Complex molecular logic gates from simple molecules

Molecular logic gates (MLGs) are compounds that can solve Boolean logic operations to give an answer (OUTPUT) upon receiving a stimulus (INPUT). These derivatives can be used as biological sensors and are promising substitutes for the present logic gates. Although MLGs with complex molecular structures have been reported, they often show stability problems. To address this problem, we describe herein six stable pseudo-hemiindigo-derived MLGs capable of solving complex logic operations. MLGs 7, 8, 9, and 10 can solve a complex logic operation connecting 4 logic gates using 2 different wavelengths (445 nm and 400 nm) and the presence of p-TsOH and triethylamine (TEA) as inputs; MLG 11 solves a complex logic operation connecting 3 logic gates and uses 3 inputs, one wavelength of 445 nm and the presence of p-TsOH and TEA; and MLG 12 can only solve one logic operation (INH) and uses only the presence of p-TsOH and TEA as an input. Each operating method of the MLGs was evaluated with several techniques; proton interactions with MLGs were screened with NMR by titrating with p-TsOH, the photochemical properties were examined with absorption ultraviolet-visible (UV-Vis) spectroscopy, and the isomerization dynamics were examined with NMR using the two wavelengths for isomerization (photostationary isomer). The results indicate that the pseudo-hemiindigo-derived MLGs described herein can be applied as multiplexers or data selectors that are necessary for the transient flow of information for biological and computer systems. Finally, to design different MLGs and a system that can treat more information as complex logic gates (demultiplexers), two and three MLGs were mixed in different experiments. In both cases, four inputs were employed (445 nm, 400 nm, p-TsOH and TEA), yielding more outputs. Detailed information about the system dynamics was obtained from NMR experiments.

S50-S55 Scheme S1. Operation cycle of MLG 8. S50 Scheme S2. Operation cycle of MLG 9. S52 Scheme S3. Operation cycle of MLG 10. S54 Table S2. Summary of results for each process performed to verify MLGs 7 and 11. Figure S34. Scheme of MLG 7 and 11 operating in the same NMR tube. S57 Table S3. Summary of method of operation for the mixture of MLGs 7 and 11, outputs and corresponding labels S58 Scheme S4. Diagrams of the logical operation solved with the mixture of MLGs 7 and 11 with the theoretical labels S59-S60 Table S4. Summary of results of each process performed to verify for the mixture of MLGs 7, 10 and 11. Figure S35. Operating process for MLG 7, 10 and 11 S62 Table S5. Digitalization responses of the process and the logical operation solved for the mixture of MLGs 7, 10 and 11  Table S5. Comparation data of DFT methods for E-7 simulate UV/Vis absorption spectra S69 Figure S37. Theory UV/Vis absorption spectra of E-7. Figure S38. Orbitals of transitions states of E-7

General information
All reagents were purchased from Sigma-Aldrich and used without previous purification. The NMR spectra were recorder using a JEOL ECA-500 spectrometer with a magnetic field of 11.75 T (1H,500.160 MHz;13C,125.765 MHz). The unified scale was used with TMS as reference diluted (volume fraction ϕ < 1%) in chloroform for 1 H resonance ((CH3)4Si 1 H, 13 C = 0). The UV/Vis absorption spectra were recorder with a Perkin Elmer Lambda 2S UV-VIS spectrophotometer. Low resolution mass spectra were acquired with HPLC/coupled mass Agilent Technologies (ESI).

Methodology 1
In a round-bottom flask, 1.12 mmol of potassium hydroxide were dissolved in 12 mL of methanol under stirring and a solution of ketone 6 (0.37 mmol in 3 mL of methanol) was added dropwise followed by dropwise addition of a solution of 2-pyrrole carboxaldehyde (0.48 mmol in 3mL of methanol). The reaction mixture was maintained under vigorous stirring for 3 days at room temperature. After the reaction time, the organic phase was extracted with methylene chloride/200 mL of brine solution and concentrated under vacuum. The product was purified by chromatographic column using mixtures of Hexane:Ethyl Acetate of increasing polarity as eluent.
Methodology 2 2-Pyrrole carboxaldehyde (0.48 mmol), ketone 6 (0.37 mmol) and 0.75 mmol of sodium sulphate were placed in a 15mL vial followed by dropwise addition of 0.2 mL of piperidine and the reaction mixture was maintained under vigorous stirring agitation for 1 day at room temperature. After the reaction time, the organic phase was extracted with methylene chloride/1% hydrochloric acid aqueous solution, evaporated under vacuum and the product was purified by column chromatography using Hexane:Ethyl acetate mixtures of increasing polarity as eluent.

Methodology 3
To a 10 mL vial, 0.37 mmol of 2-pyrrole carboxaldehyde, 0.37 mmol of Ketone 6 and 0.4 mL of ethanol were added, the reaction mixture was left under stirring until a homogeneous solution was formed. Subsequently, the reaction mixture was placed in an ice bath and a 5% aqueous solution of potassium hydroxide was added dropwise until formation of a precipitate. Then 5 mL of water were added and the reaction mixture was allowed to freeze. The frozen solution was allowed to reach room temperature and filtered under vacuum.

Spectra conditions
The NMR spectra were recorded at 21.1°C using a Jeol ECA-500 at B0=11.75 T ( 1 H at 500.159 MHz and 13 C at 125.76 MHz). All spectra were recorded in CDCl3 solution in 5 mm OD tubes. The compounds were assigned using the pfg-COSY, pfg-HMBC and pfg-HSQC pulse sequences.The UV-Vis spectra were determined using a Perkin Elmer Lambda 2S spectrometer; to obtain the coefficient absorptions we did a standard curve (Concentration/Absorption) with 3 concentrations (3X10 -5 M, 5X10 -5 M, 7X10 -5 M) and then S4 we could calculate the absorption epsilon with slope. The Mass spectra were recorded with ESI-API.

Operation cycle determination methodology information
In a 5 mm OD NMR tube, a solution of the corresponding MLG in the concentration indicated was prepared. The NMR tube was irradiated with an Aldrich® Micro Photochemical Reactor. The NMR tube was placed close to the LEDs using an aluminum tube to maintain a firm vertical position. Blue irradiation with a wavelength of 445 nm and violet irradiation with a wavelength of 400 nm were used. All 1 H NMR spectra were recorded at 500 MHz.            We did not print the spectrum of E-12 after de irradiation because we did not observe any signal in the spectrum. Therefore, the irradiation of E-12 was studied by NMR 1 H, to understand why we did not observe changes in the UV/Vis spectrum.

Titration procedure
For the titration protocol, apart of MLG solutions in deuterated chloroform were prepared. These solutions were prepared with a concentration range of 0.003-0.005 M using 0.3mL of CDCl3. MLGs 7 and 9 were dissolved in 0.1 mL of DMSO d6 and measured using 0.3mL of CDCl3. The solutions of pTsOH were prepared using two different concentrations, 0.05 M and 0.5M in 2.5mL of CDCl3. Addition of pTsOH solution was performed using a micropipette and immediately recorded in a 500MHz NMR spectrophotometer and/or in 200 MHz NMR spectrophotometer.

Experimental determination of MLG operation
MLG operation process was studied using 0.002M solutions in CDCl3 (for MLG 7 and 9 it was necessary to add 0.05mL of DMSO d6). In several experiments, the solutions were prepared using 0.0015M of each MLG in CDCl3:DMSO d6 (3:0.2 volume relation). pTsOH was added to the same titration solutions. Irradiation of the solutions in the NMR tube was performed using an Aldrich® Micro Photochemical Reactor; it is important to highlight that the NMR tube had intimate contact with the LED. To support the NMR tube firmly, aluminum cans were used as shown in image S18. Two photoreactors were set in the experiments, one with LEDs of 435-445nm wavelength and other with LEDs of 400-410nm wavelength.
All spectra were recorded with a 500MHz NMR spectrophotometer at room temperature.         The operation cycle of MLG 8 was tested starting with irradiation at 445 nm for 10 minutes where no response was observed. This was followed by irradiation at 400 nm for 10 minutes observing E to Z isomerization (entry 3, Scheme S2), followed by a second 10 min irradiation which had no effect on the isomer ratio (entry 4, Scheme S2). Next, the compound was irradiated at 445 nm for 10 minutes that led of Z to E isomerization where the E isomer was the major product (entry 6, Scheme S2) followed by further irradiation of the Z isomer at 400 nm 10 minutes (entry 7, Scheme S2). Then, two equivalents of pTsOH were added achieving Z to E reversal as confirmed by observation of the corresponding shifts in the NMR signals of both isomers (entry 8, Scheme S2). To verify that protonation slows down the isomerization of this MLG, the solution was irradiated at 400 nm light for 10 minutes observing minimal E to Z isomerization (entry 9, Scheme S2). The processes of entries 8 and 9 were repeated with the same isomerization pattern; the only difference was that when the 3.5 equivalents of pTsOH were added, the isomerization with 400 nm irradiation was minor in comparison with entry 9. To return to the initial isomer pattern, the solution was neutralized with 6.16 equivalents of triethylamine and irradiated with 400 nm light for 10 minutes to obtain the Z isomer as major product (entry 13, Scheme S2), with a pattern similar to entry 3. Finally, irradiation was performed at 445 nm for 10 minutes to complete the function of MLG 8.

Process
Molar Fraction Scheme S1. Operation cycle of MLG 8. Each process was recorded by NMR at 500 MHz in CDCl3 at room temperature. The concentration of MLG used in these experiments was 2 mM.

S52
The study of MLG 9 started with irradiation of a solution with 400 nm light for 10 minutes to produce E to Z isomerization (entry 2, Scheme S3). To verify that the isomerization was caused by irradiation, the mixture with the Z isomer as main compound was irradiated with 400 nm light for 10 minutes, observing a slight increase of the compound (entry 3, Scheme S3). To the mixture with the Z isomer as the major isomer were added 2.3 equivalents of p TsOH to achieve Z to E isomerization. The first isomerization gave low proportion ofthe expected isomer, however, with additional 3.1 equivalents of pTsOH, an increase in the E isomer which was confirmed by a shift of the signals detected by NMR for both isomers (entries 5 and 6 Scheme S3). To verify the protonation effect, the sample was irradiated with 400 nm light for 10 minutes, and the isomerization decreased (entry 7, Scheme S3). With reference to the isomerization capability of MLG 9, the solution was neutralized with 13.3 equivalents of triethylamine, irradiated with 400 nm light for 10 minutes observing E to Z isomerization (entry 9, Scheme S3). It should be mentioned that up to 103 equivalents of triethylamine were added, and no effect was detected with the excess of the base. The isomerization was slowed down but without a significant response (entry 11, Scheme S3).
After the addition of excess triethylamine, the solution was irradiated at 400 nm for 10 minutes, evidencing a E to Z isomerization pattern similar to entry 2. The solution was irradiated at 445 nm for 10 minutes, and Z to E isomerization was observed (entry 13, Scheme S3). Finally, E to Z transformation was attained with 400 nm light for 10 minutes (entry 14, Scheme S3) to close the operation cycle of MLG 9.

Process
Molar Fraction

E-9
Molar Fraction Z-9 δH1 E-9 δH1 Z-9 Scheme S2. Operation cycle of MLG 9. Each process was recorded by NMR at 500 MHz in CDCl3 at room temperature. The concentration of MLG used in these experiments was 2 mM.

S54
The investigation of MLG 10 started with irradiation with 400 nm light for 20 minutes (to analyze and compare the effect of time with the same MLG and others) to achieve E to Z isomerization (entry 3, Scheme S4), giving the Z compound as the main product. This mixture was treated with 3 equivalents of pTsOH until the 1 H NMR signals shifted toward higher frequency, confirming Z to E isomerization (entry 5, Scheme S4). When the solution had the E isomer as the major product, the effect of protonation by irradiation with 400 nm light for 10 minutes resulted in no further isomerization of the Z isomer (entry 6, Scheme S4). Next, in our study, we tested the compound using 445 nm light and determined the complete transformation to the E isomer (entry 7, Scheme S4). Subsequently, the solution was neutralized with 7 equivalents of triethylamine, and it was irradiated again with 400 nm light for 10 minutes, obtaining a high ratio of the Z isomer (entry 9, Scheme S4), thus completing the operation cycle.     I II III IV  1 Initial mixture 0 0 0 0 a 2 445nm X 10min 0    Scheme S5. Diagrams show how the logical operation is solved with MLGs 7, 10 and 11 of each process carried out in the experiments with the theoretical labels. The concentrations of the two MLGs were 1.5 mM in CDCl3:DMSO-d6 (3:0.2 volume ratio), and NMR recorded each process at 500 MHz at room temperature.

Theory calculation of MLG 7
The geometry structure of the MLGs that was used for the theorical calculate, was the structure determinated with NMR results and corroboration with literature molecules; and then the MLG series was refined used DFT b3lyp method with basis set 6-31 + g (d, p). The next part to obtain the theorical absorption spectrum of E-7, was tested methods and basis set to find the better match of the experimental results with theorical data. The screening was tested with 3 different DFT methods; b3lyp, b3pw91 and cam-b3lyp and the same basis set 3-21+g*, with b3lyp 3 different basis set 3-21+g*, 6-31+g(d), and Sto-3g. Accordingly with these data, the best method is the b3lyp with the 3-21+g* basis set and the solvation model iefpcm/CH3Cl.  Figure S37. Theory UV/Vis absorption spectra, and data. Method b3lyp with the 3-21+g* basis set and the solvation model iefpcm/CH3Cl.