Cooperative Transport and Selective Extraction of Sulfates by a Squaramide-Based Ion Pair Receptor: A Case of Adaptable Selectivity

The use of a squaramide-based ion pair receptor offers a solution to the very challenging problem of extraction and transport of extremely hydrated sulfate salt. Herein we demonstrate for the first time that a neutral receptor is able not only to selectively extract but also to transport sulfates in the form of an alkali metal salt across membranes and to do so in a cooperative manner while overcoming the Hofmeister bias. This was made possible by an enhancement in anion binding promoted by cation assistance and by diversifying the stoichiometry of receptor complexes with sulfates and other ions. The existence of a peculiar 4:1 complex of receptor 2 with sulfates in solution was confirmed by UV–vis and 1H NMR titration experiments, DOSY and DLS measurements, and supported by solid-state X-ray measurements. By varying the separation technique and experimental conditions, it was possible to switch the depletion of the aqueous layer into extremely hydrophilic or less lipophilic salts, thus obtaining the desired selectivity.


General information
Unless specifically indicated, all other chemicals and reagents used in this study were purchased from commercial sources and used as received. If necessary purification of products was performed using column chromatography on silica gel (Merck Kieselgel 60, with mixtures of chloroform/methanol. Thin-layer chromatography (TLC) was performed on silica gel plates (Merck Kieselgel 60 F254). 1 H and 13 C NMR spectra used in the characterization of products were recorded on Bruker 300 spectrometer using a residual protonated solvent as internal standard. DOSY experiments were conducted at 298 K on Varian VNMRS 600 MHz instruments with a residual solvent signal as an internal standard.
Mass spectra were measured on Quattro LC Micromass or Shimadzu LCMS-IT-TOF unit.
Dynamic Light Scattering analyses were performed using a Malvern Zetasizer instrument (Nano ZS, UK) equipped with a 4-mW helium-neon laser of light wavelength 632.8 nm was used. The scattering angle was set to 170 o and at 25ºC. The hydrodynamic diameter distributions were obtained by volume using the software package of the apparatus. Each curve represents the average of 3 measurements (16 runs each). Prior to analysis, all solutions were filtered and degassed.

UV-vis titration experiments
General procedure. UV-vis titration experiments were performed on a Thermo Spectronic Unicam UV 500 spectrophotometer in CH3CN solution at 298K. To 10 mm cuvette was added 2.5 mL of freshly prepared (receptor 1 -2.85×10 -5 M, receptor 2 -2.63×10 -5 M, receptor 3 -2.16×10 -5 M ) solution of studied receptor and in case of ion pair binding studies 1 mol equivalent of cation (KPF6 or NaClO4) was added prior titrations. Small aliquots of ca. 1.5×10 -3 M TBAX solution containing receptor 1, receptor 2 or receptor 3 at the same concentration as in cuvette, were added and a spectrum was acquired after each addition. The resulting titration data were analyzed using BindFit (v0.5) package, available online at http://supramolecular.org.

NMR Titration
The 1 H NMR titration was performed on a Bruker 300 spectrometer, at 298K in CD3CN. In each case, a 500 μL of freshly prepared ca. 3 mM solution of receptor was added to a 5mm NMR tube. In the case of ion pair titration receptor was firstly pretreated with one equivalent of NaClO4 or KPF6 (refers to receptor). Then small aliquots of solution of TBAX, containing receptor at constant concentration, were added and a spectrum was acquired after each addition.

EXTRACTION EXPERIMENTS -UV-vis
The ability of receptors 1 and 2 to extract ions from aqueous solution was qualitatively investigated using a liquid-liquid extraction technique monitored by UV-vis. A solution of receptor 1 or 2 in chloroform (4 ml, 3×10 -5 M) and (4 ml, 50 mM) aqueous solution of the appropriate salt were mixed thoroughly for 5 minutes in a vial (prolongation of the extraction time had no effect on the results obtained). The vial was allowed to stand to fully separate the two phases. The organic layer was then separated and screened by UV-vis.

EXTRACTION EXPERIMENTS -ATOMIC EMISSION SPECTROSCOPY
General procedure: A solution of receptor 2 in chloroform (1 ml, 1 mM) and 0.5 M aqueous solution of the appropriate potassium salt (KCl, KBr, KI, KNO2, KNO3, K2SO4, KHSO4) were mixed thoroughly for 5 minutes in a vial (prolongation of the extraction time had no effect on the results obtained). The vial was allowed to stand to fully separate the two phases. Then 0.5 mL of organic phase was taken and diluted to 5 mL with ethyl acetate. The potassium concentration in organic phase was determined by atomic emission spectroscopy (AES). there is no phase separation probably due to the receptor deprotonation. This eliminates direct use of basic salts such as carboxylates, hydrogen phosphates or phosphates) of suitable salts 5 mM each (2 ml) for 5 minutes. Then 1 mL of aqueous phase was taken and tenfold diluted.
The concentration of chloride, bromide, nitrite, nitrate, dihydrogenphosphate and sulfate anions in aqueous phase was determined by high performance ion chromatography (HPIC).

Dynamic Light Scattering (DLS)
The changes in the size of the receptor 2 and supramolecular complexes formed in wet chloroform after extraction were also performed in means of DLS measurements. The value of the solvodynamic diameter was found to be circa 1.2 nm for receptor 2 (1 mM) in wet chloroform and after extraction with aqueous solution of KBr or K2SO4 increased to circa 1.4 or 2.0 nm, respectively.       Source phase and receiving phase: 6 mL. The chloroform phase (6 mL) was stirred at 1 000 rpm during the experiment to ensure efficient diffusion. Salt concentration was measured every day in the receiving phase, and after 7 days the experiment was stopped. The sulfate or chloride concentration was analysed by conductometric method. Source phase and receiving phase: 6 mL. The chloroform phase (6 mL) was stirred at 1 000 rpm during the experiment to ensure efficient diffusion. Samples were collected every day from receiving phase, and after 7 days, the experiment was stopped. Ion concentration was analysed by using Ion Chromatography technique.   were collected with Bruker APEX3 program. S2 The frames were integrated with the Bruker SAINT software package S3  The structure was solved and refined using SHELXTL Software Package S4,S5 using the space group P21/c, with Z = 2 for the formula unit, C56H60F10K2N6O23. The final anisotropic fullmatrix least-squares refinement on F 2 with 523 variables converged at R1 = 4.82%, for the observed data and wR2 = 14.42% for all data. The goodness-of-fit was 1.033. The largest peak in the final difference electron density synthesis was 0.308 e -/Å 3 and the largest hole was -0.318 e -/Å 3 with an RMS deviation of 0.052 e -/Å 3 . On the basis of the final model, the calculated density was 1.521 g/cm 3 and F(000), 1500 e -. Crystal data and refinement parameters for Receptor 2+KNO3 are collected in Table S7.
In the ligand the F-substituted phenyl ring is disordered over two positions with refined occupancy ratio yielding 0.816(9):0.184(9). Also NO3moiety is disordered over two sites with refined occupancy equal 0.931(6):0.069(6). In addition the crystal contains disordered diethyl ether solvent molecule located on centre of symmetry, equally sharing two locations. To model the disorder a number of geometrical restraints was used. Some restrains for atomic displacements in the disordered fragments were also applied.

S52
All ordered non-hydrogen atoms and major component disordered moieties were refined anisotropically. Most of hydrogen atoms were placed in calculated positions and refined within the riding model. Positions of two hydrogen atoms engaged in intramolecular hydrogen bonds (N-H moieties) were refined. The temperature factors of hydrogen atoms were not refined (except the atoms engaged in hydrogen bonds) and were set to be1.2 times larger than Ueq of the corresponding heavy atom. The atomic scattering factors were taken from the International Tables. S6 Molecular graphics was prepared using program Mercury 4.1. S7 Atomic displacement parameters are presented in Fig. S87 a) at 50% probability level, packing diagrams are shown in Fig. S 88 a).
Receptor 2+KCl/NaCl. The X-ray measurement of Receptor 2+KCl/NaCl was performed at 100.0(5) K on a Bruker D8 Venture PhotonII diffractometer equipped with a TRIUMPH monochromator and a MoKα fine focus sealed tube (λ = 0.71073 Å). A total of 3794 frames were collected with Bruker APEX3 program. S1 The frames were integrated with the Bruker SAINT software package S2 using a narrow-frame algorithm. The integration of the data using a The structure was solved and refined using SHELXTL Software Package S4,S5 using the space group P21/c, with Z = 2 for the formula unit, C52H50Cl2F10K1.31N4Na0.69O16. The final S53 anisotropic full-matrix least-squares refinement on F 2 with 455 variables converged at R1 = 3.63%, for the observed data and wR2 = 9.44% for all data. The goodness-of-fit was 1.032. The largest peak in the final difference electron density synthesis was 0.284 e -/Å 3 and the largest hole was -0.208 e -/Å 3 with an RMS deviation of 0.042 e -/Å 3 . On the basis of the final model, the calculated density was 1.541 g/cm 3 and F(000), 1349 e -. Crystal data and refinement parameters for Receptor 2+KCl/NaCl are collected in Table S7.
In the structure the crown ether fragment is disordered over three sites with refined occupancy ratio equal to 0.655 (2) Tables. S6 Molecular graphics was prepared using program Mercury 4.1. S7 Atomic displacement parameters are presented in Fig. S87 b) at 50% probability level, packing diagrams are shown in Fig. S88 b).
Receptor 2+Na2SO4. The X-ray measurement of Receptor 2+Na2SO4 was performed at 100.0(5) K on a Bruker D8 Venture PhotonII diffractometer equipped with a TRIUMPH monochromator and a MoKα fine focus sealed tube (λ = 0.71073 Å). A total of 2400 frames were collected with Bruker APEX3 program. S1 The frames were integrated with the Bruker S54 SAINT software package S3 using a narrow-frame algorithm. The integration of the data using a monoclinic unit cell yielded a total of 119249 reflections to a maximum θ angle of 25.05° (0.84 Å resolution), of which 11108 were independent (average redundancy 10.735, completeness = 99.9%, Rint = 3.64%, Rsig = 1.82%) and 9555 (86.02%) were greater than 2σ(F 2 ).
The final cell constants of a = 12.0787 (9)  Receptor 2+Na2SO4 are collected in Table S7.
The structure in the asymmetric part contains one fully ordered ligand molecule; another ligand with Na + cation located in disordered over three positions crown ether fragment and coordinated by two acetonitrille molecules; halve of disordered over two sites SO4 2moiety located on 2fold axis of symmetry. In the structure there are also disordered acetonitrille, water and methanol molecules all located on 2-fold axis of symmetry.  Fig. S87 c) at 50% probability level, packing diagrams are shown in Fig. S88 c).
During structure refinement of all discussed crystals the same numbering scheme of the atoms in 2, presented in Scheme S2, was used.

Fig. S87.
Numbering scheme of all non-C atoms and atomic displacement parameters presented at 50% probability level for structure Receptor 2+KNO3 a); Receptor 2+KCl/NaCl b) and Receptor 2+Na2SO4 c). All hydrogen atoms and notcoordinating solvent molecules omitted for clarity. For the structure Receptor 2+Na2SO4 molecules A and B displayed separately.