Molecular engineering of quinoxaline dyes toward more efficient sensitizers for dye-sensitized solar cells

Abstract

D–A–π–A-featured organic dyes incorporating diphenylquinoxaline unit (such as IQ4) have shown great potential in anti-aggregation and broadening spectral response in the field of dye-sensitized solar cells (DSSCs). The crucial restriction for quinoxaline-based cell to attain higher efficiency is the relatively low photocurrent density (J_SC). In the present work, three novel push–pull dyes only differing in electron donors, have been designed based on the IQ4 backbone, in order to further improve the light-harvesting capability of quinoxaline dyes and to examine the donor influence on dye performance. Theoretical analysis of the factors correlated with the J_SC and open-circuit photovoltage (V_OC) demonstrate that, relative to the parent IQ4 dye, the NIQ4 dye bearing the elegant N-annulated perylene donor shows a good performance in light harvesting, electron injection, and dye regeneration, indicating an increased J_SC potential for the related cell. Furthermore, despite possessing a smaller vertical dipole moment, the improved blocking effect of NIQ4 not only prevents unfavorable self-aggregation, but also effectively inhibits the parasitic back-recombination. Therefore, the NIQ4 is proposed to be a potential dye in DSSC applications.

---

1. Introduction

Dye-sensitized solar cells (DSSC) are attracting widespread attention^1–5 as an alternative inexpensive and environmentally friendly method for sunlight-to electricity conversion since the seminal work of O'Regan and Grätzel in 1991.⁶ The performance of DSSCs depends on the nature of sensitizer(s), photoanode, counter electrode, electrolyte, and the complicated interactions between the above components. In the past decades, great efforts have been made to explore the structure–property relationship^7–12 and screen optimal materials for efficient DSSCs.^13,14

The vital role in light harvesting and electron injection at dye/semiconductor interface has led to an explosion in dye sensitizer research. There are two types of dyes commonly used, namely, metal–organic¹⁵ and organic dyes.^2,16 Particularly, organic dyes with D–π–A configuration are popular for their non-toxicity, high molar extinction coefficients, flexible molecular engineering, and material abundance. Experimentally, the organic dyes C259 and C239 co-sensitized DSSC has been reported to achieve the record efficiency of 12.8%.¹⁷ In addition, the D–A–π–A-featured organic dyes incorporating an additional electron-withdrawing unit have been designed to solve the self-aggregation problem between dyes and further broaden the spectral response.^18–27 In this field, dyes with quinoxaline unit seem to be impressive.^28–31 For instance, the DSSC based on IQ4 exhibits desirable short-circuit current density (J_sc) of 17.55 mA cm⁻² as well as high open-circuit voltage (V_oc) of 0.74 V with a fill factor (FF) of 0.71 under AM 1.5 illumination. This should be mainly ascribed to the strong anti-aggregation ability, the improved photostability and thermal stability of the 2,3-diphenylquinoxaline unit in IQ4.³² However, the relatively low photocurrent density restricts such cell to attain higher efficiency.³³ As well-known, the nature of the electron donor within an organic dye not only affects the regeneration rate of an oxidized dye, the interaction between the dye-TiO₂ layer and the electrolyte, but also plays a significant role in modulating molecular energy levels and light-harvesting capability of dyes.^34,35 Therefore, the scrutiny of efficient organic dyes based on judicious selection of electron donors is much beneficial.^36–41 Besides the commonly used bulky triarylamine-based donor^2,42,43 and indoline moiety,⁴⁴ favorable electron-donating ability of the donor has also been observed for JD21 containing the ullazine group,⁴⁵ and WW-6 functionalized by the N-annulated perylene (NP) building block.⁴⁶

In the present work, three donor moieties (triarylamine, ullazine, and NP groups, respectively) were introduced based on the IQ4 backbone, in order to further improve the light-harvesting ability of quinoxaline dyes, and to examine the donor influence on dye performance (Scheme 1). Theoretical analysis demonstrates that the solar cell based on NIQ4 with NP group is expected to exhibit a higher J_SC and a comparable V_OC, compared to the reference IQ4 dye. Therefore, the NIQ4 dye is proposed to be a promising candidate in DSSC applications.

Scheme 1 Chemical structures of the investigated dyes.

2. Computational details

All calculations have been performed with Gaussian 09 program package.⁴⁷ The ground (S₀) and the first excited-state (S₁) geometries of the free and adsorbed dyes were optimized using the density functional theory (DFT) B3LYP functional^48–50 coupled with the 6-31G(d, p) basis set for C, H, O, N, S atoms, and a LANL2DZ basis set for Ti atom. Frequency calculations were then carried out to confirm the nature of obtained minima at the same theory level as geometry optimizations. To get the reorganization energy (λ) which could affect the efficiency of electron injection, the cationic state of free dyes were optimized at the UB3LYP/6-31G(d, p) level. It has been reported that the B3LYP functional works very well in the description of the electronic structure^51–56 but could not give an accurate representation of charge–transfer excited states,⁵⁷ and this can be significantly improved by means of the long-range corrected functionals (e.g. CAM-B3LYP⁵⁸).^54–56 Therefore, the absorption spectra associated with the excited state of free and adsorbed dyes were simulated using the CAM-B3LYP functional with 6-31G(d, p) and LANL2DZ basis sets based on the optimized ground-state geometries, and the spectral curves were obtained within the SWizard program (http://www.sg-chem.net/swizard/)⁵⁹ using the pseudo-Voigt model (a convolution of the Gaussian and Lorentzian functions) with full-width at half-height of 3000 cm⁻¹. The charge transfer process intra-dyes and inter-dye/titanium surface at the maximum absorption (λ_max) were qualitatively displayed by electron density differences (EDD) maps and quantitatively described by natural bond orbital (NBO) analysis. The corresponding charge transfer parameters were obtained by Multiwfn 3.3.6 program.⁶⁰ All calculations above were carried out with bulk solvent effect of dichloromethane (CH₂Cl₂) included by using the conducting polarizable continuum model (C-PCM).^61–63

3. Results and discussion

The power conversion efficiency (η) of a solar cell is determined by J_SC, V_OC, and fill factor (FF), as compared to the incident solar power (P_inc):

J_SC in DSSCs can be described by:^64–66

In which, SI denotes the solar radiation intensity, e is the unit charge, h is the Planck constant, c is the light speed in vacuum, and IPCE is monochromatic incident phonon to current conversion efficiency, it can be produced by using the following equation:^65,67

IPCE(λ) = LHE(λ)Φ_injη_regη_coll (3)

Where, LHE(λ) is the light-harvesting efficiency at a given wavelength, Φ_inj represents the electron injection efficiency, η_reg is the efficiency for dye regeneration, and η_coll denotes the charge collection efficiency. Note that IPCE is sometimes expressed as the product of LHE, injection efficiency, and charge collection efficiency.^37,68 Here, we separate the regeneration efficiency from the whole charge collection efficiency, because the complex charge collection efficiency is determined by the time constant for transport and recombination of the conduction band electrons injected into the nanocrystalline TiO₂ film,⁶⁵ which are not the scope of this work. We aim at figuring out the structural aspects related to regeneration efficiency.

As for V_OC in DSSCs, it can be calculated by:⁶⁹

Here q is the unit charge, E_CB is the conduction band edge (CBE) of the semiconductor, k is the Boltzmann constant, T is the absolute temperature, n_c is the number of electrons in the conduction band, N_CB is the density of accessible states in the conduction band,⁶⁸ and E_redox is the electrolyte redox potential. The shift of E_CB (ΔCB) upon dye adsorption can be roughly expressed as:⁷⁰In this expression, γ is the surface concentration of dyes, μ_normal denotes the dipole moment of individual dye molecules perpendicular to the surface of the semiconductor, ε₀ and ε are the vacuum permittivity and the dielectric permittivity, respectively.

Based on these theoretical criterions, the parameters regulating J_SC and V_OC were separately evaluated in the following section to examine the electron donor influence on dye performance.

3.1 Evaluating J_sc

3.1.1 Light-harvesting efficiency of dye molecules. It is known that the electron-donating capacity of donor group plays an important role in determining the energetic positions of frontier molecular orbitals of dyes, which is correlated with the light-harvesting efficiency of a dye and the charge transfer kinetics at the titanium/dye/electrolyte interface, such as dye regeneration and charge recombination. Thus, at the first beginning of our study, the electron-donating capability of the four electron-donor groups (i.e. indoline, triarylamine, ullazine, and N-annulated perylene (NP) units) was evaluated by computing the energy levels of their highest occupied molecular orbitals (HOMOs). As depicted in Fig. S1,† the general order of the HOMO level for the four donors is NP (−4.93 eV) < indoline (−4.87 eV) < ullazine (−4.81 eV) < triarylamine (−4.77 eV). The HOMO in NP is the lowest in energy, indicating the weakest electron-donating ability of the NP unit. This characteristic of NP may be helpful in dye regeneration. Meanwhile, the triarylamine group exhibits the strongest electron-donating ability, which may narrow the energy gap of the frontier molecular orbitals and further red-shift the absorption spectrum of the corresponding dye, in comparison with dyes based on the other three groups.

It is conceivable that the molecules investigated here should have several possible conformations, particularly in the π-linker of quinoxaline dyes. Taking IQ4 as the model dye, three possible trans-/cis-conformations (A, B, C) have been optimized at the B3LYP/6-31G(d, p) level. With the same theory level, frequency analysis was performed to confirm the nature of obtained minima. It is shown in Fig. 1 that IQ4-B with a linear π-linker is the most stable conformer with the lowest relative energy (ΔE = 0.00 kcal mol⁻¹). This is consistent with an original intention in the design of dye sensitizers, that is, a developed sensitizer should stand perpendicularly on (rather than lie on) the surface of semiconductor. Based on this, the geometry optimizations of the newly developed push–pull dyes were carried out with the replacement of indoline unit in IQ4 by triarylamine (TIQ4), ullazine (UIQ4), and NP (NIQ4) donor moieties, respectively. The bond lengths and dihedral angles between the electron-donor group and the 2,3-diphenylquinoxaline segment are listed in Table S1.† There exist small differences (within 0.02 Å) in the distance between the two link atoms, and the largest torsional angle of −51.0° between the NP donor group and quinoxaline segment. This twist in NIQ4 may result in a larger localization of π-electron at donor moiety from HOMO and a decreased electron density overlap between HOMO and the lowest unoccupied molecular orbital (LUMO), endowing an adverse contribution on the absorption coefficient of the related dye.³⁴

Fig. 1 The optimized geometries and relative energies of IQ4 isomers with respect to the IQ4-B species (ΔE is in kcal mol⁻¹).

Based on the optimized geometries, time-dependent DFT (TDDFT) calculations were done at the CAM-B3LYP/6-31G(d, p) (LANL2DZ) level to obtain UV-Vis spectra and light-harvesting efficiency (LHE) of dyes. LHE at the certain wavelength (LHE(λ)) can be calculated according to:^71,72

LHE(λ) = 1−10^−ε(λ)bc (6)

Where, ε(λ) is the molar absorption coefficient at given wavelength, b is the thickness of TiO₂ film, and c is the dye concentration on TiO₂ surface. Here, b and c are assumed to be 10 μm and 300 mmol L⁻¹, respectively.^64,73,74 The obtained results were shown in Table 1 and Fig. 2. Though the calculation (λ^cal_max = 468 nm) significantly underestimates the absorption of the reference IQ4 dye (λ^exp_max = 529 nm), we believe that the results obtained under the same level should give a useful indication to the optical property of several similar dyes. Furthermore, it is shown in Fig. 2 that all designed dyes (TIQ4 and NIQ4) except for UIQ4 exhibit broad absorption in the visible region. This may eventually lead to a broad spectral response with high LHE values for TIQ4 and NIQ4. In this case, the NIQ4 dye functionalized by N-annulated perylene group seems to be the winner molecule with the most red-shifted (λ^cal_max = 477 nm) and broadest spectrum (the LHE reaches close to 100% in the range of 300–570 nm). The light-harvesting capability of dyes follows an order of NIQ4 > TIQ4 > IQ4 > UIQ4, which is not in line with the electron-donating ability of donor groups.

Table 1 The calculated maximum absorption wavelengths (λ^cal_max), oscillator strengths (f), transition natures of dyes (H = HOMO, L = LUMO), and electron density difference (EDD) maps between ground and the first excited state (the blue and purple zones correspond to density decrement and increment, respectively)

Dye	λ^calmax/nm	f	Assignment	S0 → S1
a The experimental value was taken from ref. 32.
UIQ4	426	0.31	H → L (66%)
TIQ4	459	1.33	H → L (59%)
IQ4	468	1.28	H → L (69%)
IQ4^a	529	—	—	—
NIQ4	477	1.43	H→ L (66%)
IQ4-TiO2	519	1.09	H → L (77%)
NIQ4-TiO2	544	0.55	H → L (81%)

---

Fig. 2 The absorption spectra (solid line) and light-harvesting efficiency curves (dash line) for IQ4 and its derivatives.

Further investigation shows that the λ^cal_max for all dyes is of S₀ → S₁ character (Table 1), which mainly corresponds to the electron transition from HOMO to LUMO. The composition⁷⁵ and the spatial distribution of the related frontier orbitals have been shown in Table 2 and Fig. 3. It can be seen in Table 2 that the electron-donors contribute over 80% to HOMOs of dyes, and the π-linkers and acceptor parts are mainly responsible for LUMO. This may be ascribed to the relatively larger twisted angles between donor group and quinoxaline segment in dyes (Table S1†). Moreover, the donors of the newly designed dyes contribute more (>90%) to the HOMO than that of IQ4. The spatial distribution of the frontier orbitals in Fig. 3 is found to be obviously consistent with the orbital composition analysis, where the HOMOs of all dyes show localization mainly at electron-donor moiety, while LUMOs delocalize over the π-conjugated bridge and acceptor part. This will result in efficient charge separations, and can be visually described by the electronic density difference (EDD) plots in Table 1. From the EDD maps, we can see that the electron densities decrease particularly in electron-donor components and increase in the π-linkers and 2-cyanoacrylic acid anchor. In order to quantitatively describe the intramolecular charge transfer (CT) of dyes at λ^cal_max, the distance of charge transfer (D), the amount of transferred charge (Δq), and t parameter (defining the through space CT character of a transition) were calculated within Multiwfn 3.3.6 program, and have been summarized in Table 3. The optimal CT performance is obtained for NIQ4 that allows the transfer of 0.95 e⁻ over 3.88 Å with good through space character (t = 3.33 Å).⁷⁶ Furthermore, the computed energetic positions of LUMOs for all dyes are more positive than the CB of TiO₂ (−4.0 eV vs. vacuum), ensuring an effective electron injection from the excited dye to titanium surface, and the HOMO of NIQ4 shows the largest amount of lowering compared to I⁻/I₃⁻ redox potential (−4.8 eV vs. vacuum), indicating a favorable dye regeneration process (Fig. 4).⁷⁷

Table 2 Molecular orbital composition (in %) of the HOMO and LUMO orbitals for the investigated dyes

Dye	Orbital	Donor	Quinoxaline	Thiophene	Anchor
UIQ4	HOMO	99.5	0.50	0.00	0.00
	LUMO	2.68	32.6	29.1	35.6
TIQ4	HOMO	90.5	6.50	1.84	1.15
	LUMO	4.34	31.5	28.9	35.2
IQ4	HOMO	85.0	10.1	3.00	1.88
	LUMO	4.36	31.1	29.2	35.4
NIQ4	HOMO	91.0	6.19	1.72	1.08
	LUMO	4.61	31.9	28.8	34.6

---

Fig. 3 The frontier molecular orbitals of dyes corresponding to the maximum absorption.

Table 3 Ground and excited-state oxidation potentials, vertical transition energies, driving forces for dye regeneration and electron injection, reorganization energies, and charge transfer parameters of dyes

Dye	E^dyeOX/eV	Eλmax/eV	E^dye*OX/eV	ΔGreg/eV	λreg/eV	ΔGinj/eV	q^CT/e^−	d^CT/Å	t/Å
a The experimental values were taken from ref. 32.
UIQ4	−4.86	−2.91	−1.95	0.06	0.22	2.05	0.80	2.10	6.83
TIQ4	−4.85	−2.70	−2.15	0.05	0.25	1.85	0.83	3.72	4.80
IQ4	−4.91	−2.65	−2.26	0.11	0.11	1.74	0.76	4.12	2.70
IQ4^a	−5.07	−2.34	−2.98	—	—	—	—	—	—
NIQ4	−4.94	−2.60	−2.34	0.14	0.14	1.66	0.95	3.88	3.33

---

Fig. 4 The energy-level arrangements of dyes together with the experimental conduction band edge of TiO₂ and the redox potential of I⁻/I₃⁻.

Based on the discussions for free dyes, the reference dye IQ4 and the designed dye NIQ4 were chosen to explore the interfacial interactions between dye molecules and semiconductor surface. To save the cost, a simple dye-Ti(OH)₃·H₂O model was adopted, which has been successfully employed to simulate the photoelectric properties of dye-TiO₂ complex.^38,68,78,79 Using the same methods as that for free dyes, the electronic and optical properties of dye-TiO₂ complex were simulated. The related results have been given in Table 1 and Fig. 5. It can be found that both IQ4 and NIQ4 in bound state red-shift the absorption spectra and broaden the spectral response compared to those in isolated state. Though weaker in strength at λ^cal_max, the adsorbed NIQ4 exhibits more red-shift at λ^cal_max (544 nm) and absorbs more efficiently in solar spectrum (LHE close to 100% from 300 to 700 nm), suggesting a stronger electron coupling between the excited NIQ4 and the chelated Ti atom. The EDD plots in Table 1 give a clear picture for the charge carrier flux of adsorbed dyes at λ^cal_max, where electron densities accelerate mostly at the dye/titanium interface. The titanium surface was also expanded to a (TiO₂)₉ cluster to confirm the reliability of the simple Ti(OH)₃·H₂O model. It is revealed that the trend of absorption spectra for dye-(TiO₂)₉ complexes is similar to that obtained with the simple dye-Ti(OH)₃·H₂O model (Fig. S2 and Table S2†).

Fig. 5 The absorption spectra (solid line) and light-harvesting efficiency curves (dash line) for adsorbed IQ4 (black) and NIQ4 (red).

3.1.2 Electron injection efficiency. Electron injection efficiency (Φ_inj) associated with J_SC can be indirectly characterized by the electron injection driving force (ΔG_inj) from excited dyes to the CB band of semiconductor. ΔG_inj can be determined by the following equation, assuming that the electron injection occurs from the unrelaxed excited state:⁸⁰In which, E^SC_CB is the conduction band edge of TiO₂ (E^SC_CB = −4.0 eV vs. vacuum), and E^dye*_OX is the excited state oxidation potential and estimated by:

In which, E^dye_OX is the ground state oxidation potential and E_λmax is the vertical transition energy from the ground state to excited state. We list these values in Table 3. It can be seen that the E^dye_OX of IQ4 (−5.07 eV) is well reproduced by our calculation (−4.91 eV), whereas, the overestimation of the vertical transition energy for IQ4 directly leads to a larger deviation of the calculated E^dye*_OX (−2.26 eV) from experimental measurements (−2.98 eV), and thus a larger ΔG_inj (1.74 eV). Moreover, it reveals in Table 3 that the replacement of the indoline unit in IQ4 by triarylamine (TIQ4), ullazine (UIQ4), and NP (NIQ4) moieties, respectively, displays a more significant effect on E^dye*_OX than that on E^dye_OX. Though the ΔG_inj value of IQ4 is the smallest in all investigated dyes, it is sufficient to inject electrons from excited dyes to CB edge of TiO₂.⁷⁷

3.1.3 Dye regeneration efficiency. For an effective DSSC, the dye regeneration completed by the reducing agent is vital in achieving a higher J_SC. A simplified, but informative, way to investigate dye regeneration relies on the analysis of the spin density distributions for the oxidized systems.³⁵ Obtained by the (U)DFT/B3LYP calculations, the contour plots of the spin densities for such systems together with the contributions from the donor moiety have been summarized in Table 4. As depicted in Table 4, the spin density for all oxidized dyes mainly localizes at the donor moiety of the dye, and the contribution from the electron donor shows a higher value for the newly designed dyes (UIQ4˙⁺ (99.7%), TIQ4˙⁺ (97.2%), and NIQ4˙⁺ (95.0%)), relative to IQ4˙⁺ (89.2%). This effect could enhance the interaction between the designed dyes and the I⁻/I₃⁻ redox couple, and therefore the rate of dye regeneration, compared to IQ4˙⁺, where holes reside on the π-bridge in a larger proportion.

Table 4 Contour plots of the spin density for the oxidized species of the investigated dyes resulting from (U)DFT/B3LYP/C–PCM(CH₂Cl₂) calculations (values given in parentheses represent the spin density contribution of the donor moiety)

---

The quantitative way to measure the efficiency of dye regeneration is computing the internal reorganization energy (λ_reg)³⁴ and the regeneration driving force (ΔG_reg) for dyestuff. The λ_reg can be calculated by:

λ_reg = [E⁺₀ − E⁺₊] + [E⁰₊ − E₀] (9)

In which, E⁺₊ and E₀ are optimized energies of the cationic and neutral forms of a dye, E⁺₀ is the energy of the cation at the neutral geometry, and E⁰₊ is the energy of the neutral molecule at the cationic geometry. The ΔG_reg is determined by the energy difference between the redox potential of I⁻/I₃⁻ (E^electrolyte_redox = − 4.8 eV) and E^dye_OX:

ΔG_reg = E^electrolyte_redox − E^dye_ox (10)

One can find in Table 3 that the NIQ4 dye presents the largest driving force of 0.14 eV for dye regeneration as well as the comparable reorganization energy (0.14 eV) to IQ4 (0.11 eV), which together with the favorable spin density distribution of the cationic radical, sufficient ΔG_inj, and better light-harvesting ability may lead to a higher J_sc for NIQ4-sensitized solar cell. As for UIQ4 and TIQ4, the relatively larger λ_reg (UIQ4/0.22 eV, TIQ4/0.25 eV) and smaller ΔG_reg (UIQ4/0.06 eV, TIQ4/0.05 eV), despite the favorable spin density contribution from donor moieties may decrease the efficiency of dye regeneration, and thus result in a lower J_sc for the related cell. In this context, it is safe to say that the N-annulated perylene unit instead of indoline in quinoxaline dyes is beneficial in enhancing the short-circuit current density.

3.2 Evaluating V_oc

Besides the J_sc, a potential cell should also possess a high open-circuit photovoltage (V_oc) value to reach high power conversion efficiency (η). The V_oc is directly relevant with the CB energy (E_CB). It is known that the adsorption of a dye molecule on TiO₂ surface can induce a shift of the conduction band profile of TiO₂ (ΔCB), and this effect can be divided into two aspects, namely, charge transfer (CT) and electrostatic (EL) potential.⁸¹ In the present work, the investigation of electrostatic effect was simplified by calculating the dipole moment of the free dyes at the geometries of dyes adsorbed onto TiO₂ surface (μ_normal),⁸² while the exploration of CT at dye/TiO₂ interface can be qualitatively or semi-quantitatively realized through the analysis of the difference of charge distribution in the semiconductor between the ground and the first excited state of dye-TiO₂ complex.^68,79 A larger μ_normal pointing outward the semi-conductor surface and a more charge distribution in TiO₂ will lead to a larger ΔCB. Considering the bidentate chelating mode between IQ4/NIQ4 and TiO₂, the C₂ axis of the carboxylate in the dye is set parallel to the x-axis, and the yz plane is parallel to the semiconductor surface. Thus, the μ_x of dye molecules is referred to as μ_normal. The calculated results are shown in Table 5. The μ_normal of IQ4 (10.3 D) is almost the double value of NIQ4 (5.78 D). As for CT effect, the natural population analysis (NPA) demonstrates that the number of photo-injected electrons in TiO₂ for NIQ4 (0.041 e) is slightly more than that for IQ4 (0.037 e), which says in favor of NIQ4 as a good electron-injection precursor.

Table 5 The natural bond orbital analysis (atomic charge) of the ground (S₀) and the first excited state (S₁) of dye-Ti(OH)₃·H₂O adducts (in e)

Dye	S0					S1				Δq^b
	donor	π-spacer	Anchor	Ti(OH)3·H2O	μnormal^a	donor	π-spacer	anchor	Ti(OH)3·H2O
a μnormal is the vertical dipole moment (Debye) of the free dyes at the geometries of dyes adsorbed onto TiO2 surface.b Δq is the charge density differences on Ti(OH)3·H2O between S0 and S1.
IQ4	0.102	0.125	−0.657	0.430	10.3	0.0495	0.231	−0.673	0.393	0.037
NIQ4	0.0434	0.163	−0.642	0.436	5.78	0.0454	0.229	−0.669	0.395	0.041

---

On the other hand, charge recombination between TiO₂-injected electrons and the oxidized dye or the electrolyte (I⁻/I₃⁻) can substantially decrease the charge density in the semiconductor, and thus leads to a reduced V_oc value. Differences in the recombination kinetics between TiO₂-injected electrons and the dye cationic radicals can be reflected from the twisted dihedral angle between the dye donor and π-conjugated moiety. Looking at the geometry parameters of oxidized dyes reported in Table S1,† the selected dihedral angle is 38.9° for IQ4˙⁺ and 49.3° for NIQ4˙⁺. The more twisted angle between NP unit and quinoxaline in NIQ4 is believed to hamper the back-recombination from the injected-electrons to dye donor, exerting a desirable effect on charge density in TiO₂. In addition, the improved blocking effect of NIQ4 (the bulky N-annulated perylene moiety and bis(2,4-dihexyloxy)benzene substituted quinoxaline³⁶) compared to IQ4 not only prevents unfavorable aggregation between dyes, but also can form a compact dye layer to effectively inhibit the approach of the redox couple to the TiO₂ surface,⁴ consequently an enhanced device performance. The retarded electron–hole recombination in NIQ4 is supposed to compensate well the undesired smaller μ_normal, and therefore makes NIQ4 endow a comparable contribution on the photovoltage compared to the reported IQ4 dye.

To this point, we can conclude that NIQ4 is a potential candidate for DSSC applications. A solar cell sensitized by NIQ4 is expected to exhibit a higher J_SC and a comparable V_OC to IQ4, and thus an improved DSSC photovoltaic performance.

4. Conclusions

In this study, with IQ4 as the prototype dye, three novel quinoxaline sensitizers (UIQ4, TIQ4, and NIQ4) were designed with electron donor modification, aiming to identify a promising dye for DSSC applications. Factors correlated with the short-circuit current density (J_SC) and open-circuit photovoltage (V_OC), including light-harvesting efficiency (LHE), electron injection efficiency (ΔG_inj), dye regeneration efficiency (ΔG_reg), the vertical dipole moment (μ_normal), and the number of photo-injected electrons in TiO₂, respectively, have been studied in detail by means of DFT and TDDFT approaches. It is found that compared to the reported IQ4 dye bearing indoline donor, the NIQ4 dye functionalized by N-annulated perylene (NP) donor displays the more red-shifted (λ_max = 477 nm) absorption and broader spectral domain (the LHE reaches close to 100% in the range of 300–570 nm). The good light-harvesting and electron injection ability together with the comparable regeneration efficiency of NIQ4 compared to those of IQ4 may eventually enhance the J_SC potential of NIQ4-based cell. Furthermore, despite possessing a smaller μ_normal, the improved blocking effect of NIQ4 not only prevents unfavorable aggregation between dyes, but also effectively impedes the parasitic back-recombination. Therefore, we can conclude that a solar cell fabricated with NIQ4 is expected to exhibit a higher J_SC and a comparable V_OC to that with IQ4, and thus an improved DSSC photovoltaic performance.

The results of this paper emphasize the significance of the electron donor in controlling the kinetic parameters, which are responsible for the device performance. We hope our work will provide a rational guideline for the future design of push–pull organic dyes towards more efficient dye-sensitized solar cells.

Acknowledgements

The authors gratefully acknowledge financial support from the Major State Basic Research Development Programs of China (2011CBA00701), the National Natural Science Foundation of China (21473010, 21303007), the Excellent Young Scholars Research Fund of Beijing Institute of Technology (2013YR1917), and Beijing Key Laboratory for Chemical Power Source and Green Catalysis (2013CX02031). This work is also supported by the opening project of State Key Laboratory of Explosion Science and Technology (Beijing Institute of Technology) (ZDKT12-03).

References

1. B. E. Hardin, H. J. Snaith and M. D. McGehee, Nat. Photonics, 2012, 6, 162–169.
2. M. F. Xu, M. Zhang, M. Pastore, R. Z. Li, F. De Angelis and P. Wang, Chem. Sci., 2012, 3, 976–983.
3. A. K. Chandiran, M. Abdi-Jalebi, M. K. Nazeeruddin and M. Grätzel, ACS Nano, 2014, 8, 2261–2268.
4. Z. Ning, Y. Fu and H. Tian, Energy Environ. Sci., 2010, 3, 1170–1181.
5. H. N. Tsao, C. Yi, T. Moehl, J.-H. Yum, S. M. Zakeeruddin, M. K. Nazeeruddin and M. Grätzel, ChemSusChem, 2011, 4, 591–594.
6. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737–740.
7. L.-N. Yang, Z.-Z. Sun, S.-L. Chen and Z.-S. Li, ChemPhysChem, 2014, 15, 458–466.
8. S. Feng, Q.-S. Li, L.-N. Yang, Z.-Z. Sun, T. A. Niehaus and Z.-S. Li, J. Power Sources, 2015, 273, 282–289.
9. P.-P. Sun, Q.-S. Li, L.-N. Yang, Z.-Z. Sun and Z.-S. Li, Phys. Chem. Chem. Phys., 2014, 16, 21827–21837.
10. L.-N. Yang, H.-Y. Zhou, P.-P. Sun, S.-L. Chen and Z.-S. Li, ChemPhysChem, 2015, 16, 601–606.
11. Q. Chai, W. Li, Y. Wu, K. Pei, J. Liu, Z. Geng, H. Tian and W. Zhu, ACS Appl. Mater. Interfaces, 2014, 6, 14621–14630.
12. X. Xu, D. Huang, K. Cao, M. Wang, S. M. Zakeeruddin and M. Grätzel, Sci. Rep., 2013, 3, 1489.
13. J.-Y. Chang, S. C. Chang, S.-H. Tzing and C.-H. Li, ACS Appl. Mater. Interfaces, 2014, 6, 22272–22281.
14. J. Mao, J. Yang, J. Teuscher, T. Moehl, C. Yi, R. Humphry-Baker, P. Comte, C. Grätzel, J. Hua, S. M. Zakeeruddin, H. Tian and M. Grätzel, J. Phys. Chem. C, 2014, 118, 17090–17099.
15. A. Yella, H. W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin, E. W. Diau, C. Y. Yeh, S. M. Zakeeruddin and M. Grätzel, Science, 2011, 334, 629–634.
16. M. Xu, D. Zhou, N. Cai, J. Liu, R. Li and P. Wang, Energy Environ. Sci., 2011, 4, 4735.
17. M. Zhang, Y. Wang, M. Xu, W. Ma, R. Li and P. Wang, Energy Environ. Sci., 2013, 6, 2944–2949.
18. W. Zhu, Y. Wu, S. Wang, W. Li, X. Li, J. Chen, Z.-s. Wang and H. Tian, Adv. Funct. Mater., 2011, 21, 756–763.
19. S. Qu, C. Qin, A. Islam, Y. Wu, W. Zhu, J. Hua, H. Tian and L. Han, Chem. Commun., 2012, 48, 6972–6974.
20. J.-J. Kim, H. Choi, J.-W. Lee, M.-S. Kang, K. Song, S. O. Kang and J. Ko, J. Mater. Chem., 2008, 18, 5223–5229.
21. Q. Feng, X. Lu, G. Zhou and Z.-S. Wang, Phys. Chem. Chem. Phys., 2012, 14, 7993–7999.
22. Y. Wu, X. Zhang, W. Li, Z.-S. Wang, H. Tian and W. Zhu, Adv. Energy Mater., 2012, 2, 149–156.
23. S. Haid, M. Marszalek, A. Mishra, M. Wielopolski, J. Teuscher, J.-E. Moser, R. Humphry-Baker, S. M. Zakeeruddin, M. Grätzel and P. Bäuerle, Adv. Funct. Mater., 2012, 22, 1291–1302.
24. W. Li, Y. Wu, Q. Zhang, H. Tian and W. Zhu, ACS Appl. Mater. Interfaces, 2012, 4, 1822–1830.
25. M. Velusamy, K. R. Justin Thomas, J. T. Lin, Y.-C. Hsu and K.-C. Ho, Org. Lett., 2005, 7, 1899–1902.
26. Y. Cui, Y. Wu, X. Lu, X. Zhang, G. Zhou, F. B. Miapeh, W. Zhu and Z.-S. Wang, Chem. Mater., 2011, 23, 4394–4401.
27. Z.-M. Tang, T. Lei, K.-J. Jiang, Y.-L. Song and J. Pei, Chem.–Asian J., 2010, 5, 1911–1917.
28. J. Shi, Z. Chai, J. Su, J. Chen, R. Tang, K. Fan, L. Zhang, H. Han, J. Qin, T. Peng, Q. Li and Z. Li, Dyes Pigm., 2013, 98, 405–413.
29. K. Pei, Y. Wu, W. Wu, Q. Zhang, B. Chen, H. Tian and W. Zhu, Chem.–Eur. J., 2012, 18, 8190–8200.
30. D. W. Chang, H. J. Lee, J. H. Kim, S. Y. Park, S.-M. Park, L. Dai and J.-B. Baek, Org. Lett., 2011, 13, 3880–3883.
31. X. Lu, Q. Feng, T. Lan, G. Zhou and Z.-S. Wang, Chem. Mater., 2012, 24, 3179–3187.
32. K. Pei, Y. Wu, A. Islam, Q. Zhang, L. Han, H. Tian and W. Zhu, ACS Appl. Mater. Interfaces, 2013, 5, 4986–4995.
33. K. Pei, Y. Wu, A. Islam, S. Zhu, L. Han, Z. Geng and W. Zhu, J. Phys. Chem. C, 2014, 118, 16552–16561.
34. J. Zhang, L. Yang, M. Zhang and P. Wang, RSC Adv., 2013, 3, 6030.
35. A. Yella, R. Humphry-Baker, B. F. E. Curchod, N. Ashari Astani, J. Teuscher, L. E. Polander, S. Mathew, J.-E. Moser, I. Tavernelli, U. Rothlisberger, M. Grätzel, M. K. Nazeeruddin and J. Frey, Chem. Mater., 2013, 25, 2733–2739.
36. J. Yang, P. Ganesan, J. Teuscher, T. Moehl, Y. J. Kim, C. Yi, P. Comte, K. Pei, T. W. Holcombe, M. K. Nazeeruddin, J. Hua, S. M. Zakeeruddin, H. Tian and M. Grätzel, J. Am. Chem. Soc., 2014, 136, 5722–5730.
37. J. Zhang, H.-B. Li, Y. Geng, S.-Z. Wen, R.-L. Zhong, Y. Wu, Q. Fu and Z.-M. Su, Dyes Pigm., 2013, 99, 127–135.
38. A. K. Biswas, S. Barik, A. Sen, A. Das and B. Ganguly, J. Phys. Chem. C, 2014, 118, 20763–20771.
39. A. Dualeh, F. De Angelis, S. Fantacci, T. Moehl, C. Yi, F. Kessler, E. Baranoff, M. K. Nazeeruddin and M. Grätzel, J. Phys. Chem. C, 2011, 116, 1572–1578.
40. A. Baheti, K. R. Justin Thomas, C. Li, C. Lee and K. Ho, ACS Appl. Mater. Interfaces, 2015, 7, 2249–2262.
41. Q. Chai, W. Li, S. Zhu, Q. Zhang and W. Zhu, ACS Sustainable Chem. Eng., 2013, 2, 239–247.
42. S. M. Feldt, E. A. Gibson, E. Gabrielsson, L. Sun, G. Boschloo and A. Hagfeldt, J. Am. Chem. Soc., 2010, 132, 16714–16724.
43. L.-N. Yang, Z.-Z. Sun, S.-L. Chen and Z.-S. Li, Dyes Pigm., 2013, 99, 29–35.
44. D. Kuang, S. Uchida, R. Humphry-Baker, S. M. Zakeeruddin and M. Grätzel, Angew. Chem., Int. Ed., 2008, 47, 1923–1927.
45. J. H. Delcamp, A. Yella, T. W. Holcombe, M. K. Nazeeruddin and M. Grätzel, Angew. Chem., Int. Ed., 2013, 52, 376–380.
46. J. Luo, M. Xu, R. Li, K. W. Huang, C. Jiang, Q. Qi, W. Zeng, J. Zhang, C. Chi, P. Wang and J. Wu, J. Am. Chem. Soc., 2014, 136, 265–272.
47. M. J. Frisch. G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Revision D. 01, Gaussian, Inc., Wallingford CT, 2009.
48. A. D. Becke, J. Chem. Phys., 1993, 98, 1372–1377.
49. A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
50. C. T. Lee, W. T. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785–789.
51. H. Li, Y. Li and M. Chen, RSC Adv., 2014, 4, 57916–57922.
52. N. Nath Ghosh, A. Chakraborty, S. Pal, A. Pramanik and P. Sarkar, Phys. Chem. Chem. Phys., 2014, 16, 25280–25287.
53. K. Chaitanya, X.-H. Ju and B. M. Heron, RSC Adv., 2015, 5, 3978–3998.
54. W. Li, L. G. C. Rego, F.-Q. Bai, J. Wang, R. Jia, L.-M. Xie and H.-X. Zhang, J. Phys. Chem. Lett., 2014, 5, 3992–3999.
55. H.-Q. Xia, C.-P. Kong, J. Wang, F.-Q. Bai and H.-X. Zhang, RSC Adv., 2014, 4, 50338–50350.
56. H.-H. G. Tsai, C.-J. Tan and W.-H. Tseng, J. Phys. Chem. C, 2015, 119, 4431–4443.
57. N. Martsinovich and A. Troisi, Energy Environ. Sci., 2011, 4, 4473–4495.
58. T. Yanai, D. P. Tew and N. C. Handy, Chem. Phys. Lett., 2004, 393, 51–57.
59. S. I. Gorelsky, SWizard program, University of Ottawa, Ottawa, Canada, 2012, http://www.sg-chem.net/.
60. J. Liu, X. Yang and L. Sun, Chem. Commun., 2013, 49, 11785–11787.
61. V. Barone and M. Cossi, J. Phys. Chem. A, 1998, 102, 1995–2001.
62. R. Cammi, B. Mennucci and J. Tomasi, J. Phys. Chem. A, 1999, 103, 9100–9108.
63. J. Tomasi, B. Mennucci and R. Cammi, Chem. Rev., 2005, 105, 2999–3093.
64. W. Ma, Y. Jiao and S. Meng, J. Phys. Chem. C, 2014, 118, 16447–16457.
65. A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin, E. W.-G. Diau, C.-Y. Yeh, S. M. Zakeeruddin and M. Grätzel, Science, 2011, 334, 629–634.
66. H.-B. Li, J. Zhang, Y. Wu, J.-L. Jin, Y.-A. Duan, Z.-M. Su and Y. Geng, Dyes Pigm., 2014, 108, 106–114.
67. J. Zhang, H.-B. Li, J.-Z. Zhang, Y. Wu, Y. Geng, Q. Fu and Z.-M. Su, J. Mater. Chem. A, 2013, 1, 14000–14007.
68. J. Zhang, H.-B. Li, S.-L. Sun, Y. Geng, Y. Wu and Z.-M. Su, J. Mater. Chem., 2012, 22, 568–576.
69. T. Marinado, K. Nonomura, J. Nissfolk, M. K. Karlsson, D. P. Hagberg, L. Sun, S. Mori and A. Hagfeldt, Langmuir, 2009, 26, 2592–2598.
70. S. Rühle, M. Greenshtein, S. G. Chen, A. Merson, H. Pizem, C. S. Sukenik, D. Cahen and A. Zaban, J. Phys. Chem. B, 2005, 109, 18907–18913.
71. S. Ardo and G. J. Meyer, Chem. Soc. Rev., 2009, 38, 115–164.
72. G. M. Hasselman, D. F. Watson, J. R. Stromberg, D. F. Bocian, D. Holten, J. S. Lindsey and G. J. Meyer, J. Phys. Chem. B, 2006, 110, 25430–25440.
73. Y. Jiao, W. Ma and S. Meng, Chem. Phys. Lett., 2013, 586, 97–99.
74. J.-Z. Zhang, J. Zhang, H.-B. Li, Y. Wu, H.-L. Xu, M. Zhang, Y. Geng and Z.-M. Su, J. Power Sources, 2014, 267, 300–308.
75. T. Lu and F. W. Chen, Acta Chim. Sin., 2011, 69, 2393–2406.
76. T. Le Bahers, T. Pauporté, P. P. Lainé, F. Labat, C. Adamo and I. Ciofini, J. Phys. Chem. Lett., 2013, 4, 1044–1050.
77. B. Lim, G. Y. Margulis, J.-H. Yum, E. L. Unger, B. E. Hardin, M. Grätzel, M. D. McGehee and A. Sellinger, Org. Lett., 2013, 15, 784–787.
78. B. Peng, S. Yang, L. Li, F. Cheng and J. Chen, J. Chem. Phys., 2010, 132, 034305.
79. W.-L. Ding, D.-M. Wang, Z.-Y. Geng, X.-L. Zhao and W.-B. Xu, Dyes Pigm., 2013, 98, 125–135.
80. J. Preat, C. Michaux, D. Jacquemin and E. A. Perpete, J. Phys. Chem. C, 2009, 113, 16821–16833.
81. E. Ronca, M. Pastore, L. Belpassi, F. Tarantelli and F. De Angelis, Energy Environ. Sci., 2013, 6, 183–193.
82. L.-N. Yang, Z.-Z. Sun, Q.-S. Li, S.-L. Chen, Z.-S. Li and T. A. Niehaus, J. Power Sources, 2014, 268, 137–145.

---

Footnote

† Electronic supplementary information (ESI) available: Energy-level graphic of the donor moiety, optical property of dye-Ti(OH)₃·H₂O and dye-(TiO₂)₉ complexes, selected geometry parameters for the neutral and oxidized forms of dyes, cartesian coordinates of the optimized structures. See DOI: 10.1039/c5ra00587f

---