Supporting Information Dynamic Covalent Organocatalysts Discovered from Catalytic Systems through Rapid Deconvolution Screening

The first example of a bifunctional organocatalyst assembled through dynamic covalent chemistry (DCC) is described. The catalyst is based on reversible imine chemistry and can catalyze the Morita–Baylis–Hillman (MBH) reaction of enones with aldehydes or N-tosyl imines. Furthermore, these dynamic catalysts were shown to be optimizable through a systemic screening approach, in which large mixtures of catalyst structures were generated, and the optimal catalyst could be directly identified by using dynamic deconvolution. This strategy allowed one-pot synthesis and in situ evaluation of several potential catalysts without the need to separate, characterize, and purify each individual structure. The systems were furthermore shown to catalyze and re-equilibrate their own formation through a previously unknown thiourea-catalyzed transimination process.

mm). Concentration in vacuo was performed at ≈10 mbar and 30-40°C, drying at the high vacuum at ≈ 0.5 torr and room temperature. HRMS was performed at the Institute of Chemistry at University of Tartu, Estonia. ATR-IR spectroscopy was performed on a Thermo Scientific Nicolet iS10 spectrophotometer. A Bruker Avance DMX 500 MHz NMR spectrometer was used for recording 1 H-NMR, 13 C-NMR and 31 P-NMR, while a Bruker Avance 400 spectrometer and a Bruker Ascend 400 spectrometer (both 400 MHz) were used for the kinetic studies and 19 F NMR spectra. Chemical shifts are reported as δ values (ppm) relative to tetramethylsilane (Me 4 Si) with residual undeuterated CHCl 3 ( 1 H NMR δ 7.26, 13 C NMR δ 77.16) or DMSO ( 1 H NMR δ 2.50, 13 C NMR δ 39.52) as internal standards. Shifts for 31 P-and 19 F-NMR are reported relative to H 3 PO 4 and CFCl 3 , respectively. All J values are given in Hertz (Hz).

1-((±)-2-aminocyclohexyl)-3-(3,5-bis(trifluoromethyl)phenyl)thiourea (B) [1]
To a solution of (±)-trans-diaminocyclohexane (0.582 g, 5.1 mmol) in anhydrous CH 2 Cl 2 (10 mL) was added a solution of 3,5-bis(trifluoromethyl)phenyl isothiocyanate (0.93 mL, 1.38 g, 5.1 mmol) in CH 2 Cl 2 (10 mL) dropwise over 20 minutes under N 2 . After addition was finished, the solution was stirred for 3 h at room temperature, until all starting material was consumed according to TLC. The slightly yellow solution was concentrated in vacuo to yield the crude product as a yellow solid.  [2] To a solution of (±)-trans-diaminocyclohexane (2.14 g, 18.7 mmol) in anhydrous CH 2 Cl 2 (250 mL) was added phenyl isothiocyanate (1.50 mL, 1.69 g, 12.5 mmol) dropwise over 20 minutes at 0 °C under N 2 . The reaction was allowed to warm to room temperature overnight, and after 17 h the starting material was consumed according to TLC analysis. The yellow solution was concentrated in vacuo to yield the crude product as a yellow foam. Further purification by column chromatography (CH 2 Cl 2 /MeOH/NEt 3 40:1:0.1→9:1:0.1) yielded the product ( (±)-2-((phenoxycarbonyl)amino)cyclohexan-1-aminium iodide and phenyl((±)-2aminocyclohexyl)carbamate (D•HI) [3] To a solution of (±)-trans-diaminocyclohexane (1.713 g, 15 mmol) in isopropanol (15 mL) was added diphenyl carbonate (3.213 g, 15 mmol) and the resulting solution was stirred at reflux under N 2 . After 30 minutes, a white precipitate was formed and the solvent was removed in vacuo. To ensure complete isopropanol removal, the residual solid was redispersed in anhydrous toluene and the solvent removed. This process was repeated three times, followed by drying under high vacuum. The obtained white solid was mixed with methanesulfonic acid (0.97 mL, 1.44 g, 15.0 mmol) and the mixture turned homogenous upon heating. After stirring the red solution under argon at 110 °C for 1 h the mixture was allowed to cool to room temperature. Distilled H 2 O (15 mL) was slowly added, leading to precipitation. The phenol obtained as a byproduct was removed by extracting with ethyl acetate (3 x 15 mL, complete separation of the phases requires prolonged waiting time) and the aqueous phase was cooled to 0 °C. Addition of potassium iodide (2.74 g, 16.5 mmol) led to the precipitation of the product, which was filtered off, washed with cold H 2 O and dried in a vacuum desiccator to obtain the pure product as an off-white solid (2.66 g, 7.35 mmol, 49% yield). As the desired free base D decomposed upon prolonged storage, the compound had to be generated fresh from the ammonium iodide salt prior to use. The salt was suspended in CH 2 Cl 2 and equivalent volume of 10% aqueous K 2 CO 3 solution was added. The mixture was shaken until the solid had dissolved, after which the organic phase was removed, the aqueous phase extracted twice more with CH 2 Cl 2 and the combined organic phases were concentrated in vacuo to obtain the free base in quantitative yield. 1 H NMR (500 MHz, CDCl 3 /CD 3

2-(dimethylamino)benzaldehyde (2) [4]
To a suspension of 2-fluorobenzaldehyde (0.55 mL, 646 mg, 5.0 mmol) and K 2 CO 3 (691 mg, 5.0 mmol) in anhydrous DMF (10 mL), dimethylamine (5.6 M solution in EtOH, 1.2 mL, 6.7 mmol) was slowly added under N 2 atmosphere at room temperature. The reaction mixture was subsequently stirred at 110 °C for 20 h, at which point the starting material was consumed according to TLC. The bright yellow solution was diluted with ethyl acetate (12 mL) and distilled. H 2 O (20 mL) was added, and upon separation of the phases, the aqueous layer was extracted with ethyl acetate (2 x 20 mL) and the combined organic phases were washed with H 2 O (40 mL) and brine (40 mL). Drying with MgSO 4 , filtration and concentration yielded the crude product as a yellow oil. Purification by column chromatography (hexane/ethyl acetate 9:1 → 7:1) yielded the product as a yellow liquid (684 mg, 92% yield).

2-(diphenylphosphanyl)benzaldehyde (3) [5]
To a vigorously stirred solution of 2-bromobenzaldehyde (2.70 g, 14.6 mmol) and Pd(PPh 3 ) 4 (0.169 g, 0.146 mmol) in anhydrous toluene (50 mL) were sequentially added NEt 3 (2.0 mL, 1.47 g, 14.6 mmol) and diphenylphosphine (3.21 mL, 3.44 g, 19.0 mmol) under N 2 at room temperature. The bright red solution was stirred at reflux for 2.5 h, upon which time the starting material was consumed according to TLC analysis. The milky suspension was filtered through a glass filter, followed by washing with saturated aq. NH 4 Cl solution (3 x 40 mL) and brine (40 mL). The solvent was removed in vacuo, yielding the crude product as a yellow solid that was recrystallized twice from MeOH to obtain the pure product as yellow crystals (2.63 g, 62% yield).

General synthetic procedure for imine condensation reactions
To a mixture of amine (0.4 mmol) and pre-activated 4 Å molecular sieves (ca 0.8 g) in anhydrous CH 2 Cl 2 (10 mL) was added the aldehyde (0.4 mmol) under N 2 . The reaction mixture was slowly stirred overnight, and monitored by sampling of a small aliquot of the reaction mixture which was analyzed by NMR spectroscopy. Upon completion of reaction, the mixture was filtered through a pad of Celite and concentrated in vacuo to obtain the imine product with a typical purity >98%. (E)-N-cyclohexyl-1-phenylmethanimine (A1) [9] Colorless oil, 87% yield.

Phenyl ((±)-2-(((E)-2-(diphenylphosphanyl)benzylidene)amino)cyclohexyl)carbamate (D3)
The compound was synthesized from the hydroiodide salt of compound D by a modified general procedure. To a suspension of D•HI (74.6 mg, 0.206 mmol) and 4 Å molecular sieves (ca 400 mg) in anhydrous CH 2 Cl 2 (5 mL) was added the aldehyde 3 (58.1 mg, 0.20 mmol) followed by addition of freshly distilled NEt 3 (27.9 µL, 20.2 mg, 0.20 mmol). The milky solution immediately turned clear, and the mixture was stirred at rt for 19 h. The solution was then filtered through a pad of celite and washed with H 2 O (2x10 mL) and brine (10 mL). Drying with MgSO 4 , filtration and concentration yielded a white mixture which was recrystallized from MeOH to obtain the product as white crystals (52.7 mg, 52%).

Equilibrium manipulation
To probe the dynamic system equilibrium, the drying agent was initially removed by filtration. To the resulting solution, H 2 O (0.01 equiv.) and benzoic acid (0.05 equiv.) dissolved in anhydrous THF (60 µL) were added and the system was stirred under N 2 at room temperature. Subsequent composition controls by NMR spectroscopy according to the description above revealed that no significant change in system composition (only minor hydrolysis occurred) had taken place after either 24 or 48 h. Similar results were also reached when Sc(OTf) 3 (0.05 equiv.) was used as equilibration catalyst. The following system distribution was recorded after 48 h: To investigate whether the system had reached equilibrium, or if the imine exchange had been inhibited, salicylaldehyde (10.5 µL, 12.2 mg, 0.10 mmol) was added to the colorless solution. Instant imine exchange occurred, as evidenced by the appearance of a strongly yellow color corresponding to the newly generated salicylimines. The system was monitored by NMR spectroscopy and reached a new stable equilibrium within 48-72 hours, showing evident dynamic behavior. Figure S1. Model system obtained after direct condensation (top), after reequilibration with benzoic acid for 24 h (middle) and 96h after addition of salicylaldehyde to the reaction mixture (bottom).

Thiourea-catalyzed condensation and equilibration
To confirm that the system was capable of catalyzing the equilibration of its own members during the condensation phase, the final equilibrium investigated above was also approached in a single step by mixing the three aldehydes directly with the two amines. Cyclohexylamine (11.4 µL, 9.9 mg, 0.10 mmol), thiourea B (38.5 mg, 0.10 mmol), 2-dimethylaminobenzaldehyde (14.9 mg, 0.10 mmol), benzaldehyde (10.1 µL, 10.6 mg, 0.10 mmol) and salicylaldehyde (10.5 µL, 12.2 mg, 0.10 mmol) were dissolved in anhydrous THF (2.5 mL) with pre-activated 4 Å MS (300 mg), and the mixture was stirred slowly at room temperature for 20 h. The system generation was monitored by sampling of an aliquot (20 µL) of the reaction mixture, which was dissolved in anhydrous CDCl 3 (0.55 mL) and analyzed by NMR spectroscopy. The system composition was highly similar to that obtained through the two-step condensation/re-equilibration procedure, and no change in the ratio of imines was seen after an additional 48 h of stirring. These results thus implicate that equilibration is fast enough to occur during the imine formation phase at this concentration. Figure S2. Equilibrium composition from system with benzaldehyde, salicylaldehyde, 2dimethylaminobenzaldehyde, cyclohexylamine and thiourea-amine B, reached in a direct (top) or indirect two-step (bottom) manner.

H 2 O-dependence on thiourea-transimination
To investigate if dynamic exchange was still possible in absence of water, the equilibrated system generated in the two-step fashion described above was stirred with 4 Å MS for a further 24 h to ensure complete drying. Afterwards, p-nitrobenzaldehyde (15.1 mg, 0.10 mmol) was added to the equilibrated system and NMR measurements were performed after 4 and 24 h. No exchange was observed, suggesting that thiourea-catalyzed transimination is only possible while water is freely available in the system.

Further tests for transimination catalysis by thiourea
To investigate whether only the strongly H-bond-donating, electron-deficient thiourea B and its imine derivatives was capable of catalyzing the imine equilibration, or if the catalytic functionality could be extended to other H-bond donors, a series of investigations on the privileged catalyst structures B3, C3 and D3 found in the dynamic deconvolution process was performed. The imine (0.015 mmol) was mixed with p-nitrobenzaldehyde (4.5 mg, 0.03 mmol) in anhydrous CDCl 3 (0.55 mL) in an NMR tube, and the colorless mixture was regularly monitored by NMR. The residual NMR tube moisture was in this case enough to initiate transimination for both thioureaimines B3 and C3, as new characteristic NMR signals in the aldehyde and imine regions could be recorded after 4 h. After 24-72 h, an apparent equilibrium was established, yielding a product ratio of approximately 30:70 in favor of the newly formed imine. In contrast, no transimination was observed for the urethane D3 even after 24 h. This suggests that the thiourea scaffold is necessary for transimination to occur. Furthermore, it also concludes that nucleophilic catalysis by primary amines is at least not the sole reason for the self-correcting abilities of the systems observed during condensation.
Scheme S1. Thiourea-catalyzed transimination by compounds B3 and C3, and absence of catalytic activity for D3.

Experimental procedure for system generation
All aldehydes and amines were dissolved in anhydrous THF (0.5 mL) in an Eppendorf vial and the solution was transferred to a dry vial containing pre-activated 4 Å MS (300 mg, bead diameter ca 2 mm) under N 2 . The mixture was stirred slowly at room temperature under N 2 for 20 hours, after which time the equilibrated system was obtained. Tests for thiourea system equilibration were performed (as described in the section above), showing that the systems were at equilibrium after condensation. Also, sensitivity tests were performed, showing that similar system compositions (to within ±5%) were obtained regardless of the order of reagent addition or system member concentration (c=0.04 M, 0.10 M and 0.15 M yielded identical systems, within experimental error). The content of remaining aldehyde was never higher than 5%. The catalyst system could also be isolated by filtering the solution through a pad of celite, concentrating in vacuo, and drying at high vacuum for 2 h. The system composition did not change measurably during this procedure.
Experimental procedure for Morita-Baylis-Hillman reactions with dynamic system catalysis A dynamic system (see Table S1 for amounts, 1 equiv. = 0.075 mmol) was generated according to the description above. After stirring the system building blocks for 20 h under N 2 together with 4 Å MS (ca 300 mg), p-nitrobenzaldehyde (18.1 mg, 0.12 mmol) in anhydrous THF (0.120 mL) was added under N 2 , followed by addition of ethyl vinyl ketone (23.9 µL, 20.8 mg, 0.24 mmol). The mixture was stirred at room temperature under N 2 .

S12
Comment: Due to the inherent complexity associated with large mixtures of compounds, it was not possible to fully quantify each library component in these reduced libraries. Tests for equilibration were performed and all the mixtures were shown to be at equilibrium. Also, the imine peak region in the 1 H NMR spectra of the systems could be used as a spectral fingerprint to determine approximate compound concentrations. This data indicated that no major changes in the relative component distributions seemed to occur upon removal of any of the components in table S1. Figure S3. 1 H-NMR spectrum of the 16-component reference system, with imine region highlighted.

Kinetic analysis of Morita-Baylis-Hillman reactions
The reaction mixtures were prepared as above and monitored continuously by microsyringe sampling. An aliquot of the reaction mixture (30.0 µL) was added to CDCl 3 (0.550 mL) in an NMR tube, with PhSiMe 3 (0.020 µL/mL CDCl 3 ) as internal standard. NMR measurements were performed within 5 minutes, though control experiments indicated that the aliquot composition was stable for several hours in anhydrous CDCl 3 . Product formation was monitored by integrating the two characteristic peaks at around 5.66 and 6.00 and comparing to the integral of the internal standard. As the product concentration in the initial rate measurements were quite low, special precautions for the NMR measurements had to be undertaken. To increase sensitivity, 128 scans for each sample were performed, and the sample was manually processed using TopSpin software by Fourier transforming the FID twice, then performing an automated baseline correction, followed by manual final calibration to make sure the relevant signals were properly aligned. Each integration event was repeated three times and averages are reported.

MBH Catalyst evaluation experiments
To evaluate singular catalyst performance, the 16 different catalysts from the system were synthesized individually and MBH reactions were carried out in situ: To a vial under N 2 , equipped with 4 Å molecular sieves (200 mg), was added aldehyde (0.02 mmol) and amine (0.02 mmol) by mixing stock solutions (0.1 M) of the respective compound in anhydrous THF. The resulting mixtures were left with slow stirring for 20 h under N 2 , after which p-nitrobenzaldehyde (15.1 mg, 0.1 mmol) in anhydrous THF (100 µL) and ethyl vinyl ketone (30 µL, 25.3 mg, 0.3 mmol) were added. The solutions were stirred for an additional 24 h, and reaction progress was monitored by removal of an aliquot that was evaluated by NMR spectroscopy with 1,4-dimethoxybenzene as internal standard.