups track each other closely and consistently remain very close at high temperatures.

Figure 12. 19F T1 (a) and T2 (b) values for the CF3 and main CF2 groups for SPFO, the 1:1, and 2:1 β-CD/SPFO complexes under MAS 20 kHz and variable temperature.

4. CONCLUSIONS A detailed study of the solid inclusion complexes of β-CD and SPFO was conducted to understand the structure of the guest and the role of the counterion in the binding interactions by comparison of the sodium- versus hydrogen-containing guests and their complexes with β-CD. The 1:1 and 2:1 β-CD/SPFO complexes were prepared using a modified dissolution method which were further characterized using NMR and FT-IR spectroscopy, thermal analysis (DSC/TGA), and powder X-ray diffraction. 19F/1H/13C CIS values in solution and in the solid state were used to provide structural evidence for the formation of β-CD/SPFO inclusion compounds. A greater understanding of the geometry of the SPFO guest in the 1:1 versus 2:1 CD/ SPFO complexes was afforded by monitoring the 19F chemical shift changes as affected by the amount of proximal hydrate content about the guest, conformational preference (paramagnetic and γ-gauche effects) of the guest, and proximity of the guest with the host or bulk water (neighboring group effect). The 1H/19F CIS values in solution revealed that SPFO is more deeply included within the β-CD cavity, relative to the PFOA guest, in accordance with the counterion effects on the binding with β-CD. The DSC/TGA results revealed differing hydration states and variable thermal stabilities of the guest for

Figure 13. 19F T1/T2 ratios for the CF3 and main CF2 groups for the 1:1 and 2:1 β-CD/SPFO complexes under MAS 20 kHz at variable temperature.

22240


Article

The Journal of Physical Chemistry C the 1:1 and 2:1 β-CD/SPFO complexes. The DSC decomposition exotherms for the β-CD/SPFO complexes are distinctive from the endotherms reported for the β-CD/PFOA complexes. The DSC results provide further support that the sodium counterion is associated with the β-CD/SPFO complex and enhances the stability of such complexes relative to the βCD/PFOA system. Counterion effects on host−guest structure are further supported by PXRD results where SPFO is bound within a head-to-head channel type structure with β-CD with variable and significant loss of long-range order for the 2:1 versus the 1:1 complexes. Structural disorder may result from variable amounts of bound water due to counterion effects. The 13C/19F MAS NMR results in the solid state reveal that the CF2 groups of the PFC chain experience fast rotations in the 2:1 β-CD/ SPFO complexes and coincide with the possible formation of a pseudo-rotaxane system. Such findings are supported by the spin−lattice and spin−spin relaxation results in the solid state where chain motional dynamics were found to be attenuated more in the 1:1 complex relative to the 2:1 β-CD/SPFO complex as summarized in Table 2.

■

(5) Hoffmann, H.; Würtz, J. Unusual Phenomena in Perfluorosurfactants. J. Mol. Liq. 1997, 72, 191−230. (6) Reth, M.; Berger, U.; Broman, D.; Cousins, I. T.; Nilsson, E. D.; McLachlan, M. S. Water-to-Air Transfer of Perfluorinated Carboxylates and Sulfonates in a Sea Spray Simulator. Environ. Chem. 2011, 8, 381−388. (7) Guo, W.; Fung, B. M.; Christian, S. D. NMR Study of Cyclodextrin Inclusion of Fluorocarbon Surfactants in Solution. Langmuir 1992, 8, 446−451. (8) Onori, G.; D'Angelo, M.; Fisica, D.; Perugia, U.; Pascoli, V. A. Study of Micelle Formation in Aqueous Sodium n-Octanoate Solutions. Prog. Colloid Polym. Sci. 1994, 97, 154−157. (9) Rayne, S.; Forest, K. Theoretical Studies on the pKa Values of Perfluoroalkyl Carboxylic Acids. J. Mol. Struct.: THEOCHEM 2010, 949, 60−69. (10) Becker, A. M. Perfluorooctanoic Acid (PFOA) and Perfluorooctane Sulfonate (PFOS) in an Aquatic Ecosystem - Distribution and Fate; University of Bayreuth: Germany, 2008. (11) Washburn, S. T.; Bingman, T. S.; Braithwaite, S. K.; Buck, R. C.; Buxton, L. W.; Clewell, H. J.; Haroun, L. a; Kester, J. E.; Rickard, R. W.; Shipp, A. M. Exposure Assessment and Risk Characterization for Perfluorooctanoate in Selected Consumer Articles. Environ. Sci. Technol. 2005, 39, 3904−3910. (12) Butenhoff, J.; Costa, G.; Elcombe, C.; Farrar, D.; Hansen, K.; Iwai, H.; Jung, R.; Kennedy, G.; Lieder, P.; Olsen, G.; et al. Toxicity of Ammonium Perfluorooctanoate in Male Cynomolgus Monkeys after Oral Dosing for 6 Months. Toxicol. Sci. 2002, 69, 244−257. (13) Kitamura, S.; Fujimura, T.; Kohda, S. Interaction between Surface Active Drug (FK906:Rennin Inhibitor) and Cyclodextrins in Aqueous Solution. J. Pharm. Sci. 1999, 88, 327−330. (14) Nicolle, G. M.; Merbach, A. E. Destruction of Perfluoroalkyl Surfactant Aggregates by β-Cyclodextrin. Chem. Commun. 2004, 1, 854−855. (15) Szejtli, J.; Osa, T. Comprehensive Supramolecular Chemistry; Cyclodextrins; Pergamon: Oxford, 1996; Vol. 3. (16) Karoyo, A. H.; Borisov, A. S.; Wilson, L. D.; Hazendonk, P. Formation of Host-Guest Complexes of β-Cyclodextrin and Perfluorooctanoic Acid. J. Phys. Chem. B 2011, 115, 9511−9527. (17) Karoyo, A. H.; Sidhu, P.; Wilson, L. D.; Hazendonk, P. Characterization and Dynamic Properties for the Solid Inclusion Complexes of β-Cyclodextrin and Perfluorooctanoic Acid. J. Phys. Chem. B 2013, 117, 8269−8282. (18) Karoyo, A. H.; Sidhu, P.; Wilson, L. D.; Hazendonk, P.; Karoyo, A. H.; Sidhu, P.; Wilson, L. D.; Hazendonk, P. Characterization and Dynamic Properties for the Solid Inclusion Complexes of BCyclodextrin and Perfluorobutyric Acid. J. Phys. Chem. C 2014, 118, 15460−15473. (19) Palepu, R.; Richardson, J. E.; Reinsborough, V. C. Binding Constants of β-Cyclodextrin/Surfactant Inclusion by Conductivity Measurements. Langmuir 1989, 5, 218−221. (20) Palepu, R.; Reinsborough, V. C. Solution Inclusion Complexes of Cyclodextrins with Sodium Perfluorooctanoate. Can. J. Chem. 1989, 67, 1550−1553. (21) Jobe, D. J.; Verrall, R. E.; Reinsborough, V. C. Ultrasonic Absorption Studies in Aqueous Solutions of Modified β-Cyclodextrins. Can. J. Chem. 1990, 68, 2131−2136. (22) Jobe, D. J.; Verrall, R. E.; Junquera, E.; Aicart, E. Effects of βCyclodextrin/Surfactant Complex Formation on the Surfactant Monomer-Micelle Exchange Rate in Aqueous Solutions of Sodium Perfluorooctanoate and B-Cyclodextrin. J. Phys. Chem. 1994, 98, 10814−10818. (23) Wilson, L. D.; Siddall, S. R.; Verrall, R. E. A Spectral Displacement Study of the Binding Constants of CyclodextrinHydrocarbon and Cyclodextrin-Fluorocarbon Surfactant Inclusion Complexes. Can. J. Chem. 1997, 75, 927−933. (24) Junquera, E.; Tardajos, G.; Aicart, E. Effect of the Presence of βCyclodextrin on the Micellization Process of Sodium Dodecyl or Sodium Perfluoroctanoate in Water. Langmuir 1993, 9, 1213−1219.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcc.5b07302. Thermogram weight percent loss; 2D 19F−19F COSY NMR of SPFO and its complexes in D2O; 1H CIS values for CD and complexes with PFCs in solution at 295 K; 13 C CIS values for CD/PFC complexes in the solid stated at 295 K; deconvolution parameters for the CF3 lineshape of SPFO and its complexes (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (Lee D. Wilson). *E-mail: [email protected] (Paul Hazendonk). Author Contributions

The manuscript was written through joint contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge Natural Sciences and Engineering Research Council of Canada (NSERC), the University of Saskatchewan, and the University of Lethbridge for support of this research. AHK wishes to acknowledge Esther Aluri at the University of Saskatchewan for her assistance with PXRD measurements.

■

REFERENCES

(1) Key, B. D.; Howell, R. D.; Criddle, C. S. Fluorinated Organics in the Biosphere. Environ. Sci. Technol. 1997, 31, 2445−2454. (2) Rao, N. S.; Baker, B. E. Organofluorine Chemistry Principles and Commercial Applications; Banks, R. E., Smart, B. E., Tatlow, J. C., Ed.; Plenum Press: New York, 1994. (3) Kissa, E. Fluorinated Surfactants and Repellents, 2nd ed.; Marcel Dekker, Inc.: New York, 2001. (4) Giesy, J. P.; Kannan, K. Perfluorochemical Surfactants in the Environment. Environ. Sci. Technol. 2002, 36, 146A−152A. 22241


Article

The Journal of Physical Chemistry C (25) Xing, H.; Lin, S.-S.; Yan, P.; Xiao, J.-X.; Chen, Y.-M. NMR Studies on Selectivity of β-Cyclodextrin to Fluorinated/Hydrogenated Surfactant Mixtures. J. Phys. Chem. B 2007, 111, 8089−8095. (26) Wenz, G. Cyclodextrins as Building-Blocks for Supramolecular Structures and Functional Units. Angew. Chem., Int. Ed. Engl. 1994, 33, 803−822. (27) Wilson, L. D.; Verrall, R. E. 1H NMR Study of Cyclodextrin Hydrocarbon Surfactant Inclusion Complexes in Aqueous Solutions. Can. J. Chem. 1998, 76, 25−34. (28) Wilson, L. D.; Verrall, R. E. 19F and 1H NMR Investigation of Cyclodextrin/Fluorocarbon Alkyl Carboxylate Surfactant Inclusion Complexes. Langmuir 1998, 14, 4710−4717. (29) Wilson, L. D.; Verrall, R. E. A Volumetric and NMR Study of Cyclodextrin-Inhalation Anesthetic Complexes in Aqueous Solutions. Can. J. Chem. 2015, 93, 1−7. (30) Taulier, N.; Chalikian, T. V. Hydrophobic Hydration in Cyclodextrin Complexation. J. Phys. Chem. B 2006, 110, 12222− 12224. (31) Dias, A. M. A.; Andrade-Dias, C.; Lima, S.; Coutinho, J. A. P.; Teixeira-Dias, J. J. C.; Marrucho, I. M. How Does B-Cyclodextrin Affect Oxygen Solubility in Aqueous Solutions of Sodium Perfluoroheptanoate? J. Colloid Interface Sci. 2006, 303, 552−556. (32) Messina, P. V.; Prieto, G.; Salgado, F.; Varela, C.; Nogueira, M.; Dodero, V.; Ruso, J. M.; Sarmiento, F. The Influence of Sodium Perfluorooctanoate on the Conformational Transitions of Human Immunoglobulin. J. Phys. Chem. B 2007, 111, 8045−8052. (33) Yao, Y.; Liu, X.; Liu, T.; Zhou, J.; Zhu, J.; Sun, G.; He, D. Preparation of Inclusion Complex of Perfluorocarbon Compound with β-Cyclodextrin for Ultrasound Contrast Agent. RSC Adv. 2015, 5, 6305−6310. (34) Ando, S.; Harris, R. K.; Scheler, U.; Grant, D. M. Fluorine-19 NMR of Solids Containing Both Fluorine and Hydrogen. Encycl. Nucl. Magn. Reson. 2002, 9, 531−550. (35) Levitt, M. H.; Suter, D.; Ernst, R. R. Spin Dynamics and Thermodynamics in Solid-State NMR Cross Polarization. J. Chem. Phys. 1986, 84, 4243. (36) Brouwer, E. B.; Gougeon, R. D. M.; Hirschinger, J.; Udachin, K. A.; Harris, R. K.; Ripmeester, J. A. Intermolecular Distance Measurements in Supramolecular Solids: 13C−19F REDOR NMR Spectroscopy of P-Tert-butylcalix[4]arene−fluorobenzene. Phys. Chem. Chem. Phys. 1999, 1, 4043−4050. (37) Periasamy, R.; Rajamohan, R.; Kothainayaki, S.; Sivakumar, K. Spectral Investigation and Structural Characterization of Dibenzalacetone: β-Cyclodextrin Inclusion Complex. J. Mol. Struct. 2014, 1068, 155−163. (38) Hu, J. Z.; Zhou, J.; Deng, F.; Feng, H.; Yang, N.; Li, L.; Ye, C. High Resolution 1H Spectra of Powdered Solids Observed by Hahn Echo Pulse Sequence with Magic-Angle Spinning. Solid State Nucl. Magn. Reson. 1996, 6, 85−94. (39) Samson, A. O.; Chill, J. H.; Anglister, J. Two-Dimensional Measurement of Proton T1ρ Relaxation in Unlabeled Proteins: Mobility Changes in A-Bungarotoxin upon Binding of an Acetylcholine Receptor Peptide. Biochemistry 2005, 44, 10926−10934. (40) Riga, A.; Collins, R.; Mlachak, G. Oxidative Behavior of Polymers by Thermogravimetric Analysis, Differential Thermal Analysis and Pressure Differential Scanning Calorimetry. Thermochim. Acta 1998, 324, 135−149. (41) Yang, X.; Zhao, Y.; Chen, Y.; Liao, X.; Gao, C.; Xiao, D.; Qin, Q.; Yi, D.; Yang, B. Host-Guest Inclusion System of Mangiferin with B-Cyclodextrin and Its Derivatives. Mater. Sci. Eng., C 2013, 33, 2386− 2391. (42) Cwiertnia, B.; Hladon, T.; Stobiecki, M. Stability of Diclofenac Sodium in the Inclusion Complex with β-Cyclodextrin in the Solid State. J. Pharm. Pharmacol. 1999, 51, 1213−1218. (43) Cursaru, B.; Stanescu, P. O.; Teodorescu, M. The States of Water in Hydrogels Synthesized from Diepoxy-Terminated Poly(ethylene Glycol)s and Aliphatic Polyamines. U.P.B. Sci. Bull., Ser. B 2010, 72, 99−114.

(44) Mancinelli, R.; Botti, A.; Bruni, F.; Ricci, M. A.; Soper, A. K. Hydration of Sodium, Potassium, and Chloride Ions in Solution and the Concept of Structure/Breaker. J. Phys. Chem. B 2007, 111, 13570− 13577. (45) Sambasevam, K. P.; Mohamad, S.; Sarih, N. M.; Ismail, N. A. Synthesis and Characterization of the Inclusion Complex of BCyclodextrin and Azomethine. Int. J. Mol. Sci. 2013, 14, 3671−3682. (46) Tatsuno, H.; Ando, S. Structure and Dynamics of Perfluoroalkane/β-Cyclodextrin Inclusion Compounds as Studied by SolidState 19F MAS and 1H→19F CP/MAS NMR Spectroscopy. J. Phys. Chem. B 2006, 110, 25751−25760. (47) Pawsey, S.; Reven, L. 19F Fast Magic-Angle Spinning NMR Studies of Perfluoroalkanoic Acid Self-Assembled Monolayers. Langmuir 2006, 22, 1055−1062. (48) Lamcharfi, E.; Kunesch, G.; Meyer, C.; Robert, B. Investigation of Cyclodextrin Inclusion Compounds Using FT-IR and Raman Spectroscopy. Spectrochim. Acta, Part A 1995, 51, 1861−1870. (49) Crupi, V.; Majolino, D.; Venuti, V.; Guella, G.; Mancini, I.; Rossi, B.; Verrocchio, P.; Viliani, G.; Stancanelli, R. Temperature Effect on the Vibrational Dynamics of Cyclodextrin Inclusion Complexes: Investigation by FTIR-ATR Spectroscopy and Numerical Simulation. J. Phys. Chem. A 2010, 114, 6811−6817. (50) Albinsson, B.; Michl, J. Anti, Ortho, and Gauche Conformers of Perfluoro-N-Butane: Matrix-Isolation IR Spectra and Calculations. J. Phys. Chem. 1996, 100, 3418−3429. (51) Moynihan, R. E. The Molecular Structure of Perfluorocarbon Polymers. Infrared Studies on Polytetrafluoroethylene. J. Am. Chem. Soc. 1959, 81, 1045−1050. (52) Ellis, D. A.; Denkenberger, K. A.; Burrow, T. E.; Mabury, S. A. The Use of 19F NMR to Interpret the Structural Properties of Perfluorocarboxylate Acids: A Possible Correlation with Their Environmental Disposition. J. Phys. Chem. A 2004, 108, 10099−10106. (53) Knochenhauer, G.; Reiche, J.; Brehmer, L.; Barberka, T.; Woolley, M.; Tredgold, R.; Hodge, P. Do Perfluorinated Chains Always Have to Be Twisted? J. Chem. Soc., Chem. Commun. 1995, 1619−1620. (54) Erkoç, Ş.; Erkoç, F. Structural and Electronic Properties of PFOS and LiPFOS. J. Mol. Struct.: THEOCHEM 2001, 549, 289−293. (55) Lo Nostro, P.; Santoni, I.; Bonini, M.; Baglioni, P. Inclusion Compound from a Semifluorinated Alkane and β-Cyclodextrin. Langmuir 2003, 19, 2313−2317. (56) Okumura, H.; Kawaguchi, Y.; Harada, A. Preparation and Characterization of Inclusion Complexes of Poly(dimethylsiloxane)s with Cyclodextrins. Macromolecules 2001, 34, 6338−6343. (57) Jiao, H.; Goh, S. H.; Valiyaveettil, S.; Zheng, J. Inclusion Complexes of Perfluorinated Oligomers with Cyclodextrins. Macromolecules 2003, 36, 4241−4243. (58) Gao, Y.-A.; Li, Z.-H.; Du, J.-M.; Han, B.-X.; Li, G.-Z.; Hou, W.G.; Shen, D.; Zheng, L.-Q.; Zhang, G.-Y. Preparation and Characterization of Inclusion Complexes of Beta-Cyclodextrin with Ionic Liquid. Chem. - Eur. J. 2005, 11, 5875−5880. (59) Liu, J.; Alvarez, J.; Ong, W.; Roman, E.; Kaifer, A. Tuning the Catalytic Activity of Cyclodextrin-Modified Palladium Nanoparticles through Host-Guest Binding Interactions. Langmuir 2001, 17, 6762− 6764. (60) Abbate, V.; Bassindale, A. R.; Brandstadt, K. F.; Taylor, P. G. Inclusion Complexes between Cyclic Siloxanes and Cyclodextrins: Synthesis and Characterization. J. Inclusion Phenom. Mol. Recognit. Chem. 2012, 74, 223−230. (61) Saenger, W. Crystal Packing Patterns of Cyclodextrin Inclusion Complexes. J. Inclusion Phenom. 1984, 2, 445−454. (62) Caira, M. R. On the Isostructurality of Cyclodextrin Inclusion Complexes and Its Practical Utility. Rev. Roum. Chim. 2001, 46, 371− 386. (63) Wilson, L. D.; Verrall, R. E. 19F and 1 H NMR Investigation of Cyclodextrin/Fluorocarbon Alkyl Carboxylate Surfactant Inclusion Complexes. Langmuir 1998, 14, 4710−4717. 22242


Article

The Journal of Physical Chemistry C (64) Schneider, H.; Hacket, F.; Rüdiger, V.; Ikeda, H. NMR Studies of Cyclodextrins and Cyclodextrin Complexes. Chem. Rev. 1998, 98, 1755−1785. (65) Goecke, C. M.; Jarnot, B. M.; Reo, N. V. A Comparative Toxicological Investigation of Perfluorocarboxylic Acids in Rats by Fluorine-19 NMR Spectroscopy. Chem. Res. Toxicol. 1992, 5, 512− 519. (66) Immel, S.; Lichtenthaler, F. W. The Hydrophobic Topographies of Amylose and Its Blue Iodine Complex. Starch 2000, 52, 1−8. (67) Wilson, L. D.; Verrall, R. E. Volumetri Study of Modified Cyclodextrin/hydrocarbon and /Fluorocarbon Surfactant Inclusion Complexes in Aqueous Solutions. J. Phys. Chem. B 1998, 102, 480− 488. (68) Ribeiro, A. A. 19F, 13C Single- and Two-Bond 2D NMR Correlations in Perfluoroheptanoic Acid. J. Fluorine Chem. 1997, 83, 61−66. (69) Newmark, R. A. Vicinal Fluorine-Fluorine Coupling Constants in Perfluoropropyl Groups. J. Fluorine Chem. 2009, 130, 389−393. (70) Emsley, J. W.; Feeney, J.; Sutcliffe, L. H. In Progress in Nuclear Magnetic Resonance Spectroscopy; Pergamon Press: New York, 1971; Vol. 7. (71) Bunn, C. W.; Howells, E. R. Structures of Molecules and Crystals of Fluoro-Carbons. Nature 1954, 174, 549−551. (72) Rüdiger, V.; Schneider, H. The Use of Complexation Induced Proton NMR Chemical Shifts for Structural Analysis of Host-Guest Complexes in Solution. Chem. - Eur. J. 2000, 6, 3771−3776. (73) Lu, J.; Mirau, P. a.; Tonelli, A. E. Chain Conformations and Dynamics of Crystalline Polymers as Observed in Their Inclusion Compounds by Solid-State NMR. Prog. Polym. Sci. 2002, 27, 357−401. (74) Koito, Y.; Yamada, K.; Ando, S. Solid-State NMR and WideAngle X-Ray Diffraction Study of Hydrofluoroether/β-Cyclodextrin Inclusion Complex. J. Inclusion Phenom. Mol. Recognit. Chem. 2013, 76, 143. (75) Ceccato, M.; LoNostro, P.; Rossi, C.; Bonechi, C.; Donati, A.; Baglioni, P. Molecular Dynamics of Novel Alpha-Cyclodextrin Adducts Studied by C-13-NMR Relaxation. J. Phys. Chem. B 1997, 101, 5094−5099. (76) James, T. L. Fundamentals of NMR; University of California; 1998; pp 1−31. (77) Karoyo, A. H. Structural Studies of Supramolecular Host-Guest Systems; University of Saskatchewan: Saskatoon, Saskatchewan, August 2014.

22243


Counterion Anchoring Effect on the Structure of the Solid-State

Recommend Documents