Effect of π-Conjugated Spacer in N-Alkylphenoxazine-Based Sensitizers Containing Double Anchors for Dye-Sensitized Solar Cells

A series of novel double-anchoring dyes for phenoxazine-based organic dyes with two 2-cyanoacetic acid acceptors/anchors, and the inclusion of a 2-ethylhexyl chain at the nitrogen atom of the phenoxazine that is connected with furan, thiophene, and 3-hexylthiophene as a linker, are used as sensitizers for dye-sensitized solar cells. The double-anchoring dye exhibits strong electronic coupling with TiO2, provided that there is an efficient charge injection rate. The result showed that the power conversion efficiency of DP-2 with thiophene linker-based cell reached 3.80% higher than that of DP-1 with furan linker (η = 1.53%) under standard illumination. The photovoltaic properties are further tuned by co-adsorption strategy, which improved power conversion efficiencies slightly. Further molecular theoretical computation and electrochemical impedance spectroscopy analysis of the dyes provide further insight into the molecular geometry and the impact of the different π-conjugated spacers on the photophysical and photovoltaic performance.


Introduction
Dye-sensitized solar cells (DSSCs) have cultivated a huge interest owing to its ease of fabrication and low cost since the report by Grätzel et al. in 1991 [1]. They imitate natural photosynthesis by means of a molecular absorber, which transforms light into electrical energy [2]. The high power conversion efficiencies of 13.0%, 11.50%, and 12.5% has been reached for porphyrin- [3], polypyridyl Ru(II) complex- [4], and metal-free organic dye-based [5] DSSCs, correspondingly. Many current reports are seen that suggest metal free organic sensitizers might have power conversion efficiency (PCE) cell efficiencies cheaply with that of Ru-based DSSCs [6]. Although Ru-based complexes maintain the record of validated efficiency of over 11%, the Ru-based dyes are costly and environmental unfriendly. Compared to the Ru dyes, metal-free organic dyes have more than a few favors: (i) they take greater molar extinction coefficients; (ii) simple synthesis; (iii) lower cost and more flexible in molecular design. As such, metal-free dyes are also considered hopeful for applications in DSSCs, and several organic dyes has been used as the sensitizers of DSSCs [7][8][9][10][11]. Besides dye-sensitized solar cells, development of new technologies with renewable energy (e.g., sunlight) for the production of H 2 , CO, CH 4 solar fuels for fuel cell applications also have attracted attention, since they help effective reductions of energy shortage and environment pollution [12,13].
Organic sensitizers usually construct with a D-π-A structure configuration, comprising of a donor (D), a π-linker (π), and an acceptor (A), which also functions as the anchor. The heteroaromatic rings such as thiophene and furan that have been demonstrated to be useful for red shifting of the absorption wavelength were widely used as π-conjugated spacers in sensitizers. Furan also has lower resonance energy than thiophene, which would favor formation of the quinoid structure during intramolecular charge transfer. For example, Hua et al. reported the benzotriazole-based sensitizers containing furan moiety nitrogen atmosphere with sodium and benzophenone. Other chemicals were bought and used without further purification.
The 1 H NMR and 13 C spectra were recorded using a Bruker Advance NMR 300 Hz spectrometer with CDCl 3 and DMSO-d 6 that were purchased from Cambridge Isotope Laboratories Inc. (Tewksbury, MA, USA). Absorption spectra were recorded on a JASCO V-730 probe UV−vis spectrophotometer. All chromatographic separations were carried out on silica gel (45-75 um mesh). Mass spectra (FAB) were verified on a Micro TOF-II mass spectrometer.
The photoelectrochemical characterizations on the solar cells were carried out using an Oriel Class AAA solar simulator (Oriel 94043 A, Newport Corp., Irvine, CA, USA). Photocurrent-voltage characteristics of the DSSCs were recorded with a potentiostat/galvanostat (CHI650B, CH Instruments, Inc., Bee Cave, TX, USA) at a light intensity of 100 mWcm −2 calibrated by an Oriel reference solar cell (Oriel 91150, Newport Corp., Irvine, CA, USA). The monochromatic quantum efficiency was recorded through a monochromator (Oriel 74100, Newport Corp., Irvine, CA, USA) at short circuit condition. The intensity of each wavelength was in the range of 1 to 3 mWcm −2 . Electrochemical impedance spectra (EIS) were recorded for DSSCs in the dark at −0.65 V potential at room temperature, whose frequency travelled ranged from 10 mHz to 100 kHz.

Fabrication of DSSCs
The photoanode used was the TiO 2 thin film (12 µm of 20 nm particles as the absorbing layer and 6 µm of 100 nm particles as the scattering layer) coated on FTO glass substrate with a dimension of 0.5 × 0.5 cm 2 , and the film thickness measured by a profilometer (Dektak3, Veeco/Sloan Instruments Inc., Santa Barbara, CA, USA). A platinized FTO produced by thermopyrolysis of H 2 PtCl 6 was used as a counter electrode. The TiO 2 thin film was dipped into the THF solution containing 3 × 10 −4 M dye sensitizers for at least 12 h. For the co-adsorbed solar cell, chenodeoxycholic acid (CDCA) was added into the dye solutions at a concentration of 10 mM. After rinsing with THF, the photoanode, adhered with a polyimide tape of 30 µm in thickness and with a square aperture of 0.36 cm 2 , was placed on top of the counter electrode and tightly clipped them together to form a cell. A 0.6 × 0.6 cm 2 of cardboard mask was clipped onto the device to constrain the illumination area. An electrolyte was then injected into the space and then the cell was sealed with the Torr Seal cement (Varian, MA, USA). The electrolyte was composed of 0.5 M lithium iodide (LiI), 0.05 M iodine (I 2 ), and 0.5 M 4-tert-butylpyridine that was dissolved in acetonitrile.

Quantum Chemistry Computation
Q-Chem 4.0 software was used for the computations. B 3 LYP/6-31G* basis set was used for geometry optimization of the molecules. For each molecule, a number of possible conformations were examined and the one with the lowest energy was used. The same function was also applied for the calculation of excited states using time-dependent density functional theory (TD-DFT). There exist a number of previous works that employed TD-DFT to characterize excited states with charge-transfer character. In some cases, underestimation of the excitation energies was seen. Therefore, in the present work, we use TD-DFT to visualize the extent of transition moments as well as their charge-transfer characters, and avoid drawing conclusions from the excitation energy.

Synthesis
All the new dyes were prepared via Knoevenagel condensation reaction by reacted corresponding aldehyde derivatives and cyanoacetic acid in the presence of the catalytic amount of ammonium acetate.

Optical Properties
The photophysical properties of the three new dyes DP-1, DP-2, and DP-3 that were investigated by the UV-vis absorption spectra of dyes in THF are shown in Figure 2a and the consistent data are summarized in Table 1. As shown, DP-1, DP-2, and DP-3 exhibited similar absorption bands, having two distinct bands at around 350 nm and 520 nm in THF, respectively. The former was assigned to the localized aromatic π-π* transitions, and the latter was attributed to an intramolecular charge-transfer (ICT) transition from the phenoxazine donor to the anchoring moiety. The ICT absorption maximum peaks of DP-1, DP-2, and DP-3 were found at 524, 526, and 528 nm, respectively, while the emission bands were displayed at 636 nm, 653 nm, and 647 nm. The Stokes shifts between the absorption and the emission bands were also supported for transfer characteristics in these dyes. Figure 2b shows the photoluminescence spectrum and the consistent data are summarized in Table 1. The maximum absorption peaks of DP-2 and DP-3 are slightly red-shifted compared to that of DP-1 with furan linker. It indicated that the delocalization of electrons over whole molecules with different π-spacers decreased in the order of nhexylthiophene > thiophene > furan. In addition, the molar extinction coefficients (ε) of these ICT bands significantly increased in the order of DP-1 (1.5 . The organic dyes have higher molar extinction coefficients that were beneficial for device fabrication afforded to use thinner TiO 2 film. Amongst these dyes, DP-3 exhibits the broadest and the most intense absorption spectra.

Optical Properties
The photophysical properties of the three new dyes DP-1, DP-2, and DP-3 that were investigated by the UV-vis absorption spectra of dyes in THF are shown in Figure 2a and the consistent data are summarized in Table 1. As shown, DP-1, DP-2, and DP-3 exhibited similar absorption bands, having two distinct bands at around 350 nm and 520 nm in THF, respectively. The former was assigned to the localized aromatic π-π* transitions, and the latter was attributed to an intramolecular charge-transfer (ICT) transition from the phenoxazine donor to the anchoring moiety. The ICT absorption maximum peaks of DP-1, DP-2, and DP-3 were found at 524, 526, and 528 nm, respectively, while the emission bands were displayed at 636 nm, 653 nm, and 647 nm. The Stokes shifts between the absorption and the emission bands were also supported for transfer characteristics in these dyes. Figure 2b shows the photoluminescence spectrum and the consistent data are summarized in Table 1. The maximum absorption peaks of DP-2 and DP-3 are slightly red-shifted compared to that of DP-1 with furan linker. It indicated that the delocalization of electrons over whole molecules with different π-spacers decreased in the order of nhexylthiophene > thiophene > furan. In addition, the molar extinction coefficients (ε) of these ICT bands significantly increased in the order of DP-1 (1.5 × 10 4 M −1 cm −1 ) > DP-2 (4.2 × 10 4 M −1 cm −1 ) > DP-3 (4.8 × 10 4 M −1 cm −1 ). The organic dyes have higher molar extinction coefficients that were beneficial for device fabrication afforded to use thinner TiO2 film. Amongst these dyes, DP-3 exhibits the broadest and the most intense absorption spectra.
(a) (b) Scheme 1. Synthetic pathways of the DP dyes.

Optical Properties
The photophysical properties of the three new dyes DP-1, DP-2, and DP-3 tha investigated by the UV-vis absorption spectra of dyes in THF are shown in Figure the consistent data are summarized in Table 1. As shown, DP-1, DP-2, and DP-3 exh similar absorption bands, having two distinct bands at around 350 nm and 520 nm i respectively. The former was assigned to the localized aromatic π-π* transitions, a latter was attributed to an intramolecular charge-transfer (ICT) transition fro phenoxazine donor to the anchoring moiety. The ICT absorption maximum peaks 1, DP-2, and DP-3 were found at 524, 526, and 528 nm, respectively, while the em bands were displayed at 636 nm, 653 nm, and 647 nm. The Stokes shifts betwe absorption and the emission bands were also supported for transfer characteris these dyes. Figure 2b shows the photoluminescence spectrum and the consistent d summarized in Table 1. The maximum absorption peaks of DP-2 and DP-3 are s red-shifted compared to that of DP-1 with furan linker. It indicated that the delocal of electrons over whole molecules with different π-spacers decreased in the orde hexylthiophene > thiophene > furan. In addition, the molar extinction coefficients these ICT bands significantly increased in the order of DP-1 (1.5 × 10 4 M −1 cm −1 ) > DP × 10 4 M −1 cm −1 ) > DP-3 (4.8 × 10 4 M −1 cm −1 ). The organic dyes have higher molar ext coefficients that were beneficial for device fabrication afforded to use thinner TiO Amongst these dyes, DP-3 exhibits the broadest and the most intense absorption s  The absorption spectra of the three dyes (DP-1, DP-2, and DP-3) adsorbed onto TiO 2 films with and without CDCA are shown in Figure 3. Generally, the absorption maxima of organic dyes on TiO 2 films would change due to the effect of the deprotonation in the adsorption process and the aggregation state of dyes on TiO 2 films. When DP-1, DP-2, and DP-3 adsorbed on the TiO 2 surface, the absorption spectra of the three dyes were broadened and absorption bands at the long-wavelength side are blue-shift compared to that of the solution spectra, which could be ascribed to the H-type aggregation or the deprotonation of the carboxylic acid upon being adsorbed on TiO 2 [36,37]. In addition, it was distinguished that the absorption spectra of the three dyes anchored onto the TiO 2 film exposed a slightly broad outline compared to those in solution, which was helpful for light-harvesting. CDCA was added to check the dye aggregation with DP dyes on TiO 2 film, as displayed in Figure 3.  The absorption spectra of the three dyes (DP-1, DP-2, and DP-3) adsorbed films with and without CDCA are shown in Figure 3. Generally, the absorptio of organic dyes on TiO2 films would change due to the effect of the deprotona adsorption process and the aggregation state of dyes on TiO2 films. When D and DP-3 adsorbed on the TiO2 surface, the absorption spectra of the three d broadened and absorption bands at the long-wavelength side are blue-shift com that of the solution spectra, which could be ascribed to the H-type aggregat deprotonation of the carboxylic acid upon being adsorbed on TiO2 [36,37]. In a was distinguished that the absorption spectra of the three dyes anchored onto film exposed a slightly broad outline compared to those in solution, which w for light-harvesting. CDCA was added to check the dye aggregation with DP dy film, as displayed in Figure 3. After the addition of CDCA, there was no obvious change for λmax of the t but the absorption intensity showed a decrease for DP-1 and DP-2, which may increased surface coverage of TiO2 with CDCA. In contrast, DP-3, with hexyllinker, displayed a slightly decreased absorption intensity when 10 mM C added, indicating that the hexyl substituent of the thiophene entity h suppression of dye aggregation.

Electrochemical Properties
The energetic arrangement of the highest occupied molecular orbital (HO lowest unoccupied molecular orbital (LUMO) energy levels is fundamental for a operation of the organic sensitizer in DSSCs. The electrochemical properties o DP-3 were analyzed by cyclic voltammetry in THF solution. The representa voltammograms of the dyes are shown in Figure 4 and the relevant electrochem After the addition of CDCA, there was no obvious change for λ max of the three dyes, but the absorption intensity showed a decrease for DP-1 and DP-2, which may be due to increased surface coverage of TiO 2 with CDCA. In contrast, DP-3, with hexyl-thiophene linker, displayed a slightly decreased absorption intensity when 10 mM CDCA was added, indicating that the hexyl substituent of the thiophene entity helps with suppression of dye aggregation.

Electrochemical Properties
The energetic arrangement of the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) energy levels is fundamental for an efficient operation of the organic sensitizer in DSSCs. The electrochemical properties of DP-1 to DP-3 were analyzed by cyclic voltammetry in THF solution. The representative cyclic voltammograms of the dyes are shown in Figure 4 and the relevant electrochemical data are presented in Table 1. All redox potentials are referenced to ferrocene utilized as an internal standard for calibrating the potential and calculating the HOMO levels. The excited state potential (E 0−0* ) of the sensitizer was estimated from the first oxidation potential (E ox ) at the ground state and the zero−zero excitation energy (E 0−0 ) estimated from the absorption onset. The assumed E 0−0* values (−1.10 to −1.09 V vs. NHE, see Table 1) are more negative than the conduction band edge energy level of the TiO 2 electrode (−0.5 V vs. NHE) [38], and the first oxidation potentials of the dyes DP-1, DP-2, DP-3 were measured to be 1.02, 1.01, and 1.00 V vs. NHE are more positive than the I − /I 3 − redox couple (~0.4 V vs. NHE) [39]. These results ensure favorable electron injection upon photoexcitation and regeneration of the dye after electron injection. mers 2021, 13, x FOR PEER REVIEW excited state potential (E0−0*) of the sensitizer was estimated from the firs potential (Eox) at the ground state and the zero−zero excitation energy (E0−0) esti the absorption onset. The assumed E0−0* values (−1.10 to −1.09 V vs. NHE, see more negative than the conduction band edge energy level of the TiO2 electr vs. NHE) [38], and the first oxidation potentials of the dyes DP-1, DP-2, measured to be 1.02, 1.01, and 1.00 V vs. NHE are more positive than the I − /I3 − r (~0.4 V vs. NHE) [39]. These results ensure favorable electron inje photoexcitation and regeneration of the dye after electron injection.

Photovoltaic Properties
The DSSCs of DP dyes were fabricated using these three dyes as photosen measured under AM 1.5 G irradiation (100 mW cm −2 ). The current density−v curves under illumination and in the dark are shown in Figure 5. The p performance statistics under a solar condition (AM 1.5) illumination are collec 2. All the devices exhibited power conversion efficiencies ranging from 1.5 Under the same condition, the power conversion efficiency of reference dye cell showed 7.38%. DSSCs based on DP-1 and DP-3 dyes exhibited ov conversion efficiencies of 1.53% and 2.92%, with JSC of 3.31 mA cm −2 and JS cm −2 , VOC of 0.64 V and VOC of 0.68 V, and FF of 0.72 and FF of 0.76, resp comparison, the device based on DP-2 gave a short-circuit photocurrent den 8.14 mA cm −2 , an open-circuit voltage (VOC) of 0.68 V, and a fill factor ( consistent with an overall conversion efficiency (η) of 3.80%. The VOC value o DP-3 was 40 mV higher than that of DP-1, which could be attributed to the s current for DP-2 and DP-3. In other words, DP-1 with furan moiety had th value and poor photovoltaic performance. The device efficiencies are in the o 2 > DP-3 > DP-1 and the performances of the devices based on DP-2 are standard cell based on ruthenium dye N719. The adsorbed dye densities of th on TiO2 were measured to be 3.70 × 10 −7 , 4.2 × 10 −7 , and 2.1 × 10 −7 mol/cm 2 for and DP-3, respectively. The low dye density on TiO2 of DP-3 might be attribu steric congestion of n-hexylthiophene linker owing to the presence of an ex chain. The higher cell efficiency of DP-2 than DP-3 and DP-1 is ascribed to the

Photovoltaic Properties
The DSSCs of DP dyes were fabricated using these three dyes as photosensitizers and measured under AM 1.5 G irradiation (100 mW cm −2 ). The current density−voltage (J−V) curves under illumination and in the dark are shown in Figure 5. The photovoltaic performance statistics under a solar condition (AM 1.5) illumination are collected in Table 2. All the devices exhibited power conversion efficiencies ranging from 1.53 to 3.96%. Under the same condition, the power conversion efficiency of reference dye N719-based cell showed 7.38%. DSSCs based on DP-1 and DP-3 dyes exhibited overall power conversion efficiencies of 1.53% and 2.92%, with J SC of 3.31 mA cm −2 and J SC = 5.62 mA cm −2 , V OC of 0.64 V and V OC of 0.68 V, and FF of 0.72 and FF of 0.76, respectively. In comparison, the device based on DP-2 gave a short-circuit photocurrent density (J SC ) of 8.14 mA cm −2 , an open-circuit voltage (V OC ) of 0.68 V, and a fill factor (FF) of 0.69, consistent with an overall conversion efficiency (η) of 3.80%. The V OC value of DP-2 and DP-3 was 40 mV higher than that of DP-1, which could be attributed to the smaller dark current for DP-2 and DP-3. In other words, DP-1 with furan moiety had the lowest Jsc value and poor photovoltaic performance. The device efficiencies are in the order of DP-2 > DP-3 > DP-1 and the performances of the devices based on DP-2 are~51% of the standard cell based on ruthenium dye N719. The adsorbed dye densities of the sensitizers on TiO 2 were measured to be 3.70 × 10 −7 , 4.2 × 10 −7 , and 2.1 × 10 −7 mol/cm 2 for DP-1, DP-2, and DP-3, respectively. The low dye density on TiO 2 of DP-3 might be attributed to more steric congestion of n-hexylthiophene linker owing to the presence of an extra n-hexyl chain. The higher cell efficiency of DP-2 than DP-3 and DP-1 is ascribed to the better light-harvesting of DP-2 and its slightly higher dye density amount on TiO 2 film compared to the other two dyes. DP-2 has better light-harvesting efficiency in the film among the three DP dyes; it also has greatly higher incident monochromatic photo-to-current conversion efficiency (IPCE) values than the other two dyes in the range of 400-700 nm. The result indicates that DP-2 has the highest photocurrent due to higher dye absorption density on TiO 2 and faster and more effective electron injection efficiency.
around 520 nm (60%) is higher than that of DP-1 (39%) or DP-3 (17%). IPCE is the light-harvesting efficiency of the photoelectrode and electron injections charge collection efficiency. The higher IPCE values of the DSSC devices base or DP-3 with thiophene or hexylthiophene linkers have higher electron transfer the dye with furan linker. This indicated that the introduction of heteroaromatic ring into phenoxazine-based sensitizer structure has a positive    The incident monochromatic photo-to-current conversion efficiency (IPCE) plots of the cells are shown in Figure 6, respectively. Well-consistent with adsorption spectra in TiO 2 , DP-2 exhibited broader and higher IPCE efficiencies. The IPCE value for DP-2 at around 520 nm (60%) is higher than that of DP-1 (39%) or DP-3 (17%). IPCE is related to the light-harvesting efficiency of the photoelectrode and electron injections yield and charge collection efficiency. The higher IPCE values of the DSSC devices based on DP-2 or DP-3 with thiophene or hexylthiophene linkers have higher electron transfer yield than the dye with furan linker. This indicated that the introduction of thiophene heteroaromatic ring into phenoxazine-based sensitizer structure has a positive effect.
Suppression of dye aggregation was also supported by the blue shift of the absorption spectra of the DP dyes on the TiO 2 film when CDCA was added. The cell performance data with CDCA co-adsorbent are summarized in Table 2. The three dyes have slight improvements in the cell performance upon addition of CDCA: PCE = 1.80%, J SC = 3.64 mAcm −2 , V OC = 0.65 V, FF = 0.76 for DP-1; PCE = 3.96%, J SC = 8.52 mAcm −2 , V OC = 0.69 V, FF = 0.67 for DP-2; PCE = 3.10%, J SC = 6.23 mAcm −2 , V OC = 0.68 V, FF = 0.74 for DP-3. For DP-1 to DP-3, the cell performance improved only marginally upon addition of CDCA 10 mM, V OC stayed almost the same, whereas J SC continued to increase. Hence, anti-aggregation of the dyes was alleviated by CDCA adsorption. By adding 10 mM CDCA, devices with DP-2 showed the better J SC , V OC , and conversion efficiency of 3.96% (

Electrochemical Impedance Spectroscopy Analysis
The electrochemical impedance spectroscopy (EIS) is a very useful technique to understand the electron injection and recombination processes in DSSCs [40]. Electrochemical impedance spectroscopy (EIS) was used to further evaluate the important interfacial charge transfer processes in a DSSC. Generally, there are three semicircles showed in the EIS spectrum, which correspond to electron recombination resistances (Rrec) at the interfaces of photoanode/dye/electrolyte, charge-transfer resistance at the photoanode/dye/electrolyte interface (RCT), and Warburg diffusion process of electrolyte (Zw), as typically reported for other DSSC devices [41][42][43]. The electrochemical impedance spectra (EIS) of DSSCs were obtained under a forward bias of −0.70 V in the dark to elucidate correlation of V OC with those dyes, and the Nyquist plots for DSSCs based on DP-1 to DP-3 are shown in Figure 7. The large semicircle in the Nyquist plots is attributed to the charge recombination resistance between the TiO 2 and the electrolyte (R rec ), where the larger R rec value suggests the smaller dark current. The radius of the biggest semicircle increases in the order of DP-1 < DP-3 < DP-2. The cell of DP-2 and DP-3 exhibits a much larger resistance value than that of DP-1, which is consistent with its smaller dark current and larger V OC measured.
Polymers 2021, 13, x FOR PEER REVIEW Suppression of dye aggregation was also supported by the blue shift absorption spectra of the DP dyes on the TiO2 film when CDCA was added. Th performance data with CDCA co-adsorbent are summarized in Table 2 For DP-1 to DP-3, the cell performance improved only marginally upon addit CDCA 10 mM, VOC stayed almost the same, whereas JSC continued to increase. Hence aggregation of the dyes was alleviated by CDCA adsorption. By adding 10 mM C devices with DP-2 showed the better JSC, VOC, and conversion efficiency of 3.96% (Ta

Electrochemical Impedance Spectroscopy Analysis
The electrochemical impedance spectroscopy (EIS) is a very useful techniq understand the electron injection and recombination processes in DSSCs Electrochemical impedance spectroscopy (EIS) was used to further evaluate the imp interfacial charge transfer processes in a DSSC. Generally, there are three semi showed in the EIS spectrum, which correspond to electron recombination resis (Rrec) at the interfaces of photoanode/dye/electrolyte, charge-transfer resistance photoanode/dye/electrolyte interface (RCT), and Warburg diffusion process of elect (Zw), as typically reported for other DSSC devices [41][42][43]. The electroche impedance spectra (EIS) of DSSCs were obtained under a forward bias of −0.70 V dark to elucidate correlation of VOC with those dyes, and the Nyquist plots for D based on DP-1 to DP-3 are shown in Figure 7. The large semicircle in the Nyquist p attributed to the charge recombination resistance between the TiO2 and the elect (Rrec), where the larger Rrec value suggests the smaller dark current. The radius biggest semicircle increases in the order of DP-1 < DP-3 < DP-2. The cell of DP-2 an 3 exhibits a much larger resistance value than that of DP-1, which is consistent w smaller dark current and larger VOC measured.

Computational Calculation
In order to gain insight into the relationship between the geometrical and electronic properties of the DP dyes, the dyes DP-1 to DP-3 were further investigated through theoretical calculations. The results for theoretical computation are included in Table 3. Figure 8 shows the ground-state geometries of the dyes with the dihedral angles between the two neighboring conjugated segments indicated. In the optimized structure, DP-1 has a nearly planar structure where the torsion angle between phenoxazine and the furan entity is almost 0 • . The planarity structure of DP-1 can increase the stacking of the dye molecules, inducing more dye aggregation. In comparison, the dihedral angle between the phenoxazine and the thiophene entity is larger than 20 • for DP-2 and DP-3. The smaller planarity of DP-2 and DP-3 leads to better charge separation between the phenoxazine unit and accepter unit. The electron distributions of the HOMOs and LUMOs for DP-1, DP-2, and DP-3 are illustrated in Figure 9. The HOMOs of DP-1, DP-2, and DP-3 are delocalized on the entire molecule including acceptor, whereas the LUMO and LUMO+1 of these molecules are mainly distributed from the π-spacer acceptor to the acceptor. In Table 3, the S 0 → S 1 transition is nearly a HOMO → LUMO transition. Therefore, the lowest energy absorption has charge transfer character for these dyes. The more intense electronic absorptions in DP-2 and DP-3 than DP-1 are supported by its larger computed oscillation strength (f). The Mulliken charges variation for the S 1 and S 2 states were calculated from the TD-DFT results. Differences in the Mulliken charges in the excited and the ground states were calculated and gathered into several segments, heteroaromatic ring (F, T, T 1 ), phenoxazine (Poz), heteroaromatic ring (F', T', T 1 ), and 2-cyanoacrylic acid (Ac), 2-cyanoacrylic acid (Ac') to estimate the extent of charge separation upon excitation. Figure 10 displays the changes in Mulliken charges of the dyes for the S 0 → S 1 and S 0 → S 2 transitions. In DP-1 to DP-3, the positive charges exist at phenoxazine in the dyes for S 0 →S 1 and S 0 →S 2 transitions. On the other hand, DP-1 to DP-3 have prominent negative charges at both acceptors for both S 0 →S 1 and S 0 →S 2 transitions, indicating that both acceptors can function as the electron injection channels.

Conclusions
In summary, we reported the new phenoxazine-based organic dyes containing a 2ethylhexyl substituent at the nitrogen atom of the phenoxazine and two 2-cyanoacrylic

Conclusions
In summary, we reported the new phenoxazine-based organic dyes containing a 2ethylhexyl substituent at the nitrogen atom of the phenoxazine and two 2-cyanoacrylic acids as the acceptors in addition to anchors. The influences of the various π bridges on the photophysical, electrochemical, and photovoltaic properties of these sensitizers were investigated. The results of calculation and experiments clearly demonstrate that photophysical properties can be regulated by introducing different π-conjugated bridges. Then, the performance of the DSSCs based on these dyes were tested and analyzed, and, among which, the DP-2 cell shows the best PCE of 3.80% without CDCA among all. DSSCs using these three DP dyes as the sensitizers showed efficiencies ranging from 1.53 to 3.80% without CDCA under simulated AM1.5G irradiation. Upon addition of CDCA as a co-adsorbent, the cell efficiency has been further improved to 3.96% for DSSCs based on DP-2, which is about 54% of the N719-based standard cell. Our future work will focus on optimization of molecular structure to fine-tune the energy levels of the dye toward higher V OC , J SC , and panchromatic DSSCs.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.