A tetrahedral molecular cage with a responsive vertex† †Electronic supplementary information (ESI) available. See DOI: 10.1039/c9sc02047k

The first system to combine the orthogonality of alkyne metathesis and dynamic imine exchange is reported for the preparation of a molecular cage with a reversibly removable vertex.


1-bromo-4-(prop-1-yn-1-yl)benzene (8):
In a flame-dried 500 mL round bottom flask equipped with stirbar, 1-bromo-4-iodobenzene (15 g, 53 mmol, 1 equiv), Pd(PPh 3 ) 2 Cl 2 (1.86 g, 2.65 mmol, 5 mol%) and CuI (1 g, 5.3 mmol, 10 mol%) were combined and the reaction flask was backfilled three times with argon. Toluene (130 mL) and Et 3 N (130 mL) were then added and the mixture was stirred briefly. The reaction mixture was then cooled to -78 o C using a dry ice/acetone bath and propyne gas was added to the reaction mixture. (Note: excess propyne was used. The propyne gas was bubbled through the mixture and into the headspace of the reaction from a small lecture bottle via an 18-gauge metal needle with an argon balloon as pressure release. The gas was added in approximately 4 ten-second bursts.) The reaction mixture was allowed to stir at -78 o C for 5 minutes after which the bath was removed, and the mixture was stirred at RT for 24 hours. The dark black reaction mixture was then filtered through a short plug of silica gel with Et 2 O and DCM and concentrated directly onto silica gel. The reaction was purified via silica gel flash chromatography with a gradient of 2 CV hexane to 2 CV 1% EtOAc/Hexane (CV = column volume) to afford 2 as a clear yellow oil (8.38 g, 42.9 mmol, 81% yield).

4-(2,4,6-triethyl-3,5-bis(4-(prop-1-yn-1-yl)benzyl)benzyl)aniline (2):
In an oven-dried 40 mL reaction vial equipped with a septum-cap and a stirbar, magnesium turnings (126 mg, 5.2 mmol, 4 equiv) were dried under vacuum with a heat gun. After cooling to room temperature and backfilling with argon, THF (6.5 mL) was added followed by 50 L of 1,1dibromoethane and the mixture was stirred for five minutes. 4-Bromo-N,N-bis(trimethylsilyl)aniline (0.75 mL, 2.65 mmol, 2 equiv, Aldrich) was then added in one portion and the mixture was stirred briefly at room temperature then at 70 o C for 1 hour. In a separate oven-dried reaction vial, 9 (665 mg, 1.3 mmol, 1 equiv) and CuI (25 mg, 0.115 mmol, 0.1 equiv) were combined and backfilled twice with argon. THF (8.5 mL) was then added, followed by the Grignard solution and the mixture was stirred overnight at 70 o C. The reaction was quenched with 1M HCl (10 mL) and stirred vigorously for one hour. After diluting with 50 mL of 1M NaOH and extracting 3x with EtOAc, the combined organic layers were washed with brine, dried with Na 2 SO 4 and filtered through a short silica plug. The mixture was then concentrated onto silica gel and purified via silica gel flash chromatography using a gradient of 10%-15%-20%-25%-30%-40%-50%-100% Et 2 O/pentane. The purified product was isolated as an off-white foam (485.3 mg, 0.92 mmol, 71% yield).   The aqueous layer was separated and the organic layer extracted x3 with Et 2 O, then dried with Na 2 S 2 O 4 , filtered and concentrated onto silica gel. The crude reaction was purified using silica gel flash chromatography using an acetone/hexanes gradient. Compound 5 was isolated as a white solid (625.18 mg, 1.21 mmol, 54% yield).

(1E,1'E,1''E)-1,1',1''-(((2,4,6-triethylbenzene-1,3,5-triyl)tris(methylene))tris(benzene-4,1diyl))tris(N-methylmethanimine) (6):
In a 1-dram vial, aldehyde 1 (25 mg, 43 mol, 1 equiv) was dissolved in CHCl 3 and a small drop of glacial acetic acid was added. To the solution was then added methylamine (2M in THF, 0.65 mL, 1.29 mmol, 30 equiv) and the reaction was stirred overnight at room temperature. The mixture was then filtered through a small plug of neutral alumina, and dried with Na 2 SO 4 . After filtration, the mixture was concentrated to dryness and dried under vacuum to afford 6 as an off-white foam (31.6 mg, quantitative yield).  35 mL) was then added which resulted in the formation of a white precipitate. The mixture was vortexed thoroughly and the solid was collected via vacuum filtration through a 0.45 m nylon membrane filter. The solid was washed thoroughly with methanol, followed by a 30% CHCl 3 /MeOH solution and finally methanol. The solid was dried under vacuum to afford an off-white solid which was re-dissolved in minimal CHCl 3 . Ethanol (ca. 35 mL) was added to re-precipitate. The solid was again collected via vacuum filtration and washed thoroughly with ethanol (ca. 50 mL), followed by a 30% CHCl 3 /EtOH solution and finally ethanol. The material was dried under vacuum to afford 6 as an off-white powder (382.3 mg, 0.188 mmol, 83% yield). We recommend storing 3 long term in a -20 o C freezer over Drierite. Note: NMR/MALDI-analysis indicate the presence of a bis-imine isomer and trace residual 2. The purity of 3 was determined to be 96% by 1   General Procedure for Synthesis of Imine/Alkyne Cage (4): In an argon-filled glovebox, precursor 3 (100 mg, 49 mol, 1 equiv) was weighed into an oven-dried 40 mL reaction vial equipped with a septum cap. 5 Å MS (300 mg, 1g/mmol of propynyl functionality) were then added, followed by 14 mL of CCl 4 . In a separate oven-dried 2-dram vial, molybdenum precatalyst A (3.3 mg, 4.9 mol, 10 mol%) and triphenol ligand B (2.3 mg, 4.9 mol, 10 mol%) were combined in 2 mL CCl 4 . Both solutions were stirred for 30 minutes at room temperature. To the solution of precursor 3 was then added the catalyst solution via Pasteur pipette, and the reaction mixture (3 mM total in 3) was capped and stirred at 40 o C in an aluminum block overnight (ca. 23 hours). The reaction was removed from the glovebox and filtered through a small plug of neutral alumina with excess CHCl 3 . The mixture was concentrated in-vacuo and the resulting solid was dissolved in CHCl 3 and filtered through a small plug of Celite into a 40 mL vial. The solution was then concentrated and the residue dissolved in minimal (ca. 2 mL) CHCl 3 . Excess ethanol (ca. 35 mL) was then added to afford a fine white precipitate. The solid was collected via filtration through a 0.45 m nylon membrane filter and washed with ethanol and a solution of 30% CHCl 3 /EtOH and finally methanol. The resulting off-white solid was dried under high-vacuum to afford cage 4 as a beige solid (80% average yield, 96% purity by GPC, average of two runs). Run 1: 75.9 mg, 0.0406 mmol, 83% yield; Run 2: 69.5 mg, 0.0372 mmol, 76% yield.  158.22, 149.05, 144.60, 141.67, 141.59, 141.45, 141.26,  139.09, 134.70, 134.04, 133.59, 133.45, 131.88, 129.13, 128.37, 128.10, 127.66, 121.61, 121.10,  89.31, 35.17, 34.82, 34.55, 23.93, 23.81, 15.13, 15.10, 14.99.  Figure S1. MALDI-TOF spectrum of isolated cage 4 (positive ion mode, DCTB matrix). Note the observed mass of a dimeric product (3743.93). We propose that the observed mass at m/z 2122.12 is an adduct of cage 4 with the DCTB matrix (see ref. 5).

S3. Reaction Time-Course Experiments
Following the general procedure for synthesis of cage 4, precursor 3 (50 mg, 24.5 mol, 1 equiv) was reacted with [Mo] catalyst A (1.6 mg, 2.45 mol, 10 mol%) and ligand B (2.45 mol, 1.1 mg, 10 mol%) in 8.2 mL CCl 4 (3 mM) at 40 o C in an argon-filled glovebox. Aliquots (200 µL) were removed at the appropriate timepoints and placed in sealed, air-filled ½ dram vials to quench. The aliquots were dried under high vacuum to remove CCl 4 and diluted with 1 mL of THF. After filtration through a 0.45 µm syringe filter, the samples were analyzed by GPC to characterize the product distribution. S9 Figure S2. GPC traces of precursor 3 and isolated cage 4. Percent purity of 4 was estimated by dividing the peak integration of higher molecular weight species (~17 min) by the total peak integration from retention time of 16-18 minutes. Traces were normalized by area. Figure S3. GPC time course of the AM reaction of precursor 3. Note the formation of higher molecular weight products which are consumed over the course of the reaction. Traces normalized by area.

S4. Cage Disassembly and Reassembly Experiments A. Cage Disassembly/Reassembly Using Scandium Triflate
Cage Disassembly: In a 1-dram reaction vial, cage 4 (29.7 mg, 15.8 µmol, 1 equiv) was dissolved in CDCl 3 (3 mL, 5 mM) and stirred for five minutes to dissolve. In an argon-filled glovebox, Sc(OTf) 3 (6.2 mg, 12.7 µmol, 80 mol%, 4 mol% relative to MeNH 2 ) 4 was weighed into a ½ dram vial, removed from the glovebox and dissolved in MeCN-D 3 (160 µL). To the solution of cage 4 was then added MeNH 2 as a 2M THF solution (160 µL, 0.318 mmol, 20 equiv), followed by the MeCN solution of Sc(OTf) 3 . The mixture was stirred vigorously for 5 h at RT. Note: some cloudiness was observed in the reaction mixture. After stirring for 5 h, a 100 µL aliquot was removed from the reaction mixture, diluted with 500 µL CDCl 3 and analyzed by 1 H NMR and MALDI-MS. The reaction mixture was then transferred to a 20 mL vial with CDCl 3 and concentrated in vacuo. The solid was dried under high-vacuum for 1 h to remove residual MeNH 2 .
Cage Reassembly: The residue was re-dissolved in 4.8 mL of CDCl 3 and 0.52 mL of MeCN-D 3 (3 mM total) (Note: some cloudiness was observed) and 200 mg of powdered 5 Å molecular sieves were added. The solution was stirred overnight at RT (ca. 17 hours), at which point a 150 µL aliquot was filtered through a small plug of neutral alumina, concentrated to dryness, diluted with 600 µL of CDCl 3 and analyzed by 1 H NMR and MALDI-MS. (1) Cage 4 (2) Imine 6 Figure S5. 1 H NMR spectra of disassembly/reassembly. Spectra were zoomed-in for clarity. Mestrenova GSD integration mode used for integration of the disassembly mixture spectrum. S12 In a 20 mL reaction vial, cage 4 (16.9 mg, 9 µmol, 1 equiv) was dissolved in CDCl 3 (2 mL, 5 mM) and stirred until dissolved. To the solution was then added n-propylamine (15 µL, 0.18 mmol, 20 equiv), followed by TFA (0.2 µL, 2.7 µmol, 30 mol%). The mixture was then stirred vigorously at room temperature for 5 hours. After 5 h, a 50 µL aliquot was removed, diluted with 0.5 mL CDCl 3 , and analyzed by 1 H NMR.