Two-photon isomerization properties of donor–acceptor Stenhouse adducts

Donor–acceptor Stenhouse adducts (DASAs) are important photo-responsive molecules that undergo electrocyclic reactions after light absorption. From these properties, DASAs have received extensive attention as photo-switches with negative photochromism. Meanwhile, several photochemical applications require isomerization events to take place in highly localized volumes at variable depths. Such focused photoreactions can be achieved if the electronic excitation is induced through a non-linear optical process. In this contribution we describe DASAs substituted with extended donor groups which provide them with significant two-photon absorption properties. We characterized the photo-induced transformation of these DASAs from the open polymethinic form to their cyclopentenic isomer with the use of 800 nm femtosecond pulses. These studies verified that the biphotonic excitation produces equivalent photoreactions as linear absorbance. We also determined these DASAs' two-photon absorption cross sections from measurements of their photoconverted yield after biphotonic excitation. As we show, specific donor sections provide these systems with important biphotonic cross-sections as high as 615 GM units. Such properties make these DASAs among the most non-linearly active photo-switchable molecules. Calculations at the TDDFT level with the optimally tuned range-separated functional OT-CAM-B3LYP, together with quadratic response methods indicate that the non-linear photochemical properties in these molecules involve higher lying electronic states above the first excited singlet. This result is consistent with the observed relation between their two-photon chemistry and the onset of their short wavelength absorption features around 400 nm. This is the first report of the non-linear photochemistry of DASAs. The two-photon isomerization properties of DASAs extend their applications to 3D-photocontrol, non-linear lithography, variable depth birefringence, and localized drug delivery schemes.


Synthesis and characterization
Synthetic methods 5-(furan-2-ylmethylene)-1,3-dimethylpyrimidine-2,4,6(1H,3H,5H)-trione: Freshly distilled 0.83 ml (10 mmol) of furfural was added to a solution of 1.56 g of barbituric acid in 40 ml of water. Color change to green was observed. The reaction was stirred for 3h at room temperature, time at which a thick slurry was formed. The reaction mixture was filtered. The filtered cake was redissolved in 100 ml CH2Cl2 and washed with 30 ml saturated NaHSO4, 30 ml saturated NaHCO3 and 30 ml brine. The organic layer was dried and the solvent was evaporated to yield 1.95 g of pure activated furane (83%). 1  Indoline: 117.1 mg (1 mmol) of indole was dissolved with 3 ml of acetic acid in a 10 ml dry roundbottom flask suited with a water trap. Then 188 mg (3 mmol) of sodium cyanoborohydride were added. The solution was diluted with 15 ml of water two hours after stirring at room temperature and sodium hydroxide was added until pH=14. A white precipitate was formed. The aqueous solution was extracted with 3x35 ml CH2Cl2. The organic layer was washed with 2x50 ml water and 1x 50 ml brine, and dried with Na2SO4. The solvent was evaporated to yield 109 mg of pure indoline as a transparent oil. 1 H NMR (301 MHz, CDCl3, 298 K) δ 3.04 (t, J = 8.3 Hz, 2H), 3.56 (t, J = 8.3 Hz, 2H), 6.66 (dq, J = 7.7, 1.2, 0.6 Hz, 1H), 6.71 (td, J = 7.4, 1.1 Hz, 1H), 7.03 (tt, J = 7.7, 1.2 Hz, 1H), 7.13 (dq, J = 7.4, 1.2, 0.6 Hz, 1H). 13  General procedure for Suzuki-Miyaura cross-coupling reactions: A clean Schlenk flask was charged with 5-bromoindoline (1 mmol, 1 eq.), boronic derivative (1.2 eq.), [Pd] (1 mol%), SPhos (4 mol%) and K3PO4 (1.2 eq. for boronic acid and 2.4 eq. for pinacol ester derivative). The reaction vessel was evacuated and backfilled with nitrogen gas 3 times. Then, previously distilled dioxane from sodium (4 ml) and water (0.8 ml) were added by syringe. The flask was placed in a preheated oil bath at 85 °C and stirred 24 h. The reaction mixture was cooled to room temperature, filtered through a short pad of celite, filter cake was then washed with 15 ml ethyl acetate. Then, 50 ml of water were added, and the aqueous layer were extracted with 3x20 ml ethyl acetate. The organic layer was washed with 2x30 ml water and 1x30 ml brine, dried with Na2SO4 and solvent evaporated. The pure product was obtained by silica column chromatography.

Determination of the open-closed isomer proportions
The relative populations of the two isomers in DASA solutions were determined by integration of specific sets of peaks assigned to each isomer in freshly prepared solutions. The relative proportions of the two isomers was used to determine the absorption coefficient at the visible band maximum (open isomer) which is required to quantify the two-photon conversion properties of molecules 4-6 as explained in the main text. Peak assignments were made from 2D (COSY) and 1 H-NMR experiments acquired in order to determine which peaks increase (closed form) and which diminish (open form) when the sample is irradiated with visible light. The full set of spectra are included in the respective section of this Supporting Information file.

DASA 4
Integration of the peaks at 6.20 and 6.

Steady-state spectroscopy Molar absorptivity coefficients ε(λ)
The equation used for the determination of ε(λ) of DASAs 4 to 6 for each isomer is: Where: A= total initial absorbance at each wavelength.
CTotal= total concentration of each DASA.  Table S1. Absorption coefficients at the first band maxima for molecules 4 to 6. DASA 5 Figure S8. Absorption spectra of an initial sample (blue) and sample fully photo-converted to the closed form (red) for DASA 5 in toluene. Total initial concentration: 11 μM.
DASA 6 Figure S9. Absorption spectra of thermally equilibrated sample (blue) and sample fully photo-converted to the closed form (red) for DASA 6 in toluene. Total initial concentration: 11.7 μM.

Linear photochemistry Photoisomerization quantum yields
Absorption data after linear excitation of DASA solutions were fitted according to the following equation. The yield was obtained from the slope after a linear regressions.
Phabs: photons absorbed by the sample.
The equation was used in the following form: Where: P= average beam intensity      Figure S16 in the isosbestic point region (∿325 nm).

Two-photon absorption cross section measurements
Data obtained from the biphotonic intensity dependence was fitted according to the following equation to measure the TPA-cross section 2 as described in the main text: Table S3. Data for the calculation of 2 of DASA 3. The last column shows the 2 value considering the specific average power in the row.     Figure S31. Electronic absorption spectra for molecular systems 1 and 6, simulated using different density functional models and the def2-SVP basis. As can be seen, the experimental spectra are best predicted by the OT-CAM-B3LYP functional.    Figure S33. 1