Synthesis, structures and photophysical properties of luminescent cyanoruthenate(II) complexes with hydroxylated bipyridine and phenanthroline ligands

Abstract

Treatment of (PPh₄)₂[Ru^II(PPh₃)₂(CN)₄] (1) with bpyOH and phenOH in DMF afforded two mononuclear compounds fac-(PPh₄)[Ru^II(bpyOH)(PPh₃)(CN)₃] (2) and fac-(PPh₄)[Ru^II(phenOH)(PPh₃)(CN)₃] (3), respectively (bpyOH = 6-hydroxy-2,2′-bipyridine; phenOH = 2-hydroxy-1,10-phenanthroline). These complexes have been characterized by various spectroscopic techniques. The structures of 1 and 2 have also been determined by X-ray crystallography. Their photophysical and electrochemical properties have been investigated. Compound 3 displays intense emission with much higher quantum efficiency (Φ_em = 19.4%) in solution, compared with other related ruthenate(II) diimine complexes. In addition, their solvatochromism, pH effects, and ion perturbation have also been investigated. The different photoluminescent behaviors between these complexes and the previously reported complex [Ru^II(phen)(PPh₃)(CN)₃]⁻ suggest that the introduction of the hydroxyl group into the diimine ligand would significantly affect their emission properties. Through spectrophotometric titrations, the ground state pK_a and excited state pK_a* values, as well as the dynamic pH response ranges of these complexes have been determined. Their photophysical response towards various metal ions have also been studied.

---

Introduction

The utilization of luminescent transition metal complexes in the design of chemosensors and bioprobes has attracted considerable interest owing to their advantageous photophysical properties such as long emission lifetime, high photostability and large Stokes shifts.^1,2 Luminescent transition metal complexes have been extensively studied as potential sensors/probes for O₂ and CO₂,^3,4 pH,⁵ chloride,⁶ temperature,⁷ various anions and cations.⁸ In particular, Ru(II) complexes bearing various α,α′-diimine ligands have received much attention owing to their outstanding chemical and photochemical stability as well as their rich electrochemical and photophysical properties.⁹ [Ru^II(bpy)(CN)₄]²⁻, one of the best luminescent ruthenium(II) compounds was firstly reported in 1986.¹⁰ Its photophysical properties have been thoroughly investigated in aqueous media and in acidic solutions.¹¹ This complex shows highly environmentally sensitive metal-to-ligand charge-transfer (MLCT) absorption and emission due to the interaction between the cyanide ligands and the surrounding microenvironment including solvent molecules, protons and metal ions, this property has been utilized in the development of luminescent chemosensors, probes, and stimuli-responsive devices.¹² However, the relatively low luminescent quantum yields of the tetracyanoruthenate(II) complexes limit their further development in various applications.¹³ Recently, we have prepared a series of derivatives of cyanoruthenate(II) diimine complexes [Ru^II(N^N)(PR₃)(CN)₃]⁻, with improved luminescent properties through the substitution of a cyanide ligand by a better π-accepting phosphine ligand.¹⁴ Similar to [Ru^II(N^N)(CN)₄]²⁻, remarkable solvatochromic effect was observed in the absorption and emission spectra of [Ru^II(N^N)(PPh₃)(CN)₃]⁻.¹⁴

Herein, we attempt to modify the photophysical properties of the [Ru^II(N^N)(PPh₃)(CN)₃]⁻ chromophore by introducing a hydroxyl group to the α-position of the diimine ligand. We anticipate that the electronic and H-bonding effects of the hydroxyl group would have a significant effect on the properties of these complexes. The keto–enol tautomerism of hydroxyl-substituted diimine ligands could also be utilized in the development of pH/metal ion sensors.¹⁵ In this context, two new tricyanoruthenate(II) complexes with hydroxyl-substituted diimine ligands have been prepared. The photophysical and electrochemical properties of these complexes have been investigated. In addition, the effects of the microenvironment on the photophysical properties such as solvatochromism, pH effects as well as cationic effect on the photophysical properties of these complexes have also been studied.

Results and discussion

Synthesis and characterization

Reaction of [Ru^II(PPh₃)₃Cl₂] with excess KCN in MeOH afforded the complex trans-[Ru^II(PPh₃)₂(CN)₄]²⁻, which was isolated as PPh₄ salt (1) with high yield. The IR spectrum of 1 shows the characteristic ν(C≡N) stretching bands at 2059 and 2100 cm⁻¹. The electrospray ionization mass spectrum (ESI/MS) of 1 in MeOH (−ve mode) shows two predominant peaks at m/z 365.0 and 1069.0, which are assigned to [M]²⁻ and [M + PPh₄]⁻, respectively. Treatment of 1 with bpyOH or phenOH in DMF afforded fac-(PPh₄)[Ru^II(bpyOH)(CN)₃(PPh₃)] (2) or fac-(PPh₄)[Ru^II(phenOH)(CN)₃(PPh₃)] (3), respectively (Scheme 1). These complexes have been characterized by ¹H NMR, ³¹P{¹H} NMR, IR spectroscopy, mass spectrometry and elemental analysis. The crystal structures of 1 and 2 have also been determined by X-ray crystallography.

Scheme 1 Synthetic route to 2 and 3.

In the IR spectra, 2 and 3 are characterized by strong ν(C≡N) stretches at 2093, 2057, 2045 cm⁻¹ and 2089, 2082, 2075 cm⁻¹, respectively. The ESI-mass spectra of 2 and 3 in methanol (−ve mode) show a predominant parent peak [M]⁻ at m/z 614 and 638, respectively. In addition, 2 and 3 show ¹H NMR and ³¹P{¹H} NMR signals with chemical shifts and integral ratios consistent with their predicted structures. The NMR signals for phenyl protons in PPh₃ and PPh₄⁺ and aromatic protons in diimine ligands are comparable with those in the other related complexes.¹⁶ The singlet at δ 10.15 and 10.69 ppm are observed for 2 and 3, respectively which are ascribed to the exchangeable OH proton in the diimine ligands. This is further confirmed by the observation of disappearance of ¹H NMR signal with chemical shift > 10 ppm when few drops of D₂O were added. In their ³¹P NMR spectra, two singlet at δ 41.02 and 22.32 ppm in 2 and δ 41.84 and 22.32 ppm in 3 are observed, which are ascribed to the phosphorus nuclei in PPh₃ and PPh₄⁺, respectively.

X-ray crystal structures determination

The crystal structures of 1 and 2 have been determined by X-ray crystallography. The crystal data, data collection and structure refinement details are summarized in Table S1† and selected bond parameters about these two compounds are listed in Table S2.† As shown in Fig. 1, the ruthenium center of 1 adopts a distorted octahedral geometry and is coordinated by four cyanides in equatorial positions and two triphenylphosphine ligands in the axial positions. The average Ru–C and Ru–P bonds are 2.047(2) and 2.334(6) Å, respectively.

Fig. 1 The ORTEP drawing of the anionic structure of 1.

The ORTEP drawing of 2 is shown in Fig. 2a. The asymmetric unit of 2 contains two octahedral [Ru^II(bpyOH)(PPh₃)(CN)₃]⁻ anions whose charge are balanced by two PPh₄⁺ cations. The ruthenium center adopts a distorted octahedral geometry and is coordinated by two N atoms from a bpyOH ligand and two C atoms from two cyanides in the equatorial plane. The axial positions are occupied by a triphenylphosphine and a cyanide ligand. The Ru–P bond distance is 2.365(4) Å, which is comparable with those reported for related compounds.¹⁶ The bond length of Ru–CN trans to phosphine is 2.054(14) Å, which is longer than the two cis-Ru–CN bonds [2.001(15), 1.979(15) Å] in the equatorial positions. This is attributable to the stronger trans influence and π-accepting ability of the phosphine ligands compared with the diimine ligands. The Ru–N bond lengths are in the range of 2.116(12) to 2.139(12) Å, which are consistent with those observed in related Ru(II) complexes with bipyridine ligands.¹⁷ The hydroxyl group in bpyOH ligand shows a high degree of disorder, which is probably due to the presence of a pseudo-symmetry plane passing through the principle axis. The calculated structure of 2 is shown in Fig. 2b. The calculated structure parameters of 2 are compared with the X-ray crystallography data in Table S2.† The results obtained from the DFT B3LYP method with LANL2DZ/6-31G (d) basis set correspond well with the experimental data, as reflected from the deviation of bond lengths between the calculated and experimental data of ≤0.05 Å.

Fig. 2 (a) The ORTEP drawing and (b) optimized structure of anionic structure of 2.

Electrochemical properties

The electrochemical properties of these cyanoruthenate(II) complexes have been studied by cyclic voltammetry (CV) in deoxygenated MeCN solution (0.1 M ⁿBuNPF₆). The CV of 1 shows a reversible redox couple centered at E_1/2 = 0.08 V (ΔE_p = 84 mV, i_pc/i_pa = 0.9) (versus Cp₂Fe^+/0, Cp = cyclopentadienyl), which is tentatively assigned to the metal centered Ru^3+/2+ couple. In addition, 2 and 3 show an reversible metal-centered Ru^3+/2+ oxidation couple at E_1/2 = 0.45 (ΔE_p = 80 mV, i_pc/i_pa = 0.9) and 0.41 V (ΔE_p = 76 mV, i_pc/i_pa = 0.9) and an irreversible ligand-based reduction couple at E_pc = −2.50 and −2.54 V.

UV/vis absorption and emission properties

The UV/vis absorption properties of all complexes have been investigated (Fig. 3, Table 1). In CH₃CN solution, 2 and 3 exhibit intense absorptions in the range of 250–350 nm with molar extinction coefficients on the order 10⁴ dm³ mol⁻¹ cm⁻¹, which are ascribed to ligand-centered (LC) ππ* absorptions of phosphine and diimine ligands. In the visible region, both complexes show moderately intense absorptions with molar absorptivities on the order of 10³ dm³ mol⁻¹ cm⁻¹, which is not observed in 1. According to the previous spectroscopic studies of related complexes,^12,14 the lowest-energy absorptions of these complexes are tentatively assigned to ¹MLCT [dπ(Ru) → π*(N^N)] transition. In order to gain further insights into the electronic structure of 2, DFT calculations have been performed. As shown in the contour plots of the calculated frontier molecular orbitals (MOs) of 2, the HOMO of 2 shows a mixing of dπ(Ru) and π(CN) orbitals (Fig. 4a), while the LUMO is mainly composed of π*(bpyOH) orbital (Fig. 4b). This result is consistent with the ¹MLCT assignment for the lowest-energy transition and is suggestive of a mixing of LLCT character in the lowest-energy transition for 2 and 3.

Fig. 3 Overlaid UV/Vis spectra for 1–3 in CH₃CN solution.

Table 1 UV/Vis absorption data of 2 and 3 in various solvent media

Complex	Medium	Absorption λabs/nm (ε/M^−1 cm^−1)
2	MeOH	269 (11 890), 276 (10 970), 332 (5770), 406 (1850)
	EtOH	268 (5800), 276 (5470), 307 (5800), 414 (1010)
	CH2Cl2	269 (11 670), 276 (10 190), 314 (10 720), 441 (1670)
	DMSO	316 (46 260), 447 (7330)
	CH3CN	260 sh (9790), 268 (9740), 276 (9540), 307 (12 680), 330 sh (5820), 433 (2200)
	DMF	315 (54 260), 446 (10 300)
3	MeOH	268 (20 080), 276 (17 620), 415 (3030)
	EtOH	268 (14 780), 275 (12 950), 414 (2010)
	CH2Cl2	269 (35 350), 447 (4570)
	DMSO	270 (35 000), 381 sh (3880), 446 (4560)
	CH3CN	268 (29 020), 274 sh (26 170), 295 sh (10 290), 383 (3000), 433 (3580)
	DMF	268 (28 740), 381 (3070), 446 (3640)

---

Fig. 4 (a) Highest occupied and (b) lowest unoccupied molecular orbitals from energy-minimized calculated (Gaussian) anionic structure of 2.

Upon optical excitation, 2 and 3 exhibit solvent-dependent greenish yellow to orange emission in various degassed solutions at 298 K with sub-microsecond emission lifetimes (Fig. 5a and Table 2). According to previous spectroscopic studies of related ruthenium(II) diimine complexes,^11,13 these emissions are ascribed to the ³MLCT [dπ(Ru) → π*(N^N)] excited state origin. The quantum yield of 2 increases from acetone (1.5%), CH₂Cl₂ (3.2%) to MeOH (5.8%), while 3 exhibits a different trend in these solvents [acetone (6.0%), MeOH (7.7%), and CH₂Cl₂ (19.4%)]. It is notable that 3 displays the highest emission quantum yield in CH₂Cl₂ solution; moreover, the quantum yield of 3 is much higher than those reported for other related cyanoruthenate(II) diimine complexes.¹⁴ The much higher quantum efficiency of 3 is probably due to two important factors: (1) a strong π-accepting ligand PPh₃ replacing the CN ligand that will stabilize the dπ(Ru) orbital; (2) a strong electron-donating −OH group on the diimine ligand that will significantly raise the π* level of the diimine ligand, resulting in the remarkable increase in the MLCT emission energy as compared with related compounds.¹⁴ Both of these factors will subsequently raise energy of the ³MLCT emission state, leading to the higher quantum efficiency through effective impediment of the thermal deactivation pathway.

Fig. 5 (a) Overlaid emission spectra of 2 and 3 in CH₂Cl₂ solution at 298 K and (b) in EtOH/MeOH (4 : 1 v/v) glassy medium at 77 K.

Table 2 Emission data of 2 and 3 in various solvent media

	Medium (T/K)	Emission λem/nm (τo/μs)^a	ϕem × 10^3^b
a Excitation at 400 nm.b Luminescence quantum yield on excitation at 436 nm.c In EtOH/MeOH (4 : 1, v/v).
2	CH2Cl2 (298)	610 (0.50)	31.5
	MeOH (298)	578 (0.58)	58.2
	Acetone (298)	629 (0.23)	15.4
	Glass^c (77)	535, 511 (11.17)
3	CH2Cl2 (298)	596 (5.72)	194.2
	MeOH (298)	569 (3.85)	77.2
	Acetone (298)	617 (1.63)	60.5
	Glass^c (77)	546, 521 (56.90)

---

In EtOH/MeOH (4 : 1 v/v) glassy medium at 77 K (Fig. 5b, Table 2), the emissions of these complexes are hypsochromically shifted with respect to those recorded at room temperature. 2 and 3 show broad emission with peak maximum at ca. 511 nm (τ = 11.17 μs) and 521 nm (τ = 56.90 μs), respectively. These luminescence profiles feature the characteristic vibrational progressions with spacing ∼870 cm⁻¹, which are similar to those reported in other [Ru(bpy)₃]²⁺-type emitter.¹⁸ In view of their structured emission and the longer excited state lifetime, these emissions are derived from the ³MLCT [dπ(Ru) → π*(N^N)] excited state origin probably with some mixing of ³LC character. This is due to the rigidochromic effect, which is commonly observed in MLCT emitters.¹⁹

Solvent effect

Similar to the reported cyanoruthenate(II) diimine complexes,^10,13 2 and 3 also display strong solvatochromism, owing to their solvent-sensitive MLCT transition and the interactions between the solvent molecules and −CN/−OH moieties of the complexes (Table 1, Fig. 6).¹² In the absorption spectra of 2 in different solvents, the lowest-energy MLCT absorption shifts from acetone (456 nm) to MeOH (406 nm) with the energy difference of ∼2700 cm⁻¹ (Fig. 6a). It is noteworthy that the solvatochromic shift of MLCT transitions of 2 in acetone, DMF, CH₂Cl₂ and MeOH is in line with the Gutmann acceptor number of solvents. The absorption peaks shift to the lower energy region with decreasing acceptor numbers of the solvent molecules. However, deviation of the absorptions in MeCN and DMSO from the correlation is clearly observed (Fig. 7a). This is attributable to the presence of hydrogen-bonding between these solvent molecules and −OH group in bpyOH. In the bpyOH ligand, the sp² hybridized hydroxyl group has a lone pair electron and a δ⁺ H atom, which could act as H-bonding acceptor (A) or H-bonding donor (D) with respect to the solvent molecules. These two contrasting interactions could result in two opposite effects on the π* orbit of the diimine ligand: interaction of the lone pair with solvent will stabilize the π* orbits of the ligand; whereas H-bonding interaction via O–H⋯A_(solvent) will raise the π* level. In MeCN and DMSO, the hydroxyl group effectively acts as a D that will raise the energy of π*(bpyOH) orbitals, which will lead to a blue-shift in absorption energies from those expected from their acceptor number. However, the hydroxyl moiety simultaneously acts as D and A in protic MeOH/EtOH, hence, these contrasting interactions may cancel out each other. As a result, the solvatochromic effects in MeOH/EtOH are in agreement with their Gutmann acceptor numbers.

Fig. 6 Overlaid UV/vis spectra of 2 in various solvent media.

Fig. 7 (a) The plot of the lowest MLCT absorption energy (E_abs) and (b) the plot of emission energy (E_em) of 2 versus the acceptor number of the solvents.

Similar to the trend observed in the absorption spectra, the emission of 2 is also solvent-dependent, the-peak shifts from acetone (628 nm) to MeOH (573 nm) (Table 2 and Fig. 8) with the energy difference of about 1520 cm⁻¹, which is much less pronounced than that in the absorption spectra. This may be due to the reduced electron density in the cyanide ligands in the MLCT excited state, resulting in lower basicity of the ligand. On the other hand, the increased basicity on the excited state, compared to the ground state, is typical for complexes with a pendant basic site. The MLCT state of 2 obviously involves bpyOH ligand rather than PPh₃ or CN⁻ ligands, therefore the pendant −OH site of bpyOH has an increased affinity for protons. The net effect is that the hydroxyl group in bpyOH acts as a better A than D. This assumption is corroborated by the emission energies of 2 in MeCN and DMSO, which are essentially in line with their respective acceptor numbers, since the increased basicity of hydroxyl group will reduce its interaction with these H-bonding acceptor solvents. Moreover, the H-bonding effect between hydroxyl group and the protic solvents MeOH/EtOH in the ground state is also disequilibrated. The deviation of the ³MLCT emission of 2 in MeOH/EtOH also arises from the hydroxyl group acting as a better A than D in its excited state (Fig. 7b).

Fig. 8 Overlaid emission spectra of 2 in various solvent media.

Similar to the trend observed in the absorption and emission spectra of 2 in various solvent media, 3 shows the lowest-energy absorptions from acetone (445 nm) to MeOH (408 nm) with the energy difference ∼2030 cm⁻¹ (Fig. S5†) and the emissions from acetone (616 nm) to MeOH (568 nm) with the energy difference ∼1370 cm⁻¹ (Fig. S6†). Compared with the energy difference of 2 and 3 in various solvent media, both absorption and emission properties of 3 are found to be less sensitive to change of solvent media. The solvatochromism of [Ru^II(phen)(PPh₃)(CN)₃]⁻ has been investigated in different solvents¹⁴ and such a variation trend is not observed in this system, which suggest that the introduction of hydroxyl group into the diimine ligand would significantly affect their emission properties.

pH effect

Previous studies show that the photophysical properties of cyanoruthenate complexes are sensitive to the pH of the microenvironment.^10,12 Upon protonation of cyanide ligands, the MLCT [dπ(Ru) → π*(N^N)] energy would increase due to the better stabilization of dπ(Ru) orbitals in the presence of better π-accepting CNH ligands.²⁰ As a result, blue-shift of the lowest-energy absorption and emission would be expected in acidic medium. In contrast, protonation at the −OH group of the diimine ligand will produce two opposite effects on the ¹MLCT energy, as the effect of −OH group protonation on the complexes is two-fold: (1) withdrawal of the electron density from the diimine ligand will reduces its π-donating character and will stabilize the dπ(Ru) orbit, and (2) substantial stabilization of the diimine ligand LUMO. The net effects of these two processes produces an overall decrease in the MLCT [dπ(Ru) → π*(N^N)] energy. Moreover, diimine–OH moiety could be deprotonated under basic media. Though complicated response towards the environmental pH values may be expected due to the presence of two distinct and spatially separate sites, 2 and 3 exhibit simple response towards the external pH value change with no obvious oscillating behaviour as reflected in the absorption and emission spectra of 2 and 3 in the pH range of 2–13.

Due to solubility limitation of 2 and 3 in pure water, MeCN/H₂O (1 : 4, v/v) solution was employed as the solvent for pH titration studies. The absorption spectral change of 2 with pH of the media is shown in Fig. 9a. The region pertaining to ¹MLCT and ¹LC transitions in 2 remain nearly constant over the pH range of 2–8. However, on further increase in pH from 8–12, there are gradual spectral changes in absorption, consistent with a deprotonation step. In this pH range, the ¹MLCT maximum is red-shifted with a stabilization of the energy level by ca. 510 cm⁻¹. The absorption band at ca. 304 nm decreases and the band at 356 nm increases with an isosbestic point at 339 nm. The evolved absorption band at 356 nm is tentatively assigned to a ligand-centered process involving the deprotonated bpyOH ligand, which occurs at relatively low energy possibly due to the presence of a more conjugated π-system. The presence of a well-defined isosbestic point suggests the existence of two detectable species in equilibrium, i.e. the deprotonated and protonated forms. The fact that the change in the ¹LC absorption maximum arise solely from deprotonation at −OH group allows determination of an apparent pK_a value. Plot of absorbance at 304 nm versus pH, gives the expected sigmoidal shape. According to eqn (1), the pK_a value of 2 is determined to be 10.5 ± 0.2.

Fig. 9 (a) Absorption and (b) emission spectral change of 2 (1.0 × 10⁻⁴ mol L⁻¹) in pH 2–13 in MeCN/H₂O (1 : 4, v:v). Inset shows the variation of (a) absorption at λ_abs = 304 nm and (b) emission intensity at λ_em = 565 nm.

The emission spectral change of 2 in MeCN/H₂O solution as a function of pH has also been studied (Fig. 9b). At pH < 8, the emission of 2 does not exhibit appreciable pH dependence, whereas the luminescence intensity is progressively quenched as the pH is increased from 8 to 12. The plot of emission intensity vs. pH is sigmoidal with an inflexion point at pH 10.2, which corresponds to 50% of quenching (pH_i). The pH_i value is essentially identical with the measured pK_a value by UV/vis absorption method. The monotonic trend of the sigmoidal curve also offers a clear proof of the existence of one excited-state protonation/deprotonation process of hydroxyl group in 2. In principle, the excited state pK_a* needs to be corrected for the differences in the excited-state lifetime of protonated (τ) and deprotonated form (τ′) according to eqn (2).

pK_a* = pH_i + log τ/τ′ (2)

As the deprotonated species is essentially non-luminescent, the excited state pK_a* for 2 could be determined by an empirical method based on Forster's cycle with the eqn (3), where ν_B and ν_BH are energies in cm⁻¹ of the lowest-energy ¹MLCT band maxima in deprotonated and protonated states.^21,22

pK_a* = pK_a + (0.625/T)(ν_BH − ν_B) (3)

For 2, we have pK_a = 10.5 and ν_BH − ν_B = 510 cm⁻¹, which gives pK_a* ≈ 11.6. These values corroborate the obvious increased in basicity of the excited state as compared to that of the ground state. The UV/vis spectra of 2 shows that the ¹LC transition remains unchanged at pH 2–8, which indicates that there is negligible interactions of −OH and −CN groups with H⁺ in this pH range. At higher pH values, deprotonation of −OH group occurs and the negative charge is delocalized onto the proximal pyridyl N atom. In this form, the ligand undergoes virtually a tautomerization to give a deprotonated amide ligand that is a better π-donor and will weaken the ligand-field strength around the metal center and destabilize the filled d_π(Ru) orbital. The luminescent intensity decreases significantly, possibly due to rapid radiationless decay resulting from thermal equilibrium between MLCT and d–d excited states. Similar emission quenching upon deprotonation of the appended phenolic OH groups was reported for related systems.²³

Fig. S7a and S7b† show the ground state absorption and emission spectra for 3, respectively, at various pH values, and the trend is similar to 2. However, spectrophotometric titrations show that the pK_a of 3 is lower than that of 2, the UV/vis spectra remain nearly constant over the pH range 2–7, whereas the deprotonation step is clearly observed over the pH range of 7–11. When the pH is increased from 7 to 11, the absorption band at 268 nm decreases and the band at 290 nm increases with an isosbestic point at 282 nm. Plot of pH as a function of the absorption at 290 nm gives a sigmoidal curve, as in the case of 2. The pK_a value of 3 was determined to be 9.2 ± 0.2 according to eqn (1). The uncorrected emission spectrum of 3 has a maximum at 560 nm, which does not show an appreciable dependence on the pH values at pH < 7 (Fig. 10b). On increasing the pH from 7 to 11, the luminescence intensity is progressively quenched, but the emission wavelength remains change. The pK_a* of the excited state of 3 is also determined by the approximate method according to eqn (3) with a value of 11.7 ± 0.2. In order to confirm that the pH effect of compounds 2 and 3 arise from the protonation/deprotonation equilibrium of the pendent hydroxyl group on the diimine ligands, the pH effect of [Ru^II(phen)(PPh₃)(CN)₃]⁻ towards its absorption and emission spectra has also been carried out. As shown in Fig. S8 and S9,† both its absorption and emission spectra remain nearly unchanged in a wide pH range from 3–12, indicating that the pH effects do not originate from the cyano ligands.

Fig. 10 (a) The photograph on the colour changes taken under daylight (above) and the luminescence colour changes taken under UV illumination (λ_ex = 365 nm) on 2 (5.0 × 10⁻⁵ mol L⁻¹) (below) upon addition of 10 equiv. of various cations; (b) The absorption and (c) emission spectra of 2 upon addition of 10 equiv. of different metal ions in MeOH.

Effects of cations

The absorption and emission spectra of complexes towards different metal ions have been recorded and shown in Fig. 10. Since the behaviours of 2 and 3 towards the examined ions are quite similar, discussion will be mainly focused on 2. The UV/vis spectra of 2 display obvious changes upon addition of Fe³⁺ and Cu²⁺, while there no significant changes are observed for Mg²⁺, Ca²⁺, Na⁺, K⁺, Ni²⁺, Mn²⁺, and Co²⁺. Upon addition of Fe³⁺, the absorbance of the system increases remarkably and the absorption position also change significantly with the appearance of a strong absorption band at λ_abs = 367 nm. Upon addition of Cu²⁺, the absorbance of solution increases remarkably in the UV region. However, there is no significant change in the visible region. The significant change in higher-energy regions may indicate that the stronger interactions of −OH group with Cu²⁺ and Fe³⁺ ions.

Dramatic changes of the absorption and emission spectra in 2 in MeOH solution are also observed when ZnCl₂ is added. The MLCT absorption of 2 moves from 405 nm to 371 nm in the presence of two mole equivalent of Zn²⁺. This constitutes a considerable blue-shift of about 2260 cm⁻¹ in the MLCT absorption maximum upon complex formation. However, the LC (ππ*) transition observed at 303 nm undergoes a much less pronounced shift to 294 nm. These spectral changes may result from replacement of solvent molecules around the nitrogen atoms of cyanide ligands upon coordination to Zn²⁺, while the interaction between the −OH group and Zn²⁺ is negligible possibly due to the significant steric repulsion. The emission intensity and wavelength of 2 remain nearly unchanged upon addition of 10 mole equivalents of cations such as Na⁺ and K⁺, while there is a blue-shift of less than 5 nm with Ca²⁺ and Mg²⁺; The more significant blue-shift by addition of Zn²⁺ is most possibly due to that Zn²⁺ is a more stronger Lewis acid with stronger RuCN–Zn interactions. On the other hand, the emission intensity is nearly fully quenched by cations such as Fe³⁺, Cu²⁺, Co²⁺, Mn²⁺ and Ni²⁺, possibly due to photo-induced electron transfer (PET) effect from the unpaired electrons in these ions.^8,24

Titration experiment

In order to further investigate the interaction of Zn²⁺ with 2, titration of 2 with Zn²⁺ has been carried out by UV/vis absorption and emission methods in MeOH (Fig. 11). Upon addition of Zn²⁺, the absorption band at ca. 405 nm decreases and the band at 371 nm increases. An isosbestic point is observed at relatively low mole ratios of Zn²⁺/2 from 0 to 0.5. Further increase of [Zn²⁺] results in an additional slight blue-shift in MLCT transition without any isosbestic points. The donor–acceptor interaction of Zn²⁺ with cyano ligands leads to an increase in the π-acceptor ability of the cyano ligands, hence, the MLCT band assigned as spin-allowed t_2g⁶ → t_2g⁵π* transition is shifted to higher energy.²⁵ The emission of 2 with its maximum at λ_em = 578 nm is readily quenched by Zn²⁺, with concomitant increase in luminescence increase at λ_em = 522 nm. The emission profiles also exhibit a temporary isoemissive point at 544 nm with the mole ratio of Zn²⁺/2 up to 0.5. The emission intensity of λ_em = 522 nm is further enhanced by increasing mole ratio of Zn²⁺/2 from 0.5 to 1.0, and then the emission intensity remains nearly unchanged even when the mole ratio of Zn²⁺/2 is increased to 1.5. Both the UV/vis and emission spectra exhibit a similar response towards Zn²⁺. These spectral changes upon titration with Zn²⁺ can be attributed to the successive formation two new species in this process. An initial 1 : 2 stoichiometric adduct is formed at the mole ratio of Zn²⁺/2 < 0.5 which is then progressively converted into a 1 : 1 adduct. In addition, the Job's plot, which is the plot of absorbance at 384 nm, also shows a 1 : 1 stoichiometry between 2 and Zn²⁺ (Fig. 11a inset).

Fig. 11 (a) UV/vis absorption and (b) emission spectral changes on titration of Zn²⁺ into a methanol solution of 2. Inset shows the Job's plot corresponds to a 1 : 1 receptor–metal complex formation and the absorbance at 384 nm was plotted as a function of the molar ratio [Zn²⁺]/{[Zn²⁺] + [2]}.

Under similar condition, addition of Cd²⁺ to a methanol solution of 2 also results in absorption and emission spectral changes. Upon addition of only 0.2 equiv. of Cd²⁺, a significant blue-shift from 575 nm to 520 nm is observed. It is noteworthy that the emission intensity of 2 exhibits a pronounced enhancement upon addition of about 0.4 equiv. Cd²⁺, however, since a white precipitate is formed in the process, further investigation is not carried out.

The effects of meal ions on [Ru^II(phen)(PPh₃)(CN)₃]⁻ has also been investigated in MeOH in order to compare with those of 2 and 3. As shown in Fig. S10,† the emission spectra of this complex reveal that its MLCT band exhibits little change towards monovalent cations such as Na⁺ and K⁺ and is slightly blue-shifted in the presence of the divalent Mg²⁺. However, the effect of Zn²⁺ is more pronounced. These variations are similar to those of 2 and 3 under similar conditions. However, titration experiment involving addition of Zn²⁺ to an methanolic solution of [Ru^II(phen)(PPh₃)(CN)₃]⁻ shows different results (Fig. S11†). Upon the addition of Zn²⁺, the complex shows a new emission band at λ_em = 521 nm, while the initial emission band at λ_em = 582 nm gradually blue shifts and becomes a shoulder peak at λ_em = 552 nm (Fig. S11†). This result indicates that there are at least two species upon addition of Zn²⁺ in this system. The existence of a broadened emission band is the result of at least two emitting species that leads to an apparent blue shifting of the λ_em. Moreover, the intensity of the new emission band at λ_em = 521 nm exhibits a remarkable enhancement as compared with that at λ_em = 582 nm; while in 2, the intensity of these emission bands remains nearly equal.

Experimental

Materials and reagents

The ligands, 6-hydroxy-2,2′-bipyridine (bpyOH), 2-hydroxy-1,10-phenanthroline (phenOH)²⁶ and the complex [Ru^II(PPh₃)₃Cl₂]²⁷ were synthesized according to literature procedures. Triphenylphosphine and RuCl₃ were purchased from Strem Chemical Company and were used without further purification. All other reagents and solvents were of analytical grade and were used without further purification.

Physical measurements and instrumentation

¹H NMR and ³¹P{¹H} NMR spectra were recorded on a Bruker AV300 (300 MHz) FT-NMR spectrometer. Chemical shifts (δ) are reported relative to tetramethylsilane (Me₄Si). Negative-ion ESI mass spectra were recorded on a PE-SCIEX API 150 EX single-quadruple mass spectrometer. Elemental analysis was performed on an Elementar Vario MICRO Cube elemental analyzer. IR spectra of the solid samples as KBr discs were obtained within the range 4000–400 cm⁻¹ on an AVATAR 360 FTIR spectrometer.

Electronic absorption spectra were recorded on a Hewlett-Packard 8453 or a Hewlett-Packard 8452A diode-array spectrophotometer. Steady-state emission and excitation spectra were measured at RT and at 77 K on a Horiba Jobin Yvon Fluorolog-3-TCSPC spectrofluorometer. The solutions were rigorously degassed on a high-vacuum line in a two-compartment cell with not less than four successive freeze–pump–thaw cycles. The measurements at 77 K were carried out on dilute solutions of the samples in EtOH/MeOH (4 : 1 v/v) loaded in a quartz tube inside a quartz-walled Dewar flask that contained liquid nitrogen. The luminescence quantum yields were determined by using the optical-dilution method as described by Demas and Crosby²⁸ with an aerated aqueous solution of [Ru(bpy)₃Cl₂](ϕ_em = 0.040 (ref. 29) with excitation at 436 nm) as the reference. Luminescence lifetimes were measured by using the time-correlated single-photon-counting (TCSPC) technique on a Fluorolog-3-TCSPC spectrofluorometer in a fast MCS mode with a Nano LED-375 LH excitation source, which had a peak excitation wavelength at 375 nm and a pulse width of less than 750 ps. The photon-counting data were analysed on Horiba Jobin Yvon Decay Analysis Software.

Cyclic voltammetric (CV) measurements were performed on a CH Instruments, Inc. model CHI 620 Electrochemical Analyser in MeCN solution with [ⁿBu₄N]PF₆ (0.1 M) as a supporting electrolyte at room temperature. The reference electrode was a Ag/AgCl (0.1 M in MeCN) electrode and the working electrode was a glassy carbon electrode (CH Instruments, Inc.) with platinum wire as the counter electrode. The surface of the working electrode was polished with a 1 μm a-alumina slurry (Linde) and then with a 0.3 μm a-alumina slurry (Linde) on a microcloth (Buehler Co.). The ferrocenium/ferrocene^+/0 couple (FeCp₂) was used as an internal reference. Solutions for electrochemical studies were de-aerated with pre-purified argon gas prior to the measurements.

X-ray crystal structure determination

The crystal structures were determined on an Oxford Diffraction Gemini S Ultra X-ray single-crystal diffractometer by using graphite-monochromated CuK_a radiation (λ = 1.5417 Å). The structures were solved by using direct methods with the SHELXS-97 program.³⁰ The Ru atoms and many of the non-hydrogen atoms were located according to the direct methods. The positions of the other non-hydrogen atoms were located after refinement by full-matrix least-squares by using the SHELXL-97 program.³¹ In the final stage of the least-squares refinement, all non-hydrogen atoms were refined anisotropically. H atoms were generated by using the SHELXL-97 program.³¹ The positions of H atoms were calculated based on the riding model with thermal parameters that were 1.2 times that of the associated C atoms and participated in the calculation of the final R indices. CCDC-1487278 (1) and CCDC-1487279 (2) contain the supplementary crystallographic data for this paper.

Computer modeling

Molecular structures were optimized using DFT (hybrid Hartree-Fock DFT functional B3LYP level, which is a combination of the Becke 3-parameter exchange and Lee–Yang–Parr correlation functional, using the Gaussian 09 program package). In all cases, the LANL2DZ basis set was used for Ru atom and 6-31G(d) basis set for others. In all energy-minimized calculated structures, the optimized structure was confirmed to be a minimum from vibrational frequency calculations.³²

Syntheses. All of the reactions were carried out under strictly anaerobic conditions in an inert argon atmosphere by using standard Schlenk techniques.

Synthesis of (PPh₄)₂[Ru^II(PPh₃)₂(CN)₄] (1). [Ru^II(PPh₃)₃Cl₂] (1.0 g, 1.04 mmol) and KCN (0.41 g, 6.26 mmol) in MeOH was heated at reflux overnight. The solvent was removed under the reduced pressure. The white residue was dissolved in water (500 mL) and then PPh₄Cl (1.12 g, 3 mmol) was added to give product as white precipitate. The crystals suitable for X-ray determination were obtained by slow evaporation of a MeOH/H₂O solution of the complex. Yield: 76.9%, 1.13 g. IR (KBr, cm⁻¹): ν(C≡N) 2059, 2100. ¹H NMR (300 MHz, CDCl₃): δ 7.91–7.96 (m, 14H, Ar–H), 7.82–7.71 (m, 28H, Ar–H), 7.21–7.27 (m, 28H, Ar–H). Elemental analysis for C₈₈H₇₀N₄P₄Ru: calcd C 75.04, H 5.01, N 3.98%; found C 75.11, H 5.28, N 3.79%. UV/vis (CH₃CN): λ_max/nm (ε/mol⁻¹ dm³ cm⁻¹): 268 (12 680), 275 (10 140), 310 sh (1800). ESI-MS: m/z 365.0 [M]²⁻, 1069.0 [M + PPh₄]⁺.

Synthesis of (PPh₄)[Ru^II(PPh₃)(bpyOH)(CN)₃] (2). A mixture of 1 (100 mg, 0.07 mmol) and bpyOH (14.7 mg, 0.08 mmol) in DMF (25 mL) was heated at reflux overnight under a nitrogen atmosphere. After removal of the solvent under reduced pressure, the residue was washed with Et₂O and further purified by column chromatography on silica gel with CH₂Cl₂/MeOH (v/v 1 : 4) as the eluent to give orange band. Crystals suitable for X-ray determination were obtained by slow diffusion of Et₂O vapour into the concentrated CH₂Cl₂ solution of 2. Yield: 66.5%, 45 mg. Elemental analysis for C₅₅H₄₃N₅OP₂Ru: calcd C 69.32, H 4.55, N 7.35%; found C 69.40, H 4.70, N 7.22%; IR (KBr, cm⁻¹): ν(C≡N) 2093, 2057, 2045. ¹H NMR (300 MHz, CDCl₃): δ 10.15 (s, 1H; Ar–OH), 9.03 (d, J = 5.4 Hz, 1H; bpy-H), 8.19 (d, J = 8.2 Hz, 1H; bpy-H), 8.04–7.94 (m, 3H, bpy-H), 7.92–7.68 (m, 17H, phenyl-H), 7.65 (d, J = 9.2 Hz, 1H, bpy-H), 7.35–7.05 (m, 18H, phenyl-H), 6.63 (d, J = 8.1 Hz, 1H, bpy-H); ³¹P{¹H} NMR (162 MHz, CDCl₃): δ 41.02 (s, PPh₃), 22.32 (s, PPh₄⁺). UV/vis (CH₃CN): λ_max/nm (ε/mol⁻¹ dm³ cm⁻¹): 260 sh (9790), 268 (9740), 276 (9540), 307 (12 680), 330 sh (5820), 433 (2200). ESI-MS: m/z 614 [M]⁻.

Synthesis of (PPh₄)[Ru^II(PPh₃)(phenOH)(CN)₃] (3). The complex was synthesized using similar synthetic procedure of 2 except that phenOH was used instead of bpyOH. Yield: 57.6%, 40 mg. Elemental analysis for C₅₇H₄₃N₅OP₂Ru: calcd C 70.07, H 4.44, N 7.17%; found C 70.12, H 4.48, N 7.10%. IR (KBr, cm⁻¹): ν(C≡N) 2089, 2082, 2075. ¹H NMR (300 MHz, CDCl₃): δ 10.69 (s, 1H; Ar–OH), 9.26 (d, J = 5.1 Hz, 1H; phen H), 8.39–8.27 (m, 2H, phen H), 8.03–7.93 (m, 6H, phenyl H), 7.90–7.67 (m, 21H, phenyl H), 7.49 (dd, J = 8.2, 5.2 Hz, 1H; phen H), 7.21–6.98 (m, 13H, phenyl H); ³¹P{¹H} NMR (162 MHz, CDCl₃): δ 41.84 (s, PPh₃), 22.32 (s, PPh₄⁺). UV/vis (CH₃CN): λ_max/nm (ε/mol⁻¹ dm³ cm⁻¹): 268 (29 020), 274 sh (26 170), 295 sh (10 290), 383 (3000), 433 (3580). ESI-MS: m/z 638 [M]⁻.

Conclusion

Two mononuclear tricyanoruthenium(II) compounds fac-(PPh₄)[Ru^II(bpyOH)(PPh₃)(CN)₃] (2) and fac-(PPh₄)[Ru^II(phenOH)(PPh₃)(CN)₃] (3) have been obtained from the reactions of (PPh₄)₂[Ru^II(PPh₃)₂(CN)₄] (1) with bpyOH and phenOH, respectively, in DMF. Both complexes display impressive quantum yields in various solvents, which are much higher than those of related ruthenate(II) diimine complexes. Due to the presence of two different and spatially separate CN⁻ and −OH sites in these two complexes, solvatochromic behaviour, pH and ion effects of 2 and 3 are distinct from those of [Ru^II(phen)(PPh₃)(CN)₃]⁻. Our results demonstrate that these complexes useful for the development of new luminescent sensors/probes and sensitizers.

Acknowledgements

The authors gratefully acknowledge the financial support of Natural Science Foundation of China (21201023). This work was also supported by the Hong Kong University Grants Committee Area of Excellence Scheme (AoE/P-03-08).

Notes and references

1. (a) A. P. De Silva, H. Q. N. Gunaratne, T. Gunnlaugsson, A. J. M. Huxley, C. P. McCoy, J. T. Rademacher and T. E. Rice, Chem. Rev., 1997, 97, 1515–1566; (b) J. N. Demas and B. A. DeGraff, Coord. Chem. Rev., 2001, 211, 317–351.
2. (a) E. Baggaley, S. W. Botchway, J. W. Haycock, H. Morris, I. V. Sazanovich, J. A. G. Williams and J. A. Weinstein, Chem. Sci., 2014, 5, 879–886; (b) Q. Ju, Y. Liu, D. Tu, H. Zhu, R. Li and X. Chen, Chem.–Eur. J., 2011, 17, 8549–8554; (c) Y. Ma, S. J. Liu, H. R. Yang, Y. Q. Wu, H. B. Sun, J. X. Wang, Q. Zhao, F. Y. Li and W. J. Huang, J. Mater. Chem. B, 2013, 1, 319–329; (d) H. E. Rajapakse, N. Gahlaut, S. Mohandessi, D. Yu, J. R. Turner and L. W. Miller, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 13582–13587; (e) J. N. Demas and B. A. DeGraff, Anal. Chem., 1991, 63, 829A–837A.
3. J. N. Demas, B. A. DeGraff and P. B. Coleman, Anal. Chem., 1999, 71, 793A–800A.
4. A. Mills and K. Eaton, Quim. Anal., 2000, 19, 75–87.
5. C. Malins, H. G. Glever, T. E. Keyes, J. G. Vos, W. J. Dressick and B. D. MacCraith, Sens. Actuators, B, 2000, 67, 89–95.
6. C. Huber, I. Klimant, C. Krause and O. S. Wolfbeis, Anal. Chem., 2001, 73, 2097–2103.
7. J. N. Demas and B. A. DeGraff, Proc. SPIE-Int. Soc. Opt. Eng., 1992, 1796, 71–75.
8. (a) Y. Sun, Z. M. Hudson, Y. L. Rao and S. N. Wang, Inorg. Chem., 2011, 50, 3373–3378; (b) R. Zhang, Z. Q. Ye, B. Song, Z. C. Dai, X. An and J. L. Yuan, Inorg. Chem., 2013, 52, 10325–10331; (c) Z. Q. Ye, X. An, B. Song, W. Z. Zhang, Z. C. Dai and J. L. Yuan, Dalton Trans., 2014, 43, 13055–13060; (d) R. Zhang, X. J. Yu, Z. Q. Ye, G. L. Wang, W. Z. Zhang and J. L. Yuan, Inorg. Chem., 2010, 49, 7898–7903; (e) H. Xu, H. C. Hu, C. S. Cao and B. Zhao, Inorg. Chem., 2015, 54, 4585–4587.
9. (a) A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser and A. Von Zelewsky, Coord. Chem. Rev., 1988, 84, 85–277; (b) V. Balzani, A. Juris, M. Venturi, S. Campagna and S. Serroni, Chem. Rev., 1996, 96, 759–833 and references therein. (c) A. K. D. Mesmaeker, J. P. Lecomte and J. M. Kelly, Top. Curr. Chem., 1996, 177, 25–76; (d) V. Balzani and F. Scandola, in Comprehensive Supramolecular Photochemistry, ed. J. Atwood, D. MacNicol, E. Davies, F. Vogtle and J. M. Lehn, Pergamon Press, Oxford, U.K., 1996, vol. 10, p. 687.
10. (a) C. A. Bignozzi, C. Chiorboli, M. T. Indelli, M. A. R. Scandola, G. Varani and F. J. Scandola, J. Am. Chem. Soc., 1986, 108, 7872–7873; (b) A. Ülveczky and A. Horváth, Inorg. Chim. Acta, 1995, 236, 173–176.
11. (a) J. L. Habib Jiwan, B. Wegewijs, M. T. Indelli, F. Scandola and S. E. Braslavsky, Recl. Trav. Chim. Pays-Bas, 1995, 114, 542; (b) M. D. Ward, Coord. Chem. Rev., 2006, 250, 3128; (c) C. J. Timpson, C. A. Bignozzi, B. P. Sullivan, E. M. Kober and T. J. Meyer, J. Phys. Chem., 1996, 100, 2915.
12. (a) C. D. Borsarelli, S. E. Braslavsky, M. T. Indelli and F. Scandola, Chem. Phys. Lett., 2000, 317, 53–58; (b) M. A. Rampi, M. T. Indelli, F. Pina, F. Scandola and A. J. Parola, Inorg. Chem., 1996, 35, 3355–3361; (c) J. K. Evju and K. R. Mann, Chem. Mater., 1999, 11, 1425–1433; (d) C. E. Keller, C. Pollard, L. K. Yeung, W. D. Plessinger and C. J. Murphy, Inorg. Chim. Acta, 2000, 298, 209–215; (e) T. Lazarides, T. L. Easun, C. Veyne–Marti, W. Z. Alsindi, M. W. George, N. Deppermann, C. A. Hunter, H. Adams and M. D. Ward, J. Am. Chem. Soc., 2007, 129, 4014–4027; (f) F. Pina and A. J. Parola, Coord. Chem. Rev., 1999, 149, 185–186; (g) T. Megyes, G. Schubert, M. Kovács, T. Radnai, T. Grósz, I. Bakó, I. Pápai and A. Korváth, J. Phys. Chem. A, 2003, 107, 9903–9909; (h) M. D. Ward, Coord. Chem. Rev., 2006, 250, 3128–3141; (i) G. Bergamini, C. Saudan, P. Ceroni, M. Maestri, V. Balzani, M. Groka, S. K. Lee, J. V. Heyst and F. Vögtle, J. Am. Chem. Soc., 2004, 126, 16466–16471.
13. (a) M. D. Ward, Dalton Trans., 2010, 39, 8851–8867; (b) T. Abe, T. Suzuki and K. Shinozaki, Inorg. Chem., 2010, 49, 1794–1800; (c) S. Encinas, A. F. Morales, F. Barigelletti, A. M. Barthram, C. M. White, S. M. Couchman, J. C. Jeffery, M. D. Ward, D. C. Grills and M. W. George, J. Chem. Soc., Dalton Trans., 2001, 3312–3319; (d) M. E. Garca Posse, N. E. Katz, L. M. Baraldo, D. D. Polonuer, C. G. Colombano and J. A. Olabe, Inorg. Chem., 1995, 34, 1830–1835; (e) S. G. Baca, H. Adams, C. S. Grange, A. P. Smith, I. Sazanovich and M. D. Ward, Inorg. Chem., 2007, 46, 9779–9789; (f) L. T. –L. Lo, S. W. Lai, S. M. Yiu and C. C. Ko, Chem. Commun., 2013, 49, 2311–2313.
14. H. Feng, F. Zhang, S. W. Lai, S. M. Yiu and C. C. Ko, Chem.–Eur. J., 2013, 19, 15190–15198.
15. (a) Y.-H. Chi, L. Yu, J. M. Shi, Y. Q. Zhang, T. Q. Hu, G. Q. Zhang, W. Shi and P. Cheng, Dalton Trans., 2011, 40, 1453–1462; (b) X. M. Zhang, M. L. Tong, M. L. Gong, H. K. Lee, L. Luo, K. F. Li, Y. X. Tong and X. M. Chen, Chem.–Eur. J., 2002, 8, 3187–3194; (c) X. M. Zhang, M. L. Tong and X. M. Chen, Angew. Chem., Int. Ed., 2002, 41, 1029–1031; (d) K. Fujita, Y. Tanaka, M. Kobayashi and R. Yamaguchi, J. Am. Chem. Soc., 2014, 136, 4829–4832; (e) R. Kawahara, K. Fujita and R. Yamaguchi, Angew. Chem., Int. Ed., 2012, 51, 12790–12794; (f) T. Tomon, T. Koizumi and K. Tanaka, Eur. J. Inorg. Chem., 2005, 285–293.
16. (a) J. Xiang, L. T.-L. Lo, C.-F. Leung, S.-M. Yiu, C.-C. Ko and T.-C. Lau, Organometallics, 2012, 31, 7101–7108; (b) J. Xiang, J.-S. Wu and Z. Anorg, Z. Anorg. Allg. Chem., 2013, 639, 606–610.
17. N. Nickita, M. J. Belouscoff, A. I. Bhatt, A. M. Bond, G. B. Deacon, G. Gasser and L. Spiccia, Inorg. Chem., 2007, 46, 8638–8651.
18. (a) J. V. Houten and R. J. Watts, J. Am. Chem. Soc., 1975, 97, 3843; (b) D. P. Rillema, G. Allen, T. J. Meyer and D. Gonrad, Inorg. Chem., 1983, 22, 1617; (c) E. M. Kober, B. P. Sullivan and T. J. Meyer, Inorg. Chem., 1984, 23, 2098; (d) J. V. Caspar and T. J. Meyer, J. Am. Chem. Soc., 1983, 105, 5583; (e) A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser and A. Von Zelewsky, Coord. Chem. Rev., 1995, 33, 2438.
19. (a) M. T. Indeli, M. Ghirotti, A. Prodi, C. Chiorboli, F. Scandola, N. D. McClenaghan, F. Puntoriero and S. Camanga, Inorg. Chem., 2003, 42, 5489–5497; (b) N. R. M. Simpson, M. D. Ward, A. F. Morales, B. Ventura and F. Barigelletti, J. Chem. Soc., Dalton Trans., 2002, 2455–2461; (c) N. R. M. Simpson, M. D. Ward, A. F. Morales, B. Ventura and F. Barigelletti, J. Chem. Soc., Dalton Trans., 2002, 2449–2454; (d) D. J. Stufkens, Inorg. Chem., 1992, 13, 359; (e) T. G. Kotch and A. J. Lees, Inorg. Chem., 1993, 32, 2570.
20. M. E. Garca Posse, N. E. Katz, L. M. Baraldo, D. D. Polonuer, C. G. Colombano and J. A. Olabe, Inorg. Chem., 1995, 34, 1830–1835.
21. (a) C. Hicks, G.-Z. Ye, C. Levi, M. Gonzales, I. Rutenburg, J.-W. Fan, R. Helmy, A. Kassis and H. D. Gafney, Coord. Chem. Rev., 2001, 211, 207; (b) Md. K. Nazeeruddin and K. Kalyanasundaram, Inorg. Chem., 1989, 28, 4251; (c) G. Y. Zheng, Y. Wang and D. P. Rillema, Inorg. Chem., 1996, 35, 7118; (d) F. Casalboni, Q. G. Mulazzani, C. D. Clark, M. Z. Hoffman, P. L. Orizondo, M. W. Perkovic and D. P. Rillema, Inorg. Chem., 1997, 36, 2252; (e) Md. K. Nazeeruddin, S. M. Zakeeruddin, R. Humphry-Baker, M. Jirousek, P. Liska, N. Vlachopoulos, V. Shklover, C.-H. Fischer and M. Gratzel, Inorg. Chem., 1999, 38, 6298.
22. (a) A. M. W. C. Thompson, M. C. C. Smailes, J. C. JeVery and M. D. Ward, J. Chem. Soc., Dalton Trans., 1997, 737–744; (b) J. G. Vos, Polyhedron, 1992, 11, 2285–2299.
23. J. R. Lakowicz, Principles of Fluorescence Spectroscopy, Springer–Verlag, New York, 3rd edn, 2006.
24. W. Cao, X. J. Zheng, D. C. Fang and L. P. Jin, Dalton Trans., 2014, 43, 7298–7303.
25. (a) A. Fodor–Kardos and A. Horváth, Photochem. Photobiol. Sci., 2005, 4, 185–190; (b) T. Lazarides, T. L. Easun, C. Veyne–Marti, W. Z. Alsindi, M. W. George, N. Deppermann, C. A. Hunter, H. Adams and M. D. Ward, J. Am. Chem. Soc., 2007, 129, 4014–4027.
26. M. Zalas, B. Gierczyk, M. Cegłowski and G. Schroeder, Chem. Pap., 2012, 66, 733–740.
27. (a) T. A. Stephenson and G. J. Wilkinson, Inorg. Nucl. Chem. Lett., 1966, 28, 945–956; (b) P. S. Hallman, T. A. Stephenson and G. J. Wilkinson, Inorg. Synth., 1970, 12, 237–240; (c) S. Nishimura, O. Yumoto, K. Tsuneda and H. Mori, Bull. Chem. Soc. Jpn., 1975, 48, 2603–2604.
28. J. N. Demas and G. A. Crosby, J. Phys. Chem., 1971, 75, 991–1024.
29. J. Van Houten and R. Watts, J. Am. Chem. Soc., 1976, 98, 4853–4858.
30. G. M. Sheldrick, SHELX–97: Programs for Crystal Structure Analysis (Release 97–2), University of Göttingen, Göttingen, Germany, 1997.
31. A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, M. Burla, G. Polidori and M. J. Camalli, Appl. Crystallogr., 1994, 27, 435–436.
32. L. Salassa, C. Garino, G. Salassa, R. Gobetto and C. Nervi, J. Am. Chem. Soc., 2008, 130, 9590.

---

Footnotes

† Electronic supplementary information (ESI) available. CCDC 1487278 and 1487279. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6ra16319j

‡ These two authors contributed equally.

---