Molecular memory with downstream logic processing exemplified by switchable and self-indicating guest capture and release

Molecular-logic based computation (MLBC) has grown by accumulating many examples of combinational logic gates and a few sequential variants. In spite of many inspirations being available in biology, there are virtually no examples of MLBC in chemistry where sequential and combinational operations are integrated. Here we report a simple alcohol-ketone redox interconversion which switches a macrocycle between a large or small cavity, with erect aromatic walls which create a deep hydrophobic space or with collapsed walls respectively. Small aromatic guests can be captured or released in an all or none manner upon chemical command. During capture, the fluorescence of the alcohol macrocycle is quenched via fluorescent photoinduced electron transfer switching, meaning that its occupancy state is self-indicated. This represents a chemically-driven RS Flip-Flop, one of whose outputs is fed into an INHIBIT gate. Processing of outputs from memory stores is seen in the injection of packaged neurotransmitters into synaptic clefts for onward neural signalling. Overall, capture-release phenomena from discrete supermolecules now have a Boolean basis.


Supplementary Information Molecular Memory with Downstream Logic Processing exemplified by Switchable and Self-indicating Guest Capture and Release
Daly et al

Synthesis and characterization data of all compounds used in the study
Synthesis of 1,7,21,27-tetraoxa-[7.1.7.1]-paracyclophane (1). 4,4'-Dihydroxybenzophenone 5 (3.20 g, 14.9 mmol) and diiodopentane (4.86 g, 15.0 mmol) were dissolved in acetone (400 ml, freshly distilled HPLC grade, dried over anhydrous potassium carbonate) and slowly added, with the aid of a dropping funnel whose capillary tip is dipping below the surface, to a refluxing suspension of acetone (600 ml) and cesium carbonate (8.0 g, 25 mmol) over 12 hours. The colourless solution turned a pale yellow colour. This mixture was then refluxed for a further three days. The hot reaction mixture was then filtered through a silica/hyflo plug and the filtrate chilled to 4 °C whereupon a white solid precipitated out. The solid was purified using flash silica chromatography eluting with dichloromethane: ethyl acetate (95:5 v/v) yielding the desired white plate-like solid (0.8 g, 10%

.1]paracyclophane (2).
A suspension of sodium borohydride (100 mg, 2.64 mmol) in ethanol (5 ml) was added dropwise to a stirring solution of 1 (50 mg, 0.089 mmol) dissolved in tetrahydrofuran (10 ml) and stirred overnight. A few drops of acetic acid and water (20 ml) was added to destroy the reducing agent and the mixture extracted with 3x15 ml of dichloromethane. The organic layer was dried with sodium sulfate and evaporated to dryness to yield the desired white solid (25 mg, 50%

Synthesis of 1-[4-(1-azoniabicyclo[2.2.2]oct-1-ylmethyl)benzyl]-1azoniabicyclo[2.2.2] octane dibromide (9)
α,α'-Dibromo-p-xylene (1.19 g, 4.5 mmol) was dissolved in dimethylformamide (70 ml). Quinuclidine (1.0 g, 9.0 mmol) was then added slowly with stirring. After 60 min a white precipitate formed and the suspension was stirred for a further 20 hours. The precipitate was then filtered and dried yielding a white solid (0.71 g, 98%   1 mg, 0.007 mmol) added. An efficient double walled reflux condenser was attached and the other neck was sealed with a rubber septum, a needle was placed though this septum so that its point was just below the surface of the liquid and its end open to the air. A vacuum was attached to the top of the reflux condenser and arranged so that air was drawn through the needle and into the solution. The mixture was then heated at 90 o C overnight. Then the DMSO was removed under reduced pressure and the residue dissolved in water. The solution was acidified with 4M HCl and a precipitate formed. This was centrifuged, the supernatant was discarded, more distilled water was added and the process repeated. The pellet was then flushed from the centrifuge vial with distilled water and dried under reduced pressure at 60 o C to give a white solid. 1 H-NMR and TLC analysis confirmed that the ketone, 9,17,29,37-Tetracarboxy-1,7,21,27tetraoxa-[7.1.7.1]-paracyclophane (3), had been regenerated. 9,17,29,37-Tetracarboxy-14,34-dihydroxy-1,7,21,27tetraoxa[7.1.7.1]-paracyclophane (4) with KMnO 4 in water 9,17,29,37-Tetracarboxy-14,34-dihydroxy-1,7,21,27-tetraoxa[7.1.7.1]paracyclophane (50 mg, 0.07 mmol) was suspended in distilled water (100 ml) and sonicated. Aqueous 5M NaOH solution was added until the solution was clear (ca. 2.5 ml). KMnO 4 (100 mg, 0.67 mmol) was then added and the mixture was kept at 40 o C for 5 hours with vigorous stirring. After this time, methanol (5 ml) was added and kept until the purple color had faded. The solution was filtered while hot and the methanol removed under reduced pressure. The solution was made upto 100 ml with water before acidification with 5M HCl until a precipitate formed. This was centrifuged, the supernatant was decanted, more distilled water was added and the process repeated. The pellet was then flushed from the centrifuge vial with distilled water and dried under reduced pressure at 60 o C to give a white solid. 1 H-NMR analysis confirmed that the ketone, 9,17,29,37-Tetracarboxy-1,7,21,27-tetraoxa- Stability test of 7 under the reduction conditions which were applied to 3 with NaBH 4 in water 7 (200 mg, 0.90 mmol) was dissolved in water (10 ml), sodium borohydride (1.02 g, 27 mmol) added and stirred for 24 hours at room temperature (ca. 20 o C). A few drops of hydrobromic acid were then added to destroy unreacted sodium borohydride and evaporated to dryness under reduced pressure. 1 H-NMR analysis confirmed that 7 is the sole component.

Complexation-induced shifts in 1 H NMR spectra of 7-9 with 4
All spectra are obtained at 25 o C, after annealing the samples at 60 o C for 20 min.    [9]

Supplementary
Crystal data for C 36 H 36 O 6 (1) (M = 564.65 g/mol): monoclinic, space group C2/c (no. 15), a = 17.0438(12) Å, b = 18.5687(13) Å, c = 9.5106(7) Å, β = 106.4400(17)°, V = 2886.9(4) Å 3 , Z = 4, T = 100 K, µ(MoKα) = 0.087 mm -1 , Dcalc = 1.299 g/cm 3 , 23446 reflections measured (4.38° ≤ 2Θ ≤ 54.98°), 3300 unique (R int = 0.0982, R sigma = 0.0457) which were used in all calculations. The final R 1 was 0.0497 (>2sigma(I)) and wR 2 was 0.1495 (all data). The absence of a carbonyl and its electronic effects mean that the phenylene rings are free to rotate so that the angle between each and the macrocycle mean plane is > 80 o in all instances, and thus all of the walls are said to be 'erect'. The perpendicular distance from the macrocycle mean plane to the outermost carbon of each of the phenylene rings is now more uniform with distances from 0.990 to 1.384 Å. The rotation of the rings means that the opposing rings (A1-B2 and A2-B1) are perpendicular; however the expansion of the distances between these rings means that no intra-annular interactions are possible. Benzene is found exclusively in the intramolecular cavity and no disorder is observed in its position, this lack of disorder can be explained by edge to face contacts between the benzene ring and each of the phenylene rings of the host, as measurements show the distance from the benzene centroid to that of each ring in the macrocycle is ca 5.2 Å. The molecular volume was measured using Olex2 and found to be 563.660 Å 3 (error = 0.147%) this represents an increase in volume of over 17% through oxidation of the bridging carbonyl alone.