Theoretical investigation using DFT of quinoxaline derivatives for electronic and photovoltaic effects.

Photovoltaic properties of solar cells based on fifteen organic dyes have been studied in this work. B3LYP/6-311G (d,p) methods are realized to obtain geometries and optimize the electronic properties, optical and photovoltaic parameters for some quinoxaline derivatives. The results showed that time dependent DFT investigations using the CAM-B3LYP method with the polarized split-valence 6-311G (d,p) basis sets and the polarizable continuum model PCM model were sensibly able to predict the excitation energies, the spectroscopy of the compounds. HOMO and LUMO energy levels of these molecules can make a positive impact on the process of electron injection and dye regeneration. Gaps energy ΔE g , short-circuit current density J sc, light-harvesting efficiency LHE, injection driving force ΔGinject, total reorganization energy λtotal and open-circuit photovoltage V oc enable qualitative predictions about the reactivity of these dyes.

Introduction

The quinoxaline derivatives have also found applications in molecules [1, 2, 3, 4, 5, 6, 7, 8]. Newly, the research of synthesis, structural and energetic properties of these compounds were made by our group [9, 10, 11, 12, 13, 14]. In addition, the organic photovoltaic solar cells represent a significant potential for evolution in the study of the electricity production from low-cost compositions. Then, the past 2 years we have seen an important growth in the conversion efficiency of organic photovoltaic solar cell PVSC, pass a 1% return realized in recent years [15], at a 5% recorded in 2 years ago [16]. The long-term goal of this study is to minimize costs of PV compositions; we present here a few technical aspects. There are, to expand a technology founded on eco-friendly materials along nearly limitless availability. This goal becomes achievable day by day, in the evolution of high performance organic screens in the electronic industry [17]. We use the diode technology of organic light-emitting [18] to establish guidelines for organic photovoltaic solar cell OPVCs research. Advances in OPVCs need a distinct agreement of the particular physics of deformed organic semiconductors and devices [19]. We begin by the essential properties of organic semiconductors, as charge transport. We provide guidance for choosing efficient OPV materials and we verify a few castes of materials used in various coats of a PVCs. Then, we present an electrical representation of organic solar cell OSC. A criticism analysis of physical procedures allows us to size the maximum and minimum yields possible using several device structures.

Lately, polymer PVCs have received a lot of attention for their flexibility, ease of processing, weak weight and weak cost of production [20, 21, 22]. Their relatively large band gaps limited the Jsc, reducing the power conversion efficiencies (PCE). After improving efficiency, low band-gap conjugated polymers (CPs), were developed to better conform to solar spectra, and thus produce higher Jsc [23]. Significant advance was done in PVCs founded on bulk-heterojunction networks done of weak band gap CPs and the fullerenes at last, Nowadays the photoelectric conversion efficiency of OSC is much greater than 7% [24]. Generally, the PCE based on the Voc, the Jsc, and the fill factor (FF) of the systems. The Jsc is controlled by the hatching among the absorption of the CPs and the solar spectra [25]. The Voc is calculated by variance among the ELUMO of the fullerene derivatives, and the EHOMO of CPs [26]. So the novel improvement in PCE request development of new CPs with adequate energy levels and greater absorption along the solar spectra. Thus, the mobility of the heavy load carrier of polymer semiconductors should be considered [27].

On the other hand, with the experimental results obtained for organic compounds without metals, research is again restricted. Just some study groups have searched the photophysical and optoelectronic properties for dye sensitizers [28, 29], as well intra compound e−-dynamic process enter compound and TiO2 [30,31]. So, the investigation study of novel quinoxaline-based dyes was signalized in Figure 1. The central quinoxaline was paired by means of conjugation to an (benzene, chlorobenzene, methylbenzene …) and the group (C=O, –NO2, –C3H5, …).

Figure 1

Optimized geometries of all molecules.

The optical absorption and e−-structure properties of fifteen dye sensitizers Q i (i = 1–15) were investigated by using DFT. According to the results obtained, we analyzed the role of various e−-D groups in setting of structures, optical absorption and e−-structure properties have been analyzed. As well, we wanted to see the sensitizer D impacts on the Voc and the Jsc of the cell via discuss the main factors influencing Voc and Jsc along the purpose to discovery the potential sensitizers for use in dye-sensitized solar cell DSSC.

Theoretical methodology

Theoretical background

The design of current–voltage features of a cell below lighting and in darkness, allows evaluating more of its PV performances as well as its electrical behavior [32]. The maximum power Pmax issued by the PVC; FF is defined as:

The Incident Photon to Current Efficiency (IPCE) is determined by: [33];

Other:

The external PV yield η is described according to the illumination G and the surface S by:

The conversion yield is very important for cell productivity. This must be carefully estimated [34], and not be abashed with IPCE. The cell drives during the voltage surpass a threshold Vs. or Is is the saturation current in inverse polarization, an ideal cell can follow the thermionic injection model [35]:

For the Jsc in DSSC: : Electron injection efficiency, : Charge collection efficiency. For the even DSSC by just various molecules, so we can assume that: = Constant. Then, to make light on the link enter the Jsc and η, we calculated the LHE, and λtotal. For a high Jsc, the efficient sensitizers used in DSSC having had to a broad LHE, that determined by [36]:where f is the oscillator strength of the dye associate to the λmax. We noticed that the larger oscillator strength would have the higher light-harvesting efficiency. Thus, a large could also guarantee a high Jsc, who is linked to the ΔGinject and evaluated as [36]:where and : oxidation potential energy of the composed in the ground and excited state;

ECB: reduction potential of the conduction band of TiO2.

So, in this study, we use ECB = –4.0 eV for TiO2 [37], that is broadly used in a few works [38, 39, 40], and the can be evaluated [39, 40, 41]; and E00 is an electronic vertical transition energy associate to λmax. It has in general admitted that there are 2 diagrams to estimate the ΔGinject. Thereby, the e− injection with excited states of the compound to TiO2 (CB) is calculated by unrelax path. The small λtotal, that contains the hole and e− reorganization energy (λh and λe) could improve the Jsc. Thus, we calculated the (λh and λe) by [33, 42]:

: Cation or anion Energy obtained with the optimized neutral molecule structure,

: Cation or anion Energy obtained with the optimized cation or anion structure,

: Neutral molecule Energy calculated at the cationic or anionic state,

: Neutral molecule Energy at ground state.

Figure 2 present the equivalent circuit, who is an absolute current generator. Where IL is a current source of which intensity according to G, Rs and Rsh are the series and shunt resistances; RL is the charge resistance of the external circuit [35]. The Isc is the one who passes the cell at zero utilized voltage.

Figure 2

Idealized equivalent circuit of a real PVC under light.

The slope near zero polarization is a value of the Rsh, who matches to leaks and shorts in the diode [43]. If we assume such: (Rs = 0; Rsh = ∞) with (I = 0; IL = Isc), The Voc determined by:

A small Rsh will decrease Voc. Furthermore, within the weak illumination G the cell will not be liberated any voltage; Isc decreased by the Rs. The Voc can be calculated by the analytical equation [38]: Voc = ELUMO – ECB

Calculation methods

Every our investigations have been effected in the gas phase by the DFT optimization with the B3LYP hybrid functional [44, 45, 46] and 6-31G (d,p) Gaussian basis sets [47, 48]. The provide excited state energy and oscillator strength (ƒ) have been calculated using the TD-DFT/CAM-B3LYP methods in chloroform solvent. In this work, the integral equation formalism polarizable continuum model (IEF-PCM) [49, 50, 51, 52] has been selected in excitation energy; while to determinate the λtotal, we investigated at the B3LYP/6-311G (d,p) level the cationic and anionic states for each compound.

Results and discussion

Synthesized

We have synthesized these symmetric ligands in good yields, by varying the nature of ortho phenlene diamine substituents. It is carried out in ethanol by two steps: the first step aims to elaborate new tetraone compound. In the second step we condensate the precursor with one equivalent of the 1,2-o-phenylenediamine [53, 54, 55, 56, 57].

Geometrical structures

In this paper, the quinoxaline dyes [53, 54, 55, 56, 57, 58] chemical structures presented in Figure 1. In other papers, the DFT-investigated geometries have been a good according with the result observed in x-ray analysis [59, 60]. The calculated for every compounds indicate that they have a similar coplanar structure. This coplanar compound-conformation should ameliorate the e− transfer from the e−-D to the e−-A across the π-spacer unit for our molecules. The optimized critical bond lengths (d1, d2) are compared from Qi (i = 1–15) in ground and excited state (S0, S1). The corresponding structural parameters in S0 are similar to each other. Thus, that both of them, substituent introduction and conjugate chain elongating, have a minor impact on these structural parameters as though the compounds have like π-conjugated linker. That the bond lengths parameters d1 and d2 of all molecules is in the range of (1.326 ± 0.03 Å) in S0, and in S1 is in the range of (d (S0) ± 0.018 Å), who is likely due to the conjugation extension. Furthermore, the linkage among the e−-D and π-conjugated bridge is between 1.278–1.374 Å showing in particular more C=C character who favors ICT. Effectively, the π-conjugated group, is used as a deck from ICT of the e−-D to e−-A group. As we know, about to the S1 photoexcitation, the bond lengths for all compounds notably decreased in similitude along those in S0, in particular the linkage among the π-conjugated group and the A-half d2. From this, we conclude who is significant for the absorption spectrum, that linking of A-group and π-bridge (quinoxaline) is critical for the greatly enhanced ICT feature.

Intramolecular charge transfer

Using the frontier molecular orbital (FMO) we obtain the intramolecular charge transfer (ICT). Figure 3 present the e− locative distribution of LUMO and HOMO of every molecule. Usually, the parcels of the LUMO and HOMO proved the characteristics π-type MO typical. The HOMO exhibit an anti-bonding type enter two adjacent fragments and bonding type in every unit. The LUMO display the bonding type enter the two adjacent fragments, therefore the lowest lying singlet states are matching to electronic transition of π→π∗ character. The pattern of LUMOs and HOMOs are similar to each other (Figure 3). The e− distributions in HOMOs are essentially located in e−-D to π-conjugated spacing, whilst LUMOs are mainly localized on the conjugation spacing half and the e−-A fragments. Thus, the all molecules electronic transitions from HO to LU molecular orbital could lead to ICT from D-units to A/anchoring groups across a conjugated bridge, thus HOMO–LUMO passage maybe ranked as a ππ∗ ICT. The C=O/anchoring group in all molecules has an important contribution to LUMOs which give a solid electronic coupling with TiO2 area and thus enhance the e−-injection performance, and thereafter improve the Jsc.

Figure 3

Frontier orbital contour plots of all molecules.

Molecular orbitals

With a stronger e−-D group normally gives a high HOMO as compared to that with a weaker e−-D. Using that, we have calculated the e−-D impact on the electronic properties. Under HOMO, the results of these molecules are:

Q8 > Q12 > Q4 > Q2 = Q3 = Q6 = Q11 > Q5 > Q9 = Q13 > Q15 = Q1 > Q7 = Q14 > Q10. The Q8 marked by the strongest e−-D group (twofold chlorine) as it has the highest HOMO (-3.288 eV). Compounds Q7, Q14 and Q10 with investigating EHOMO respectively -5.363, -5.363 and -5.417 eV, have a low contribution in e−-D capacity because they restrain a group in the e−-donor moiety. The investigated LUMO level in Figure 4 from every sensitizers are comparatively uninfluenced by the modifications in composed geometry, due to the impaction of a few e−-A group (C=O) in these sensitizers, who is less affected by changing of the D-group. The ΔEg for Qi (i = 1–15) are listed in Figure 4, and the results of ΔEg is:

Figure 4

Schematic energy diagram of all molecules.

Q8 < Q4 < Q12 < Q5 < Q13 = Q15 < Q 1 = Q2 = Q3 = Q11 = Q14 < Q10 < Q7 = Q6. The ΔEg decrease, another photons at lengthier wavelength side so-called absorbed to excite the e− until unoccupied MO; that increases the Jsc and moreover enhances the conversion efficiency correspondent SC. The ranges of ΔEg are approximately 0.41–2.34 eV; and therefore, these molecules have the potential employing in the DSSC practice.

Optical properties

We are getting the optical properties of all sensitized compounds in chloroform solution [61, 62]. The chloroform has been used in UV-Visible absorption spectrum as solvent on quinoxaline based dyes [13, 31]. Since the function used is very economical and requires only one calculation, we took into account for the simulated spectra 20 excited states to obtain a better spectrum for the molecules studied here. The spectrum indicates a similar outline for all molecules; so a principal intense band at highest energies between 300 and 650 nm (Figure 5). The more intense participation to the principal band is an excitation between LUMO and HOMO in solvent as the 1st singlet excitation. Thus, the location (concerning to the gap HOMO-LUMO) and the largeness of 1st band in spectra are the two 1st parameters which can be concerning to compound efficiency, for absorption shift to reduce energies fosters the light harvesting process. The order of 1st vertical excitation energies (Eve) of our molecules is:

Figure 5

Absorption spectra of all dyes.

Q6 > Q7 > Q10 > Q14 > Q11 > Q3 > Q2 > Q1 > Q15 > Q13 > Q9 > Q5 > Q12 > Q4 > Q8. Therefore, when passing from Q6 to Q8, this shows that there is a bathochromic shift. For Q10, the absorption spectrum of Q7 and Q6 present less decreased oscillator strength with a slight blue shift, due probably to the heteroatom electronegativity in the e−-D groups. The absorption spectrum of Q8, Q4 and Q9 show the main peak at 524.91, 522.23 and 483.07 nm, respectively (Figure 5), which are well shifted to smaller wavelengths relative to those correspondent derivatives. Every data of absorption spectrum are in good, according to the results of band gap and energy debated previously.

The emission spectra in adiabatic measure have been used to study the photoluminescence (PL) characters of the molecules Qi (i = 1–15). We present in Table.1 the investigated fluorescence wavelengths with the strongest oscillator. We obtained the Stokes shift (SS) for all molecules, the emission spectrum arising from the S1 state is assigned to π∗→π transition and LUMO→HOMO orbital molecule character for all dyes. We establish that the investigated fluorescence emission is only the reverse process of lowest lying absorption. Furthermore, the order of red shift fluorescence observed of PL spectra is:

Table 1

Emission spectra results for all molecules.

Dye	EHOMO (eV)	ELUMO (eV)	ΔEg (eV)	E00 (eV)	Eve (eV)	λmax (nm)	f	SS (nm)	ET (eV)	μ (Debye)
Q1	-5.336	-3.240	2.095	2.26	2.70	515.54	1.755	89.3	-38452.23	6.4136
Q2	-5.308	-3.213	2.095	2.25	2.71	505.36	1.752	84.5	-32067.99	6.5906
Q3	-5.308	-3.213	2.095	2.26	2.71	503.17	1.716	80.4	-56960.52	3.3557
Q4	-3.594	-3.159	0.435	0.61	2.41	572.23	1.968	103.8	-32128.91	8.8931
Q5	-5.254	-3.213	2.040	2.10	2.48	553.95	1.960	101.7	-37600.21	2.9054
Q6	-5.308	-2.968	2.340	2.22	2.88	473.92	1.541	57.8	-25949.01	7.8137
Q7	-5.363	-3.023	2.340	2.07	2.85	479.96	1.587	59.7	-30202.29	5.3867
Q8	-3.288	-2.880	0.408	0.33	2.40	575.81	1.994	104.5	-57021.77	5.6837
Q9	-5.281	-3.213	2.068	2.01	2.57	533.07	1.955	99.2	-33132.02	7.0973
Q10	-5.417	-3.268	2.149	2.38	2.84	480.73	1.601	63.1	-25943.16	5.9401
Q11	-5.308	-3.213	2.095	2.29	2.72	500.15	1.709	70.3	-34196.06	7.2559
Q12	-3.485	-2.887	0.598	0.44	2.44	563.40	1.964	95.9	-26007.53	6.6097
Q13	-5.281	-3.213	2.068	2.08	2.61	531.12	1.931	97.5	-35260.09	7.7362
Q14	-5.363	-3.268	2.095	2.37	2.72	495.14	1.703	57.6	-27007.25	6.4986
Q15	-5.336	-3.268	2.068	2.20	2.66	529.05	1.901	96.8	-31475.38	2.2708

Q6 < Q7 < Q10 < Q14 < Q11 < Q3 < Q2 < Q1 < Q15 < Q13 < Q9 < Q5 < Q12 < Q4 < Q8; thus the good according with the obtained data of absorption when passing from Q6 to Q8. Moreover, the SS of our molecules has established to be in the range 57.8 and 104.5 nm. The Q8 emitted at higher wavelengths (575.81 nm) with strongest intensity (f = 1.994), and larger SS (104.5 nm). We conclude that Q8 with two fold-chlorine e−-D group will be the best candidate in the DSSC.

We presented the dipole moment (μ) vectors in the space of our fifteen quinoxaline molecules (Figure 6). The dipole moment for Q2 (6.590 D) is decreased in (Q1, Q7, Q10) and increased in (Q6, Q9, Q13) with a weak orientation, and even effect has been noted in the compound Q4, Q5, Q12 and Q14 with an orientation important. Furthermore, the μ variations in these ten cases is greatly smaller than that establish in Q3 and Q8 where the existence of electronegative twofold chlorine in the opposite side along respect to carbonyl, not only reduces or increased dramatically the μ, but also modifies the vector orientation along large angle (180° in Q11) (Figure 6). The indication providing by NPA charges as good as this change with account to dipole moment are in according with recognized low eligibility impact of chlorine [46, 50, 63], as likened to its inductive impact.

Figure 6

Dipole moments μ (Debye) in a three-dimensional representation, for quinoxalin-2(1H)-one derivatives in Figure 1.

Photovoltaic properties

Oxidation potentials energies computed Edye could be optimized as negative EHOMO [64]. is estimated, of all molecules is increasing as following:

Q8 < Q4 < Q14 < Q11 < Q10 < Q12 = Q3 < Q2 < Q1 < Q6 < Q15 < Q5 < Q13 < Q9 < Q7. Thus, the most practical oxidizing species is Q8 by cons Q7 is the worst. All ΔGinject obtained is negative for every dyes, so the e− injection from the composed to TiO2 is impulsive.

In Table.2 and Figure 6(a) as seen, the estimated ΔGinject are decreased as following:

Table 2

Calculated electrochemical parameters for all molecules.

Dye	Edye (eV)	E^dye∗(eV)	ΔGinject (eV)	LHE	λh (eV)	λe (eV)	λtotal (eV)	Voc (eV)
Q1	5.33	3.07	-0.93	0.9824	0.282	0.321	0.603	0.760
Q2	5.30	3.05	-0.95	0.9822	0.273	0.317	0.590	0.787
Q3	5.30	3.04	-0.96	0.9807	0.280	0.332	0.612	0.787
Q4	3.59	2.98	-1.02	0.9892	0.221	0.280	0.501	0.841
Q5	5.25	3.15	-0.85	0.9890	0.240	0.302	0.542	0.787
Q6	5.30	3.08	-0.92	0.9712	0.263	0.321	0.584	1.032
Q7	5.36	3.29	-0.71	0.9741	0.280	0.352	0.632	0.977
Q8	3.28	2.95	-1.05	0.9898	0.213	0.281	0.494	1.120
Q9	5.28	3.27	-0.73	0.9889	0.270	0.300	0.570	0.787
Q10	5.41	3.03	-0.97	0.9749	0.273	0.331	0.604	0.732
Q11	5.30	3.01	-0.99	0.9804	0.285	0.333	0.618	0.787
Q12	3.48	3.04	-0.96	0.9891	0.241	0.272	0.513	1.113
Q13	5.28	3.20	-0.8	0.9882	0.241	0.323	0.564	0.787
Q14	5.36	2.99	-1.01	0.9801	0.293	0.332	0.625	0.732
Q15	5.33	3.13	-0.87	0.9874	0.231	0.313	0.544	0.732

Q8 > Q4 > Q14 > Q11 > Q10 > Q12 = Q3 > Q2 > Q1 > Q6 > Q15 > Q5 > Q13 > Q9 > Q7. This presents that Q8 has the greatest ΔGinject while Q7 has the littlest value. The result according to LHE of the molecules, must be the highest possible to extend the photocurrent reply. The LHE for every compounds are in close range 0.9712–0.9898, but growth lightly with growthing the conjugation length Figure 7(b); thus which all the dye sensitizers give comparable photocurrent.

Figure 7

Jsc along of calculated sensitizers: (a) the LHE, (b) the ΔGinject, (c) the λtotal and (d) the Voc.

In addition to the free energy reaction, the λtotal could equally allocate the kinetics of e− injection. Thus, the investigated λtotal is equally significant to analyze the relation among the e− structure and the critical parameters influenced Jsc. Thus, in Table.2 and Figure 7(c) the investigated λtotal of all molecules are increased as follows:

Q8 < Q4 < Q12 < Q5 < Q15 < Q13 < Q9 < Q6 < Q2 < Q1 < Q10 < Q3 < Q11 < Q14 < Q7. It shows that composed Q8 possesses the smallest λtotal, while composed Q7 has the largest. Consequently, composed Q8 presents a favorable Jsc because of the relative similar LHE, larger ΔGinject and smaller λtotal. Therefore, ΔGinject and λtotal are more important to govern the Jsc mostly.

We know that in addition the Jsc the overall power conversion efficiency η also could be influenced by the Voc. Thus, between 2 compounds of similar conformations, the e−-injection is more effective for this compound with the higher excited state linked to the semiconductor conduction band edge, that is to say higher Voc. It was found that Voc of all compounds is in the range 0.732–1.120 eV and as following by:

Q8 > Q12 > Q6 > Q7 > Q2 > Q9 > Q4 > Q5 = Q3 = Q2 = Q11 = Q13 > Q1 > Q14 > Q15. It shows that Q8 and Q12 have the higher Voc than other compounds, while Q14 and Q15 have the smallest. Consequently, we conclude that the high (LHE, -ΔGinject and Voc) and as well small λtotal can give an efficiency important. Therefore, the effectiveness of DSSC sensitized by the compound Q8 might be preferable to the other molecules, due to its favorable effectiveness of these factors presented on our calculated results.

Conclusions

The results are determined by the DFT investigations using the CAM-B3LYP method.

The calculated absorption maximums are in the range 473–576 nm;

The obtained band gap (ΔEg) of the studied compounds was in the range 0.408–2.340 eV;

The obtained values of Voc of the studied compounds range from 0.732 to 1.120 eV;

Since Q8 has the largest value of (LHE, -ΔGinject, Voc) and has the smallest value of (λtotal), the compound Q8 was found to be the best photo-sensitizer for use in dye-sensitized solar cell DSSC, in comparison with other dyes, as the investigation results exhibit its good oxidation potential energy and e− injection force which are above the ECB of TiO2 and in reduction potential energy of electrolyte.

All the studied compounds can be employed as organic solar cells, because these results are sufficient for an efficient electron injection. To finish, the process of investigated calculations can be used to predict the opto-electronic properties on the other molecules, and more to designing novel materials for organic solar cell.

Declarations

Author contribution statement

A. El Assyry: Conceived and designed the experiments; Performed the experiments; Wrote the paper.

M. Lamsayah: Conceived and designed the experiments; Performed the experiments.

I. Warad & F. Bentiss: Conceived and designed the experiments; Contributed reagents, materials, analysis tools or data.

R. Touzani: Conceived and designed the experiments; Analyzed and interpreted the data.

A. Zarrouk: Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Competing interest statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.