The amplified electrochemiluminescence response signal promoted by the Ir(III)-containing polymer complex†

Yayun Fang ^a, Ziyu Wang ^b, Yang Li ^a, Yiwu Quan *^a and Yixiang Cheng *^b
aDepartment of Polymer Science & Engineering, State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210093, China. E-mail: quanyiwu@nju.edu.cn
^bKey Lab of Mesoscopic Chemistry of MOE and Jiangsu Key Laboratory of Advanced Organic Materials, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023, China. E-mail: yxcheng@nju.edu.cn

First published on 7th April 2018

---

---

Introduction

Electrochemiluminescence (ECL) can be regarded as a light-emitting phenomenon through the electron transfer (ET) process between the reduced and oxidized species generated at the surface of electrodes.¹ Generally, the pathway of the ECL reaction has been proposed to undergo the following stages: (1) electrode oxidation or reduction, (2) generation of excited species, and (3) returning to the ground state and ECL emission.^2,3 ECL, as one of the most promising bioanalytical detection methods, is superior to traditional photoluminescence (PL) analytical technology and has many advantages including rapidity, high sensitivity and selectivity as well as a low detection limit due to the absence of excitation background light.^1b,4 Therefore, more and more attention has been paid to this simplified technique for various applications in immunoassays, diagnostics and environmental assays.⁵

It is well known that tris(2,2′-bipyridine)ruthenium(II) (Ru(bpy)₃²⁺) and its derivatives have been regarded as one of the most successful ECL luminophores since the pioneering work of Bard's group.⁶ Recently, more and more novel non-ruthenium photoactive ECL materials have emerged and experienced rapid development, which can be classified into inorganic materials, organic materials and nanomaterials.⁷ Normally, inorganic materials mainly involve a variety of cyclometalated complexes. Since strong metal-to-ligand charge transfer (MLCT) was mediated by the metal–ligand orbital mixing, cyclometalated complexes (for example, Ru, Os and lr) exhibit MLCT excited states which usually show long excited-state lifetimes and high quantum yields independent of the excitation wavelength.^8,9 Among various transition metal complexes, cyclometalated Ir(III) complexes have been proposed as one of the most excellent ECL materials due to their high photoluminescence quantum yields (PLQYs), stable electrochemical characteristics, efficient intersystem crossing between the singlet and triplet excited states caused by the strong spin–orbit coupling of Ir(III), long luminescence lifetimes and tunable wavelength range.¹⁰ For instance, the most simple tris(2-phenylpyridine)iridium(III) complex [Ir(ppy)₃³⁺] can not only exhibit high luminescence efficiency (φ_PL ∼ 0.4) and good electrochemical stability at room temperature, but also show strong ECL emission via the annihilation reaction between its redox precursors.¹¹ In addition, Kim's group designed two Ir(III) complexes as ECL chemodosimetric probes for highly sensitive and selective detection of sulfide (S²⁻).^4a Cola and his co-workers also reported the strong PL and the amplified ECL signal from the bis-cyclometalated Ir(III) complex, [Ir-(ĈN)₂(X)] (ĈN = cyclometalated ligand, X = picolinate (pic) or acetylacetonate (acac)), both in organic solvents and aqueous buffer solution.¹²

Although a series of small molecular cyclometalated iridium complexes can exhibit excellent photophysical, electrochemicaly and ECL properties by decorating the main ligand or changing the ancillary ligand, the ECL properties of Ir(III)-containing polymer complexes have been seldom investigated. As is well known, much effort has been devoted to the conjugated polymer based ECL since the delocalized π-electron system along the polymer backbone can provide an efficient pathway for charge migration and energy transportation, which leads to the advantageous electrochemical and photo-physical properties. Recently, Tefashe et al. designed block copolymers by incorporating the Ir(ppy)₂(bpy)⁺ complex moiety to the polymer backbone and soon afterwards they demonstrated the application of these Ir(III)-containing micelles to ECL-based detection assays.¹³ Therefore, developing polymer-based Ir(III) complexes as ECL materials is of great significance and is challenging.

In the past few decades, polymers incorporated with the 9,9-dialkyl fluorene moiety have been successfully applied as optoelectronic materials for polymer light-emitting diodes (PLEDs) and photodetectors because of their highly efficient fluorescence emission, high thermal and oxidative stability and good solubility.¹⁴ Herein, we found that the Ir(III)-containing polymer complex P2 incorporating the 9,9-dibutylfluorene moiety as a polymer backbone linker can emit a strong ECL signal compared with P1 and the Ir(III) model complex by using TPrA as a co-reactant in CH₃CN solution, which can be attributed to the effective intramolecular metal–ligand charge transfer (MLCT) from the Ir(III)-complex centre to the polymer backbone. This work can provide a new strategy for the development of excellent ECL materials based on Ir(III)-containing polymer complexes.

Experimental section

Materials and reagents

All the reagents including tetrabutylammonium hexafluorophosphate (TBAPF₆, ≥99%), anhydrous acetonitrile (CH₃CN, ≥99%), tripropylamine (TPrA, ≥98%) and iridium(III) chloride hydrate were purchased from Sigma-Aldrich and used as received.

Apparatus

UV-vis absorption spectra were obtained using a Nanodrop-2000C UV-vis spectrophotometer (Thermo, USA). Fluorescence measurements were conducted on an F-7000 fluorescence spectrometer (Hitachi Co., Japan) equipped with a xenon lamp. NMR spectra were obtained from a Bruker Advance 400 spectrometer (Bruker, German) at 400 MHz for ¹H NMR and 100 MHz for ¹³C NMR reported as parts per million (ppm) from the internal standard tetramethylsilane (TMS). The molecular weight was determined by gel permeation chromatography (GPC) with a Waters 244 HPLC pump, and THF was used as the solvent relative to polystyrene standards. The content of Ir(III) in P2 was detected by using inductively coupled plasma mass spectrometry (ICP-MS, PE, NEXION300).

Preparation of electrodes

The glassy carbon electrodes (GCE, 5 mm in diameter) were polished to a mirror using a 0.02–0.05 μm alumina slurry (Gaoss Union, Wuhan), followed by sonication in water, ethanol and water, individually. After the electrodes were rinsed thoroughly with ultrapure water and dried in a N₂ flow, we detected the ECL of the Ir(III) model complex, P1, and P2 in the presence of 0.1 M TBAPF₆ and 25 mM TPrA by potential scan experiments.

Electrochemical and ECL measurements

The cyclic voltammograms of 9,9-dibutylfluorene, the Ir(III) model complex, P1, and P2 in CH₃CN solvent were studied on a CHI 660E electrochemical workstation with a conventional three-electrode configuration containing a platinum wire counter electrode and an Ag/Ag⁺ (0.1 M AgNO₃) electrode as the reference electrode. Cyclic voltammetry measurements were carried out in CH₃CN solution that was deoxygenated with N₂ for 20 min and 0.1 M Bu₄NBF₄ as an electrolyte.

The ECL properties of the Ir(III) model complex (the concentration was fixed at the same concentration as the Ir(III)-complex moiety of P2), P1 and P2 were studied in CH₃CN solution. To prepare solutions of P2 and the Ir(III) model complex with an equal concentration of the Ir, the content of Ir in P2 was examined by ICP-MS (5‰, wt), assuming that polymerization did not affect the extinction coefficient of the Ir(III) complex. Thus, for a particular Ir(III) concentration, there would be 71.4 times more P2 than the Ir(III) model complex. ECL measurements were carried out in a self-made cell on a MPI-E multifunctional electrochemical and chemiluminescent analytical system (Xi'an Remex Analytical Instrument Co. Ltd, China) with the prepared glass carbon working electrode, a platinum wire counter electrode and a Ag/AgCl (saturated KCl) reference electrode in CH₃CN solution containing 0.1 M TBAPF₆ as the electrolyte and 25 mM TPrA as the co-reactant. Unless specifically noted, the ECL window was placed in front of the photomultiplier tube (PMT) biased at 400 V with a scan rate of 100 mV s⁻¹ to record the ECL signals.

ECL spectrum

The ECL emission spectrum of P2 was acquired by collecting the light signals during cyclic voltammetric (CV) scans from 0 to 1.6 V with a series of long filters (460, 480, 500, 520, 540, 560, 580, and 600 nm) in a CH₃CN solution containing 25 mM TPrA and 0.1 M TBAPF₆. Different wavelength filters were placed in front of the transparent window of PMT (600 V) and the corresponding ECL signals were recorded.

The relative ECL efficiency was calculated using the below equation:^7c,15

where Φ_ECL and Φ⁰_ECL are the ECL efficiency, I and I⁰ are the integrated ECL intensities (integrating ECL intensity vs. wavelength), and Q and Q⁰ are the consumed charges (integrating cyclic voltammogram vs. time) of the target and standard, respectively. Here, the ECL spectrum of 0.5 mM Ru(bpy)₃²⁺ at +1.1 V with a Φ⁰_ECL value of 1 was used as the standard in a CH₃CN solution containing 25 mM TPrA and 0.1 M TBAPF₆. The Φ_ECL of P2 relative to the Ru(bpy)₃²⁺/TPrA system was calculated to be 0.14.

Results and discussion

Synthesis

The synthesis routes of P1, P2 and the Ir(III) model complex are outlined in Scheme 1, and the detailed procedures and characterization of all the molecules are presented in the ESI.† The monomer 9,9-dibutyl-2,7-diethynyl-9H-fluorene (M1) and the monomer N,N‘-(ethane-1,2-diyl)bis(1-(4-bromophenyl) methanimine) (M2) were synthesized according to the reported literature, respectively.¹⁶ [Ir₂(2-ppy)₄(μ-Cl)₂] (M3) and N,N‘-(ethane-1,2-diyl)bis(1-phenylmethanimine) (M4) could be obtained by the same synthesis procedures.¹⁷ The polymer P1 could be synthesized by the Pd-catalyzed Sonogashira coupling polymerization reaction of the monomer M1 and monomer M2 in 84% yield to give M_w = 10090, M_n = 6470 and PDI = 1.56. P1 can show good solubility in common solvents, such as THF, CH₂Cl₂, and CHCl₃, due to the flexible n-butyl substituents on the fluorene moiety. In addition, the N-containing bidentate ligand (–CHN–CH₂–CH₂–NCH–) can orient the well-defined spatial arrangement in the polymer main chain and also further coordinate with M3 to form the corresponding Ir(III)-containing polymer complex P2 in CH₃CN–CH₂Cl₂ mixed solvents (v/v 1:1). The Ir(III) model complex could be obtained by reaction of M3 with M4 with similar conditions to P2. Meanwhile, we surprisingly found that the Ir(III) content (about 5‰, wt) is very low in P2 according to the result of ICP-MS detection, indicating that rather few N-containing bidentate chelating ligands undergo the coordination reaction with M3.

Photophysical properties

The UV-vis absorption and fluorescence emission spectra of P1 and P2 were obtained in CH₃CN solution (at the fixed concentration of 1.0 × 10⁻⁵ mol L⁻¹ corresponding to the fluorene moiety) (Fig. 1). The absorption spectrum of P1 shows a maximum at 433 nm with a shoulder peak at about 398 nm as shown in Fig. 1a, which can be assigned to the conjugated π–π* transition of the entire conjugated backbone.¹⁸ Notably, P2 exhibits strong absorption (π–π*) inside the polymer below 320 nm and weaker MLCT transition bands from 330 to 450 nm indicating the successful coordination of P1 with M3 according to the reported Ir(III) complexes.^19,20

As is illustrated in Fig. 1b, the fluorescence emission of P2 was observed around 462 nm. The emission wavelengths of ionic Ir(III) complexes, especially the emission colour, were mainly dominated by the auxiliary ligand structures.²¹ So P2 exhibits a similar emission wavelength to P1 because of the same conjugated polymer backbone. In addition, P1 shows well-resolved vibrionic structure emission peaks at 410, 442, and 476 nm and a shoulder at 506 nm, which is similar to the fluorescence emission of poly(9,9-dioctylfluorene).^22,23 However, a typical emission spectrum of the organometallic complex, a broad emission with a full width at half maximum (FWHM) about 57 nm can be observed for P2 (quantum yield: 9.9%, quinine sulphate as a reference).²⁴ Thus, we argue that P1 has successfully coordinated with M3 and the relaxation of the excited state for P2 can be mainly attributed to the MLCT transition from dπ(Ir) to π* (N^N). The average fluorescence lifetimes are 0.56 ns for P1 and 0.92 ns for P2 (Fig. 1c), which indicates almost no phosphorescence emission for P2 due to a very small amount of Ir(III) in P2. But we did not obtain the lifetime of the Ir(III) model complex because of its weak emission signal (Fig. S1†).

Electrochemistry and ECL

The CV studies of 9,9-dibutylfluorene, the Ir(III) model complex, P1 and P2 were carried out in CH₃CN solution at a concentration of 1 × 10⁻⁴ M containing 0.1 M Bu₄NBF₄ as the supporting electrolyte. The 9,9-dibutylfluorene shows an irreversible oxidation wave at a peak potential of +1.40 V with an onset oxidation potential of +1.28 V (Fig. 2a). It can also be observed that P1 shows a broad irreversible oxidation wave with an onset potential of +1.30 V due to the oxidation of the 9,9-dibutylfluorene part (Fig. 2c). Additionally, the Ir(III) model complex has two quasi-reversible reduction waves at −0.96 V and −1.15 V from the reduction of the ancillary ligand (M4) and two oxidation peaks at 0.65 V and 1.76 V, respectively (Fig. 2b).²⁴ Meanwhile, the CV of P2 can match well with the combination of the Ir(III) model complex and 9,9-dibutylfluorene (Fig. 2d). In the positive potential range, except for the two oxidation peaks at 0.80 V and 1.62 V generated by the oxidization of the Ir(III) complex moiety, P2 also has an oxidation wave of 9,9-dibutylfluorene. Thus, we can draw a conclusion that the Ir(III) complex moiety and the 9,9-dibutylfluorene in P2 can be oxidized sequentially under an anodic scan. Moreover, P2 shows two reduction waves at −0.95 V and −1.40 V corresponding to the reduction of the polymer ligand.²⁵

Furthermore, the molecular energy levels can be estimated from the CVs and are presented in Table 1. The HOMO and LUMO energy levels of P2 can well coincide with that of the Ir(III) model complex and 9,9-dibutylfluorene, respectively. The band gaps of P1 and P2 can be evaluated by the difference between the HOMO and LUMO energy levels to give the value as 2.55 eV for P1 and 1.84 eV for P2. Their optical band gaps in CH₃CN solution can also be calculated as 2.60 eV for P1 (λ_em = 476 nm) and 2.68 eV for P2 (λ_em = 462 nm) (Fig. 1b) by using the formula: E_g = 1239.8/λ (eV). The electrochemical gap of P1 is close to its optical energy gap, demonstrating that the LUMO and HOMO energy levels are reliable from the electrochemical measurement. Compared with P1, the electrochemical gap of P2 shows an obvious decline and is much lower than its optical energy gap, which can be ascribed to its strong intramolecular MLCT on the electrode surface.

---

In general, ECL can be achieved via ion annihilation and co-reactant processes.²⁶ In the ion annihilation approach, both of the oxidized and reduced ECL precursors are generated at an electrode by a two-direction potential step.^1c In the co-reactant process, the co-reactant first produces the oxidizing or reducing intermediates and then reacts with the ECL luminophore to form excited states. As illustrated in Fig. S2,† no obvious ECL signal could be observed in the ion annihilation process for P2. So the ECL behaviors of the Ir(III) model complex (containing the equivalent concentration of Ir(III) with P2), P1 and P2 in degassed CH₃CN solution including 0.1 M TBAPF₆ as the electrolyte and 25 mM TPrA as the co-reactant were evaluated from ECL emission intensities and peak potentials, time and emission wavelength in the co-reactant process. In the presence of TPrA, each compound has two ECL emission peaks, a weak one at a lower potential (0.80–1.04 V) and another strong emission at a higher potential (Fig. 3b). The ECL emission of P2 at +1.04 V can match well with that of the Ir(III) model complex at +1.02 V. But P2 can exhibit a much stronger ECL signal than the Ir(III) model complex and P1 at +1.55 V with an onset potential of 1.18 V, which originates from the redox of the polymer ligand.

Herein, we assume that the amplified ECL response behaviour of P2 is induced by the effective intramolecular MLCT from Ir(III) to the polymer ligand backbone. The probable mechanism for the generation of ECL is shown as the following steps.^27,28 Firstly, TPrA itself undergoes a redox process (steps 1 and 2), and then the Ir(III) complex moiety in P2 reacts with TPrA˙ to give an excited state P2[Ir(ppy)₂(M4)]* (step 3). Afterwards, the electron transfer occurs from this excited state to the polymer backbone of P2 (step 4). The formation of an excited state [P2 (P1)*] involves the reaction of TPrA˙⁺ with [P2 (P1⁻)] (step 5).

TPrA → TPrA˙+ + e−	(1)

TPrA˙+ → TPrA˙ + H+	(2)

	(3)

P2[Ir(ppy)2(M4)]* + P2(P1) → P2[Ir(ppy)2(M4)]+ + P2(P1)−	(4)

P2(Pl)− + TPrA˙+ → P2(P1)* + TPrA	(5)

P2(Pl)* → P2(P1) + hv	(6)

This mechanism can also be supported by the CV as well as ECL measurements that the Ir(III) model complex shows a lower LUMO energy level (Table 1) and ECL emission potential in comparison with P1. In other words, the Ir(III) complex moiety in P2 tends to grab an electron and generate an excited state, preferentially. Simultaneously, owing to the reversible redox of the Ir(III) complex moiety, P2 exhibits excellent ECL stability without any decline in the multicycle as shown in Fig. 3c.

ECL spectrum

The ECL emission spectrum of P2 can be obtained by drawing a plot of wavelength vs. ECL intensity (Fig. 4). The ECL spectrum of P2 displays a weak wave centred at 485 nm, which correlates well with the PL spectrum in acetonitrile solution. Upon comparison with the fluorescence, the main ECL emission peak of P2 is shifted approximately 55 nm to a lower energy. As illustrated in eqn (3) to (5), the Ir(III) complex moiety in P2 first forms an excited state preferentially, and then transfers electrons to the polymer ligand to form the emissive excited state [P2(P1)*] on the surface of the electrode. Thus, the effective intramolecular MLCT from the Ir(III) complex centre to the polymer backbone results in a more stabilized HOMO level and a narrower bandgap (Table 1) which leads to the redshifted and drastically enhanced ECL emission.

Conclusions

In summary, the obtained Ir(III)-containing polymer complex material can exhibit a strong ECL response and act as an excellent ECL emitter. The amplified ECL response behaviour is induced from the effective intramolecular MLCT from the Ir(III) complex center to the polymer backbone even at a rather low Ir(III) content. Our findings are of great significance in the amplification of the ECL signal and the design of novel promising ECL materials based on Ir(III)-containing polymers.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21474048, 21674046, and 51673093).

Notes and references

1. (a) J. Suk, P. Natarajan and J. N. Moorthy, J. Am. Chem. Soc., 2012, 134, 3451 CrossRef CAS PubMed ; (b) Electrogenerated chemiluminescence, ed. A. J. Bard, CRC Press, 2004 Search PubMed ; (c) M. M. Richter, Chem. Rev., 2004, 104, 300 CrossRef PubMed .
2. T. Wang, D. Wang and J. W. Padelford, J. Am. Chem. Soc., 2016, 138, 6380 CrossRef CAS PubMed .
3. K. N. Swanick, S. Ladouceur and E. Zysman-Colman, RSC Adv., 2013, 3, 19961 RSC .
4. (a) S. Y. Kim, H. J. Kim and J. I. Hong, RSC Adv., 2017, 7, 10865 RSC ; (b) W. Miao, Chem. Rev., 2008, 108, 2506 CrossRef CAS PubMed ; (c) E. Rampazzo, S. Bonacchi, D. Genovese, R. Juris, M. Marcaccio, M. Montalti, F. Paolucci, M. Sgarzi, G. Valenti, N. Zaccheroni and L. Prodi, Coord. Chem. Rev., 2012, 256, 1664 CrossRef CAS .
5. (a) T. Y. Wang, S. J. Fan, R. Erdmann and C. Shannon, Langmuir, 2013, 29, 16040 CrossRef CAS PubMed ; (b) M. Sentic, M. Milutinovic and F. Kanoufi, Chem. Sci., 2014, 5, 2568 RSC .
6. (a) N. E. Tokel and A. J. Bard, J. Am. Chem. Soc., 1972, 94, 2862 CrossRef CAS ; (b) L. Hu and G. Xu, Chem. Soc. Rev., 2010, 39, 3275 RSC ; (c) L. Dennany, R. J. Forster, B. White, M. Smyth and J. F. Rusling, J. Am. Chem. Soc., 2004, 126, 8835 CrossRef CAS PubMed .
7. (a) L. Li, Y. Chen and J. J. Zhu, Anal. Chem., 2016, 89, 358 CrossRef PubMed ; (b) W. Liang, Y. Zhuo and C. Xiong, Biosens. Bioelectron., 2017, 94, 568 CrossRef CAS PubMed ; (c) Y. Feng, C. Dai and J. Lei, Anal. Chem., 2015, 88, 845 CrossRef PubMed .
8. R. Golnak, J. Xiao and M. Pohl, Inorg. Chem., 2015, 55, 22 CrossRef PubMed .
9. Y. Ma, H. Zhang and J. Shen, Synth. Met., 1998, 94, 245 CrossRef CAS .
10. (a) M. J. Li, P. Jiao and M. Lin, Analyst, 2011, 136, 20 RSC ; (b) E. M. Gross, N. R. Amstrong and R. M. Wightman, J. Electrochem. Soc., 2002, 149, 137 CrossRef ; (c) Y. Zhou, H. Gao, X. Wang and H. Qi, Inorg. Chem., 2015, 54, 1446 CrossRef CAS PubMed ; (d) Y. Zhou, W. Li, L. Yu, Y. Liu, X. Wang and M. Zhou, Dalton Trans., 2015, 44, 1858 RSC ; (e) S. Zhang, M. Hosaka, T. Yoshihara, K. Negishi, Y. Iida, S. Tobita and T. Takeuchi, Cancer Res., 2010, 70, 4490 CrossRef CAS PubMed .
11. I. S. Shin, J. I. Kim and T. H. Kwon, J. Phys. Chem. C, 2007, 111, 2280 CAS .
12. J. M. Fernandez-Hernandez, E. Longhi, R. Cysewski, F. Polo, H. P. Josel and L. De Cola, Anal. Chem., 2016, 88, 4174 CrossRef CAS PubMed .
13. (a) K. L. Metera, K. D. Hanni, G. N. Zhou, M. K. Nayak, H. S. Bazzi, D. Juncker and H. F. Sleiman, ACS Macro Lett., 2012, 1, 954 CrossRef CAS ; (b) N. B. Sankaran, A. Z. Rys, R. Nassif, M. K. Nayak, K. Metera, B. Chen, H. S. Bazzi and H. F. Sleiman, Macromolecules, 2010, 43, 5530 CrossRef CAS ; (c) U. M. Tefashe, K. L. Metera and H. F. Sleiman, Langmuir, 2013, 29, 12866 CrossRef CAS PubMed .
14. (a) Z. He, G. Lai and Z. Li, Chin. J. Chem., 2015, 33, 550 CrossRef CAS ; (b) F. Malvolti, C. Rouxel and A. Triadon, Organometallics, 2015, 34, 5418 CrossRef CAS .
15. (a) Y. Liao, J. Hu and C. Zhu, J. Mol. Struct., 2016, 1108, 467 CrossRef CAS .
16. M. Hesari, M. S. Workentin and Z. Ding, ACS Nano, 2014, 8, 8543 CrossRef CAS PubMed .
17. (a) S. Sprouse, K. A. King and P. J. Spellane, J. Am. Chem. Soc., 1984, 106, 6647 CrossRef CAS ; (b) C. Talotta, L. Rubino and C. Gaeta, Supramol. Chem., 2016, 28, 403 CrossRef CAS .
18. C. W. Wu, C. M. Tsai and H. C. Lin, Macromolecules, 2006, 39, 4298 CrossRef CAS .
19. Y. Zhou, J. Jia, W. Li, H. Fei and M. Zhou, Chem. Commun., 2013, 49, 3230 RSC .
20. B. D. Stringer, L. M. Quan, P. J. Barnard, D. J. D. Wilson and C. F. Hogan, Organometallics, 2014, 33, 4860 CrossRef CAS .
21. Q. Zhao, M. X. Yu, L. X. Shi, S. J. Liu, C. Y. Li and M. Shi, Organometallics, 2010, 29, 1085 CrossRef CAS .
22. J. Y. Qiang, Y. Q. Xu and B. Tong, Inorg. Chim. Acta, 2013, 394, 184 CrossRef CAS .
23. I. Prieto, J. Teetsov and M. A. Fox, J. Phys. Chem. A, 2001, 105, 520 CrossRef CAS .
24. J. Truong, K. B. Spilstead and G. J. Barbante, Analyst, 2014, 139, 6028 RSC .
25. W. Yang, H. Fu, Q. J. Song, M. Zhang and Y. Q. Ding, Organometallics, 2011, 30, 77 CrossRef CAS .
26. D. Jiang, X. Du and Q. Liu, Chem. Commun., 2015, 51, 4451 RSC .
27. E. H. Doeven, E. M. Zammit and G. J. Barbante, Chem. Sci., 2013, 4, 977 RSC .
28. K. N. Swanick, M. Sandroni and Z. Ding, Chem. – Eur. J., 2015, 21, 7435 CrossRef CAS PubMed .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8an00426a

This journal is © The Royal Society of Chemistry 2018