*2.3. DFT Analysis*

TDDFT approach at DFT level, using adiabatic local density approximation and ethanol as the simulated solvent, was used to run excitation energies calculations. Table 3 shows the most relevant optoelectronic properties calculated for the compounds A1 and A2. For A1, HOMO delocalization covers the entire conjugated backbone, extending to the terminal groups. LUMO is delocalized over the central rings and diazo groups. The main transition for both compounds is HOMO → LUMO, at 425 nm for A1 and 419 nm for A2. For A2, HOMO is mainly delocalized over the central ring and the oxygen atoms (Figure 5). A2 LUMO shows the electron-withdrawing effect of the nitro groups, with a higher delocalization of the orbitals over the terminal groups compared to the diethylamino derivative A1. The HOMO-LUMO gap is also higher for A2 than for A1 (Table 3). Because of the electron-withdrawing effect of the nitro groups, A2 shows a higher oxidation potential compared to A1, whereas the hole and electron reorganization energies (HRE and ERE) are typical for these class of compounds. A2 shows a decrease in the ERE due to the presence of the nitro groups. Reorganization energy (RE) is one of the parameters involved in the hopping rate, and HRE and ERE are strongly correlated to cation and anion geometries. A compound with a small RE usually shows high carrier mobility, and the energies are proportional to the deformation of the geometry during the process of charge transfer. Both derivatives show an ERE higher than their HRE, which means that there is less deformation upon electron injection compared to hole injection. Furthermore, the nitro derivative shows a smaller electron extraction potential, meaning that electron injection into A2 is easier than into

A1. The electron-withdrawing effect of the nitro groups affects the electron-transporting features of the system.


**Table 3.** Electro-optical properties calculated on A1 and A2 in vacuum.

(**b**)

**Figure 5.** Frontiers orbitals HOMO and LUMO calculated for A1 (**a**) and A2 (**b**).

The significant PL performance of A1 blended in PFO with 57% PLQY must be underlined. This result appears striking, keeping in mind that PFO itself is a blue emitter and A1-PFO blend

**λ**

− −

− − − −

−

− −

emit in the orange-red region. Moreover, this solid-state emission is due to an azobenzene dye. For this reason, we analyzed the behavior of A1 in the PFO blend, the entire discussion reported in the Supplementary Material section. The formation of stacked dimers of the azo dye and PFO resulted in modifying the electro-optical properties significantly with the lowering of HOMO and LUMO values for both systems. The new HOMO energy values are even closer to the HOMO values of the PFO (Figure S1). Overall, stacking the dyes with PFO provides a shortcut for electronic transitions and is responsible for some degree of PFO distortion. Therefore, it must be underlined the ability of the azo dyes to modulate the optoelectronic characteristics of the PFO through two effects: the steric hindrance operated by the octyl chains, and the stacking between the dyes and the polymer. Especially in the case of A1-PFO, the calculations show a significant reduction of the electronic hopping energy barrier and justify the high quantum yield observed. Analysis of the oxidation potentials of the substituted dyes revealed that the oxidation potential of A1 dyes was about 460 mV easier to oxidize (more negative potentials) than the A2. Analysis of the reduction potentials revealed that A2 was about 400 mV easier to reduce (more positive potentials) than A1 dye. The diminished ability to reduce A1 dye is attributed to the presence of electron-withdrawing groups. Compared with other azo dyes like the 4-methoxylazobenzene [54] with a Ered of −1.44 V, or diarylaminoazobenzenes with a reduction potential between −1.36 V and −1.50V [55], both A1 and A2 have a lower reduction tendency and higher stability (see Tables S1 and S2 in Supplementary Material).

#### **3. Materials and Methods**

Commercially available starting products were supplied by Sigma Aldrich (Sigma-Aldrich Corporation, St. Louis, MO, USA). 2-hydroxy-4-nitrobenzaldehyde was obtained as described in [56]. 4,4′ -((1,1′ )-(2,5-dimethoxy-1,4-phenylene)bis(diazene-2,1-diyl))dianiline (AB-NH2) was obtained as described in [13]. <sup>1</sup>H NMR spectra were recorded in DMSO-d6, with a Bruker Advance II 400 MHz apparatus (Bruker Corporation, Billerica, MA, USA). Mass spectrometry measurements were performed using a Q-TOF premier instrument (Waters, Milford, MA, USA) equipped by an electrospray ion source and a hybrid quadrupole-time of flight analyzer.

Zeiss Axioscope polarizing microscope (Carl Zeiss, Oberkochen, Germany) equipped with an FP90 Mettler hot stage (Mettler-Toledo, LLC-Columbus, OH, USA). DSC/TGA Perkin Elmer TGA 4000 (PerkinElmer, Inc., Waltham, MA, USA), scanning rate 10 ◦C/min, provided phase transition temperatures and enthalpies. The decomposition temperatures (the temperature at 5 wt.% weight loss, Td) were measured under nitrogen flow. UV-Visible and fluorescence spectra were recorded by JASCO F-530 and FP-750 spectrometers (JASCO Inc., Mary's Court, Easton, MD, USA). Thin films of the neat samples and the polymeric blends were obtained using an SCS P6700 spin coater (Specialty Coating Systems Inc., Indianapolis, IN 46278, USA) operating at 600 rpm for 1 min (first step) and at 1200 rpm for 1 min (second step). 10 wt% NMP solution of the chromophores in commercially available PS (molecular weight 18,700 Da), in PVK (molecular weight 1100 Da) or PFO (molecular weight ≥20,000 Da), were employed.

Photoluminescence quantum yield (PLQY) measurements were conducted with a setup similar to that of de Mello et al. [57]. It considers not only the excitation laser and direct photoluminescence but also the scattering of the integrating sphere that is part of the setup. The setup consists of a 405 nm laser, whose emission does not overlap with the photoluminescence spectrum, an integrating sphere (Stellarnet Inc, Tampa, FL, USA) and a photo spectrometer (BLACK Comet Stellarnet Inc, Tampa, FL, USA). The emission was measured at five different points on the sample.

Field emission scanning electron microscopy (FESEM) images were obtained with a FEI Nova NanoSEM 450 emission SEM (Thermo Fisher Scientific Inc., Waltham, MA, USA) at an accelerating voltage of 10 kV (range of acceleration voltage: 50 V–30 kV) equipped with an Everhart Thornley detector (ETD) and a Through Lens Detector (TLD). Samples were mounted on Al specimen mounts and coated with a thin layer of Au-Pd in order to eliminate any undesirable charge effects during the SEM observations.
