Reversible Photoisomerization in Thin Surface Films from Azo-Functionalized Guanosine Derivatives

Two novel azo-functionalized guanosine derivatives were synthesized, and their photoisomerization process was investigated in molecular monolayers at the air–water interface and in the Langmuir–Blodgett (LB) films on solid substrates. Measurements of surface pressure vs area isotherms, surface potential measurements, UV–visible (vis) absorption spectroscopy, Brewster angle microscopy (BAM), and atomic force microscopy (AFM) were performed. Despite not having a typical amphiphilic molecular structure, the derivatives formed stable films on the water surface. They could also undergo repeated photoisomerization in all of the investigated thin-film configurations. The observations suggest that in the films at the air–water interface, the molecules first exhibit a conformational change, and then they reorient to an energetically more favored orientation. In the LB films transferred onto solid substrates, the isomerization process occurs on a similar time scale as in solution. However, the isomerization efficiency is about an order of magnitude lower than that in solution. Our results show that DNA nucleobases functionalized with azobenzene moieties are suitable candidates for the fabrication of photoactive two-dimensional (2D) materials that can provide all beneficial functionalities of DNA-based compounds.


2',3',5' -tri-O-[(E)-4-(tert-butoxycarbonyl)phenyl)diazenyl)benzoyl]guanosine (GAzo 3 )
Acid 1 (0.351 g, 1.08 mmol) was dissolved in THF (10 mL) and the resulting solution was cooled to 0°C. Triethylamine (280 L, 2.0 mmol) and methanesulfonyl chloride (88 L, 1.1 mmol) were added and the mixture was stirred for 1 h at 0°C then allowed to warm to r.t. Guanosine (85 mg, 0.30 mmol) and a catalytic amount of 4-(dimethylamino)pyridine were added and the reaction was monitored by TLC (CH 2 Cl 2 /MeOH 96:4). After 48 h the solvent was removed by distillation under reduced pressure. The crude reaction mixture was partitioned between CHCl 3 and sat. Na 2 CO 3 . The aqueous phase was washed several times with CHCl 3 and the combined organic fractions were dried over MgSO 4 . Solvent was removed by distillation and the residue was purified by column chromatography (CHCl 3 /MeOH, gradient from 99:1 to 97:3). The product thus obtained was further crystalized from MeOH, affording 0.176 g (48%) of the title compound as a bright orange solid.

COMPARISON OF SURFACE PRESSURE, SURFACE POTENTIAL, AND LIGHT ABSORPTION OF LANGMUIR FILMS DURING PHOTOISOMERIZATION
Langmuir films from GAzo and GAzo 3 were prepared by depositing 75 µL (GAzo) or 60 µL (GAzo 3 ) of 1 mM chloroform solution to the air water surface and compressing the barriers so that the final surface area was equal to 177.5 cm 2 . This resulted in mean molecular areas of 39 Å 2 in the case of GAzo film and 49 Å 2 in the case of GAzo 3 film. Figure S7: Changes in light absorption (blue), surface potential (green), and surface pressure (red) of a GAzo film during irradiation with UV and blue light. The intervals with gray background indicate UV irradiation, while white background indicates irradiation with visible light. Light absorption is shown in terms of the change of the ratio of the voltage signals recorded on the two photoiodes. A brigther blue line is drawn on top of the data points for the photoiode signal: this represents the same data but filtered to reduce measurement noise.

S10
The recorded changes in light absorption, surface potential, and surface pressure in GAzo film during irradiation with blue and UV light are shown in Figure S. While irradiation induced a significant drop in surface pressure (from 30 mN/m at the end of compression to below 10 mN/m at the end of the measurement), surface potential and absorbance both remained mostly constant -the slight drop in the signal on the photodiode seen in Figure S amounts to only a 0.3 % increase in signal with respect to the start of the measurement. Had the drop in surface pressure been caused by the loss of molecules into the subphase, we would expect a proportional drop in absorption of light and surface potential. The fact that this was not observed leads us to believe that the drop in surface pressure is not a consequence of the dissolution of molecules in the subphase but rather the consequence of the molecules rearranging themselves at the film surface, possibly forming multilayered structures. This is also consistent with the continued evolution of the appearance of the water surface after compression, as observed under BAM.
A single cycle of blue and UV irradiation is shown in greater detail in Figure S. The data in this image are fitted with exponential functions to better illustrate the characteristic times for the change in each of the measured quantities. A linear drift term was added when fitting the data for surface pressure to account for the drop in surface pressure over the entire course of the measurement.
When the film is irradiated with blue light, the change in light absorption happens the fastest, with surface pressure and surface potential changing at slower rates. During UV irradiation, however, surface pressure follows the change in absorption, while the change in surface potential still lags behind the two. The fact that the fitted characteristic time for the change in surface pressure is actually shorter than the one for absorption is likely a consequence of the fact that a simple exponential function with a linear drift term does not accurately describe the behavior of surface pressure during irradiation 3 . Figure S8: Surface pressure, surface potential, and light absorption of GAzo Langmuir film during a single cycle of blue and UV irradiation. The dashed lines are exponential fits -the obtained characteristic times are written next to the lines. In the case of surface pressure, an additional linear drift term was added to the fitting function: in this case, the dotted line represents the fit without the added linear drift term.
Photoinduced changes in light absorption, surface potential, and surface pressure in a GAzo 3 film are shown in Figure S9. Similar observation as for the data measured in GAzo films can be made here as well, however, there is a slight downward trend in surface potential, consistent with loss of molecules from the film S11 surface. However, the relative change in surface potential is only 4%, which is still small in comparison to the much larger drop in surface pressure. Figure S10 shows a more detailed view of a single blue and UV irradiation cycle of GAzo 3 Langmuir film. Just as in the case of the GAzo film, the change in light absorption occurs at a faster rate than the change in surface pressure and surface potential. The characteristic time for the change in light absorption during irradiation with blue light is remarkably short, especially when compared to the characteristic time for the change in surface potential: 40 s vs. 1300 s. Figure S9: Changes in light absorption (blue), surface potential (green), and surface pressure (red) of a GAzo 3 film during irradiation with UV and blue light. The intervals with gray background indicate UV irradiation, while white background indicates irradiation with visible light. Light absorption is shown in terms of the change of the ratio of the voltage signals recorded on the two photoiodes. A brigther blue line is drawn on top of the data points for the photoiode signal: this represents the same data but filtered to reduce measurement noise.
Since a change in surface potential indicates a rotation of the molecular dipole moment, the slower change in surface potential in comparison to the change in light absorption would suggest that the molecules in the film first undergo isomerization and then slowly rotate to an energetically more favorable orientation. The mismatch between the rate of change in light absorption, surface pressure, and surface potential is in S12 contrast to what was reported by Maack et al., where all the quantities appeared to be changing at the same rate 4 . Figure S10: Surface pressure, surface potential, and light absorption of GAzo 3 Langmuir film during a single cycle of blue and UV irradiation. The dashed lines are exponential fits -the obtained characteristic times are written next to the lines. In the case of surface pressure, an additional linear drift term was added to the fitting function: in this case, the dotted line represents the fit without the added linear drift term.