Norbornene chaotropic salts as low molecular mass ionic organogelators (LMIOGs)

A humble norbornene functions as an ionic organogelator, forms aqueous biphasic and triphasic systems and assembles to form chiral helices.


Reagents
All listed reagents were obtained from specialist suppliers and use without further purification. Pet. Spirits refers to the fraction of Petroleum Spirits boiling between 40 -60 °C.

Instrumentation
All 1 H NMR and 13 C NMR spectra were recorded on either a JEOL JNM-EX 270 MHz FT-NMR, a JEOL JNM-Eclipse+ 400 MHz FT-NMR or a as indicated. Samples were dissolved in deuterated chloroform (CDCl 3 ) and the relevant solvent peak (CDCl 3 : δH 7.26 ppm) used as an internal reference. The template for reporting proton spectra is as follows: chemical shift δ (ppm), (multiplicity (s = singlet, br s = broad singlet, d = doublet, t = triplet, q = quartet, m = multiplet), coupling constant J (Hz), integral, assignment). All IR spectra were recorded using a Bruker ALPHA FTIR spectrometer in the range of 4000-500 cm -1 . The TGA experiments were performed using a TA Instruments TGA Q50 V20. 13 Build 39. For the crystals of 9:Na the heating was performed using a platinum pan, under nitrogen gas, at a heating rate of 10 degrees per minute up to 400 degrees.

General procedure for synthesis of amino acid functionalised norbornenes
Norbornene anhydride (1 equiv.), amino acid (1.1 equiv.) and NEt 3 (0.4 equiv.) were combined in a round bottom flask with PhMe (10 mL per 1 g of anhydride). The reaction setup was fitted with a Dean-Stark apparatus and heated at reflux for 16 h. After this time, the reaction mixture was allowed to cool before the solvent was removed under reduced pressure. The residue was then taken up in EtOAc and transferred to a separatory funnel before being washed with 0.2M HCl (2 × 15 mL) and sat. NaCl (2 × 15 mL) before being dried (MgSO 4 ), filtered and the solvent removed under reduced pressure.

Nomenclature and NMR assignment
All norbornane and fused [n]polynorbornane-based compounds are named according to the Von-Baeyer system of nomenclature as (multiplier)cyclo[X,Y,Z]hydrocarbon where X ≥ Y > Z > 0. 1 The stereodescriptors syn/anti are used to refer to substituents on the Z bridge that are located on the same/opposite side as the X bridge. The stereodescriptors α/β are used to refer to the configuration of substituents on the X and Y bridges. The norbornane system is numbered to allow for the longest possible chain before numbering bridges.
The amino acid side chain is assigned and labelled following Pople notation for AMX pairs. 2

General Procedure for salt formation
All sodium and potassium salts were formed by dissolving a suitable quantity of the acid in EtOAc and transferring to a separatory funnel. The organic phase was then extracted with excess sat. Na 2 CO 3 or K 2 CO 3 . From the resultant triphasic mixture the middle layer was separated and concentrated to dryness. The resultant white solid was the desired salt (quantitative). Figure S1.1: 1 H NMR of (S)-9:Na in D 2 O All other salts were formed by dissolving (S)-9 (1 equiv.) in i-PrOH and adding an aqueous solution of the cation hydroxide (1 equiv.). The combined solution was stirred overnight before being concentrated to dryness to afford the desired salt.
All salts were observed to be pure through 1 H NMR spectroscopy The melting point of all salts were observed to be over 300 °C (the limits of our instrumentation) except for 10:Na which melted from 75.6 -82.5 °C.

Aqueous biphasic system
To identify compounds capable of forming aqueous biphasic systems an aqueous (5 M) solution of the compound was added to a test tube then the equivalent volume of sat. Na 2 CO 3 added. After stoppering and shaking if the aqueous biphasic system was not immediately apparent the solution was heated to near boiling and upon cooling again inspected for a biphasic system. Figure S1.4: Aqueous biphasic system formed by combining equal volumes of aqueous 9:Na (5M) and sat. aqueous Na 2 CO 3 .

S2. Diffusion
1 H relaxation measurements were carried out at 11.7 T on a Bruker Avance III 500 MHz wide-bore spectrometer and a 5 mm static probe. Samples of (S)-9:Na were made up to the desired concentration in 500 μL of D 2 O.
The pulsed field gradient stimulated echo (PFG-STE) pulse sequence was utilised with the magnitude of the applied gradient varied in 16 increments and the maximum gradient strength, gradient pulse length and diffusion time (D) optimised in each case. The 1H 90° pulse length was 9.5 ms and recycle delays were set to > 5 × T 1 in all cases. The data were fitted to the Stejskal-Tanner equation (Equation 1 below) using a non-linear leastsquares method within the Bruker TopSpin software (v3): In this expression, I is the observed signal intensity, I 0 is the maximum signal intensity, γ is the gyromagnetic ratio of the nucleus being observed, g is the gradient strength and δ is the gradient pulse duration. 7 Diffusion constants were determined for multiple integral regions for each sample and averaged to provide the final diffusion constant plotted in Figure 2 in the manuscript.

Solid gelator
Gelation tests were carried out by adding the gelator to the organic solvent as a solid powder before heating to the boiling temperature of the solvent (table S3.1). The solutions were then allowed to cool to room temperature. All apparent gels were then subjected to the six hour inversion test. If the six hour inversion test was passed, the gel classification was maintained. A turbid gel was designated to all gels not opaque in nature.
The following terms were used to classify the remaining samples: Weak Gel: initial gel formation noted but the six hour inversion test was failed. Partial Gel: the gel failed to entrap all the solvent. Dissolved: sample was fully dissolved, resulting in a clear solution. Insoluble: the sample failed to fully dissolve resulting in a precipitate. Suspension: the sample remained dispersed throughout the solvent upon standing. Aggregate: large accumulations of fine solid material encapsulated in small gel-like formations. Compounds 10-12 displayed no gelation behaviour in any of the tested solvents and so have been omitted from these results.
Gels were formed upon sonication. For THF, no sonication was required to elicit complete gelation.

S4. SEM images Instrumentation
Gel samples were prepared by placing a small amount of gels (~10 -15 mL by vol.) on silicon wafers and making a thin film by smooth smearing. The samples were dried initially in air for 12 hours followed by drying in high vacuum for 2 -3 hours. Before recording the images, the samples were gold coated using sputter. These samples were imaged in Carl Zeiss ULTRA Plus, field emission scanning electron microscope (FE-SEM) using beam current of 5 -10 kV with an SE2 or InLens detector.

Preparation samples for SEM analysis
A small scoop of gel (approx. 10 μL by vol) was placed on a Silica wafer and spread into a thin film. This was subsequently dried in air overnight then under vacuum for a further 2 hrs. These samples were then coated with Au using sputter coating.         Figure S4.11. SEM images of (S)-9:Li (2 wt%) 1,4-Dioxane partial gel. TOP: When a thin SEM sample was prepared, a film was formed; BOTTOM: When a thick SEM sample was used, micron sized crystals were observed.

S5. DSC calorimetry Instrumentation
All DSC measurements were obtained using a TA Instruments DSC Q200 instrument using a Tzero aluminum hermetic pan and lid. A small sample of the gel (approx. 8-10 mg) was placed into the pan and the lid was then crimped on. A small hole was punctured into the lid, carefully avoiding disturbing the sample, and the pan was then placed in the sample port. An air flow rate of 50.0 mL/min was employed. The samples were allowed to equilibrate at 20 °C before being heated at 1 °C/min.  Thermal stability (T gel ) of 9:K and 9:Na gels prepared by increasing the gelator concentrations was measured by ball drop method. T gel values of gels increased with increasing concentration of gelator. Variation of Tgel vs concentration is shown in Figure S5.12. 9:K and 9:Na gels in EtOH, i-PrOH, n-BuOH, were formed by heating solid powder in organic solvents and cooling to room temperature. Whereas, 9:Na gels in THF and 1,4-dioxane were formed by adding aqueous solution (c = 100%) of gelator to organic solvents.

Instrumentation
Rheological measurements were conducted on a Discovery HR-3 hybrid rheometer with a 40 mm 1.995° cone plate geometry at 25 °C. The set gap was approximately 0.5 mm for all samples. Steady shear measurements (flow sweep) were carried out in the shear rate range of 0.1-5000 s−1 which is the upper limit of the instrument.         Note that there are no sudden "drops" or "shifts" in the stress measured (these would indicate slippage). The magnitudes and trends in the stress measured are consistent and correspond well to the experimental design as would be expected for reversible gel behaviour.

Instrumentation and data refinement
Crystal and refinement parameters are shown below (Table S7.1). The data for 9:Na (R and S enantiomers) were collected using microfocus Cu Kα radiation (λ = 1.54184 Å) with an Oxford Diffraction SuperNova instrument equipped with an Atlas CCD detector, with sample maintained at a temperature of 130 K. Data reduction was carried out using Crysalis Pro software package, with empirical absorption correction performed using SCALE3 ABSPACK. 8,9 The data for 9:TEA were collected using a Bruker APEX-II DUO instrument using microfocus Cu Kα radiation (λ = 1.54184 Å), with the sample maintained at 100 K using a Cobra cryostream. These data were processed using Bruker APEX3 software, with multi-scan absorption correction carried out with SADABS. 10, 11 All datasets were solved using SHELXT, and refined with SHELXL, operated within the OLEX2 software package. [12][13][14][15] The functions minimized were Σw(F 2 o -F 2 c ), with w=[σ 2 (F 2 o )+aP 2 +bP] -1 , where P=[max(F o ) 2 +2F 2 c ]/3. All non-hydrogen atoms were refined with anisotropic displacement parameters. All carbon-bound hydrogen atoms were placed in calculated positions and refined with a riding model, with isotropic displacement parameters equal to either 1.2 or 1.5 times the isotropic equivalent of their carrier atoms. Where appropriate, hydrogen atoms involved in hydrogen bonding were located from Fourier residuals and refined with loose distance restrains and riding U iso dependencies. In the case of 9:Na, both enantiomers exhibited crystallographic disorder on the carboxylate groups, which were modelled in separate orientation with the occupancies refined as free variables, with similar treatment applied to the disordered aqua ligands. The crystallographic formulae were amended to include the hydrogen atoms for these groups which could not be explicitly modelled. In the case of 9:TEA, in order to achieve acceptable Friedel pair coverage, two datasets were collected on the same crystal with the sample manually reoriented between scans. and the data were merged using Bruker XPREP to achieve >98% Friedel pair coverage. 16 Full details are provided in the combined crystallographic information file. In all cases, the absolute configuration of each material was unambiguously confirmed by anomalous dispersion effects.CCDC 1826290.
Single crystals of both the R and S enantiomers of 9:Na were prepared by first forming a gel in i-PrOH and allowing the solvent to evaporate slowly. Analysis by single crystal X-ray diffraction was performed and the data for each were solved and refined in the hexagonal space group P6 3 . Both compounds crystallised in their enantiomerically pure forms (Table S7.1) and exhibited identical (inverted) behaviour to one another. The asymmetric unit of 9:Na contains one molecule of 9, deprotonated and coordinating to one unique sodium ion ( Figure S7.1). One aqua ligand is also present within the structure which exhibits minor crystallographic disorder, the coincidence of which with the crystallographic sixfold axis also prevents unambiguous location of the associated hydrogen atoms in the structural model. The sodium ion adopts an irregular six-coordinate geometry; the carboxylate group of 9 (which also exhibits minor crystallographic disorder) coordinates in a μ 3 -κO;κO,O′;κO′ coordination mode, with symmetry equivalents occupying four coordination sites on each sodium ion (Figure 7.1). The remaining sites are occupied by the aqua ligand and coordination through one imide oxygen atom O2 from the 9 anion. Coordination through the imide oxygen atom is relatively unusual, though is more frequently observed for alkali metal systems than for transition metal or lanthanide complexes; [17][18][19][20][21][22][23]   Each sodium ion in the structure of 9:Na connects to three 9 ligands, and the μ 3 -bridging behaviour of the 9 ligand gives rise to a one-dimensional polymeric assembly oriented parallel to the crystallographic c axis. This takes the form of a one-dimensional tubular assembly with a hydrated sodium-carboxylate core and with the periphery decorated by the phenyl and norbornenyl groups of 9. The co-incidence of these chains to the 6 3 symmetry axis imparts a helical twist to these assemblies, with M helicity in the R enantiomer and P helicity in the S form. Considering the helices as consisting of alternating phenyl and norbornenyl fragments, which further interact via myriad weak C-H···π interactions (e.g., C6-H6···C15, d C···C 3.689(7) Å, C-H···C angle 143.0(4) °), the overall structure is best visualised as a triple-stranded helix ( Figure S7.2). Adjacent stacks in the structure of 9:Na associate in a hexagonal rod packing fashion with hydrophobic interactions between the outwards facing organic periphery of each chain; no significant voids or crystallographically localised guests were observed in the interstitial spaces, indicating efficient packing of the homochiral helical stacks. This observation may offer some justification as to the lack of any observed crystallisation of the racemate; given the significant undulation in the external surface of the columns and numerous intermolecular contacts occurring in the vicinity of these grooves (clearly visible using Hirshfeld surface mappings, Figure S7.3), co-crystallisation of helices with alternate handedness would be expected to lead to a significant decrease in packing efficiency. Single crystals of 9:TEA were prepared by recrystallization from THF and analysed by single crystal X-ray diffraction, and the data were solved and refined in the triclinic space group P1. The asymmetric unit contains two deprotonated 9 anions, two tetraethylammonium cations, and four lattice water molecules. Although the two unique 9 molecules within the unit cell are geometrically similar, the pseudo-inversion symmetry is disrupted by the homochiral nature of the structure, unambiguously confirmed through anomalous dispersion effects (Table S7.1). The 9 anions in 9:TEA adopt a near-identical conformation to that observed in 9:Na, with a gauche conformation for the phenyl and norbornenyl substituents (given by the torsion angle N1-C10-C12-C13 of 49.2(3) °). This leads to a similar mode o f intramolecular C-H···π interaction to that observed in 9:Na, with C···C distance of 3.635(4) Å and C-H···C angle 150.3(2)° for the interaction C2-H2···C16. The structure of 9:TEA is shown in Figure 7.4. The extended structure of 9.TEA is dominated by hydrogen bonding interactions between the 9 anion and the lattice water molecules. These interactions take the form of a undulating one-dimensional hydrogen-bonded ribbon consisting of alternating R 4 4 (8) planar water tetramers and R 6 6 (16) loops involving both water molecules and carboxylate acceptors, oriented parallel to the crystallographic a axis. Adjacent ribbons associate via faceto-face π-π interactions between phenyl rings (minimum interatomic distance 3.584(4) Å for C16···C31). Cationic {9·2H 2 O} layers are separated by layers of loosely associated tetraethylammonium cations, which contribute numerous weak C-H···O hydrogen bonding interactions within the structure.