New tetrazole-based organic dyes for dye-sensitized solar cells

New tetrazole-based organic dyes for dye-sensitized solar cells

Accepted Manuscript New tetrazole-based organic dyes for dye-sensitized solar cells Zahra Jafari Chermahini , Alireza Najafi Chermahini , Hossein A D...

1MB Sizes 1 Downloads 82 Views

Accepted Manuscript

New tetrazole-based organic dyes for dye-sensitized solar cells Zahra Jafari Chermahini , Alireza Najafi Chermahini , Hossein A Dabbagh , Abbas Teimouri PII: DOI: Reference:

S2095-4956(15)00099-6 10.1016/j.jechem.2015.10.015 JECHEM 61

To appear in:

Journal of Energy Chemistry

Received date: Revised date: Accepted date:

29 June 2015 8 September 2015 28 September 2015

http://www.journals.elsevier.com/ journal-of-energy-chemistry/

Please cite this article as: Zahra Jafari Chermahini , Alireza Najafi Chermahini , Hossein A Dabbagh , Abbas Teimouri , New tetrazole-based organic dyes for dye-sensitized solar cells, Journal of Energy Chemistry (2015), doi: 10.1016/j.jechem.2015.10.015

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

New tetrazole-based organic dyes for dye-sensitized solar cells Zahra Jafari Chermahinia, Alireza Najafi Chermahinib,*, Hossein A Dabbagha, Abbas Teimouria a

Department of Chemistry, Isfahan University of Technology, 84154-83111 Isfahan, Iran b

Chemistry Department, Payame Noor University (PNU), Tehran 19395-4697, Iran

Article history: Received 29 June 2015

CR IP T

Revised 8 September 2015 Accepted 28 September 2015 Available online Abstract

AN US

A series of new metal-free organic dyes that contain donors with triphenylamine or its derivatives and tetrazole-based acceptors were synthesized and characterized by photophysical, electrochemical, and theoretical computational methods. They were applied in nanocrystalline TiO2 solar cells (DSSCs). It is found that the introduction of diphenylamine units as antennas in the as-synthesized dyes could improve

M

photovoltaic performance compared with phenothiazine and carbazole units as antennas in DSSCs. The dye with (2H-tetrazol-5-yl) acrylonitrile electron acceptor also displayed the highest solar-to-electrical

ED

energy conversion efficiency.

Key words: Tetrazole, Dye sensitized solar cells, DFT, Synthesis, Anchoring group

PT

* Corresponding author. Tel: +983133913256; Fax: +983133913250; E-mail: [email protected]

CE

This work was supported by Isfahan University of Technology. 1. Introduction

AC

Since the discovery by Grätzel et al. in 1991, dye-sensitized solar cells (DSSCs) on organic

and organometallic complexes have attracted considerable attention [1]. A typical DSSC is constructed using wide band gap semiconductor oxides such as TiO 2 or ZnO, a molecular sensitizer, a redox electrolyte and a Pt-coated counter electrode [2,3]. In the DSSCs system, the photoexcited electrons from dye inject into the conduction band (CB) of the TiO 2. The oxidized dye is reduced back to the ground state by electron transfer from the electrolyte. The

1

ACCEPTED MANUSCRIPT

photosensitizer is thus one of the critical components in DSSC to efficient light harvesting and electron generation/transfer [4]. Organic dyes have been used as photosensitizers in DSSCs, because of their many advantages, such as large absorption coefficients due to intramolecular π-π* transitions, inexpensiveness with no transition metals contained, environment-friendliness [5]. Generally, donor-linker-acceptor

Electron-withdrawing

anchoring

groups

such

as

CR IP T

(D-π-A) structure is the common design of the organic sensitizer [6 –10]. the

nitro

group,

aldehyde,

2-(1,1-dicyanomethylene) rhodanine, pyridine and 8-hydroxylquinoline as an alternative to conventional carboxyl groups had already been used in DSSCs [11]. Recently, the effectiveness of the tetrazole functional group as a serious alternative anchoring group for organic

AN US

photosensitizers in DSSCs has also been reported by Massin et al. [12]. Because of its highest value of ionization potential among azoles and its low HOMO value, this functional group is extraordinarily stable to both acids and bases as well as to oxidizing and reducing conditions [13–15]. They are used in coordination chemistry as a ligand, in pharmaceuticals as a

M

metabolically stable surrogate for a carboxylic acid group, in materials science applications including propellants and explosives [16 –18].

ED

In this paper, we report on the synthesis and characterization of novel metal-free organic dyes (Figure 1) that contain donors with triphenylamine or its derivatives and tetrazole-based acceptors and their

PT

application as sensitizers in DSSCs. In this design, TD1 is the basic model, in which the triphenylamine unit was connected to the ethyl 2-(1H-tetrazol-5-yl) acetate by one double bond. Phenothiazine,

CE

carbazole and diphenylamine units as antenna groups were attached to the phenyl rings of TD1, leading to the structures of TD2, TD3, TD4. Therefore, new starburst dyes with D-D-π-A structure were

AC

synthesized by introducing starburst donor units into the D-π-A molecule. Such a starburst conformation probably minimizes the charge recombination processes of injected electrons with triiodide in the electrolyte [19]. The dyes TD5 and TD6 containing (2H-tetrazol-5-yl) acrylonitrile and 1H-tetrazole-5-acetic acid moieties as acceptor, for the purpose of comparison with TD1 containing ethyl 2-(1H-tetrazol-5-yl) acetate were also synthesized and the effect of different tetrazole-based motifs as anchoring group on the performance of the dye-sensitized solar cells was studied.

2

ACCEPTED MANUSCRIPT

TD4

TD5

TD3

CR IP T

TD2

AN US

TD1

TD6

Figure 1. Molecular structures of triphenylamine dyes.

M

2. Experimental 2.1. General methods

ED

All solvents and chemicals used in this work were reagent grade and used without further purification and were obtained from commercially available sources. Phenothiazine and carbazole were purchased

PT

from Merck. All solvents and other chemical used in this work were purchased from Sigma-Aldrich.

CE

Solvents were dried from appropriate drying agents (calcium sulfate for dimethylformamide, calcium chloride for 1,2-dichlorobenzene, calcium hydride for acetonitrile and dichloromethane) and freshly

AC

distilled before use. The 1H and 13C NMR spectra were measured using Bruker 400 MHz spectrometers. Mass spectra (EI-MS) were measured with 5973 Network Mass Selective Detector (Agilent Technology (HP)). FTIR absorption spectra were recorded on a FTIR Jasco 680 plus spectrometer using KBr pellets. FTIR spectrum of TiO2 nanoparticles was subtracted from the spectra of dyes on TiO2 nanoparticles. The absorption spectra of the dyes in solution and adsorbed on TiO2 films were measured with a Jasco-570 UV/vis spectrophotometer. The precursor compounds 1-3 (Scheme 1) and the acceptor moieties were synthesized according to the literature procedures [20–23]. Cyclic voltammetry experiments of the dyes were performed in SAMA 500 electroanalyser system, SAMA research center, Iran, with a Pt working

3

ACCEPTED MANUSCRIPT

electrode, a Pt counter electrode, an Ag/AgCl reference electrode, and a scan rate of 0.1 V/s at room temperature under nitrogen in a CH2Cl2 solution that included 0.1 M tetrabutylammonium hexafluorophosphate. The potentials were calibrated against the ferrocene internal standard. DSSCs were constructed by Sharif Solar, with an effective area of 0.25 cm2 and an iodine-based liquid electrolyte. The photoanodes were prepared by doctor blade and two layers of nanocrystalline TiO2 particles. A transparent layer composed of 20 nm TiO2 particles (Sharif Solar) with thickness of 7 µm and a

CR IP T

scattering layer of 150–300 nm TiO2 particles with a thickness 5 µm to enhance the light harvesting and improve the device performance. The TiO2 electrodes were immersed into a dichloromethane solution of dyes (4×10-4 M) for 20 h. Solar cells were illuminated by solar simulator (SIM-1000, Sharif Solar) to provide an incident irradiance of 100 mW/cm2 at the surface of the device. I-V measurements were

AN US

performed using a potentio state (Palmsense, Netherlands). Action spectra of the incident photon-to-current conversion efficiencies (IPCE) were measured using a 100 W halogen lamp associated with a monochromator (Jarrel Ash monochromator) in DC mode. The amount of each dye loaded onto TiO2 is calculated from the difference in concentration of each solution before and after TiO2 film

M

immersion. UV-visible absorption spectra of each solution before and after sensitization of TiO2 film were recorded. The difference in concentration of each solution before and after TiO2 film immersion

ED

was deduced from the difference between the two sets of data. 2.2. Synthesis

PT

The dyes (TD1, TD5 and TD6) were prepared in two steps involving the Vilsmeier-Haack formylation

CE

of the triphenylamine followed by the Knoevenagel condensation of the resulting aldehydes with ethyl 2-(1H-tetrazol-5-yl) acetate, (2H-tetrazol-5-yl) acrylonitrile, or 1H-tetrazole-5-acetic acid (Scheme 1).

AC

The dyes (TD2-TD4) were synthesized from 4-(diphenylamino) benzaldehyde 1 following a three-step procedure. Iodination of aldehyde 1 led to 4-(bis(4-iodophenyl)amino) benzaldehyde 2. Cu-catalysed ullman coupling reaction of 2 with phenothiazine, carbazole or diphenylamine provided 3 which was subsequently reacted with ethyl 2-(1H-tetrazol-5-yl) acetate by a Knoevenagel condensation.

4

AN US

CR IP T

ACCEPTED MANUSCRIPT

Scheme 1. Preparation route to the dyes TD1-TD6a. (i) POCl3, DMF, 95–100°C, 20 h; (ii) Ethyl 2-(1H-tetrazol-5-yl) acetate, (2H-tetrazol-5-yl) acrylonitrile, or 1H-tetrazole-5-acetic acid, NH(Et)2, CH3CN, reflux, 24 h; (iii) KI, KIO3,CH3COOH, H2O, 80° C, 5 h; (iv) Phenothiazine, carbazole, or diphenylamine, Cu, K2CO3, 18-crown-6, 1,2-dichlorobenzene, reflux, 48 h; (v) Ethyl 2-(1H-tetrazol-5-yl) acetate, NH(Et)2,

M

CH3CN, reflux, 48 h.

ED

2.3. General procedures for the synthesis of TD1-TD6 Aldehyde (1equiv) and ethyl 2-(1H-tetrazol-5-yl) acetate, (2H-tetrazol-5-yl) acrylonitrile, or

PT

1H-tetrazole-5-acetic acid (1.5 equiv) were added to 5 mL of acetonitrile and refluxed in the presence of diethylamine (20 equiv). After cooling to room temperature, the mixture acidified with 5 M hydrochloric

CE

acid. The solid was filtered and washed by water. Finally, the precipitate was purified by column

AC

chromatography (hexan/ethylacetate). 2.3.1. Ethyl (E)-3-(4-(diphenylamino)phenyl)-2-(1H-tetrazol-5-yl) acrylate (TD1) TD1 was prepared according to the general procedure from 1 (273 mg, 1 mmol), ethyl

2-(1H-tetrazol-5-yl) acetate (234 mg, 1.5 mmol) and diethylamine (2 mL, 20 mmol) in CH3CN (5 mL). Yield was 82% (deep yellow solid). 1H NMR (400 MHz, CDCl3): δ 1.38 (t, J = 7.2 Hz, 3H), 4.38 (q, J = 7.2 Hz, 2H), 6.90 (d, J = 8.8 Hz, 2H), 7.16 (m, 6H), 7.34 (t, J = 8Hz, 4H), 7.56 (d, J = 8.4 Hz, 2H), 8.18 (s, 1H). 13C NMR (400 MHz, CDCl3): δ 14.27, 62.33, 119.26, 124.75, 124.99, 125.26, 125.93, 126.23, 129.48, 129.66, 133.74, 146.05, 149.15, 166.62. Ms (EI): Calcd for C24H21N5O2, 411.17; Found: 411.00.

5

ACCEPTED MANUSCRIPT

Anal. Calcd for C24H21N5O2: C, 70.06; H, 5.14; N, 17.02; Found: C, 69.06; H, 5.24; N: 16.04. 2.3.2. Ethyl (E)-3-(4-(bis(4-(10H-phenothiazin-10-yl)phenyl)amino)phenyl)-2-(1H-tetrazol-5-yl)acrylate (TD2) TD2 was prepared according to the general procedure from 3 (334 mg, 0.5 mmol), ethyl 2-(1H-tetrazol-5-yl) acetate (117 mg, 0.75 mmol) and diethylamine (1 mL, 10 mmol) in CH3CN (2.5

CR IP T

mL). Yield was 80% (yellow solid). 1H NMR (400 MHz, CDCl3): δ 1.31 (t, J = 7.2 Hz, 3H), 4.32 (q, J = 7.2 Hz, 2H), 6.38 (d, J = 8 Hz, 4H), 6.86 (t, 7.2 Hz, 4H), 6.96 (t, J = 7.2 Hz ,4H), 7.05 (m, 6H), 7.28 (d, 8.4 Hz , 2H), 7.36 (m, 8H), 8.16 (s, 1H). 13C NMR (400 MHz, CDCl3): δ 13.94, 61.73, 116.52, 116.62, 121.16, 121.47, 122.82, 126.13, 126.73, 126.96, 126.98, 127.20, 131.37, 132.93, 137.37, 144.09, 145.43,

AN US

147.75, 149.93, 165.71. Ms (EI): Calcd for C48H35N7O2S2: 805.23; Found: 805.00. Anal. Calcd for C48H35N7O2S2: C, 71.53; H, 4.38; N, 12.17; S, 7.96. Found: C, 70.55; H, 3.38; N, 11.21; S, 7.02. 2.3.3. Ethyl (E)-3-(4-(bis(4-(9H-carbazol-9-yl)phenyl)amino)phenyl)-2-(1H-tetrazol-5-yl)acrylate (TD3) TD3 was prepared according to the general procedure from 3 (302 mg, 0.5 mmol), ethyl

M

2-(1H-tetrazol-5-yl) acetate (117 mg, 0.75mmol) and diethylamine (1 mL, 10 mmol) in CH3CN (2.5 mL). Yield was 77% (orange solid). 1H NMR (400 MHz, CDCl3): δ 1.30 (t, J = 7.2 Hz, 3H), 4.30 (q, J = 7.2

ED

Hz, 2H), 7.38 (m, 12 H), 7.42 (m, 6H), 7.50 (m, 6H), 8.10 (d, J = 7.6 Hz, 4H), 8.14 (s, 1H). 13C NMR (400 MHz, CDCl3): δ 14.18, 62.49, 120.01, 120.11, 120.40, 120.45, 121.11, 125.95, 125.97, 126.04,

PT

126.57, 127.01, 128.06, 128.17, 128.29, 128.32, 132.96, 133.77, 134.24, 166.39. Ms (EI): Calcd for C48H35N7O2: 741.29; Found: 741.00. Anal. Calcd for C48H35N7O2: C, 77.71; H, 4.76; N, 13.22. Found: C,

CE

76.77; H, 3.87; N, 12.72.

2.3.4. Ethyl (E)-3-(4-(bis(4-(diphenylamino)phenyl)amino)phenyl)-2-(1H-tetrazol-5-yl)acrylate (TD4)

AC

TD4 was prepared according to the general procedure from 3 (304 mg, 0.5 mmol), ethyl

2-(1H-tetrazol-5-yl) acetate (117 mg, 0.75 mmol) and diethylamine (1 mL, 10 mmol) in CH3CN (2.5 mL). Yield was 75% (red solid). 1H NMR (400 MHz, CDCl3): δ 1.35 (t, J = 7.2 Hz, 3H), 4.32 (q, J = 7.2 Hz, 2H), 6.96 (m, 8H), 7.6 (m, 8H), 7.20 (m, 17H). 13C NMR (400 MHz, CDCl3): 14.02, 61.74, 124.28, 126.48, 127.15, 129.31, 129.35, 129.38, 129.52, 130.00, 134.64, 145.41, 145.55, 146.86, 148.26, 166.04. Ms (EI): Calcd for C48H39N7O2: 745.32; Found: 745.00. Anal. Calcd for C48H39N7O2: C, 77.29; H, 5.27; N, 13.15. Found: C, 76.44; H, 4.6; N, 12.77.

6

ACCEPTED MANUSCRIPT

2.3.5. (E, Z)-3-(4-(diphenylamino)phenyl)-2-(1H-tetrazol-5-yl)acrylonitrile (TD5) TD5 was prepared according to the general procedure from 1 (273 mg, 1 mmol), (2H-tetrazol-5-yl)-acrylonitrile (164 mg, 1.5 mmol) and diethylamine (2 mL, 20 mmol) in CH3CN (5 mL). Yield was 87% (dark yellow). 1H NMR (400 MHz, CDCl3): δ 6.92 (d, J = 8.8 Hz, 2H), 7.04 (d, J = 8.8 Hz, 2H), 7.21 (m, 12H), 7.38 (m, 8H), 7.60 (s, 1H), 7.70 (d, J = 8.8Hz, 2H), 7.88 (d, J = 9.2 Hz, 2H), 13

C NMR (400 MHz, CDCl3): 116.62, 118.84, 119.23, 125.48, 125.60, 125.82, 126.47,

CR IP T

8.37 (s, 1H).

126.49, 129.80, 129.86, 132.60, 133.69, 145.62, 145.64, 151.90, 152.08, 152.40. Ms (EI): Calcd for C22H16N6: 364.14; Found: 364.00. Anal. Calcd for C22H16N6: C, 72.51; H, 4.43; N, 23.06. Found: C, 71.70; H, 5.4; N, 23.97.

AN US

2.3.6. (E)-3-(4-(diphenylamino)phenyl)-2-(1H-tetrazol-5-yl)acrylic acid (TD6)

TD6 was prepared according to the general procedure from 1 (273 mg, 1 mmol), 1H-tetrazole-5-acetic acid (192 mg, 1.5 mmol) and diethylamine (2 mL, 20 mmol) in CH3CN (5 mL). Yield was 85% (yellow solid). 1H NMR (400 MHz, DMSO-d6): δ 6.93 (d, J = 8.8 Hz, 2H), 7.12 (m, 6H), 7.35 (t, 7.6 Hz, 4H), 7.55 (s, 1H), 7.59 (m, 3H). 13C NMR (400 MHz, DMSO-d6): δ 107.61, 121.42, 123.98, 124.92, 128.07,

M

128.75, 129.71, 137.36, 146.46, 148.53. Ms (EI): Calcd for C22H17N5O2: 383.14; Found: 383.00. Anal.

ED

Calcd for C22H17N5O2: C, 68.92; H, 4.47; N, 18.27; Found: C, 69.80; H, 5.00; N, 19.1. 2.4. Theoretical calculations

PT

To model the electronic state of the dyes absorbed on the TiO2 surface, we employed a simple dye-titanium dioxide model. It is verified that this model is sufficient for the examination of geometries,

CE

excitations, and electronic structures of organic dyes after binding to TiO2 [24]. The geometries of the dyes before and after linked to TiO2 were optimized using the B3LYP method [25,26] with the 6-31+G(d)

AC

for C, H, O, N, and S atoms, effective core potential LANL2DZ basis set for Ti atom. The stable geometries were verified by no imaginary frequency. The excitation transitions of the dyes were calculated using time-dependent density functional theory (TD-DFT) calculation with B3LYP/6-31+G(d). The solvent (dichloromethane) effects were considered using the polarizable continuum model (PCM) [27,28]. All the calculations were performed using Gaussian 09 package [29]. 3. Results and discussion 3.1. UV-Vis absorption spectra

7

ACCEPTED MANUSCRIPT

Figure S1 (Supporting Information) shows the absorption spectra of the as-synthesized dyes in dichloromethane solution, and the corresponding data are presented in Table 1. All the dyes exhibit two distinct absorptions bands. The absorption peaks at around 280 nm correspond to the π-π* electron transition. The absorption peaks in the visible part can be assigned to an intramolecular charge transfer (ICT) between the TPA donor part of the molecule and the acceptor end group. As depicted in Figure 2, among the dyes TD1, TD5 and TD6, the dye TD6 show the highest extinction coefficients (ε) of 29700

AN US

CR IP T

L/mol/cm at 384 nm.

M

Figure 2. Absorption spectra of TD1-TD6 in CH2Cl2 solution (1 × 10-5 M).

ED

While dye TD5 exhibits the charge transfer band at 434 nm, the corresponding peak is slightly broader than those for dyes TD1 and TD6. Compared with the absorption of dye TD1, the absorption spectrum of dye TD2 shows a 4 nm hypsochromic shift. This may be due to the nonplanar structure of phenothiazine,

PT

which decreases the coplanarity between the electron donor and the electron acceptor in the ground state

CE

[30]. The maximum absorption of TD4 is redshifted in comparison with that of TD1, which is due to the expansion of π-conjugation systems by introduction of the diphenylamine unit to the phenyl rings of

AC

TD1. The extinction coefficient of the charge transfer transition band of TD1-TD4 increases in the order of TD1 > TD4 > TD3 > TD2. The spectra of the as-synthesized dyes at different concentrations are shown in Figure S2 (Supporting

Information). The spectra shape and maximum absorption aren’t changed, illustrating that there is not any stacked aggregation [31] in the solution state due to the nonplanar structure of TPA, which prevents of the π-π interactions between two adjacent parallel phenyl rings. Figure S3 (Supporting Information) shows the absorption spectra of the dye TD4 in different solvents. TD4 shows negative solvatochromism for the charge transfer band in more polar solvents, which is due

8

ACCEPTED MANUSCRIPT

to the partial deprotonation of the tetrazole unit in the excited state, which leads to smaller dipole moment in the excited state than that in the ground state. Figure S4 (Supporting Information) shows the absorption spectrum of the dyes adsorbed on the TiO2 surface. Compared to the spectrum in dichloromethane, a broadening of the absorption peak is observed in all dyes on the TiO2 surface, indicating that the oxide layer is photosensitized by the dyes effectively. The absorption spectrum of TD2 adsorbed on the TiO2 is similar to that of the corresponding solution

CR IP T

spectra. TD4 shows slight blue shift of the absorption (4 nm). The blue-shift of the absorption peak on the TiO2 was also observed in TD1 (14 nm), TD5 (28 nm) and TD3 (14 nm).

The blue-shifts of the absorption spectra by adsorption on the TiO2 are due to the deprotonation of the dye molecules and to their interaction with TiO2 [32] or H-aggregation of the dyes [33]. Compared with

AN US

the dye TD1, very little blue-shift of the absorption spectrum of TD2 and TD4 may be rationalized by the prevention of the dyes to aggregate on the TiO2 due to the starburst antenna groups linked to the triphenylamine. The dyes TD1 and TD5 with a flatter structure have a greater tendency to form aggregates; therefore, the blue shifts of the absorption spectra are probably due to the formation of

M

H-aggregates on the TiO2 surface. The different spectral changes in TD1 and TD5 can be attributed to the different interaction modes with TiO2. The absorption peak of the dye TD6 is red-shifted from 384

ED

nm to 408 nm by adsorption on the TiO2 surface, which can be ascribed to the formation of J-type

PT

aggregates [34].

CE

Table 1 Optical and electrochemical properties of dyes TD1-TD6 in CH2Cl2 solutions and adsorbed on TiO2. Dye λmaxa λmaxb E0-0c Eoxd E0-0*e HOMOexpf (calc) LUMOexpf(calc.) εmax (nm) (nm) (eV) (V) (V) (eV) (eV) (L/M/cm)

AC

TD1 422 20400 408 2.51 0.88 -1.63 -4.90 (-5.60) -2.40 (-2.48) TD2 418 12200 418 2.52 0.56 -1.96 -4.32 (-5.29) -1.80 (-2.73) TD3 423 13100 406 2.48 0.84 -1.64 -4.74 (-5.58) -2.26 (-2.66) TD4 429 17200 418 2.43 0.64 -1.79 -4.45 (-5.10) -2.01 (-2.38) TD5 434 26000 404 2.46 0.53 -1.93 -4.30 (-5.82) -1.85 (-2.77) TD6 384 29700 408 2.90 0.45 -2.45 -4.20 (-5.69) -1.30 (-2.62) a Absorption maxima of dyes TD1-TD6 in CH2Cl2 solutions (1×10-5 M at 298 K). bAbsorption maxima of dyes adsorbed on TiO2. cThe band gap, E0–0, was estimated from the observed optical edge. dOxidation potential measured in CH2Cl2 containing 0.1 M Bu4NPF6 as supporting electrolyte, Ag/AgCl as a reference electrode, Pt as working electrode. Potentials calibrated with ferrocenium/ferrocene as an internal reference were converted to normal hydrogen electrode (NHE) by addition of +0.63 V [35]. eE0–0*: the excited-state oxidation potential vs NHE. fHOMO =–(Eoxonset+4.2) [36], LUMO = E0-0 + HOMO.

9

ACCEPTED MANUSCRIPT

3.2. FTIR spectra Figure S5 shows FTIR spectra for the free TD1-TD6 powders and the compounds adsorbed on the TiO2 nanoparticles. For the free dye TD5, the stretching peak for cyano is observed at 2212 cm-1. After the dye is adsorbed on the TiO 2 nanoparticles, the IR peak for cyano is observed at the same frequency, indicating that the nitrile group is not involved in the anchoring process and

CR IP T

TD5 is anchored on the TiO 2 surface through the tetrazole group. The bands appearing at 1239, 1307, 1342 and 1438 cm -1 in free TD5 and shifted to 1222, 1296, 1333 and 1429 cm -1 in the dye adsorbed on the TiO 2 surface are probably associated with the N=N and C=N bonds. For TD6, the carbonyl stretching peak is unchanged after adsorbing on the TiO 2 nanoparticles (at 1644

AN US

cm-1), while the 2195–3235 cm-1 stretching peak for O–H is disappeared. This suggests that the OH group of the carboxylic acid is involved in adsorption on the TiO 2 surface. When the dye TD1-TD4 are adsorbed on the TiO 2 nanoparticles, the ester stretching peaks are shifted by 14 –46 cm-1 to lower wavenumbers, showing the interaction of the ester groups with the TiO 2 nanoparticles. For the dyes TD1-TD4 and TD6, the stretching bands for C=N and N=N show

M

only very few changes after adsorbing on the TiO 2 surface, indicating that ring vibrations are

ED

little affected upon grafting. This may be due to the interaction of TD1-TD4 and TD6 with the TiO2 surface through N1 of the tetrazole and the carbonyl group of the ester (TD1-TD4) or the OH group of the carboxylate (TD6) (Figure 3). The calculated IR spectra of TD1, TD5 and TD6

AC

CE

PT

are shown in Figure S6 (Supplementary Information).

TD5

TD1-TD4

TD6

Figure 3. Adsorption modes of TD1-TD6 on the TiO2 nanoparticles. 3.3. Electrochemical properties Cyclic voltammetry measurements of the dyes are summarized in Table 1. Cyclic voltammograms recorded for the dyes exhibit a reversible process which is required for sustaining dye-sensitized solar

10

ACCEPTED MANUSCRIPT

cells (Figure 4). The Comparison of the oxidation potentials of the dyes and the conduction band of TiO2 layer and the redox potential of the redox electrolyte used in the DSSC is shown in Figure 5. The oxidation potential of the dyes TD1-TD6 (0.45–0.88 V versus NHE) is more positive than the standard potential of iodine/iodide (0.4 V versus NHE) that the photooxidized dyes could be expected to be

AN US

CR IP T

reduced efficiently by the iodine/iodide redox couple.

M

Figure 4. Cyclic voltammograms of the dyes TD1-TD6 recorded in dichloromethane solutions (5× 10-4 M). The reduction potential of the dyes vs NHE (E0-0*) was calculated from Eox–E0-0. The optical band-gap

ED

energy (E0-0) was estimated from the absorption onset. The deduced E0–0* values (-1.96 to -2.45 V versus NHE) are more negative than the level of the TiO2 conduction band (CB) edge (-0.5 V vs NHE),

PT

indicating that the electron injection process is energetically favorable. These results clearly demonstrate that the redox potentials of the dyes are favorable for efficient electron injection to TiO2 and the dye

CE

regeneration by iodine/iodide redox system. As listed in Table 1, a negative shift of both Eox and E0–0* can be observed for TD2 vs TD1 by 0.32 V which resulted in a similar band gap. The Eox and E0–0* of

AC

TD3 aren’t different from those for TD1. In the case of TD4 vs TD1, a negative shift of both Eox and E0–0* was observed, and the shift of the Eox was found to be larger than that of E0–0*, which resulted in a decrease of the band gap between the HOMO and LUMO.

11

ACCEPTED MANUSCRIPT

CR IP T

Figure 5. Comparison of the oxidation potentials of the dyes TD1-TD6 dissolved in CH2Cl2 in the ground and excited states.

3.4. Photovoltaic properties

AN US

The incident photon-to-current conversion efficiencies (IPCE) of the dyes on TiO2 are shown in Figure 6. The photocurrent response of TD5 sensitized DSSC is the best due to the IPCE exceeding 80% in the range of 380–438 nm and reaches its maximum of 95% at 418 nm. The onset wavelength of the IPCE for DSSC based on TD5 is 510 nm. The maximum IPCE values of the DSSCs based on TD1 and TD6 are 22% and 25% at 400 nm. The maximum IPCE value of 47% at 420 nm is observed for the

M

DSSC based on TD4. The onset wavelength is shifted to longer wavelength by increasing the

ED

π-conjugated system of the dye compared with the dye TD1. TD2 (maximum IPCE, 17% at 420 nm) and TD3 (maximum IPCE, 29% at 400 nm) show the lower IPCE values than TD4. The lower IPCE values

PT

of TD2 and TD3 are probably due to their lower extinction coefficient [6]. Electron transition is easier in a dye with large conjugated structure and more excited electrons can be

CE

generated from more delocalizing π-electrons on the conjugated donor moiety and the bridge part. These factors affect the IPCE value [37]. The introduction of phenothiazine unit in TD2 brings the decrease of

AC

co-planarity between the electron donor and the electron acceptor in the ground-state. In this dye, HOMO is only localized on a phenothiazine ring, too. Thus, the fewer electrons can be excited under the light excitation and transfer to semiconductor film and a low IPCE value is resulted.

12

CR IP T

ACCEPTED MANUSCRIPT

Figure 6. IPCE spectra for DSSCs based on the as-synthesized dyes.

Figure 7 shows the I-V curves of DSSCs based on the as synthesized dyes and Table 2 presents the photovoltaic data. Among TD2, TD3 and TD4, TD4 exhibits an increase in η of about 0.59% as

AN US

compared to TD1, which can be ascribed to the improvement of the electron-donating ability by introducing diphenylamine groups. The DSSC based on TD4 also shows better open circuit voltage and short circuit photocurrent density. However, the photovoltaic parameters measured with devices employing TD2 and TD3 aren’t better than that of TD1. This result is previously reported for D-D-π-A

M

dyes. These dyes don’t always show better photovoltaic performances than their D-π-A counterparts due to the steric hindrance of two donor moieties, which decreases the effective conjugation of the molecules

ED

[38]. Compared with the overall light to electricity conversion efficiency of TD1, the overall efficiency of TD5 shows an increase of about 0.87%. TD5 also affords an open-circuit photovoltage that is 90 mV

PT

higher than that of TD1 and a short-circuit current density that is 1.86 mA/cm2 higher than that of TD1. Furthermore, the highest η was obtained when TD5 was employed as an organic sensitizer. This highest

CE

η benefits from the highest Jsc, which can be explained by the best IPCE value. These observations offer that among ethyl 2-(1H-tetrazol-5-yl) acetate, (2H-tetrazol-5-yl) acrylonitrile and 1H-tetrazole-5-acetic

AC

acid as anchoring groups, (2H-tetrazol-5-yl) acrylonitrile is the best electron acceptor, suggesting that the dye with (2H-tetrazol-5-yl) acrylonitrile is more effectively adsorbed onto the TiO2 surface. The fill factor values for all the dyes range from 63%–69% with the exception of TD2 (FF = 38%).

The relatively low fill factor of the cell based on the dye TD2 could probably be attributed to the short hole diffusion length and insufficient charge collection [39,40].

13

ACCEPTED MANUSCRIPT

Table 2 Photovoltaic performance of DSSCs sensitized with TD1-TD6 and dye loadings on TiO2. Dye

Voc (mV) 550 520 530 620 640 590

FF (%) 69 38 68 64 63 68

η (%) 0.38 0.28 0.38 0.97 1.15 0.36

Dye loading (×10-7 mol/cm2) 2.64 3.10 2.84 3.21 3.20 2.93

AN US

CR IP T

TD1 TD2 TD3 TD4 TD5 TD6

Jsc (mA/cm2) 1.00 1.44 1.05 2.43 2.86 0.91

M

Figure 7. Current-potential (I-V) curves for the DSSCs based on TD1-TD6 under AM 1.5 irradiation (100 mW/cm2)

ED

3.5. Quantum chemical calculations

PT

Theoretical calculations are performed to understand the electronic structure of the dyes. The electron distribution of the HOMO and LUMO of the dyes before and after linked to TiO2 are depicted in Figure

CE

8. The HOMOs in the isolated TD3 and TD4 are delocalized over the triarylamine and diarylamine/triarylamine units, respectively. The HOMO in TD1 is only localized on a phenothiazine

AC

ring, which can be ascribed to the plane of phenothiazine almost being vertical to the plane of benzene in triphenylamine. The HOMOs in TD1, TD5 and TD6 are to a large extent delocalized over the entire molecule. The LUMOs in all the dyes are located in the acceptor part. Therefore, there is an efficient charge separation. Thus, the HOMO-LUMO excitation induced by light illumination moves the electron-density distribution from the electron-donor moiety to the electron-acceptor moiety through π-bridge. For dyes linked to TiO2, the HOMOs remain unaltered, whereas LUMOs partly mix with TiO2 unoccupied states, leading to a expanding of the dye LUMO.

14

ACCEPTED MANUSCRIPT

The calculated absorption spectra of the dyes before and after linked to TiO2 model are shown in Figure S7, and the corresponding data are summarized in Table 3. The lowest transition of TD2 in CH2Cl2 is calculated to be 2.7382 eV, corresponding to a charge-transfer excitation of the HOMO-2 to the LUMO. The transitions of HOMO→LUMO (f = 0.0000, 506.3 nm) and HOMO-1→ LUMO (f = 0.0010, 498.0) of TD2 are prohibited and there is no obvious absorption in the corresponding theoretical absorption. The lowest absorption bands of TD1 and TD3-TD6 mainly correspond to charge-transfer

CR IP T

excitation of the HOMO to the LUMO. The red shift of λmax from theory to experiment is associated to the self-interaction error in TD-DFT arising through the electron transfer in the extended charge-transfer state [37]. The absorption spectra of TD3 and TD4 are red-shifted in comparison with that of TD1 due to the expansion of the π conjugated system. The calculated lowest-energy absorption bands for the dyes

AN US

TD1-TD6 except TD5 on TiO2 are red-shifted compared with the isolated dyes. The red-shift of the calculated absorption spectra on TiO2 is previously reported [41,42]. The red-shift is due to the interaction between the acceptor group and the 3d orbital’s of the Ti atom, resulting in the decrease of

M

the LUMO energies compared with the isolated dyes.

ED

TD1

AC

TD3

CE

PT

TD2

15

ACCEPTED MANUSCRIPT TD4

TD5

HOMO

CR IP T

TD6

LUMO

HOMO

LUMO

AN US

Figure 8. Isodensity surface plots of frontier orbitals of isolated dyes and dyes linked to TiO2.

The absorption spectrum of the dye TD5 on TiO2 is blue-shifted compared with the isolated dye. The blue-shift absorption is perhaps due to the different coordination mode between the dye and TiO2,

M

leading to a smaller contribution of the 3d orbitals of the Ti atom to the LUMO of the dye.

E (eV) 2.73 2.74 2.59 2.28 2.68

1.06 1.10 0.79 0.64 1.09

λmax (nm) 461 456 502 547 447

455

H → L ( 70.4)

2.72

1.08

563

f

Dyes/TiO2 composition (%) H → L (70.3) H-2 → L (70.2) H → L (70.4) H → L (70.4) H → L (67.8) H → L+1 (17.1) H → L (69.8)

E (eV) 2.69 2.72 2.47 2.27 2.77

f 0.96 1.04 0.56 0.60 0.56

2.20

0.19

AC

TD6

Isolated dyes composition (%) H → L (70.4) H-2 → L (70.2) H → L (70.2) H → L (70.4) H → L (70.4)

PT

TD1 TD2 TD3 TD4 TD5

λmax (nm) 454 453 478 543 462

CE

Dyes

ED

Table 3 Computed maximum absorption wavelengths (eV) and oscillator strengths (f) of isolated dyes in CH2Cl2 and adsorbed on TiO2.

Figure 9 presents the calculated permanent dipole moments of the dyes in the ground state. It was

confirmed theoretically that a dye with a large dipole moment would lead to a large Voc [43]. The results show that TD5 has the largest dipole moment (9.51 D); therefore we deduce that dye should have the largest Voc. According to the experimental results, it turns out that the Voc of TD5 is the largest. Compared to TD2 and TD3, the dipole moment of TD4 is larger (8.48 D), thus TD4 has larger Voc than TD2 and TD3.

16

ACCEPTED MANUSCRIPT µGS = 8.12 D

µGS = 6.84 D

TD2

µGS = 9.51 D

µGS = 7.93 D

AN US

µGS = 8.48 D

TD3

CR IP T

TD1

µGS = 6.90 D

TD4

TD5

TD6

M

Figure 9. Calculated dipole moment of the dyes in the ground-state at B3LYP/6-31+G(d).

ED

4. Conclusions

In this paper, we have designed metal free organic dyes that contain donors with triphenylamine

PT

and tetrazole-based acceptors bridged by a methine fragment. The dyes (TD2, TD3 and TD4) contain the same acceptors (ethyl 2-(1H-tetrazol-5-yl) acetate) as anchor groups but differ in antenna units

CE

(phenothiazine, carbazole and diphenylamine). The dyes TD1, TD5 and TD6 contain different electron acceptors,

such

as

ethyl

2-(1H-tetrazol-5-yl)

acetate,

(2H-tetrazol-5-yl)

acrylonitrile

and

AC

1H-tetrazole-5-acetic acid as anchoring groups. The influence of different anchoring groups on the photophysical, electrochemical, and photovoltaic properties was studied. The results show that the introduction of diphenylamine units as antennas in the as synthesized dyes could improve photovoltaic performance compared with phenothiazine and carbazole units as antennas in DSSCs. (2H-tetrazol-5-yl) acrylonitrile also proved to be the best anchoring group in improving high cell conversion efficiency.

Acknowledgments We thank the Isfahan University of Technology for the financial support of this work. We

17

ACCEPTED MANUSCRIPT

acknowledge the computational support by the national high performance computing center of Iran (Rakhsh). In addition, authors would like to acknowledge to Professor Hassan Haddadzadeh for his help to recording CVs.

References [1] K. R. Justin Thomas,Y.-C. Hsu, J. T. Lin, K.-M Lee, K.-C. Ho, C.-H. Lai, Y.-M. Cheng, P.-T. Chou,

CR IP T

Chem. Mater. 20 (2008) 1830–1840.

[2] A. Baheti, P. Singh, C.-P. Lee, K. R. Justin Thomas, K.-C. Ho, J. Org. Chem. 76 (2011) 4910–4920. [3] S. S. Park, Y. S. Won, Y. C. Choi, J. H. Kim, Energy & Fuels 23 (2009) 3732–3736.

[4] M. Liang, W. Xu, F. Cai, P. Chen, B, Peng, J. Chen, Z. Li, J. Phys. Chem. C 111 (2007) 4465–4472.

AN US

[5] K. Hara, Z.-S. Wang, T. Sato, A. Furube, R. Katoh, H. Sugihara, Y. Dan-oh, C. Kasada, A. Shinpo, S. Suga, J. Phys. Chem. B 109 (2005) 15476–15482.

[6] Z. Wan, C. Jia, Y. Duan, L. Zhou, J. Zhang, Y. Lin, Y. Shi, RSC Adv. 2 (2012) 4507–4514. [7] Y. Wu, W. Zhu, Chem. Soc. Rev. 42 (2013) 2039–2058.

M

[8] M. Tsuji, A. Saeki, Y. Koizumi, N. Matsuyama, C.Vijayakumar, S. Seki, Adv. Funct. Mater. 24 (2014) 28–36.

ED

[9] Z. S.Wang, Y. Cui, K. Hara, Y. Dan-oh, C. Kasada, A. Shinpo, Adv. Mater.19 (2007) 1138–1141. [10] X. Zhu, H. Tsuji, A. Yella, A. Chauvin, M. Gratzel, E. Nakamura, Chem. Commun. 49 (2013)

PT

582–584.

[11] Y. Ooyama, T. Sato, Y. Harima, J. Ohshita, J. Mater. Chem. A 2 (2014) 3293–3296.

CE

[12] J. Massin, L. Ducasse, T. Toupance, C. Olivier, J. Phys. Chem. C 118 (2014) 10677–10685. [13] T. Harel, S. Rozen, J. Org. Chem. 75 (2010) 3141–3143.

AC

[14] T. Eicher, S. Hauptmann, The chemistry of heterocycles, Thieme: New York, 1995. [15] R.G. Giles, N.J. Lewis, P.W. Oxley, J. K. Quick, Tetrahedron Lett. 40 (1999) 6093–6094. [16] Z. P. Demko, K. B. Sharpless, J. Org. Chem. 66 (2001) 7945–7950. [17] V. Rama, K. Kanagaraj, K. Pitchumani, J. Org. Chem. 76 (2011) 9090–9095. [18] F. Himo, Z. P. Demko, L. Noodleman, J. Org. Chem. 68 (2003) 9076–9080. [19] Z. Wan, C. Jia,J. Zhanga, Y. Duan, Y. Lin, Y. Shi, J. Power Sources 199 (2012) 426–431. [20] G. Lai, X. R. Bu, J. Santos, E. A. Mintz, Synlett (1997) 1275-1276.

18

ACCEPTED MANUSCRIPT

[21] L. Emmanuvel, R. K. Shukla, A. Sudalai, S. Gurunath, S. Sivaram, Tetrahedron Lett. 47 (2006) 4793–4796. [22] N. B. McKeown, S. Badriya, M. Helliwell, M. Shkunov, J. Mater. Chem. 17(2007) 2088–2094. [23] R. Haiges, K. O. Christe, Inorg. Chem. 52 (2013) 7249–7260. [24] B. Peng, S. Yang, L. Li, F. Cheng, J. Chen, J. Chem. Phys. 132 (2010) 034305–034309.

[26] A. D. Becke, J. Chem. Phys. 98 (1993) 5648–5652.

CR IP T

[25] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 37 (1988) 785–789.

[27] E. Cancès, B. Mennucci, J. Tomasi, J. Chem. Phys. 107 (1997) 3032–3041. [28] V. Barone, M. Cossi, J. Tomasi, J. Comput. Chem. 19 (1998) 404–417.

[29] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,M. A. Robb, J. R. Cheeseman, G.

AN US

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.

M

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.

ED

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,

PT

S. Dapprich, A. D. Daniels,O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09, Revision A.01, Gaussian Inc., Wallingford, CT, 2009.

CE

[30] D. Sun, S. V. Rosokha, J. K. Kochi, J. Am. Chem. Soc. 126 (2004) 1388–1401. [31] T. Kitamura, M. Ikeda, K. Shigaki, T. Inoue, N. A. Anderson, X. Ai, T. Lian, S. Yanagida, Chem.

AC

Mater. 16 (2004) 1806–1812. [32] G. Zhou, N. Pschirer, J. C. Schöneboom, F. Eickemeyer, M. Baumgarten, K. Müllen, Chem. Mater. 20 (2008) 1808–1815.

[33] S.-L. Li, K.-J. Jiang, K.-F. Shao, L.-M. Yang, Chem. Commun. (2006) 2792–2794. [34] Z.-S. Wang, K. Hara, Y. Dan-oh, C. Kasada, A. Shinpo, S. Suga, H. Arakawa, H. Sugihara, J. Phys. Chem. B 109 (2005) 3907–3914. [35] T.-H. Kwon, V. Armel, A. Nattestad, D. R. MacFarlane, U. Bach, S. J. Lind, K. C. Gordon, W. Tang, D. J. Jones, A. B. Holmes, J. Org. Chem. 76 (2011) 4088–4093.

19

ACCEPTED MANUSCRIPT

[36] A. J. Bard, L. A. Faulkner, Electrochemical Methods, Fundamentals and Applications, Wiley: New York, 1984. [37] Z. Tian, M. Huang , B. Zhao, H. Huang , X. Feng, Y. Nie, P. Shen, S. Tan, Dyes Pigm 87 (2010) 181–187. [38] W. Xu, B. Peng, J. Chen, M. Liang. F. Cai, J. Phys. Chem. C 112 (2008) 874–880. [39] Z. Huang, G. Natu, Z. Ji, M. He, M. Yu, Y. Wu, J. Phys. Chem. C 116 (2012) 26239–26246.

CR IP T

[40] J. Zhang, H.-B. Li, S.-L. Sun, Y. Geng, Y. Wu, Z.-M. Su, J. Mater. Chem. 22 (2012) 568–576. [41] X. L. Zhang, Z. Zhang, F. Huang, P. Bäuerle, U. Bach, Y.-B. Cheng, J. Mater. Chem. 22 (2012) 7005–7009.

[42] C.-L. Wang, J. Wang, F.-Q. Bai, J. Chen, H.-X. Zhang, Int. J. Quantum Chem 114 (2014) 560–567.

AN US

[43] P. Chen, J. H. Yum, F. De Angelis, E. Mosconi, S. Fantacci, S.-J. Moon, R. Humphr-Baker, J. Ko, M. K. Nazeeruddin, M. Grätzel, Nano Lett. 9 (2009) 2487–2492.

Graphical Abstract

AC

CE

PT

ED

M

New tetrazole based organic dyes were synthesized and applied in dye sensitized solar cells. The effect of different antenna groups and different tetrazole-based anchoring groups on the performance of solar cells was studied.

20