Design and synthesis of new Ru-complexes as potential photo-sensitizers: experimental and TD-DFT insights

Abstract

We report density functional theory (DFT) and time-dependent density functional theory (TDDFT) calculations on a novel organic ligand and a novel class of ruthenium complexes; cis-RuL₂X₂ with L = 2,2′-bipyridine-6,6′-bis ethyl ester phosphonate and phosphonic acid, X = Cl, CN or NCS. The calculations show that cis-configurations are more stable than the trans-counterparts. The DFT results have been used to help design such novel complexes for potential use as sensitizers. We demonstrate the opportunity to synthesize such complexes with high purity. The synthesis of these complexes relies on the preparation of the key intermediates cis-Ru(2,2′-bipyridine-6,6′-bisdiethyl ester phosphonate)Cl₂. These complexes were characterized by ¹H, ¹³C, and ³¹P NMR, elemental analysis and FTIR spectroscopy. The NCS complex shows the smallest optical band gap followed by the Cl and CN complexes, respectively, with the highest performance upon use as a sensitizer in dye-sensitized solar cells.

---

1. Introduction

With the looming energy crisis and the depletion of petroleum and natural-gas reserves, various ways have been investigated to convert sunlight directly into chemical fuel or electricity.¹ However, the capital cost of such devices is still inconvenient for large-scale implementation. To this end, dye-sensitized solar cells (DSSCs) have attracted considerable attention due to their high quantum efficiency giving the opportunity of low-cost conversion of solar energy into electrical power.² The ongoing research into DSSCs is mainly focused on the optimization of all cell components; namely, the photoactive material, the dye and the redox electrolyte.³

The development of optically active sensitizers over a wide region of the solar spectrum is a hot topic nowadays. In particular, ruthenium (Ru)-based complexes have been extensively used as sensitizers.^4,5 Ruthenium metal has been particularly investigated for a number of reasons:⁶ (1) it has octahedral geometry that enable the introduction of specific ligands in a controlled manner; (2) it forms very inert bonds with imine nitrogen centers; (3) it possesses various stable and accessible oxidation states. However, the main problem limiting the further development of sensitizers is the fact that dyes with high absorption coefficients has narrow bands and vice versa. Consequently, the optimization of sensitizers is mainly based on “guess-and-check” procedures. A more systematic approach is required in order to find optimum sensitizers with the required specifications. In this regard, density function theory (DFT) calculations are considered ideal to solve the “guess-and-check” problem involved in the design of DSSCs as well as solar hydrogen production systems. Specifically, it can be used to study the changes in the optical and electronic structures of the sensitizers.

Herein, we report detailed DFT calculations on new cis-ruthenium complexes with coordinated bipyridine bisphosphonate ligands. Our DFT findings were confirmed via the preparation and characterization of such sensitizers and their use in dye-sensitized solar cell devices.

2. Computational and experimental methods and materials

2.1 Density functional theory and time-dependent density functional theory

The molecular geometry for all complexes and ligands was fully optimized using 6-31+G mixed basis sets for C, H, N, O, P, Cl and S atoms and effective core potential LANL2/LANL2DZ basis set^7–10 for Ru atom with the density functional (DFT) level. The DFT calculations were carried out using the hybrid three-parameter density functional method abbreviated as B3LYP, which includes Becke's 3-parameter gradient exchange correction function (B3)¹¹ and the Lee, Yang and Parr correlation functional.^12,13 The B3LYP method typically provides energetic better than Hartree–Fock method¹⁴ and can reproduce better geometrical parameters comparable to the experimental values than any other method.¹⁵ No symmetry constrains were implemented during all the geometry optimization procedures. All possible configurations for the complexes were calculated and the one shown here is the lowest in energy or the global minimum for each complex. All the geometry optimization, vibrational frequency calculations were done using the Gaussian 09 software package.¹⁶ The optimized structures were visualized using Chemcraft version 1.6 packages.¹⁷ The bond dissociation energy of A–B bond is calculated by subtracting the total energy of the optimized geometry for AB molecule from single point energy of its constituents A and B at the same geometry, which is more accurate than substring it from optimized energy of isolated molecules as this would include some part of possible conformation relaxation energy changes. Ionization potential and electron affinity of cation and anion total energy subtracted from neutral total energy at optimized geometry without further optimization as electron movements are assumed to be faster than nuclear ones. Time-dependent DFT calculations were done on optimized geometry using number of excited states equal to 10. Exciton binding energies were estimated by the difference between LUMO and HOMO in DFT and in TD-DFT calculation. The electronic absorption spectra were created using convolution with Gaussian function using Gauss View Software¹⁸ using UV-Vis peak Half-Width at half height of 0.1 eV.

2.2 Experimental materials and synthesis

All start materials are high grade chemicals and solvents that were purchased from Sigma Aldrich. The solvents were distilled before use. ¹H, ¹³C NMR and ³¹P NMR spectra were recorded using 5 mm tube on a Bruker AC-250 (250.133 and 62.896 MHz, respectively) or Varian Gemini 2000 (199.976 and 50.289 MHz, respectively) and were referenced to tetramethylsilane (TMS) and triphenylphosphine.

Synthesis and characterization of 2,2′-bipyridine N,N′-dioxide. 2,2′-Bipyridine (4 g, 25.61 mmol) was dissolved in 50 ml chloroform. mCPBA m-chloroperbenzoic acid (22.1 g, 64.03 mmol) dissolved in 200 ml chloroform was added slowly (4 h) to this solution at zero temperature. After the completion of addition, the solution was stirred for another 2 days. The reaction was filtered off and quenched by the addition of methanol for one day. The precipitation filtered and dried at room temperature; (yield 92%).

¹H NMR (200 MHz, D₂O, ppm): 8.31–8.27 (d, 2H), 7.67–7.54 (m, 6H);

¹³C NMR (100 MHz, D₂O, ppm) 139.69, 137.60, 129.23, 126.75, 126.31.

Synthesis and characterization of 6,6′-dicyano-2,2′-bipyridine. 2,2′-Bipyridine-N,N′-dioxide (6 g, 31.88 mmol) and potassium cyanide (12.3 g, 189 mmol) were dissolved in 100 ml water. Methylene chloride 40 ml and benzoyl chloride (15.69 g, 111.6 mmol) were added slowly upon three hours at zero temperature. After completion of addition, the mixture was stirred at 0 °C for about four hours. The solution was filtered and quenched by ethanol for one day. The precipitation filtered and dried at room temperature to yield 75%.

¹H NMR (200 MHz, CDCl₃, ppm): 8.74–8.69 (dd, 2H), 8.06–7.98 (t, 2H), 7.79–7.75 (dd, 2H); ¹³C NMR (100 MHz, CDCl₃, ppm) 153.83, 138.78, 131.48, 129.01, 123.88, 116.24; EA found: C, 69.95; H, 2.83; N, 27.39%. Theoretical: 69.90; H, 2.93; N, 27.17%.

Synthesis and characterization of 6,6′-bis(ethoxycarbonyl)-2,2′-bipyridine. 6,6′-Dicyano-2,2′-bipyridine (5 g, 24.25 mmol) was dissolved in 80 ml ethanol and 34 ml of sulfuric acid was added slowly to solution. The reaction mixture was heated to reflux for one day. The solution poured over 100 g of ice and stirred for 2 h, then extracted with methylene chloride. The solution was washed with brine three times. After that the solution dried over magnesium sulfate anhydrous and reduced to yellow white solid (85%).

¹H NMR (200 MHz, CDCl₃, ppm) 8.79–8.74 (dd, 2H), 8.16–8.12 (dd, 2H), 8.02–7.94 (t, 2H), 4.55–4.44 (q, 4H), 1.50–1.43 (t, 6H);

¹³C NMR (100 MHz, CDCl₃, ppm) 163.31, 153.63, 146.01, 136.18, 123.57, 122.84, 60.04 ppm.

EA found, C, 64.22; H, 5.37; N, 9.54%. Theoretical; C, 63.99; H, 5.37; N, 9.33%.

Synthesis and characterization of 6,6′-bis(hydroxymethyl)-2,2′-bipyrine. 6,6′-Di-functionalized bipyridine was prepared from 6,6′-dibromo-2,2′-bipyridine via lithium–bromide exchange, followed by addition of suitable electrophiles such as N,N-dimethyl formamide, to give dialdehyde.¹⁹ The 6,6′-bis(hydroxymethyl)-2,2′-bipyrine was prepared following the procedure reported in literature.²⁰

¹H NMR (400 MHz, CDCl₃): 8.63–8.31 (d, 2H), 7.86–7.79 (t, 2H), 7.27–7.24 (d, 2H), 4.83 (s, 4H), 3.98 (bs, 2H) ppm;

¹³C NMR (100 MHz, CDCl₃): 160.14, 150.23, 138.69, 120.90, 119.04, 62.33 ppm;

Anal. calcd for C₁₂H₁₀N₂O₂, C, 66.65; H, 5.59; N, 12.96. Found; C, 66.46; H, 5.69; N, 13.18.

Synthesis and characterization of 6,6′-bis(bromomethyl)-2,2′-bipyrine. 6,6′-Bis(hydroxymethyl)-2,2′-bipyridine (6.39 mmol, 1.3 g) was dissolved in a mixture of 48% HBr (30 ml) aqueous solution and concentrated sulfuric acid (10.5 ml). The resulting solution was refluxed for 8 h and then allowed to cool to room temperature. Then, 75 ml of water was added. The pH was adjusted to neutral with NaOH solution. The resulting precipitate was filtered, and washed with water. The product was dissolved in chloroform (60 ml) and filtered. The solution was dried over anhydrous magnesium sulfate and evaporated under vacuum to dryness. Product was isolated as a white powder; (yield 88%).

¹H NMR (400 MHz, CDCl₃): δ 4.60 (4H, s, CH₂); 7.38 (2H, d, J = 5 Hz, aryl H on C5 and C5); 8.45 (2H, s, aryl H on C3 and C3); 8.68 (2H, d, J = 5 Hz, aryl H on C6 and C6) ppm.

¹³C NMR (100 MHz, CDCl₃): δ 165.20, 155.45, 137.91, 123.55, 120.50, 34.1 ppm.

Anal. calcd for C₁₂H₁₀N₂Br₂: C, 42.1; N, 8.18; H, 2.92. Found: C, 42.36; N, 7.86; H, 2.82.

Synthesis and characterization of 6,6′-bis(diethylmethyl phosphonate)-2,2′-bipyridine. A chloroform (15 ml) solution of 6,6′-bis(bromomethyl)-2,2′-bipyrine (0.3 g, 8.77 mmol) and 2.91 g (17.7 mmol) of triethyl phosphate was refluxed for 5 h under nitrogen. The excess phosphate was removed under high vacuum, and then the crude product was obtained as yellow-white powder; yield 85%.

¹H NMR (400 MHz, CDCl₃): δ 1.30 (12H), 3.5 (4H, CH₂P), 4.10 (8H, OCH₂), 7.35–7.38 (2H), 8.34–8.37 (2H), 8.62 (2H) ppm.

¹³C NMR (100 MHz, CDCl₃): δ 155.58, 152.08 (d, J = 7.9 Hz), 137.27 (d, J = 2.2 Hz), 124.16 (d, J = 4.6 Hz), 119.03 (d, J = 3 Hz), 62.15 (d, J = 6.8 Hz), 37.33, 35.99 and 16.31 (d, J = 6 Hz) ppm.

Elemental anal. calcd for, C₂₀H₃₀N₂O₆P₂: C, 52.63; N, 6.14; H, 6.63. Found: C, 52.5; N, 5.93; H, 6.7.

Synthesis and characterization of Ru[(6,6′-(CH₂PO₃Et₂)₂bpy)]₂Cl₂. A solution of 6,6′-bis(diethylmethyl phosphonate)-2,2′-bipyridine (295 mg, 0.636 mmol) and LiCl (286 mg, 6.8 mmol), and Ru(DMSO)₄Cl₂ (205 mg, 0.423 mmol) in dry DMF (15 ml) was refluxed for 7 h under nitrogen in the dark at (160–170 °C). After the solution was cooled to room temperature, methylene chloride was added and the precipitate was filtered and washed with methylene chloride. Finally the precipitate was washed with diethyl esther (three times 10 ml). After the compound was dried in a vacuum, and then the crude product was obtained as brown-white (yield 85%). The complex was used as such in the next step.

¹H NMR (400 MHz, D₂O): δ 1.20 (t, J = 6.8 Hz, 12H, CH₃), 1.22 (t, J = 7.2 Hz, 12H, CH₃), 3.9 (m, J = 7.2 Hz, 8H, CH₂), 3.8 (m, J = 7.2 Hz, 8H, CH₂), 3.4 (d, J = 20.8 Hz, 8H, PCH₂), 7.96 (dd, J = 7.2 Hz, 4H and J = 2.4 Hz, 4H, H3, H4), 7.54 (dd, J = 1.6 Hz and, J = 5.2 Hz, 2H, Hs) and J = 2, J = 7.2, 2H, H5′ ppm.

¹³C NMR (100 MHz, D₂O): δ 155.625, 155.6, 155.4, 155.33, 138.561, 124.727, 124.689, (d, J = 3.8 Hz), 120.573 (d, J = 3 Hz), 61.527, 61.467 (d, J = 6 Hz, CH₂), 37.71, 37.05, 36.458 (t, CH₂P) and 16.082, 16.021 (d, J = 6.1 Hz) ppm.

³¹P NMR (400 MHz, D₂O): 28.167 ppm.

Elemental anal. calcd for (C₄₀H₆₀Cl₂N₄O₁₂P₄Ru), (C₃₂H₄₄Cl₂N₄O₁₂P₄Ru), theoretical; C, 39.5, H, 4.56, N, 5.67. Found, C, 41.0, H, 4.77, N, 6.1.

¹H NMR (400 MHz, CD₃OD): δ 1.15 (t, J = 7.2 Hz, 12H, CH₃), 1.13 (t, J = 6.8 Hz, 12H, CH₃), 3.8 (d, J = 7.2 Hz, 8H, CH₂), 3.86 (m, J = 7.2 Hz, 8H, CH₂), 3.3 (d, J = 23.2 Hz, 8H, PCH₂), 7.41 (dd, J = 8 Hz, 4H), 7.8 (dd, J = 7.6 Hz and J = 8 Hz, 4H), 8.09 (dd, J = 8 Hz, 4H) ppm.

¹³C NMR (100 MHz, CD₃OD): δ 158.45, (d, J = 9.1 Hz), 155.49, 139.799, 126.35, (d, J = 6.8 Hz), 119.73, 61.72 (d, J = 6.1 Hz, CH₂) 38.40, 37.12, 17.18, (d, J = 7.6 Hz) ppm.

Synthesis and characterization of Ru[(6,6′-(CH₂PO₃H₂)₂bpy)]₂Cl₂. A solution of Ru[(6,6′-(CH₂PO₃Et₂)₂bpy)]₂Cl₂ (2.18 g, 0.212 mmol) in 15 ml of 18% HCl was refluxed for 15 h. After that, the solvent was evaporated on a rotary evaporator. The resulting yellow-brown was dissolved in a minimal amount of water and evaporated. Then dry in high vacuum to obtain a title compound in 90% yield.

¹H NMR (400 MHz, D₂O): δ 2.71 (P–OH, 8H), 3.59 (d, J = 20.8 Hz, CH₂–P, 8H), 7.80 (dd, J = 68 Hz, 4H), 8.311 (dd, J = 68 Hz, 8H), notice: when we were using D₂O as a solvent, we found two peaks was overlap in one peak (H3, H4), and for methanol we found three peaks for pyridine ring.

¹H NMR (400 MHz, CD₃OD): δ 8.39 (dd, J = 7.6 Hz, 4H), 8.27 (4H) and 7.78 (dd, J = 8 Hz, 4H), 3.64 (d, J = 21.6 Hz, CH₂–P, 8H) and 2.63 (P–OH, 8H) ppm.

¹³C NMR (100 MHz, D₂O): δ 153.04, 146.718, 143.686, 128.176, 122.028, 36.292 and 35.041 ppm.

³¹P NMR (400 MHz, D₂O): two peaks, at 7.76 ppm and 25.718 ppm (reference PPh₃).

Elemental anal. calcd for (C₂₄H₂₈Cl₂N₄O₁₂P₄Ru·H₂O), theoretical C, 33.82, H, 3.44, N, 6.38. Found, C, 32.54, H, 3.29, N, 6.15.

Synthesis and characterization of Ru[(6,6′-(CH₂PO₃H₂)₂bpy)]₂(CN)₂. A solution of Ru[(6,6′-(CH₂PO₃H₂)₂bpy)]₂Cl₂, (50 mg, 0.058 mmol) and KCN (75.5 mg, 1.0 mmol) were dissolved in a mixture of water 10 ml and methanol 10 ml in a round-bottomed flask 150 ml. And then the solution was purged with nitrogen. The flask was then covered with aluminum foil and heated at reflux for 15 h in dark. After the solution cooled at room temperature, a grade acetone was added to the crude reaction mixture and the precipitate was filtered off and washed with acetone. The result of complexes is to yield 80% brown white. To purify this complex, dissolve in a minimum amount of water and purified by column chromatography using silica-gel (LH20) as stationary phase and water as eluent.

¹H NMR (400 MHz, CD₃OD): δ 2.15 (P–OH, 8H), 3.14 (d, J = 19.6Hz, CH₂–P, 8H), 7.31 (dd, J = 7.6 Hz, 4H), 7.81 (dd, J = 7.6 Hz and J = 8 Hz, 4H), 7.88 (dd, J = 8 Hz, 4H) ppm.

¹³C NMR (100 MHz, CD₃OD) δ 171.19 (CN), 158.6 (d, J = 6.9 Hz), 154.86, 138.19, 124.49 (d, J = 3.8), 119.55, 40.51, 39.33 ppm.

³¹P NMR (400 MHz, CD₃OD), one peak at 25.23 ppm ref. pph₃.

IR (KBr) (CN) at 2052 cm⁻¹.

Synthesis and characterization of [Ru[(6,6′-(CH₂PO₃H₂)₂bpy)]₂(NCS)₂. A solution of Ru[(6,6′-(CH₂PO₃H₂)₂bpy)]₂Cl₂, (50 mg, 0.058 mmol) and KSCN (22.25 mg, 2.32 mmol), were dissolved in a mixture of water 10 ml and methanol 10 ml in a round-bottomed flask 150 ml, and then the solution was purged with nitrogen. The flask was then covered with aluminum foil and heated at reflux for 15 h in the dark. After the solution cooled at room temperature, a grade acetone was added to the crude reaction mixture and the precipitate was filtered off and washed with acetone. The result of complexes is to yield 85% brown.

¹H NMR (400 MHz, CD₃OD): δ 8.38 (d, J = 8 Hz, 1H), 8.26 (t, J = 8 Hz, 1H), 7.78 (d, J = 8 Hz, 1H), 3.57 (d, J = 20.8 Hz, 2H) ppm.

¹H NMR (400 MHz, D₂O): δ 8.15 (broad peak, 2H), 7.67 (broad peak, 1H), 3.49 (d, J = 42 Hz, 2H) ppm.

¹³C NMR (400 MHz, D₂O): δ 154.0, 153.9, 146.95, 146.93, 143.3, 127.92, 127.87, 121.57, 36.79 and 35.58 ppm.

³¹P NMR (400 MHz, CD₃OD): one peak at 25.093 ppm ref. pph₃.

IR (KBr) (NCS) at 2113 cm⁻¹.

Fabrication and testing of the DSSCs. The DSSC devices were fabricated and tested as detailed in our previous work.²¹

3. Results and discussion

The optimized geometry of the cis- and trans-configurations of the synthetized sensitizers is shown in Fig. 1 and 2, respectively with their energetic parameters are summarized in Table 1. Our calculations showed that cis-configurations are more stable than the trans-counterparts. The P=O bond length was calculated to be 1.49 Å in all complexes. The Ru–N bond (connected to the organic part) length was found to be similar in all complexes with a range from 2.15 Å to 2.20 Å. Furthermore, the difference between the cis- and trans-configurations is less than 1%, indicating that the main change in bond lengths is around the Cl, CN and NCS ligands. In the cis-configurations, the Cl–Ru–Cl, NC–Ru–CN and SCN–Ru–NCS bond angles are 86.9°, 93.7° and 91.1°, respectively. On the other hand, in the trans-configurations, the Cl–Ru–Cl, NC–Ru–CN and SCN–Ru–NCS bond angles are 176.4°, 179.5° and 179.3°, respectively. In NCS complexes, the N–C–S bond angle is 177.8° and 179.5° for the cis- and trans-configurations, respectively. In case of the CN complexes, only hydrogen bonds are formed between the N of the CN group and H of phosphate group. Note that this interaction does not exist in the other two complexes. The calculations showed that two hydrogen bonds with average length of 1.77 Å are observed in the trans-configuration while only one hydrogen bond with 1.73 Å is formed in case of the cis-configuration due to geometrical orientation of the phosphate groups.

Fig. 1 The optimized geometry of cis-Ru complexes, bond lengths are shown in angstrom.

Fig. 2 The optimized geometry of trans-Ru complexes, bond lengths are shown in angstrom.

Table 1 The total energy E_T, difference in total energy ΔE_T in Hartree, E_LUMO, E_HOMO, ΔE in eV and dipole moment μ (Debye) for complexes

Configuration	trans			cis
Complex	Cl	NCS	CN	Cl	NCS	CN
ET, (a.u.)	−5062.31702835	−4495.03286616	−3698.63057783	−5062.3544477	−4495.04200812	−3698.63603973
ΔET (kcal mol^−1)	23.4810	5.7367	3.4274	0.0	0.0	0.0
ELUMO, eV	−2.358	−2.951	−2.957	−2.207	−2.9219	−2.856
EHOMO, eV	−5.065	−5.834	−6.446	−4.983	−5.3938	−5.897
ΔE, eV	2.706	2.883	3.489	2.776	2.472	3.041
μ, D	4.39	4.02	3.59	12.89	16.30	15.93

---

The Ru–Cl, Ru–CN and Ru–NCS bond lengths are 2.18 Å, 1.98 Å and 2.11 Å and 2.49 Å, 2.05 Å and 2.06 Å for cis- and trans-configurations, respectively. It is worth mentioning that the longer bond lengths in case of Ru–Cl may be the reason for the greater total energy difference ΔE_T between cis- and trans-configuration of this complex relative to the others. The energy of the highest occupied molecular orbital (E_HOMO) and the lowest unoccupied molecular orbital (E_LUMO) for the complexes under study are summarized in Table 1. The energy gap (ΔE) between E_LUMO and E_HOMO is inversely proportional to the reactivity of the complex. The reactivity as function of ΔE is in the order Cl > NSC > CN and NSC > Cl > CN for trans- and cis-configurations, respectively. The dipole moment of the cis-configuration is larger than that of the trans-counterpart. The bond dissociation energy between Ru metal and ligand has been calculated. The Ru–Cl, Ru–NCS and Ru–CN bond energies are 75.70, 96.19 and 118.89, and 97.63, 115.19 and 151.99 kcal mol⁻¹ for trans- and cis-configurations, respectively. These numbers clearly show that the strength of the inorganic ligands is in the order CN > NCS > Cl. It further explains the greater stability of cis- over trans-configuration by at least 20%. On the other hand, the average Ru–N bond (where N is in the organic ligand) dissociation energies do not show that greater variance where they are 27.11, 32.89 and 35.09, and 34.13, 35.97 and 35.78 kcal mol⁻¹ for Cl, NSC and CN complexes with trans- and cis-configurations, respectively. The NBO charges on Ru are 0.13, −.017 and −0.32 for NCS, Cl and CN complexes, respectively. The most positive charge is observed on P atom, which is almost 2.4 in all complexes. However, the most negative charge is observed on adjacent O atom in the P=O with a value of about −1.1 in all complexes.

The optical parameters have been theoretically studied for the cis-conformer only and the main characteristics are summarized in Table 2. The calculation of ionization potential (IP) and electron affinity (EA) was done by subtracting the total energy of the ionic complex from neutral one at the same geometry. It is worth mentioning that the IP and EA seem to be shifted by about 1 eV relative to the E_HOMO and E_LUMO. In addition, time-dependent density functional theory (TD-DFT) calculations were performed to estimate the optical electronic absorption spectra and exciton binding energies, with the detailed data represented in Table 3. The NCS complex shows the smallest optical band gap, the highest oscillator strength and the lowest exciton binging energy followed by the Cl and CN complexes, respectively, indicating a better performance for use as a sensitizer in solar cells application. The electronic absorption spectra are shown for the three complexes in Fig. 3. The CN and NCS have three bands at 385 nm, 455 nm and 505 nm, and 461 nm, 610 nm and 671 nm, respectively. On the other hand, the Cl complex spectra have almost one main band at about 530 nm and other bands seem to be overlapped underneath and appearing as shoulders. As a general trend, the HOMO and HOMO−1 orbitals are either π orbital on CN and NCS or p-orbital on the Cl along with d-orbital form Ru atom, see Fig. 4. The LUMO and LUMO+1 are π* orbital localized on one of the two bipyridine ring. The LUMO+2 and LUMO+4 are also π* orbital on both of the two bipyridine ring. Furthermore, all the calculated electronic absorption spectra involve transitions from d-orbital on metal along with orbitals on Cl, CN or NCS to orbitals on bipyridine rings.

Table 2 The ionization potential IE, electron affinity EA, energy of the first excited singlet state S₁, its oscillator strength and exciton binding energy for cis-complexes

cis complexes	Cl	NCS	CN
IP, eV	5.901	6.510	7.019
EA, eV	−1.126	−1.1792	−1.704
TD-DFT S1	1.9864	1.8205	2.2186
f	0.0075	0.0171	0.0017
Eexciton	0.7896	0.6515	0.8224

---

Table 3 The TDDFT results for singlet excited states with the highest oscillator strength

Compound	State no.	Main configuration	Coefficient	f	λ, nm
Cl	S1	H → L	0.61083	0.0075	624.2
	S3	H-1 → L	0.55092	0.0284	558.9
	S4	H-2 → L	0.57511	0.0521	530.2
CN	S1	H → L	0.68420	0.0017	558.8
	S4	H-2 → L	0.57914	0.0219	499.9
	S6	H-1 → L + 1	0.60695	0.0142	454.6
	S8	H → L + 4	0.42140	0.0184	387.9
NCS	S1	H → L	0.67866	0.0171	681.0
	S2	H → L + 1	0.63450	0.0161	618.3
	S8	H → L + 2	0.41559	0.0081	471.6
	S10	H → L + 2	0.41978	0.0131	461.5

---

Fig. 3 The theoretical electronic absorption spectra of the studied complexes.

Fig. 4 The HOMO and LUMO plots for (a) Cl, (b) NCS and (c) CN complexes.

Based on the promising DFT results, herein, we show the possibility to synthesize such complexes. However, we will limit our discussion to the cis-Ru complexes as an example. The investigated complexes are of the general formula cis-RuL₂X₂ with L = 2,2′-bipyridine-6,6′-bis(diethyl methyl phosphonate) and 2,2′-bipyridine-6,6′-bis(methylphosphoric acid). The preparation of this type of complexes usually involves two steps.^22,23 The first complex consists of refluxing two equivalent of the bidentate ligand L with 1 equivalent of ruthenium trichloride in DMF to yield a cis-RuL₂Cl₂. In the second step, the diethyl phosphate was hydrolysed to phosphonic acid, and then the chloride ligands are substituted by potassium thiocyanate or potassium cyanide. This is achieved by heating in a mixture of methanol or ethanol and water to prepare complex that contains cyanide and thiocyanate. The preparation of complexes with different positions (such as 4, and 5) has been previously patented by Grätzel et al.²⁴ However, no preparation and characterization of ligand and complexes were reported in the open literature. Herein, we report on the synthesis of new ligand and new complexes, which are shown in Scheme 1 and 2.

Scheme 1 Reaction and conditions: (i) P-CPBA, CHCl₃ at 0 °C 2 d (ii) KCN, H₂O, PhCOCl, DCM, 12 h (iii) EtOH, H₂SO₄, reflux, 12 h (iv) NaOH, EtOH, reflux 12 h (v) NaBH₄, DCM, r.t. 6 h (vi) HBr, H₂SO₄, reflux, 8 h (vii) P(OEt)₂, CHCl₃, reflux 5 h.

Scheme 2 Stepwise synthesis of the Ru-based complexes.

Starting from the phosphonated bipyridine as a ligand, these complexes were prepared using Ru(DMSO)₄Cl₂ and RuCl₃ in water to get highly pure complexes. This could be clearly seen from the ¹H NMR spectrum of the crude reaction mixture that indicated a lower amount of impurities when Ru(DMSO)₄Cl₂ was used. This is probably due to the use of a ruthenium(II) precursor with weekly binding ligands (DMSO). A purification step was necessary to remove tris-bipyridine ruthenium complex that was formed during the reaction. The resulting complexes were purified and crystallized using column chromatography. The diethyl ester phosphonate groups were fully hydrolyzed by heating the complexes in hydrochloric acid solution. Chloride ligand exchange with thiocyanate or cyanide was subsequently performed in water with an excess of monodenate ligands. These complexes were found to be soluble in methanol and water with a larger solubility at basic pH.

The NMR spectra of the complexes were measured in D₂O and CD₃OD solutions to achieve sufficiently large concentrations, compatible with fast spectrum recording and high signal-to-noise ratio. The ¹H NMR spectra of all complexes show six sharp and well-resolved signals in the aromatic region, corresponding to the six magnetically inequivalent protons of the bipyridine (Fig. 5). Using D₂O as a solvent, the downfield shifted protons resonance peaks can be assigned to the protons H₃ and H₄ that are merged in one peak. On the other hand, upon using CD₃OD as solvent, H₃ and H₄ peaks were found to be very close to each other. The assignment is based on the assumption that deshielding of the protons can be due to an induced magnetic field created by the ring current on bipyridine aromatic moieties. This deshielding is only significant at short distance and therefore affects only protons that are close to the bipyridine. In this case, we found that the complexes are cis-octahedral, with two equivalent bipyridine ligands. In complex (I) in Scheme 2, the two 6,6′-methylene phosphonate ethyl ester groups (CH₂PO₃Et) are inequivalent and they give rise to four multiplets. Therefore, the observed patterns are in agreement with the cis geometry of the complexes. The results were also confirmed via ¹³C and ³¹P NMR spectra, which were recorded in D₂O and CD₃OD as solvents, see the ESI.† The trans geometry would have resulted in simpler NMR due to the higher symmetry and would have yielded only three peaks in the aromatic region due to the four magnetically equivalent pyridine units.

Fig. 5 ¹H NMR spectra of complex I and II recorded in D₂O andCD₃OD.

Infrared (IR) spectra were recorded in KBr pellets, and the v_CN bands of complex (III) and (IV) were located, respectively at 2052 cm⁻¹ and 2113 cm⁻¹, see the ESI† for more details. It was demonstrated that the band appears around 2110 cm⁻¹ can be assigned to N-bound thiocyante.^25,26 Note that it is shifted to lower energy for the S-bound complex to 2052 cm⁻¹.

To validate the results, dye-sensitized solar cell devices were fabricated using the synthesized complexes as sensitizers to TiO₂ nanotubes under AM 1.5 illumination (Fig. 6 and Table 4), as detailed in our previous report. Note that the power conversion efficiency (PCE) increases in the order –SCN > –CN > –Cl, resulting in PCE of 1.91, 2.38 and 2.9%, respectively. Note that those efficiencies are at least twice that reported for the same complexes with the substituents in the 4- and 5-positions,²⁷ confirming the importance of the position of the functional group in determining the performance of the Ru-complexes.

Fig. 6 J–V characteristics of the solar cell devices measured under the irradiance of AM 1.5G full sunlight (100 mW cm⁻²) with a cell active area of 0.4 cm².

Table 4 Comparison between photovoltaic parameters of the DSSCs prepared using different complexes

Dye	Jsc [mA cm^−2]	Voc [V]	FF	η [%]
Cl	3.78	0.67	0.75	1.91
CN	5.47	0.57	0.76	2.38
SCN	6.27	0.60	0.77	2.9

---

4. Conclusion

DFT and TDDFT calculations were performed on cis-[Ru(bipyridine)₂(phosphonic acid)₂]X₂ where X = Cl, CN complexes. Due to geometrical orientation of the phosphate groups, two hydrogen bonds with average length of 1.77 Å were observed in the trans-configuration with only one hydrogen bond with 1.73 Å is formed in case of the cis-configuration. The order of reactivity as function of ΔE was found to be in the order Cl > NSC > CN and NSC > Cl > CN for trans- and cis-configurations, respectively. The dipole moment of the cis-configuration is larger than that of the trans-counterpart. The NCS complex shows the smallest optical band gap, the highest oscillator strength and the lowest exciton binging energy followed by the Cl and CN complexes, respectively, indicating a better performance for use as a sensitizer in solar cells application. The CN and NCS have three bands at 385 nm, 455 nm and 505 nm, and 461 nm, 610 nm and 671 nm, respectively. These complexes were found to be soluble in methanol and water with a larger solubility at basic pH. The cis-configuration of such novel ruthenium complexes was prepared and characterized by several spectroscopic methods. Upon their use as photosensitizers in DSSC devices, the power conversion efficiency (PCE) increases in the order –SCN > –CN > –Cl, resulting in PCE of 1.91, 2.38 and 2.9%, respectively. We hope that our work will open a new route toward the synthesis and use of optically active Ru-complexes based on phosphonate functional groups compared to carboxylic counterparts.

Acknowledgements

The Authors acknowledge the financial support from The Science and Technology Development Fund (STDF) provided by the Egyptian Government (Grant # 5415) and the IRD provided by the French Government (Grant # 005-2013).

References

1. B. O'Regan and M. Gratzel, Nature, 1991, 353, 737–739.
2. M. K. Nazeeruddin, A. Kay, I. Rodicio, R. Humphry-Baker, E. Müller, P. Liska, N. Vlachopoulos and M. Grätzel, J. Am. Chem. Soc., 1993, 115, 6382–6390.
3. R. Argazzi, C. A. Bignozzi, T. A. Heimer, F. N. Castellano and G. J. Meyer, Inorg. Chem., 1994, 33, 5741–5749.
4. P. Ghosh and T. G. Spiro, J. Am. Chem. Soc., 1980, 102, 5543–5549.
5. D. C. Bookbinder and M. S. Wrighton, J. Electrochem. Soc., 1983, 130, 1080–1086.
6. C. J. Miller, C. A. Widrig, D. H. Charych and M. Majda, J. Phys. Chem., 1988, 92, 1928–1933.
7. T. H. Dunning Jr and P. J. Hay, in Modern Theoretical Chemistry, ed. H. F. Schaefer III, Plenum, New York, 1976, vol. 3, pp. 1–28.
8. W. R. Wadt and P. J. Hay, Ab initio effective core potentials for molecular calculations - potentials for main group elements Na to Bi, J. Chem. Phys., 1985, 82, 284–298.
9. P. J. Hay and W. R. Wadt, Ab initio effective core potentials for molecular calculations - potentials for the transition-metal atoms Sc to Hg, J. Chem. Phys., 1985, 82, 270–283.
10. P. J. Hay and W. R. Wadt, Ab initio effective core potentials for molecular calculations - potentials for K to Au including the outermost core orbitals, J. Chem. Phys., 1985, 82, 299–310.
11. A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
12. C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785.
13. B. Miehlich, A. Savin, H. Stoll and H. Preuss, Chem. Phys. Lett., 1989, 157, 200.
14. A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
15. J. M. Martin, J. El-Yazal and J. Francois, Mol. Phys., 1995, 86, 1437.
16. Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
17. http://www.chemcraftprog.com.
18. Gauss View, Version 5, R. Dennington, T. Keith and J. Millam, Semichem Inc., Shawnee Mission, KS, 2009.
19. S. G. Chen, S. Chappel, Y. Diamant and A. Zaban, Chem. Mater., 2001, 13, 4629.
20. S. Sakaki, T. Kuroki and T. Hamada, J. Chem. Soc., Dalton Trans., 2002, 840.
21. W. Sharmoukh and N. K. Allam, ACS Appl. Mater. Interfaces, 2012, 4(8), 4413–4418.
22. E. Holder, G. Schoetz, V. Schurig and E. Lindner, Tetrahedron: Asymmetry, 2001, 12, 2289–2293.
23. G. R. Newkome, W. E. Puckett, G. E. Kiefer, V. K. Yuanjiao, M. Coreil and M. A. Hackney, J. Org. Chem., 1982, 47, 4116–4120.
24. M. K. Nazeeruddin, P. Liska, N. Vlachopoulos and M. Grätzel, J. Am. Chem. Soc., 1993, 115, 6382–6390.
25. R. Argazzi, C. A. Bignozzi, T. A. Heimer, F. N. Castellano and G. J. Meyer, Inorg. Chem., 1994, 33, 5741–5749.
26. M. Gratzel, O. Kohle, M. K. Nazeeruddin, P. Pechy, F. P. Rotzinger, S. Ruile and S. M. Zakeeruddin, PCT Int. Appl., WO 95 29, 924, Ecole Polytechnique Federale de Lausanne, Switz., Nov. 1995, p. 52.
27. K. Neuthe, F. Bittner, F. Stiemke, B. Ziem, J. Du, M. Zellner, M. Wark, T. Schubert and R. Haag, Dyes Pigm., 2014, 104, 24–33.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra16458g

---