Dynamic covalent chemistry with azines

Dynamic covalent chemistry is used in many applications that require both the stability of covalent bonds and the possibility to exchange building blocks. Here we present azines as a dynamic covalent functional group that combines the best characteristics of imines and acylhydrazones. We show that azines are stable in the presence of water and that dynamic combinatorial libraries of azines and aldehydes equilibrate in less than an hour.


Reagents and solvents
All commercially available reagents were purchased from Sigma Aldrich, Merck, or Alfa Aesar and used without further purification. Deuterated solvents used for exchange reactions were used as such without being dried. Where mentioned, the drying of solvent was done with molecular sieves 3Å, for at least 2 d.

Instruments and conditions
NMR spectra were recorded on Varian VNMRS 400 ( 1 H: 400 MHz, 13 C: 100 MHz) equipped with a OneNMR probe (25°C, 2.6 µs pulse, 3 s acquisition time, 3 s relaxation delay). Chemical shifts (δ) are given in parts per million (ppm) and referenced to the corresponding residual solvent peak. Coupling constants (J) are denoted in Hertz (Hz). Abbreviations indicating multiplicity were used as follows: s = singlet, d = doublet, dd = double doublet, m = multiplet.

Synthesis and characterisation of azine compounds
The syntheses of azines were conducted via one general procedure: the aldehyde and hydrazine monohydrate were mixed in a 2:1 ratio, in ethanol. In a typical experiment, an immediate colour change is observed. The completion of the reaction was monitored by thinlayer chromatography (silica gel; ethyl acetate:n-heptane, 4:6). The product was collected by suction filtration, washed with ethanol, and dried under vacuum. All the synthesized azines were yellow solids. 1 Benzaldehyde (1.91, 18.8 mmol, 2 eq.) was added dropwise to a solution of hydrazine monohydrate (0.48 mL, 9.4 mmol, 1 eq.) in ethanol (25 mL). The mixture was heated at 50 °C for 20 minutes and then allowed to cool down to room temperature. A precipitate was formed immediately after the mixture was cooled down. The precipitate was filtered and washed with ethanol to give a yellow solid in 84% isolated yield (1.65 g, 7.9 mmol). 1 2 2-Chlorobenzaldehyde (1mL, 8.9 mmol, 2 eq.) was added dropwise over a solution of hydrazine monohydrate (0.22 mL, 4.45 mmol, 1 eq.) in ethanol (25 mL) and stirred for 30 minutes. The mixture was stirred for 2 minutes and precipitated. The precipitate was filtrated and washed with ethanol to give an as yellow solid in 89% isolated yield ( 3 4-(trifluoromethyl)benzaldehyde (0.78 mL, 5.7 mmol, 2 eq.) was added dropwise over a solution of hydrazine monohydrate (0.15 mL, 2.9 mmol, 1 eq.) in ethanol (20 mL) and stirred for one hour, until precipitation of the compound. The precipitate was filtrated and washed with ethanol to give an yellowish solid in 85% isolated yield (0.85 g, 2.4 mmol). 1 3 4-Methoxybenzaldehyde (1.26 mL, 11 mmol, 2 eq.) was added dropwise over a solution of hydrazine monohydrate (0.27 mL, 5.5 mmol, 1 eq.) in ethanol (25 mL) and stirred for one hour, until precipitation of the compound. The precipitate was filtrated and washed with ethanol to give an yellowish solid in 98% isolated yield (1.45 4 Vanilin (50 g, 328.6 mmol) was dissolved in 100 ml ethanol at 50 degrees and hydrazine monohydrate (8.23 mL, 164.3 mmol, 0.5 eq.) was added dropwise over 15 minutes. The resulting mixture was stirred for 1 hour, cooled to room temperature and put in the fridge overnight. The resulting crystals were filtered on a sintered glass funnel and washed with cold ethanol (2x 30 ml) to yield the desired azine as yellow crystals (97.7 %, 48.   1 p-Tolualdehyde (1 mL, 8.3 mmol, 2 eq) was added dropwise over a solution of hydrazine monohydrate (0.20 mL, 4.1 mmol, 1 eq) in ethanol (25 mL) and stirred for 30 minutes. A precipitate was formed immediately. The precipitate was filtrated and washed with ethanol to give an as yellow solid in 82% isolated yield (0.8 g, 3.    5 2-Bromobenzaldehyde (0.315 mL, 2.7 mmol, 2 eq) was added dropwise over a solution of hydrazine monohydrate (0.067 mL, 1.35 mmol 1 eq) in 25 mL ethanol, and stirred for 30 minutes. A precipitate was formed immediately. The precipitate was filtrated and washed with ethanol to give an as yellow solid in 85% isolated yield (0.42 g, 1.15 mmol).

Calculation of molar ratios
The percentage of the molar ratios was calculated as described in equations 1-3. For instance, in a mixture of aldehydes X and Y and azines XX, YY, and XY, the area (A) of the selected peak corresponding to each molecule is divided by the number (N) of protons contributing to the integral (1 for aldehydes, 2 for azines). After that, the total area (Atotal) was calculated as follows:

Azines stability in acidic conditions and aqueous solutions
For the formation of AA in acid conditions were mixed: A (2 eq., 28 mM), hydrazine hydrate (1 eq., 14 mM), and trifluoroacetic acid (TFA) (1 eq.). For the hydrolysis reactions of AA in acid conditions were mixed: AA (1 eq., 14 mM) and TFA (1 eq.). The DMSO-d6/D2O ratio of the solvent was varied to evaluate the azine stability in the presence of water. Table S1 summarizes the most relevant results.

Hydrolysis of CC and DD compounds
For the hydrolysis reactions of CC and DD in acid conditions were mixed: CC or DD (1 eq., 7 mM) and TFA (1 eq.) in DMSO-d6/D2O (95/5). Table S2 summarizes the most relevant results.    Table S3 summarizes the mol % of E (aldehyde) EE (azine) and EH (hydrazone) present at each pD value at the specified times.

Proof of reversibility
The equilibrium from different starting points was assessed to demonstrate that azine exchange is reversible. For this study, one equivalent of azine (7 mM), two equivalents of aldehyde 14 mM), and TFA (7 mM) in CDCl3 were mixed in glass scintillation vials (2 mL). The progress of the reaction was monitored with 1 H NMR spectroscopy after 1 h, 24 h, and 7 days, showing no significant changes in the ratios over time, as demonstrated by the spectra after 1 h and 7 days (Figures S17-S18). The protons of the methine group (-CH=) from the corresponding azines and aldehydes were integrated and assigned using their corresponding letters.

Acylhydrazone control experiments
Acylhydrazone (4) and benzaldehyde (2) were mixed in equimolar amounts (7 mM, CDCl3) and were subjected to the same exchange conditions as used for azine exchange (TFA, 1 eq.). The library was monitored by the ratio of aldehydes (2 and 6) until equilibrium was reached.

Azine exchange reactions
In an typical experiment, azine AA (initial concentration: 1 eq., 7 µmol, 14 mM), aldehyde B (initial concentration: 1 eq., 7 µmol, 14 mM) were mixed in the presence of 1 equivalent (7 µmol, 14 mM) or 0.1 equivalent (0.7 µmol, 1.4 mM) of additive (trifluoroacetic acid, ptoluenesulfonic acid, acetic acid, formic acid) in deuterated solvent solutions (Vtotal: 500 µL). The reactions were carried out in 2 mL glass vials. After the addition of all reagents, the vials were shaken properly. The reaction progress was monitored by 1 H NMR spectroscopy after 1 h, 24 h, and 7 or 10 days. The protons of the methine group (-CH=) of the azines and aldehydes were integrated and assigned using their corresponding letters.

DCLs freezing
The possibility of locking or freezing the azine exchange reaction was explored using 2 equivalents of K2CO3 (28 mM) after the reaction started. For each sample, identical amounts of azine AA and aldehyde B (14 mM) and 1 equivalent of TFA in DMSO-d6/D2O (90:10) were mixed in 2 mL glass scintillation vials. The base was added after 3 min, 20 min, and 40 min, respectively. In addition, one sample without base was studied for comparison. All the samples were examined by 1 H NMR spectroscopy five hours after adding the acid.  CD3CN) spectra of the three above experiments after 24 hours, zoomed in at the aromatic region where the aldehyde carbonyl CH signals appear. The effect of substituents on Azine stability during exchange is exemplified by comparison of relative amounts of aldehydes present at equilibrium for the above three experiments. For electron withdrawing substituents (spectra 1 and 3) more benzaldehyde (A) is present in solution, indicating that azines formed from the aldehydes featuring electron withdrawing substituents (B and C) are more favoured. Accordingly, the electron donating substituent (spectrum 2) shows more anisaldehyde (D) than benzaldehyde (A) present at equilibrium, indicating that azines from benzaldehyde are more favoured in this case.

Azine metathesis reactions
In a typical experiment, the azines AA and BB or CC and DD (2 mM, 1 eq. of each one) were dissolved DMSO-d6/H2O and mixed in a 2 mL scintillation vial. The reaction proceeded by the addition of 1 equivalent of TFA, (Vtotal: 500 µL).

Appendix 1
As a means of comparison to the stability studies carried out by Kalia and Raines 7 on oximes and hydrazones, azine PP was synthesized and its hydrolysis studied by a procedure adapted from the original one used for the oximes and hydrazones due to the solubility of azine PP.

Pivaldehyde azine (PP)
Pivaldehyde (1.26 mL, 11.61 mmol) was dissolved in 20 ml ethanol and hydrazine monohydrate (290 µL, 5.8 mmol, 0.5 eq.) was added. The resulting mixture was stirred for 2 hours, after which solvent was removed in vacuo until and oily solid was obtained. This solid was run through a plug of silica gel (3 cm in height) using pentane/Et2O (95/5

Hydrolysis studies:
A 100 mM stock solution of PP in DMSO-d6 was added to a buffered D2O solution containing formaldehyde (6 mM phosphate buffer, pD 5-9, 10 mM formaldehyde) to give a 1 mM solution of PP. The resulting solutions were monitored by 1 H NMR spectroscopy to yield kinetic traces for the hydrolysis of the azine from which first order rate constants and halflifes could be obtained. The tBu signals at 1.10 to 0.72, originating from the azine, hydrazone, aldehyde and hydrated aldehyde were integrated and used as an internal reference. The integral of the azine azomethine CH signal was compared to the original amount of PP (based on the total tBu signal) to evaluate the degree of hydrolysis. The degree of hydrolysis versus time is plotted to give the kinetic trace of the hydrolysis reaction shown in Figure S64. These kinetic traces were then fitted to the equation below: Half-lives were calculated by the following equation: t 1/2 = 0.692 k Table S5 gives the first order rate constants as well as half-lifes at each pD. A sample of NMR spectra at selected times for the hydrolysis of azine PP at pD 5 can be seen in Figure S65