Relationship between photo-physical and electrochemical properties of D-π-A compounds regarding solar cell applications. 1. Substituent type effect in photovoltaic performance

(1)

ORIGINAL PAPER

Relationship between photo-physical and electrochemical properties of D- π -A compounds regarding solar cell applications. 1. Substituent type effect in photovoltaic performance

E. Ortega¹&Jean Christian Bernède²&A. M. R. Ramírez³&G. Louarn⁴&F. R. Díaz¹&L. Cattin⁴&Maria Angelica del Valle¹

Received: 19 October 2018 / Accepted: 5 February 2019

#Springer-Verlag GmbH Germany, part of Springer Nature 2019

Abstract

Studying the electrochemical characteristics is an important step for determining interactions between molecules and the chemical environment. Moreover, the electrochemical evaluation of dyes is highly needed to establish the behavior of electro-active chemical species inside dye-sensitized solar cells (DSSCs). Four compounds, M8-1, M8-2, M8-O1, and M8-O2 (with a common organic structure (E)-2-cyano-3-(5-((E)-2-(9,9-diethyl-7-(phenylamino)-9H-fluoren-2-yl)vinyl)thiophen-2-yl)acrylic acid), are studied in two solvents, tetrahydrofuran (THF) and dimethylsulfoxide (DMSO). Among the studied compounds, M8-1 has highlighted characteristics compared with the others: its ground and excited states oxidation potential are the highest (1.14 and−1.22 V, respectively). Also, it shows the lowest energy gap between the excited state oxidation potential and the TiO2

conduction band. Relating to the substituent effect, the shorter the length, the higher the energetic difference in the electronic transition (M8-1 and 2). Comparing characteristics through quantum chemistry, the values obtained in DMSO are the most predictable. The injection energies signal that M8-1 is the best injector. The performances in solar cells are measured in three TiO2

materials: Degussa (D-TiO2), active opaque (A-TiO2), and transparent (T-TiO2). The IPCE results show the A > T > D average tendency, and the family of substituted alkyl has higher values than the alcoxyl one. Furthermore, in the first family the methyl substituent has a higher value than the ethyl one. M8-1 has the highest IPCE value, on average. In terms of efficiency, the alkyl substituted family again has higher values than the alcoxyl family. On average, the methyl substituent has a higher value than the ethyl one in both families. M8-1 has the highest efficiency value.

Keywords Dyes . Ground and excited states . Electronic injection energy . Substituent effect . Solar cell performance

Introduction

Solar energy is one of the most studied energetic resources owing to its qualities of producing clean and renewable energy [1]. Among the different kinds of cells, dye-sensitized solar cells (DSSCs) are shown as a promising alternative compared with common photovoltaic devices based on silicon [2].

Grätzel’s work team, who are pioneers in these types of devices, have created compounds and methodologies that allow positioning these cells as an interesting alternative [3].

According to their studies, the compounds named cis-dithio- cyanato-bis-(4,4′-dicarboxy-2,2′-bipyridine) ruthenium(II) and trithiocyanato-4,4′,4″-tricarboxy-2,2′:6,2-terpyridine ruthenium(II), commonly known as N3 and black dye, respectively, are in the group of compounds that paved the way for a powerful energy alternative [4]. Nowadays, many different kinds of dyes have been synthesized, and the free metal dyes are the most highlighted because they have characteristics Electronic supplementary materialThe online version of this article

(https://doi.org/10.1007/s00894-019-3955-1) contains supplementary material, which is available to authorized users.

* Jean Christian Bernède

[email protected]

* Maria Angelica del Valle [email protected]

1 Facultad de Química, Pontificia Universidad Católica de Chile, Av.

V. Mackenna 4860-Macul, BP 7820436 Santiago, Chile

2 MOLTECH-Anjou, CNRS, UMR 6200, Université de Nantes, 2 rue de la Houssinière, BP 92208, F-44000 Nantes, France

3 Facultad de Estudios Interdisciplinarios, Laboratorio de

Electroquímica, Universidad Mayor, Núcleo Química y Bioquímica, Av. Alemania 0281, 4801043 Temuco, Chile

4 Institut des Matériaux Jean Rouxel (IMN), CNRS, UMR 6502, 2 rue de la Houssinière, BP 32229, 44322 Nantes cedex 3, France

(2)

comparable to the above mentioned organometallic dyes [5–9]. Electrochemistry allows determining fundamental properties as well as estimating the possible uses of each of these compounds in solar cells devices [10–14]. In this work, the electrochemical, theoretical, and photovoltaic characterization of four dyes (Fig.1, Table1) we previously reported on [15] were realized. The photovoltaic characterization was made in three different TiO2 materials: the Degussa (D- TiO2), active opaque (A-TiO2), and transparent (T-TiO2).

Furthermore, the above-mentioned characteristics are compared with previous photo-physical data.

Experimental

All studied compounds were synthesized and characterized according to a previously reported work (Table1and Fig.1) [15], and some photo-physical data were extracted from it to compare behaviors. The electrochemical measurements were carried out in a CH Instruments potentiostat, controlled by the CH750D software. Previously dried tetrahydrofuran (THF) and dimethylsulfoxide (DMSO) solvents were used. The dye concentration used was 1 mmol·L⁻¹. Tetrabutylammonium hexafluorophosphate (n-Bu4NPF6), with a concentration of 10 mmol·L⁻¹, was used as well. The voltammetry patterns were taken using a scan rate of 100 mV·s⁻¹. The theoretical geometries were calculated with the Gaussian 09 program, using the DFT theory level, the B3LYP functional, and the 6-31G* basis set, in both studied solvents. The TiO2thick- nesses in the anodes are measured using a DEKTAK 8 BRUKER profilometer [15].

The Degussa (D-TiO2) anodes are made using a commercial TiO2powder as raw material (P25, Degussa AG, Germany, a mixture of ca. 30% rutile and 70% anatase). The paste was made as follows: TiO2(12 g) was mixed with water (4 mL) containing acetylacetone (0.4 mL) to prevent reaggregation of the particles. After the dispersion preparation, it was diluted by the slow addition of water (16 mL) under continued grinding.

Finally, a detergent (0.2 mL Triton X-100, Aldrich) was added to facilitate the spreading of the colloid on the substrate [16].

The active opaque (A-TiO2) and transparent (T-TiO2) anodes are made using commercial 18NR-AO and 18NR-T TiO2

pastes (Greatcell Solar company). The FTO glass (fluorine doped tin oxide, 15 glass plates 2.2, 15Ω/sq., Greatcell Solar company) was covered by tape, making 0.25 cm²area squares.

Then the paste was applied to the free edges of the conducting glass and distributed with a glass rod sliding over the tape- covered edges. After air drying, the electrode was fired for 30 min at 450-550 °C at air conditions. Now, with the anode prepared, the standard dyes were loaded on by soaking the electrode overnight in a 3·10⁻⁴mol·L⁻¹solutions of the ruthenium complex in dry ethanol and the organic dyes in THF [16].

The cathodes were prepared using tape-covered glasses. Then PT1 platinum paste was applied. After air drying, the electrode was fired increasing 50 °C every 10 min, until 250 °C, to ensure a regular surface. Afterwards, it wass fired at 400 °C for 20 min more. Once the electrodes were prepared, a Low Temperature Thermoplastic Sealant (Greatcell Solar Co.) was used to seal the cell. Finally, the solar cells were filled through a hole in the anode with a redox electrolyte (0.50 mol·L⁻¹LiI, 0.050 mol·L⁻¹ I2and 0.50 mmol·L⁻¹tert-butylpyridine in dry acetonitrile) and sealed again, following the procedure above [17] (Scheme1S).

Fig. 1 Molecular structures of the studied dyes.aAlkyl-substituted series, M8-n family. R = Me, Et andbAlcoxyl-substituted series, M8-On

family. R = Me, Et. On red and black, the non-variant and variant structures, respectively

Table 1 Nomenclature of the synthesized dyes

Dye IUPAC name Abbreviation

(E)-2-cyano-3-(5-((E)-2-(9,9-diethyl-7-(methyl(phenyl)amino)-9H-fluoren-2-yl)vinyl)thiophen-2-yl)acrylic acid M8-1 (E)-2-cyano-3-(5-((E)-2-(9,9-diethyl-7-(ethyl(phenyl)amino)-9H-fluoren-2-yl)vinyl)thiophen-2-yl)acrylic acid M8-2 (E)-2-cyano-3-(5-((E)-2-(9,9-diethyl-7-((4-methoxyphenyl)(methyl)amino)-9H-fluoren-2-yl)vinyl)thiophen-2-yl)acrylic acid M8-O1 (E)-2-cyano-3-(5-((E)-2-(7-((4-ethoxyphenyl)(ethyl)amino)-9,9-diethyl-9H-fluoren-2-yl)vinyl)thiophen-2-yl)acrylic acid M8-O2

(3)

The IPCE profiles were carried out on an Agilent spectro- photometer HORIBA Scientific Fluoromax-4spectrofluorome- ter, and the currents were measured with the CH Instruments potentiostat. We used a home-made system that does not allow the electrodes to touch, letting the electrolyte be the only contact between them. The photovoltaic characterization in DSSCs is made with a global solar simulatorAM1.5(Oriel 300 W), in darkness and light, using an intensity of 100 mW·cm⁻². The system is calibrated with a standard solar cell (NREL, EE.UU), with an area of 0.5 cm². All the measurements were carried out at ambient conditions. The results are compared with N3 and N719, which are two ruthenium complexes often used as stan- dards for these types of measurements.

Results and discussion

The electrochemical analysis compares behaviors of the studied dyes in two solvents, tetrahydrofuran (THF) and dimethylsulfoxide (DMSO), to get a general overview of the

characteristics in solution, determining the effects on these compounds (Table2).

The solvents correspond to the more and less polar used in a previous study, where these dyes are soluble [15]. The red- ox potentials (E1/2 or EOX) are lower in DMSO than THF;

hence, the reactions in DMSO need less energy to occur, which means that the solvent favors the reaction, owing to solvation (Table3). Additionally, the ionic product is more stable in a polar solvent like DMSO [18,19].

To calculate the diffusion coefficients, the following reaction was considered:

S→S^þþe ð1Þ

whereBS^is the neutral dye,BS⁺^the oxidized one andBe^the electron involved. SmallerΔE values favor reversible reactions, as happen in DMSO.

Voltammetric behavior in THF and DMSO

For all compounds, in both solvents THF (Suppl. material Figs.1S to4S, [a, b, c]) and DMSO (Fig.2and Suppl. material Table 2 Frontier orbital estimation (in eV), from electrochemical and theoretical measurements

Orbital HOMO LUMO HOMO LUMO HOMO LUMO HOMO LUMO

Solvent THF DMSO THF DMSO

Dye /measurement type Electrochemistry Theoretical

M8-1 −5.64 −3.90 −5.22 −2.76 −5.03 −2.85 −5.04 −2.87

M8-2 −5.52 −4.05 −5.44 −2.85 −4.90 −2.79 −4.91 −2.83

Family average −5.58 −3.98 −5.33 −2.81 −4.97 −2.82 −4.98 −2.85

M8O1 −5.40 −3.97 −5.40 −2.86 −4,83 −2.83 −4.85 −2.86

M8-O2 −5.48 −4.48 −5.40 −2.90 −4,99 −2.80 −5.02 −2.88

Family average −5.44 −4.23 −5.40 −2.88 −4,91 −2.82 −4.94 −2.87

Table 3 Electrochemical characteristics of the studied dyes

Solvent THF DMSO

Dye/ characteristic Peak EPa(V) EPc(V) ΔE (V) E1/2(V) EPa(V) EPc(V) ΔE (V) E1/2(V)

M8-1 1 1.32 0.92 0.40 1.12 1.00 0.79 0.21 0.895

2 −1.55 −1.30 0.18 −1.46

M8-2 1 1.35 0.71 0.64 1.03 0.97 0.79 0.18 0.88

2 −1.49 −1.28 0.21 −1.385

Family average 1 1.34 0.82 0.52 1.08 0.99 0.79 0.20 0.89

2 −1.52 −1.29 0.20 −1.42

M8-O1 1 1.38 0.88 0.50 1.13 0.90 0.72 0.18 0.81

2 −1.50 −1.29 0.21 −1.395

M8-O2 1 1.33 0.88 0.45 1.12 0.87 0.87 0.15 0.795

2 −1.51 −1.33 0.18 −1.42

Family average 1 1.36 0.88 0.48 1.13 0.89 0.80 0.17 0.80

2 −1.51 −1.31 0.18 −1.41

(4)

Figs.5S to 7S, [a, b, c]), a single peak appears at positive potentials. Nevertheless, only DMSO also shows a peak at negative potentials. Analyzing the peak kinetically in THF (Suppl. material Figs.8S to11S, [a]) and DMSO (Suppl.

materials Figs.12S to19S, [a]), a progressive rise with the

scan rate appears, which agrees with quasi-reversible and ir- reversible reactions.

Moreover, the peak maxima displacement in DMSO is less than that observed in THF. If theΔE values are compared in THF (Suppl. materials Table1S), M8-1 is smaller than M8-2 and M8-O1 is bigger than M8-O2, showing a different substituent effect on these two families. Also, the M8-n family tends to raise theΔE value in relation to M8-On. For E1/2, the methyl substituent is always higher. Comparing the families, M8-n values are smaller than M8-On. Now, if theΔE values are compared in DMSO (Table1S), the methyl substituent in both families (M8-1 and M8-O1) is always higher. Also, these values are smaller than the values determined in THF. For E1/2, the same effect is beholden, and in both casesΔE and E1/2of the M8-n family are higher. On the contrary, the peak obtained by scanning in the negative direction showsΔE and

-2 -1 0 1 2

-160 0 160 320

/JmcA-2

E / V vs SCE

a)

0.0 0.4 0.8 1.2 1.6

0 170 340

J mcA-2

E / V vs SCE

b)

-2.0 -1.5 -1.0 -0.5 0.0 -166

-83 0

J / Acm-2

E (V) vs SCE

c )

Fig. 2 Voltammetric profiles of M8-1 in DMSO:aFull scan,b Scan to positive potential, andc Scan to negative potential

Table 4 Dyes band gap (E_g) (eV) comparison in the used solvents

Solvent THF DMSO THF DMSO THF DMSO

Dye/measurement type

Electrochemistry UV-vis Theoretical

M8-1 1.74 2.46 2.26 2.47 2.18 2.17

M8-2 1.47 2.59 2.33 2.32 2.11 2.08

M8-O1 1.43 2.54 2.21 2.31 2.00 1.99

M8-O2 1.00 2.50 2.16 2.33 2.19 2.14

(5)

E1/2values with different tendencies. In the M8-n family,ΔE is smaller for M8-1 and higher for M8-O1, in the case of M8- On. The E1/2values show the same tendency. Relating to the families, theΔE and E1/2values follow the same trend seen in the peak registered by scanning in the positive direction; however, for E1/2, M8-On is slightly higher. In both solvents, THF (Suppl. material Figs.8S to11S, [b]) and DMSO (Suppl.

material Figs.12S to19S, [b]), the oxidation peak for all of these compounds is controlled by diffusion. To determine the substituent effect in solution, the diffusion coefficients were calculated through the Randles-Sevcik plot for THF (Suppl.

material Figs.8S to 11S, [c]) and DMSO (Suppl. material Figs.12S to 19S, [c]). The same tendency appears in THF as that obtained forΔE. Additionally, the M8-n family average value is the lowest (Table1S). In the second solvent, the tendency is also the same as that obtained forΔE. The M8-n family has bigger values than M8-On (Suppl. material Table1S).

Additional effects

The alcoxyl group effect on the redox potential is according to the reported behaviors for other molecules [20–22]. The diffusion coefficients depend on solvent viscosity, solute size, and temperature [23–25]. In this measurement, the only fixed parameter is the temperature (20 °C). According to the size, the compounds should be organized as M8-O2 > M8-O1 >

M8-2 > M8-1. Nevertheless, values signal M8-1 > M8-2 >

M8-O2 > M8-O1 in THF and for DMSO, M8-O1 > M8-1 >

M8-O2 > M8-2 (scanning in the positive direction) and M8-

O2 > M8-O1 > M8-2 > M8-1 (scanning in the negative direction).

The above results show that the molecules might adopt different geometries during the red-ox processes. In addition, an inversely proportional relation between the coefficients and the viscosity has been found (Suppl. material Table1S), showing a dense solvent as an obstacle to the molecular transit [26].

Frontier orbital

Obtaining the frontier orbitals, highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO), from the voltammetric profiles, requires an estimation through on-set potentials [27–31]. There is a clear difference between THF and DMSO orbital values (Table2). The LUMO orbital values obtained in both solvents (experimental) show a great difference near 1.0 eV. This difference is not so big for HOMO orbital values and might be an effect from the negative peak disappearance in THF. This leads to a decrease in the reduction on-set potential in the mentioned solvent.

Comparison shows that the HOMO values are more negative in THF than DMSO. The same effect is seen for LUMO values. Those effects are the same in both families. It seems that the high polarity of DMSO destabilizes HOMO and LUMO orbitals, in relation with THF whose values are more negative.

For theoretical values, this difference does not exist.

Comparing theoretical and experimental values, the data in DMSO are nearer and a correction exists between the theoretical and experimental HOMO and LUMO orbitals in DMSO.

Band gap

Contrasting several data from various measurements is an important step for determining correlations. Table4 shows the band gaps measured by different methods. Through this, it is easier to establish trends than to just analyze the HOMO and LUMO values. However, for the substituent length, none of the measurements are according to the other two. For ex- ample, the methyl group in THF (M8-1 and M8-O1) tends to increase the gap in comparison to the ethyl group (M8-2 and M8-O2). Analyzing the substituent type, all measurements signal that the M8-n family increases the gap, when it is Fig. 3 Different molecular

moieties used in the charge differences determination:aM8- 1, R1 = Me, R2 = H;bM8-2, R1 = Et, R2 = H;cM8-O1, R1 = Me, R2 = Ome;dM8-O2, R1 = Et, R2 = OEt

Table 5 Injection energies of each compound

Dye Mülliken charge

gap nitrogen

THF DMSO

M8-1 −0.02 −0.48 −0.72

M8-2 0.05 −0.60 −1.00

Average 0.015 −0.54 −0.86

M8-O1 −0.18 −0.46 −0.94

M8-O2 −0.03 −0.44 −1.07

Average −0.11 −0.45 −1.01

(6)

compared with the M8-On family, the M8-n values are always the highest in both solvents.

Energy transfer studies

Five different moieties (Fig.3), were labeled to behold the effect of the electronic transition and where the charge is contained [15]. Determining patterns requires calculating the difference in the Mülliken charges [32].

The obtained patterns in THF (Suppl. material Figs.20S [a to d]) and DMSO (Suppl. material Figs.21S [a to d]) showed several behaviors. Almost every graphic shows the charge loss over the acceptor. However, only M8-2 (in THF and DMSO) and M8-O2 (in THF) show the expected result for both moieties. Also, the transition does not only depend on those moieties, the fluorenyl and thiophenyl fractions are likely involved. Furthermore, this transition might represent the electrochemical process occurred in the positive voltammetric peak (A description of the equation used is in theSuppl. material, Eq. 2S).

Evaluation of these dyes as photo-sensitizers

Using the oxidation potential values of ground state and excited state [4,32–35] (Table2S) shows that (comparing the substituent effect) longer chains decrease the ground potential (peak 2 for the M8-n family in DMSO is an exception), and the same tendency appears for the excited potential (not for M8-On family in THF). The substituent type effect is not clear.

Of the possible chemical structures to be oxidized, it is well-known the amine moiety is the main electro-active re- gion. The above-mentioned moiety has oxidation potentials near 1.0 V (vs. NHE); because of this the positive peak is selected [19,36–38] (A description of the equation used is in theSuppl. material).

Injection energies

To complete the analysis, an efficient comparison with the support material (TiO2) requires establishing the injection energies (ΔGinj) [35]. Table5verifies that in DMSO it is only possible to make a relationship with the substituent length: a short group decreases the injection energy.

In THF the M8-On family has the best value (near 0 V). In contrast, in DMSO, the M8-n family possesses the best value.

Also, in THF the obtained values are near zero. Finally, comparing the injection energies and the Mülliken charges, a proportional relationship is shown for M8-1 to M8-2 in both solvents. Nevertheless, from M8-O1 to M8-O2, in THF there is a proportional tendency and in DMSO there is an inversely

proportional relationship. (A description of the equation used is in theSuppl. material).

Photovoltaics in three TiO2

Three TiO2 were studied, the Degussa (D-TiO2), active opaque (A-TiO2), and transparent (T-TiO2) TiO2. The D- TiO2was prepared according to a method described before [16]. A and T TiO2were bought from the Greatcell Solar company. When trying to make correlations of these three semiconductors, the first step is to determine their morphology by scattering electron microscopy (SEM) (Fig. 4a, b, c).

Fig. 4 SEM for the different TiO2 layers.aD-TiO2,bA-TiO2, andcT- TiO2

(7)

The material D-TiO2shows nano-spheres and an irregular surface as does the A material; however, A has some big formations on and over the structure, whereas the material T is more regular. Although, the surfaces show some differences, the compositions are exactly the same (Suppl. material Fig.22S [a, b]).

Incident photon-to-light efficiency (IPCE)

After the cell assembly (Scheme S1), the photovoltaic characteristics are measured. The results are compared with the N3 and N719 dyes, as the common photovoltaic studies show. The IPCE also allows determining the maximum wavelength of conversion [39–44]. It is seen that

cells with the aliphatic modified compounds (M8-1 and M8-2) have higher IPCE, which means the higher conversion from light to electricity in all materials, in other words higher currents (Jsc) (Fig. 5a, b, c).

Comparing the maximum IPCE values (Suppl. material Table3S) shows that the A-TiO2allows the highest IPCE values on average, showing a A-TiO2> T-TiO2> D-TiO2val- ue tendency. Furthermore, M8-1 and M8-2 exhibit that the methyl substituent has the highest IPCE, whereas M8-O1 and M8-O2 exhibit a contrary effect on average, with the ethoxyl substituent the highest. Specifically, in materials A and T the ethyl substituent shows a higher value than methyl.

In material A, methyl is always the highest. All of these values are less than the values shown by the standard compounds (IPCE between 70 up to 80% for both N3 and N719 dyes

400 500 600 700 800

0 10 20 30 40 50 60

N3 N719 M8-1 M8-2 M8-O1 M8-O2

)%(ECPI

Wavelength (nm)

a)

400 500 600 700 800

0 10 20 30 40 50 60 70

N3 N719 M8-1 M8-2 M8-O1 M8-O2

)%(ECPI

Wavelenght (nm)

c)

400 500 600 700 800

0 20 40 60 80 100

N3 N719 M8-1 M8-2 M8-O1 M8-O2

)%(ECPI

Wavelength (nm)

b)

Fig. 5 IPCE of the DSSCs with the different compounds.aD- TiO2,bA-TiO2, andcT-TiO2

(8)

[45–47]). Regarding the differences among the TiO2ma- terials, it is possible to make a correlation between the TiO2 morphologies and the values obtained (Figs. 4 and

5). On the one hand, and despite that the compositions are the same, materials A and T have a higher anatase/rutile rate than material D. The above seems to be one of the main influential factors [16]. On the other hand, the big formations appearing in A-TiO2 (Fig. 4b) are a positive aspect regarding the IPCE.

Photovoltaic characterization

The photovoltaic profiles were taken using a light simulator (Fig.6a, b, c).

The current is higher for the standard compounds (N3 and N719) than the organic tested ones. Moreover, the materials A and T show an increase of these values in comparison to D.

The IPCE profiles show the same effect. M8-1 has the highest current among the other dyes. M8-2 should have values sim- ilar to M8-1; however, it does not. This could be a

0 100 200 300 400 500 600 700 0

5 10 15

J (mcAm-2 )

E(mV)

M8-1 M8-2 M8-O1 M8-O2 N3 N719

a)

0 100 200 300 400 500 600 700

0 5 10 15 20

J (mcAm-2 )

E(mV)

M8-1 M8-2 M8-O1 M8-O2 N3 N719

b)

0 100 200 300 400 500 600 700 0

5 10 15 20

J (mcAm-2 )

E(mV)

M8-1 M8-2 M8-O1 M8-O2 N3 N719

c)

Fig. 6 J-V curves for all the studied dyes.aD-TiO2,bA- TiO2, andcT-TiO2

1,1

4,1

4,7

1,4

3,3

5,8

1,32

4,27

6,36

M8-1 N3 N719

0 1 2 3 4 5 6 7 8

)%(ycneiciffE

Compound

D-TiO

2

A-TiO

2

T-TiO

2

Fig. 7 Comparison between organic and standard dyes

(9)

consequence of an M8-2 mis-adsorption in the semiconductor surface (the color of the electrode is less red in this case). A detailed comparison of the VOC, JSC, FF, andηdata obtained (Suppl. material Table4S) enables us to determine that the aliphatic family (M8-1 and M8-2) has higher values than the alcoxyl family (M8-O1 and M8-O2). Furthermore, the methyl substituent contributes to increasing the different values more than the ethoxyl in both families. Nevertheless, the highest values are seen for the alcoxyl family (in T TiO2). A trend exists between IPCE and JSCvalues, the aliphatic family has higher values than the alcoxyl family, on average, showing the relationship cited in the IPCE section. However, looking at every compound there is not a trend among the values obtained, except for some coincidences.

If the average values are compared (Suppl. material Figs. 23S), the standard compounds are higher in value.

In contrast, if a proper comparison is made (Fig.7), compound M8-1 is not so far from the efficiencies found in the standard dyes.

Conclusions

The red-ox properties of four D-π-A molecules were measured. Two tested solvents, THF and DMSO, show one and two red-ox peaks, respectively. FromΔE values, reactions in DMSO turn more reversible than in THF, and the frontier orbitals are more predictable and comparable. Additionally, the oxidation potentials (ground state) are between 1.05 and 1.14 V (vs. NHE), and the oxidation potentials (excited state) vary between−1.57 and−1.22 V (vs. NHE).

M8-1 has high light characteristics compared with the other dyes. It includes the highest ground and excited state potentials, 1.14 V and−1.22 V vs. NHE, respectively, and the highest injection energy (DMSO).

The injection energies allow determining that the shorter the substituent length, the less the energetic difference in the electronic transition. Additionally, the alkyl group incorporation increases this gap. Also, a barely noticable charge in the nitrogen is an indicator of small energetic gaps. The injection energies signal that M8-1 is the best injector. The performances in solar cells were measured in three TiO2materials.

The IPCE results show the A-TiO2> T-TiO2> D-TiO2aver- age tendency, and the family alkyl substituted has higher values than the alcoxyl family. Furthermore, in the first family the methyl substituent has a higher value than the ethyl one.

However, the second family shows the contrary effect. M8-1 has the highest IPCE value on average. In terms of efficiency, the alkyl substituted family again has higher values than the alcoxyl family. On average, the methyl substituent has a higher value than the ethyl in both families. M8-1 has the highest efficiency value.

Acknowledgments The authors gratefully acknowledge the financial support from FONDECYT, project Nr. 1141158 and Ecos-Conicyt C14E05. E. Ortega thanks CONICYT-Chile for a doctoral scholarship folio 21140005.

Publisher’s noteSpringer Nature remains neutral with regard to jurisdic- tional claims in published maps and institutional affiliations.

References

1. Adedokun O, Titilope K, Awodugba AO (2016) Review on natural dye-sensitized solar cells (DSSCs). Int J Eng Technol 2(2):34–41.

https://doi.org/10.19072/ijet.96456

2. Han L, Islam A, Chen H, Malapaka C, Chiranjeevi B, Zhang S, Yang X, Yanagida M (2012) High-efficiency dye-sensitized solar cell with a novel co-adsorbent. Energy Environ Sci 5:6057–6060.

https://doi.org/10.1039/C2EE03418B

3. Hua Y, Chang S, Huang D, Zhu XZ, Zhao J, Chen T, Wong W, Wong W (2013) Significant improvement of dye-sensitized solar cell performance using simple phenothiazine-based dyes. Chem Mater 25:2146–2153.https://pubs.acs.org/doi/10.1021/cm400800h 4. Liu W, Wu I, Lai C, Lai C, Chou P, Li Y, Chen C, Hsu Y, Chi Y ( 2 0 0 8 ) S i m p l e o r g a n i c m o l e c u l e s b e a r i n g a 3 , 4 - ethylenedioxythiophene linker for efficient dye-sensitized solar cells. Chem Commun 41:5152–5154.https://doi.org/10.1039/

B808535H

5. Lee C, Lin R, Lin L, Li C, Chu T, Sun S, Lin JT, Ho K (2015) Recent progress in organic sensitizers for dye-sensitized solar cells.

RSC Adv 5:23810–23825.https://doi.org/10.1039/C4RA16493H 6. Yang L, Zheng Z, Li Y, Wu W, Tian H, Wang Z (2015) N-annulated

perylene-based metal-free organic sensitizers for dye-sensitized solar cells. Chem Commun 51:4842–4845.https://doi.org/10.1039/

C5CC00650C

7. Qu S, Qin C, Islam A, Wu Y, Zhu W, Hua J, Tian H, Han L (2012) A n o v e l D–A -π- A o r g a n i c s e n s i t i z e r c o n t a i n i n g a diketopyrrolopyrrole unit with a branched alkylchain for highly efficient and stable dye-sensitized solar cells. Chem Commun 48:

6972–6974.https://doi.org/10.1039/C2CC31998E

8. Yu C, Jiang K, Huang J, Zhang F, Bao X, Wang F, Yang L, Song Y (2013) Novel pyrene-based donor–acceptor organic dyes for solar cell application. Org Electron 14:445–450.https://doi.org/10.1016/

j.orgel.2012.12.013

9. Cong J, Yang X, Liu J, Zhao J, Hao Y, Wang Y, Sun L (2012) Nitro group as a new anchoring group for organic dyes in dye-sensitized solar cells. Chem Commun 48:6663–6665.https://doi.org/10.1039/

C2CC31516E

10. Wachter P, Zistler M, Schreiner C, Berginc M, Krasovec UO, Gerhard D, Wasserscheid P, Hinsch A, Gores HJ (2008) Characterisation of DSSC-electrolytes based on 1-ethyl-3- methylimidazolium dicyanamide: measurement of triiodide diffusion coefficient, viscosity, and photovoltaic performance. J Photochem Photobiol Chem 197:25–33.https://doi.org/10.1016/j.

jphotochem.2007.12.001

11. Qin W, Leen V, Rohand T, Dehaen W, Dedecker P, Auweraer MV, Robeyns K, Meervelt LV, Beljonne D, Averbeke BV, Clifford JN, Driesen K, Binnemans K, Boens N (2009) Synthesis, spectroscopy, crystal structure, electrochemistry, and quantum chemical and molecular dynamics calculations of a 3-Anilino difluoroboron dipyrromethene dye. J Phys Chem A 113:439–447.https://doi.

org/10.1021/jp8077584

12. Mastroianni S, Lembo A, Brown TM, Reale A, Carlo A (2012) Electrochemistry in reverse biased dye solar cells and dye/

(10)

electrolyte degradation mechanisms. Chem Phys Chem 13:2964– 2975.https://doi.org/10.1002/cphc.201200229

13. Kavan L, Tétreault N, Moeh T, Gratzel M (2014) Electrochemical characterization of TiO2 blocking layers for dye-sensitized solar cells. J Phys Chem C 118:16408–16418.https://doi.org/10.1021/

jp4103614

14. Stalder R, Xie D, Islam A, Han L, Reynolds JR, Schanze KS (2014) Panchromatic donor–acceptor–donor conjugated oligomers for dye-sensitized solar cell applications. Appl Mater Interfaces 6:

8715–8722.https://doi.org/10.1021/am501515s

15. Ortega E, Montecinos R, Cattin L, Díaz FR, del Valle MA, Bernède JC (2017) Synthesis, characterization and photophysical-theoretical analysis of compounds D-p-a. 1. Effect of alkyl-phenyl substituted amines in photophysical properties. J Mol Struct 114:615–623.

https://doi.org/10.1016/j.molstruc.2017.04.019

16. Nazeeruddin MK, Kay A, Rodicio I, Humpbry-Baker R, Miiller E, Liska P, Vlachopoulos N, Gratzel M (1993) Conversion of light to e l e c t r i c i t y b y c i s - X z B i s ( 2 , 2′- b i p y r i d y l - 4 , 4′- dicarboxylate)ruthenium(11) charge-transfer sensitizers (X = C1-, Br-, I-, CN-, and SCN-) on Nanocrystalline Ti02 electrodes. J Am Chem Soc 115:6382–6390.https://doi.org/10.1021/ja00067a063 17. Gutiérrez-Tauste D, Zumeta I, Vigil E, Hernández-Fenollosa MA,

Domènech X, Ayllón JA (2005) New low-temperature preparation method of the TiO2 porous photoelectrode for dye-sensitized solar cells using UV irradiation. J Photochem Photobiol A Chem 175:

165–171.https://doi.org/10.1016/j.jphotochem.2005.04.031 18. Gupta N, Linschitz H (1997) Hydrogen-bonding and protonation

effects in electrochemistry of quinones in aprotic solvents. J Am Chem Soc 119:6384–6391.https://doi.org/10.1021/ja970028j 19. Dubois D, Moninot G, Kutner W, Jones MT, Kadish KM (1992)

Electroreduction of buckminsterfullerene, C60, in aprotic solvents.

Solvent, supporting electrolyte, and temperature effects. J Phys Chem 96:7137–7145.https://doi.org/10.1021/j100196a056 20. Garcia P, Pernaut JM, Hapiot P, Wintgens V, Valat P, Garnier F,

Delabouglise D (1993) Effect of end substitution on electrochemical and optical properties of oligothiophenes. J Phys Chem 97:513– 516.https://doi.org/10.1021/j100104a040

21. Wang H, Hsiao S (2014) Substituent effects on electrochemical and electrochromic properties of aromatic polyimides with 4-(carbazol- 9-yl)triphenylamine moieties. J Polym Sci Pol Chem 52:1172– 1184.https://doi.org/10.1002/pola.27103

22. Seo ET, Nelson RF, Fritsch JM, Marcoux LS, Leedy DW, Adams RN (1966) Anodic oxidation pathways of aromatic amines.

Electrochemical and Electron paramagnetic resonance studies. J Am Chem Soc 88(15):3498–3503. https://doi.org/10.1021/

ja00967a006

23. Mohan S, Okumu F, Oluwafemi O, Matoetoe M, Arotiba O (2016) Electrochemical behaviour of silver nanoparticle-MWCNTs hybrid nanostructures synthesized via a simple method. Int J Electrochem Sci 11:745–753.http://www.electrochemsci.org/papers/vol11/

110100745.pdf

24. Lu JF, Tsai YY, Tsa CJ (2016) Shape dependence of the electrochemical properties ofα-Fe2O3 particles as anode materials for lithium ion batteries. RSC Adv 6:26929–26935.https://doi.org/

10.1039/C5RA27630F

25. Martinez MV, Rodrigue RC, Moncada AB, Rivarola CR, Bruno MM, Miras MC, Barbero CA (2016) Electrochemistry of Tris(1, 10-phenanthroline)iron(II) inside a polymeric hydrogel. Coupled chemical reactions and migration effects. J Solid State Electrochem 20(11):2951–2960.https://doi.org/10.1007/s10008- 016-3312-6

26. Avramov I (2009) Relationship between diffusion, self-diffusion and viscosity. J Non-Cryst Solids 355:745–747.https://doi.org/10.

1016/j.jnoncrysol.2009.02.009

27. Huang Y, Xu W, Zhou C, Zhong W, Xie R, Gong X, Ying L, Huang F, Cao Y (2016) Synthesis of medium-bandgapπ-conjugated

polymers based on isomers of 5-Alkylphenanthridin-6(5H)-one and 6-Alkoxylphenanthridine. J Polym Sci Pol Chem 54(14):

2119–2127.https://doi.org/10.1002/pola.28079

28. Woo CH, Beaujuge PM, Holcombe TW, Lee OP, Frechet JMJ (2010) Incorporation of furan into low band-gap polymers for efficient solar cells. J Am Chem Soc 132(44):15547–15549.https://

doi.org/10.1021/ja108115y

29. Wei W, Lu Y, Chen W, Chen S (2011) One-pot synthesis, photoluminescence, and electrocatalytic properties of subnanometer-sized copper clusters. J Am Chem Soc 133(7):

2060–2063.https://doi.org/10.1021/ja109303z

30. Abe R, Higashi M, Domen K (2010) Facile fabrication of an efficient oxynitride TaON photoanode for overall water splitting into H2 and O2 under visible light irradiation. J Am Chem Soc 132(34):

11828–11829.https://doi.org/10.1021/ja1016552

31. Dou L, Gao J, Richard E, You J, Chen C, Cha KC, He Y, Li G, Yang Y (2012) Systematic investigation of benzodithiophene- and diketopyrrolopyrrole-based low-bandgap polymers designed for single junction and tandem polymer solar cells. J Am Chem Soc 134(24):10071–10079.https://doi.org/10.1021/ja301460s 32. Chen C, Hsu Y, Chou H, Thomas KRJ, Lin JT, Hsu C (2010)

Dipolar compounds containing fluorene and a heteroaromatic ring as the conjugating bridge for high-performance dye-sensitized solar cells. Chem Eur J 16:3184–3193.https://doi.org/10.1002/chem.

200903151

33. Capodilupo AL, Fabiano E, Marco L, Ciccarella G, Gigli G, Mar tinel li C, Ca rdone A (2016) [1]Be nzothie no[3,2- b]benzothiophene-based organic dyes for dye-sensitized solar cells.

J Org Chem 81:3235–3245.https://doi.org/10.1021/acs.joc.

6b00192

34. Wu K, Laws DR, Nafady A, Geiger WE (2014) Substitution of CO ligand by P(OPh)3 in radical cations of the cymantrene family:

relationships of substitution rates to E1/2 values and carbonyl IR frequencies. J Inorg Organomet Polym 24:137–144.https://doi.org/

10.1007/s10904-013-9976-9

35. Obotowo IN, Obot IB, Ekpe UJ (2016) Organic sensitizers for dye- sensitized solar cell (DSSC): properties from computation, progress and future perspectives. J Mol Struct 1122:80–87.https://doi.org/

10.1016/j.molstruc.2016.05.080

36. Deinhammer RS, Ho M, Anderegg JW, Porter MD (1994) Electrochemical oxidation of amine-containing compounds: a route to the surface modification of glassy carbon electrodes. Langmuir 10:1306–1313.https://doi.org/10.1021/la00016a054

37. Huang W, Humphrey BD, MacDiarmid AG (1986) Polyaniline, a novel conducting polymer. Morphology and chemistry of its oxidation and reduction in aqueous electrolytes. J Chem Soc 82:2385–

2400.https://doi.org/10.1039/F19868202385

38. Cheng S, Hsiao S, Su T, Liou G (2005) Novel aromatic poly(amine- imide)s bearing a pendent triphenylamine group: synthesis, ther- mal, photophysical, electrochemical, and electrochromic characteristics. Macromolecules 38:307–316. https://doi.org/10.1021/

ma048774d

39. Li L, Gibson EA, Qin P, Boschloo G, Gorlov M, Hagfeldt A, Sun L (2010) Double-layered NiO photocathodes for p-type DSSCs with record IPCE. Adv Mater 22:1759–1176.https://doi.org/10.1002/

adma.200903151

40. Chiba Y, Islam A, Watanabe Y, Komiya R, Koide N, Han L (2006) Dye- sensitized solar cells with conversion efficiency of 11.1%. Jpn J Appl Phys 45(25):24–28.http://iopscience.iop.org/article/10.

1143/JJAP.45.L638/meta

41. Chai Q, Li W, Wu Y, Pei K, Liu J, Geng Z, Tian H, Zhu W (2014) Effect of a long alkyl group on cyclopentadithiophene as a conjugated bridge for D−a−π−a organic sensitizers: IPCE, Electron diffusion length, and charge recombination. Appl Mater Interfaces 6:

14621–14630.https://doi.org/10.1021/am503891q

(11)

42. Hara KSK, Mori N, Satsuki M, Suga S, Tsukagoshi S, Abe Y, Sugihara H, Arakawa H (2000) Photosensitization of a porous TiO electrode with merocyanine dyes containing a carboxyl group and a long alkyl chain. Chem Commun 0:1173–1174.https://doi.

org/10.1039/B001517M

43. Xue G, Yu X, Yu T, Bao C, Zhang J, Guan J, Huang H, Tang Z, Zou Z (2012) Understanding of the chopping frequency effect on IPCE measurements for dye-sensitized solar cells: from the viewpoint of electron transport and extinction spectrum. J Phys D Appl Phys 45:

425104 (7pp).http://iopscience.iop.org/article/10.1088/0022-3727/

45/42/425104/meta

44. Tian H, Yang X, Chen R, Hagfeldt A, Sun L (2009) A metal-free Bblack dye^for panchromatic dye-sensitized solar cells. Energy Environ Sci 2:674–677.https://doi.org/10.1039/B901238A 45. Guillén E, Casanueva F, Anta JA, Vega-Poot A, Oskam G,

Alcántara R, Fernández-Lorenzo C, Martín-Calleja J (2008)

Photovoltaic performance of nanostructured zinc oxide sensitised with xanthene dyes. J Photochem Photobiol A Chem 200:364–370. https://doi.org/10.1016/j.jphotochem.

2008.08.015

46. Lee W, Roh S, Hyung K, Park J, Lee S, Han S (2009) Photoelectrochemically polymerized polythiophene layers on ruthenium photosensitizers in dye-sensitized solar cells and their ben- eficial effects. Sol Energy 83:690–695.https://doi.org/10.1016/j.

solener.2008.10.017

47. Stergiopoulos T, Karakostas S, Falaras P (2004) Comparative studies of substituted ruthenium(II)–pyrazoyl–pyridine complexes with classical N3 photosensitizer: the influence of–NCS dye ligands on the efficiency of solid-state nanocrystalline solar cells. J Photochem Photobiol A Chem 163:331–340.https://doi.org/10.1016/j.

jphotochem.2004.01.002