A pH‐Switchable Triple Hydrogen‐Bonding Motif

Abstract A stimuli responsive linear hydrogen bonding motif, capable of in situ protonation and deprotonation, has been investigated. The interactions of the responsive hydrogen bonding motif with complementary partners were examined through a series of 1H NMR experiments, revealing that the recognition preference of the responsive hydrogen bonding motif in a mixture can be switched between two states.


Data for Condition B
The use of condition B: trifluoroacetic acid (TFA) and 1,4-diazabicyclo [2.2.2]octane (DABCO), to protonate 1 and deprotonate 1-H + HBM was studied. 1 H NMR analysis revealed a downfield shift of diagnostic resonances, Ha, Hb, Hc and Hg. on the addition of 1 equivalent of TFA to UIM 1 (5 mM in CDCl3) as observed for condition A ( Figure ESI 3 (d)-(c)). This suggests TFA is also able to protonate UIM 1 to form UIM-TFA 1-H + . To reverse the protonation, one equivalent of DABCO was added to the 5 mM solution of UIM-TFA 1-H + ( Figure ESI 3 (c)-(b)). This resulted in small changes in the chemical shifts of the diagnostic proton resonances Ha, Hb, Hc and Hg but the resonances did not fully align with those observed for neutral UIM 1 (or a 1:1 UIM:DABCO mixture) suggesting a mixture of neutral and protonated UIM. However, with addition of excess DABCO (3 eq.) the diagnostic protons shifted further upfield consistent with formation of UIM 1 as the dominant species ( Figure ESI 3 (a)). The lower pKa of DABCO in comparison to sodium hydrogen carbonate dictate that a greater concentration of DABCO for full deprotonation is to be expected. Similarly the pKa of DABCO is close to imidazolium (which forms the core of UIM). The switching 'off' behaviour of the 1·2 dimer interaction was tested using condition B. On the addition of 1 equivalent of TFA to the 1·2 dimer the diagnostic Hg resonance moved significantly downfield and became well resolved, unlike in the 1·2 dimer (Figure ESI 4 (c)-(b)). Additionally the presence of NH resonance (Hd) of UIM, not seen in neutral UIM 1 or UIM·AIC 1·2 dimer, indicated protonation of UIM 1 to give UIM-TFA 1-H + . The addition of DABCO to this mixture indicated that the ADD-DAA dimer was reformed by the upfield shift and broadening of the Hg resonance ( Figure ESI 4 (a). However, some of the resonances associated with 1 (Ha and Hc) did not fully match in the 1·2 dimer as anticipated. This could be a result of partial deprotonation using DABCO, compared to sodium hydrogen carbonate (used in condition A), leading to a mixture of species as well as TFA and DABCO salts. The recognition preferences of complementary and competing HBMs, 2 and 3, with protonated and neutral HBM 1 formed using condition B were studied. Addition of TFA to a mixture of the UIM·AIC 1·2 dimer and HBM BB1 3 disrupted the UIM·AIC 1·2 dimer generating a mixture of UIM-TFA·BB1 1-H + ·3 dimer in presence of AIC 2 ( Figure ESI 5 (b)). This switch is highlighted by a change in UIM resonances; the sharpening and downfield shift of diagnostic Hg resonance and the presence of NH resonance Hd. As well as more subtle chemical shifts in the resonances of AIC 2 and BB1 3. The addition of 3 equivalents of DABCO to this mixture did not fully switch back 'on' UIM·AIC 1·2 dimerization (Figure ESI 5 (a)); although some shifts in the spectra are observed these do not return to those observed for the spectrum of UIM·AIC 1·2 dimer and HBM BB1 3 (Figure ESI 5 (c)) before the protonation/deprotonation cycle. This likely results from greater competition between DABCO and the 1·3 complex for the proton i.e. a reduced difference in pKa between the two. Aside from the diminished reversibility, the interactional preferences for conditions B are in line with the observations for conditions A.

NMR Titration Experiments
For 1 H NMR titrations anhydrous CDCl3 was purchased from Aldrich and stored over molecular sieves (type 4A, 1 to 2mm beads). For the qualitative titration study of hydrochloric acid into UIM 1 the 1 H NMR spectrum was recorded for a solution of host (5 mM) in CDCl3 and the change in chemical shift of key proton resonances was recorded upon sequential additions of a solution of guest (0.25-3 equivalents) in CDCl3. For the quantitative titration study of UIM 1 and UIM-HCl 1-H + into BB1 3, the 1 H NMR spectrum was recorded for a solution of host (0.5 mM) in CDCl3 and the change in chemical shift of key proton resonances was recorded upon sequential additions of a solution of guest (0.2-12.3 equivalents) in CDCl3. The solutions of guests (UIM 1 and UIM-HCl 1-H + ) were made through a half dilution series and added to a solution of host (BB1 3). The equivalents were calculated by the ratio of integration of individual resonances. The data was subsequently analysed using the Supramolecular.org online bindfit program using the appropriate model to give an association constant. [1,2] Supramolecular.org uses data from multiple resonances for curve fitting. Representative 1 H NMR spectra and exported binding curves can be seen below.

Studies with Hexafluorophosphate Anion
Reflecting on the crystal structure of the intermediate I (Figure 1(a)) where the chloride ion bridges two NH groups, the chloride ion may influence the equilibria (Scheme ESI 1); although the time averaged nature of the 1 H NMR analysis reports provides information that can be related to the interactions of the HBMs, the coordination of chloride ion could alter the desired hydrogen bonding interactions between the HBMs. To assess this, the chloride ion was exchanged for a non-interacting anion, hexafluorophosphate (PF6) using silver hexafluorophosphate to create UIM-HPF6 1-H + PF6 (Figure ESI 10). 1 H NMR suggested that after the addition of silver hexafluorophosphate the species was still protonated; as the chemical shift of the imidazole CH resonance (Hg) was similar to that observed for UIM-HCl 1-H + and the NH resonances (Hd, He, Hf and Hh) were well-resolved ( Figure ESI 10 (b)). However, the NH resonances (Hd, He, Hf and Hh) shifted significantly for UIM-HPF6 1-H + PF6 compared to the resonances observed for UIM-HCl 1-H + . Next, proton dependent switching of UIM-HPF6 1-H + PF6 was explored, revealing it was possible to deprotonate motif UIM-HPF6 1-H + PF6 by washing with NaHCO3 in the same way that UIM-HCl 1-H + was deprotonated using condition A ( Figure ESI 10 (a)).  ). Efforts were made to switch this self-sorting behaviour 'off' by the addition of base to reform the UIM·AIC 1·2 dimer in the presence of BB1 3. However due to the lower solubility of BB1 3 and the process of washing with NaHCO3, sample was lost, thus it was not possible to obtain a well resolved 1 H NMR spectra with equal ratios of hydrogen bonding motifs. The limited solubility of BB1 3 in CDCl3 was more profound in the system using PF6 salts than the hydrochloric acid containing systems. Overall the recognition behaviour exhibited by UIM-HPF6 1-H + PF6 closely resembled the preferences exhibited by HBM UIM-HCl 1-H + indicating that whilst the counter anion may influence the equilibria under analysis, the 1 H NMR data reflect accurately on the recognition preference of the HBMs in these mixtures.  Figure  1 for the resonance assignment and chemical structure.

General materials and methods for synthesis
Solvents and reagents were purchased from Sigma Aldrich or Fisher Scientific and used without further purification unless otherwise stated. Where anhydrous solvents were required, dichloromethane, chloroform, tetrahydrofuran and acetonitrile were obtained from the in-house solvent purification system Innovative Inc. PureSolv®. Anhydrous dimethyl formamide and N,N-diisopropylethylamine were obtained from Sigma Aldrich equipped with Sure/Seal™. All non-aqueous reactions were carried out under a nitrogen atmosphere. Chloroform-d was dried over Linde 5Å molecular sieves or placed on CaCl2 before being distilled and stored on KOH prior to use in 1 H NMR experiments. For reactions under non-anhydrous conditions, the solvents used were HPLC quality and provided by Sigma Aldrich or Fisher. Water in aqueous solutions and used for quenching was deionised. Mixtures of solvents are quoted as ratios and correspond to a volume: volume ratio. Analytical thin layer chromatography was performed on Merck Kieselgel 60 F254 0.25 mm pre-coated aluminium plates. Product spots were visualised under UV light (λmax = 254 nm) or using a suitable stain. Flash chromatography was carried out using Merck Kieselgel 60 silica gel using pressure by means of head bellows or using disposable RediSepRf silica flash columns on an automated Biotage Isolera One system. Nuclear magnetic resonance spectra were obtained at 298 K (unless stated) using a Bruker AV500 spectrometer operating at 11.4 T (500 MHz for 1 H and 125 MHz 13 C) and JEOL ECA600ii operating at 14.1 T (150 MHz for 13 C) and NOESY spectra as stated. Infra-red spectra were obtained using a Bruker Alpha Platinum ATR where absorption maxima (νmax) are quoted in wavenumbers (cm -1 ) and only structurally relevant absorptions have been included. High Resolution Mass Spectra (HRMS) were recorded on a Bruker Daltonics Micro TOF using electrospray ionisation (ESI). Liquid Chromatography and Mass Spectrometry (LC-MS) was performed using an Agilent Technologies 1200 series LC and a Bruker HCT ultra ion-trap MS.

3, 5-diiodo-2,6-diaminopyridineamine
Compound prepared using minor adaptations to a previously described procedure. [4] To a solution of 2,6-diaminopyridine (0.99 g, 9.09 mmol) in dry DMF (28 mL) N-iodosuccineimide ( 4.50 g, 19.9 mmol) in DMF (20 mL) was added dropwise at -30 o C (cooled with dry ice) over 1 hr. After the addition was completed, the cooling bath was removed and the reaction mixture was stirred for 1 hr and then poured into ice-cold water and stirred for 30 min. Resulting precipitate was filtered and washed with water (2 x 20 mL) and pentane (2 x 15 mL) and dried in vacuum oven at 40 o C for 24 hrs. Product was obtained as grey solid with 94 % yield. 1

4-tert-Butyl-1H-imidazole-2-amine hydrochloride (I)
Tert-butyl 5-tert-1H-imidazol-2-yl-carbamate (750 mg, 3.10 mmol) was dissolved in 1M HCl in ethanol (30 mL) and refluxed for 16 hrs. The reaction was allowed to cool, concentrated and dried under pressure to give a colourless solid (400 mg, 2.28 mmol, 74%) 1  Measurements were carried out at 120K on an Agilent SuperNova diffractometer equipped with an Atlas CCD detector and connected to an Oxford Cryostream low temperature device using mirror monochromated Cu K radiation ( = 1.54184 Å) from a Microfocus X-ray source. The structure was solved by intrinsic phasing using SHELXT [6] and refined by a full matrix least squares technique based on F 2 using SHELXL2014. [7] The compound crystallised as colourless prisms from acetonitrile. The compound crystallised in a triclinic cell and was solved in the P1 space group, with two imidazolium cations and two chloride anions in the asymmetric unit. All non-hydrogen atoms were located in the Fourier Map and refined anisotropically. All carbon bound hydrogen atoms were placed in calculated positions and refined isotropically using a "riding model". All nitrogen bound hydrogen atoms were located in the Fourier Map and refined isotropically. Data and structure refinement given in Table 1 and was deposited via the joint CCDC/FIZ Karlsruhe deposition service, deposition number CCDC 1916237.

General procedure for sample preparation for NMR switching experiments Condition A
The mass of each component was calculated to make a final concentration of 5 mM in 0.6 mL of CDCl3. The required mass of the starting component(s) was dissolved in 0.6 mL of CDCl3. The sample was allowed to equilibrate for a minimum of ten minutes before acquisition. After acquisition the sample was protonated by the addition of 1 equivalent of 4M HCl in 1,4-dioxane solution directly to the sample tube. The sample was allowed to equilibrate for a minimum of ten minutes before acquisition. After acquisition the sample was transferred to a vial and deprotonated by the addition of excess basic NaHCO3 solution. The aqueous layer was separated, and the organic layer was dried and added to an NMR sample tube. The sample was allowed to equilibrate for a minimum of ten minutes before acquisition. Any additional components were added to the same sample when required and the protonation and deprotonation method was repeated as required.

Condition B
The mass of each component was calculated to make a final concentration of 5 mM in 0.6 mL of CDCl3. The required mass of the starting component(s) was dissolved in 0.6 mL of CDCl3. The sample was allowed to equilibrate for a minimum of ten minutes before acquisition. After acquisition the sample was protonated by the addition of 1 equivalent of TFA directly to the sample tube. The sample was allowed to equilibrate for a minimum of ten minutes before acquisition. After acquisition the sample was deprotonated by the addition of 1 or 3 equivalents of DABCO directly to the sample tube. The sample was allowed to equilibrate for a minimum of ten minutes before acquisition. Any additional