Negative cooperativity in the formation of H-bond networks involving primary anilines

Networks of H-bonds can show non-additive behaviour, where the strength of one interaction perturbs another. The magnitude of such cooperative effects can be quantified by measuring the effect of the presence of an intramolecular H-bond at one site on a molecule on the association constant for formation of an intermolecular H-bond at another site. This approach has been used to quantify the cooperativity associated with the interaction of a primary amine with two H-bond acceptors. A series of compounds that have an intramolecular H-bond between an aniline NH2 group and a pyridine nitrogen were prepared, using polarising substituents on the pyridine ring to vary the strength of the intramolecular H-bond. The presence of the intramolecular interaction was confirmed by X-ray crystallography in the solid state and NMR spectroscopy in n-octane solution. UV-vis absorption titrations were used to measure the association constants for formation of an intermolecular H-bond with tri-n-butyl phosphine oxide in n-octane. Electron-donating substituents on the pyridine ring, which increase the strength of the intramolecular H-bond, were found to decrease the strength of the intermolecular H-bond between the aniline and the phosphine oxide. The results were used to determine the H-bond donor parameters for the anilines, α, and there is a linear relationship between the values of α and the H-bond acceptor parameter of the pyridine group involved in the intramolecular H-bond, β. The slope of this relationship was used to determine the cooperativity parameter (κ = −0.10), which quantifies the negative allosteric cooperativity between the two H-bonding interactions. Calculated molecular electrostatic potential surfaces of the anilines quantitatively reproduce the experimental result, which suggests that effects are electrostatic in origin, either due to polarisation of the NH bonds or due to secondary electrostatic interactions between the two H-bond acceptors.


Materials and methods
All reagents were purchased from commercial sources (Sigma Aldrich UK, Acros, Tokyo Chemical Industry, Alfa Aesar, Manchester Organics and FluoroChem) and were used as received without any further purification unless stated.Dry solvents were obtained by means of a Grubbs solvent purification system.Flash chromatography was done with an automated system (Combiflash Companion) using prepacked cartridges of silica (50 µm PuriFlash® column) or basic alumina (45 μm PuriFlash® column) The LC-MS analysis of samples was performed using Waters Acquity H-class UPLC coupled with a single quadrupole Waters SQD2.ACQUITY UPLC CSH C18 Column, 130 Å, 1.7 μm, 2.1 mm X 50 mm was used as the UPLC column for all samples.The conditions of the UPLC method are as follows: Solvent A: Water +0.1% Formic acid; Solvent B: Acetonitrile +0.1% Formic acid; Gradient of 0-2 minutes 5% -100%B + 1 minute 100% B with re-equilibration time of 2 minutes.Flow rate: 0.6 ml/min; column temperature of 40°C; injection volume of 2 μL.The signal was monitored with MS-ES+, MS-ES-, and UV-vis absorption at 254 nm or at 290 nm.

UV-vis Absorption and NMR Titrations
UV-vis titrations were carried out on an Agilent Cary 60 UV-Vis spectrophotometer, using standard titration protocols.A sample of the host (1-9) was prepared at a known concentration (typically between 0.02-0.06mM) in n-octane.The UV-vis spectrum of the free host (2 mL) was recorded.The guest was dissolved (Bu3PO) in 2 mL of the host solution at a known concentration.Aliquots of the guest solution were successively added to the cuvette, and the UV-vis absorption spectrum was recorded after each addition.The UV-vis absorption spectra were analysed using a purpose-built Python script to fit the changes in the absorption at fixed wavelengths to either a 1:1 or a 1:2 binding isotherm by optimizing the association constant and absorption of the free and bound host.
NMR titrations were carried out on a Bruker 500 MHz spectrometer, using standard titration protocols.A sample of the host was prepared at a concentration of 0.3-0.8mM in n-octane.The NMR spectrum of the host solution (0.6 mL) was recorded.The guest (Bu3PO) was dissolved in 2 mL of the host solution at a known concentration.Aliquots of the guest solution were successively added to the NMR sample tube containing the host solution, and the 1 H NMR spectrum was recorded after each addition.For the dilution experiment, 0.6 mL of the host in n-octane was placed in an NMR tube and aliquots of a solution pure n-octane was added with the spectrum being recorded after every addition.The NMR spectra were analysed using a purpose-built Python script to fit the changes in the chemical shifts for different protons to either a 1:1 or a 1:2 binding isotherm by optimizing the association constant and shifts of the free and bound host.

Results from fitting of titration data UV-vis absorption titrations
Compound X NMe2 134 ± 3 75 3 ± 3 25 Table S.3 -Association constants of for the 1:1 and 1:2 complexes of numbered compounds with Bu3PO as guest and the maximum percentage of host that was present as a given 1:n complex during the course of the titration.

DFT calculations
The crystal structure of 9 was used as the starting point for the calculations for 2 to 7, with the alkyl chain replaced with a proton and the X group on the pyridine ring changed accordingly.The resulting structures were then used as a basis for DFT optimisation of the newly introduced group with the other coordinates from the crystal structure constrained.The DFT was using a 6-31G* basis set, 5-7 a B3LYP functional [8][9][10][11] and a nonrelativistic Hamiltonian (gas phase).The results were used to calculate the molecular electrostatic potential (MEP) on the 0.0104 e bohr −3 electron density isosurface in NWChem 7. The MEP was then converted to atomic interaction point (AIP) H-bond parameters using purpose-built Python and JavaScript programmes.

2
Scheme 3 -Synthesis of 9. Synthesis of starting material reported previously by our group.2

Figure S. 12 -
Figure S.12 -Partial 500 MHz 15 N-1 H HSQC spectrum of compound 4 (0.31 mM in n-octane) recorded with the 1-bond N-H coupling constant set to 90 Hz.The cross-peak is between the signal due to the aniline protons and the signal due to the aniline nitrogen.

Figure S. 13 -
Figure S.13 -Partial 500 MHz 15 N-1 H HMBC spectrum of compound 4 (0.31 mM in n-octane) recorded with the long-range N-H coupling constant set to 8 Hz.The cross-peak is between the signal due to the aniline protons and the signal due to the pyridine nitrogen.

Figure S. 14 -
Figure S.14 -Structure of the host 1 with protons labelled for NMR assignment.

Figure S. 15
Figure S.15 -UV-vis absorption spectra for the titration of Bu3PO into 1 (0.0503 mM in n-octane, at 298 K).The UV-vis spectrum of the host 1 and the final point of the titration are reported in black and in red, respectively.

Figure S. 16 -
Figure S.16 -The fit of the absorbance at selected wavelengths to a 1:2 binding isotherm for the titration of Bu3PO into 1 (0.0503 mM in n-octane, at 298 K).

Figure S. 17 -
Figure S.17 -Stack plot for the 500 MHz 1 H NMR dilution of 1 in n-octane at 298 K ranging from 0.414 mM (bottom) to 0.124 mM (top).

Figure S. 18 -
Figure S.18 -Stack plot for the 500 MHz 1 H NMR titration of Bu3PO into 1 (0.414 mM) in n-octane at 298 K.

Figure S. 19 -
Figure S.19 -Fit of the chemical shift changes to a 1:2 binding isotherm for the NMR titration of Bu3PO into 1 (0.310 mM) in n-octane at 298 K.

Figure S. 21
Figure S.21 -UV-vis absorption spectra for the titration of Bu3PO into 2 (0.0562 mM in n-octane, at 298 K).The UV-vis spectrum of the host 2 and the final point of the titration are reported in black and in red, respectively.

Figure S. 22 -
Figure S.22 -The fit of the absorbance at selected wavelengths to a 1:2 binding isotherm for the titration of Bu3PO into 2 (0.0562 mM in n-octane, at 298 K).

Figure S. 24
Figure S.24 -UV-vis absorption spectra for the titration of Bu3PO into 3 (0.0479 mM in n-octane, at 298K).The UV-vis spectrum of the host 3 and the final point of the titration are reported in black and in red, respectively.

Figure S. 25 -
Figure S.25 -The fit of the absorbance at selected wavelengths to a 1:2 binding isotherm for the titration of Bu3PO into 3 (0.0254 mM in n-octane, at 298 K).

Figure S. 26 -
Figure S.26 -Structure of the host 4 with protons labelled for NMR assignment.

Figure S. 27
Figure S.27 -UV-vis absorption spectra for the titration of Bu3PO into 4 (0.0496 mM in n-octane, at 298K).The UV-vis spectrum of the host 4 and the final point of the titration are reported in black and in red, respectively.

Figure S. 28 -
Figure S.28 -The fit of the absorbance at selected wavelengths to a 1:2 binding with guest absorption isotherm for the titration of Bu3PO into 4 (0.0496 mM in n-octane, at 298 K).

Figure S. 29 -
Figure S.29 -Stack plot for the 500 MHz 1 H NMR dilution of 4 in n-octane at 298 K ranging from 0.310 mM (bottom) to 0.0715 mM (top).

Figure S. 30 -
Figure S.30 -Stack plot for the 500 MHz 1 H NMR titration of Bu3PO into 4 (0.310 mM) in n-octane at 298 K.

Figure S. 31 -
Figure S.31 -Fit of the chemical shift changes to a 1:2 binding isotherm for the NMR titration of Bu3PO into 4 (0.310 mM) in n-octane at 298 K.

Figure S. 33
Figure S.33 -UV-vis absorption spectra for the titration of Bu3PO into 5 (0.0459 mM in n-octane, at 298 K).The UV-vis spectrum of the host 5 and the final point of the titration are reported in black and in red, respectively.

Figure S. 34 -
Figure S.34 -The fit of the absorbance at selected wavelengths to a 1:2 binding with guest absorption isotherm for the titration of Bu3PO into 5 (0.0459 mM in n-octane, at 298 K).

Figure S. 36
Figure S.36 -UV-vis absorption spectra for the titration of Bu3PO into 6 (0.0472 mM in n-octane, at 298 K).The UV-vis spectrum of the host 6 and the final point of the titration are reported in black and in red, respectively.

Figure S. 37 -
Figure S.37 -The fit of the absorbance at selected wavelengths to a 1:2 binding with a guest absorption isotherm for the titration of Bu3PO into 6 (0.0472 mM in n-octane, at 298 K).

Figure S. 38 -
Figure S.38 -Structure of the host 7 with protons labelled for NMR assignment.

Figure S. 39
Figure S.39 -UV-vis absorption spectra for the titration of Bu3PO into 7 (0.0458 mM in n-octane, at 298 K).The UV-vis spectrum of the host 7 and the final point of the titration are reported in black and in red, respectively.

Figure S. 40 -
Figure S.40 -The fit of the absorbance at selected wavelengths to a 1:2 binding isotherm for the titration of Bu3PO into 7 (0.0458 mM in n-octane, at 298 K).

Figure S. 41 -
Figure S.41 -Stack plot for the 500 MHz 1 H NMR dilution of 7 in n-octane at 298 K ranging from 0.105 mM (bottom) to 0.0243 mM (top).Quartet at 5.34 and singlet at 6.53 are impurities in the solvent. 34

Figure S. 42 -
Figure S.42 -Stack plot for the NMR titration of Bu3PO into 7 (0.468 mM) in n-octane at 298 K.

Figure S. 43 -
Figure S.43 -Fit of the chemical shift changes to a 1:2 binding isotherm for the NMR titration of Bu3PO into 7 (2.71 mM) in n-octane at 298 K.

Figure S. 45
Figure S.45 -UV-vis absorption spectra for the titration of Bu3PO into 8 (0.0248 mM in n-octane, at 298 K).The UV-vis spectrum of the host 8 and the final point of the titration are reported in black and in red, respectively.

Figure S. 47 -
Figure S.47 -Stack plot for the 500 MHz 1 H NMR dilution of 8 in n-octane at 298 K ranging from 0.825 mM (bottom) to 0.108 mM (top).

Figure S. 48 -
Figure S.48 -Stack plot for the 500 MHz 1 H NMR titration of Bu3PO into 8 (0.825 mM) in n-octane at 298 K.

Figure S. 49 -
Figure S.49 -Fit of the chemical shift changes to a 1:1 binding isotherm for the NMR titration of Bu3PO into 8 (0.825 mM) in n-octane at 298 K.

Figure S. 51
Figure S.51 -UV-vis absorption spectra for the titration of Bu3PO into 9 (0.0253 mM in n-octane, at 298 K).The UV-vis spectrum of the host 9 and the final point of the titration are reported in black and in red, respectively.

Figure S. 52 -
Figure S.52 -The fit of the absorbance at selected wavelengths to a 1:2 binding isotherm for the titration of Bu3PO into 9 (0.0503 mM in n-octane, at 298 K).

Figure S. 53 -
Figure S.53 -Stack plot for the 500 MHz 1 H NMR dilution of 9 in n-octane at 298 K ranging from 0.844 mM (bottom) to 0.110 mM (top). 40

Figure S. 54 -
Figure S.54 -Stack plot for the 500 MHz 1 H NMR titration of Bu3PO into 9 (0.844 mM) in n-octane at 298 K.

Figure S. 55 -
Figure S.55 -Fit of the chemical shift changes to a 1:2 binding isotherm for the NMR titration of Bu3PO into 9 (0.844 mM) in n-octane at 298 K.

12
6 -Comparison of the H-bond donor parameters for the anilines 2-7 calculated using DFT and those measured using UV-vis absorption titrations.

Table S .
2 -Summary of the crystal and refinement details.

Table S . 4
TableS.5 -A table of the shift change for the protons above in Figure S.57 from the limiting shift upon 1:1 complexation to the limiting shift upon 1:2 complexation with Bu3PO.Protons f, g and h don't exist in species 1 as the pyridyloxy group is replaced by a methyl group.Compound 8 not included as it only shows 1:1 complexation in NMR titrations.
-A table of the shift change for the protons above in Figure S.57 from the free species to the limiting shift upon 1:1 complexation with Bu3PO.Protons e, f, g and h don't exist in species 1 and 8 as the pyridyloxy group is replaced by a methyl group.