Near-Infrared Luminescence from Visible-Light-Sensitized Ruthenium(II)-Neodymium(III) Heterobimetallic Bridged Complexes Containing Alkoxy(silyl) Functional Groups

New infrared emitting d-f (ruthenium(II)–neodymium(III)) heterobimetallic complexes with alkoxy(silyl) functional groups have been prepared. Visible excitation evidenced energy transfer processes from the ruthenium(II) donor to neodymium(III) acceptors leading to infrared emission. Energy transfer rates (k EnT ) and efficiency of energy transfer ( η EnT ) are, respectively, 0.61 × 10 7 s -1 and 44% for RuL 1 –NdL 3 complex. Larger values of k EnT (3.04 × 10 7 s -1 ) and η EnT (84%) were detected for RuL 2 –NdL 4 complex. RuL 1 –NdL 3 and RuL 2 –NdL 4 complexes were fully characterized by elementary analysis (EA), mass spectrometry (MS), Fourier transform infrared spectroscopy (FTIR) and Fourier transform Raman spectroscopy (FT-Raman). Total correlation spectroscopy (TOCSY1D), 1 H{ 13 C} heteronuclear single quantum correlation (HSQC) and 1 H{ 13 C} heteronuclear multiple bond correlation (HMBC) nuclear magnetic resonance (NMR) analyses were also carried out to characterize NdL 3 and RuL 1 –NdL 3 complexes. The presence of trialkoxysilyl-substituted ligands would allow further grafting onto any silica or silicated surface aiming at applications as new luminescent near infrared (NIR)-emitting biosensors or biomarkers.


Introduction
The design and assembly of heterobimetallic complexes based on d-block metalloligands with near-infrared (NIR) emitting lanthanides (Ln) have attracted interest in the recent years due to potential applications in non-invasive bio-analysis and bio-imaging. [1][2][3][4][5][6] Given the inherent low molar absorption coefficients of lanthanide(III) ions (ε < 10 mol L −1 cm −1 ), d-block chromophores serve as effective light-harvesting groups to sensitize Ln-emissive centres. 1 Beyond that, most of lanthanide complexes require excitation in the ultraviolet (UV) region, which is harmful for the biological systems. The aromatic residues of proteins and deoxyribonucleic acid (DNA) absorb in competition with the chromophores. In addition, the UV excitation cause damages in these biological systems. 1 One of the strategies to overcome these limitations consists on using d-block chromophores as antenna groups to sensitize luminescence from lanthanide(III) ions with low energy f-f excited states.
Typically, ruthenium(II) complexes displaying polypyridine bridging ligands are used to sensitize NIR emitting lanthanide ions through the strong absorption of the visible light by 3 MLCT (MLCT = metal-to-ligand charge transfer) excited state. 1 These d-block chromophores, with relatively low 3 MLCT states, act as good energy donors to f-f levels of praseodymium(III), neodymium(III), erbium(III) and ytterbium(III). In particular, [Ru(bpy) 3 ] 2+ (bpy = 2,2'-bipyridine) possesses a broad and strong absorption in the visible and a low triplet state (energy of triplet state = 17400 cm -1 ) that can allow lanthanide(III) to be efficiently excited. 1 Authors 7 reported the use of distinct lengthy bridging ligands (saturated or unsaturated) between energy donor (d-block) and acceptor fragments (f-block) and concluded that the role of ligands is vital in determining the ruthenium(II) → lanthanide(III) energy transfer (EnT), and the energy level of triplet state was related with the bridging processes. 7 Pyridine derivate ligands can also serve as bridging ligands to prepare ruthenium(II)-lanthanide(III) heterobimetallic d-f complexes. Specifically, 2,2'-bipyrimidine (bpmd) is widely used to synthesize planar heterobimetallic complexes since it can easily coordinate two metal centers through four nitrogen sites leading to a relatively short metal-metal distance. As a result, an efficient channel is available for energy and electron transfer from d-block chromophores in the visible region to NIR emitting lanthanide(III) centres. [7][8][9][10][11][12][13][14] Despite the examples of ruthenium(II)-lanthanide(III) heterobimetallic complexes being reported in the last few years, their use as luminophors for biosensors or biomarkers is yet scarce. The most likely reasons are their low water solubility and relatively weak luminescence. 15 As an alternative to explore the potentialities of such heterobimetallic complexes in biological applications, these complexes need to be functionalized to be covalently anchored on silica nano and microparticles. We have demonstrated the preparation of bipyridine and diketone derivative ligands such as bpy-Si (4-methyl-4'-[methylamino-3(propyltriethoxysilyl)]-2,2'-dipyridine) 16 and TTA-Si (4,4,4-trifluoro-2-(3-(trimethoxysilyl)-propyl)-1-(2-thienyl)-1,3-butanedione) 17 appropriately substituted by a trialkoxysilyl groups. Sequentially, silylated ruthenium(II) and lanthanides(III) complexes have been isolated and grafted at the surface [16][17][18][19][20][21] or embedded 17,18,21,22 in nanosized silica. These previous results stimulated us to design d-f heterobimetallic complexes by introducing silylated appropriated ligands, bpy-Si or TTA-Si, involving NIR emitting lanthanides as neodymium(III) and bipyrimidine as bridging d-f ligand.
In this way, two routes were employed to obtain two types of dyads based on the chemistry of bipyridine ruthenium(II) and diketonate neodymium(III) derived complexes according to the position of the silylated functional group. For the sake of brevity we will use a labelling Scheme 1 whereby the complexes are referred to as: RuL 2 -NdL 4 dinuclear complexes, bearing the alkoxysilyl group on bipyridine ruthenium moiety (i), and RuL 1 -NdL 3 bearing the alkoxysilyl groups on the TTA lanthanide complex (ii). In both cases bipyrimidine is the bridging ligand on ruthenium center. Successful structural characterization of the silylated d-f heterobimetallic complexes is described and discussed in this work. Furthermore, photophysical properties of these heterobimetallic complexes are examined with respect to the efficiency of energy transfer processes from ruthenium(II) moieties to the luminescent NIR-emitting neodymium(III) ions.

Characterization methods
Fourier transform infrared (FTIR) spectra were obtained in the spectral range from 4000 to 650 cm -1 (4 cm -1 of resolution) with a Bruker Vector 22 (KBr dispersion). The Raman spectra from all complexes were collected on a RFS 100 FT-Raman Bruker spectrometer equipped with a Ge detector using liquid nitrogen as the coolant and a Nd:YAG laser emitting at 1064 nm. The laser light, with a power varying from 30 to 150 mW, was introduced and focused on the sample, and the scattered radiation was collected at 180°. For each spectrum, an average of 1024 scans was performed at a resolution of 4 cm -1 over a range from 3500 to 50 cm -1 . The OPUS 6.0 (Bruker Optik, Ettlingen, Germany) software was used for Raman data acquisition. For the lanthanide and heterobimetallic complexes, nuclear magnetic resonance (NMR) spectra were obtained using a Bruker Advance 600 MHz spectrometer at room temperature in deuterated solvents. Total correlation spectroscopy (TOCSY1D) spectra were recorded by selective irradiation of different aromatic hydrogen atoms of the complex with a selective pulse. The spectra were recorded after magnetization transfer at 120 ms of mixing time. 1 H{ 13 C} heteronuclear single quantum correlation (HSQC) NMR experiments were recorded using the spectral widths of 12 000 and 24 000 Hz for the 1 H and 13 C dimensions, respectively. The number of collected complex points was 1024 for the 1 H dimension with a recycle delay of 1.5 s. The number of transients was from 24 to 48 depending of the sample and 128 time increments were recorded in the 13 C dimension. The 1 J H-C used was 145 Hz. Prior to Fourier transformation, the data matrices were zero filled to 1024 points in the 13 C dimension. 1 H{ 13 C} heteronuclear multiple bond correlation (HMBC) NMR experiments were carried out using the spectral widths of 12000 and 24000 Hz for the 1 H and 13 C dimensions, respectively. The number of collected complex points was 1024 for the 1 H dimension with a recycle delay of 2.0 s. The number of transients was from 16 to 32 depending of the sample and 128 time increments were recorded in the 13 C dimension. The 3 J H-C used was 8 Hz. Prior to Fourier transformation, the data matrices were zero filled to 1024 points in the 13 C dimension. Data processing was performed using standard Bruker Topspin-NMR software. The central solvent line was used as an internal chemical shift reference point. Mass spectra (MS) were recorded using Q TOF 1er (Waters) spectrometer: source electrospray (ESI) or DSQ Thermo Fisher Scientific spectrometer equipped with chemical ionization source (NH 3 ). UV-Vis spectra were recorded Scheme 1. Synthesis of the heterobimetallic complexes, RuL 1 -NdL 3 (i) and RuL 2 -NdL 4 (ii). Numbering of the hydrogen and carbon atoms are identified as red and blue labels.
using a Varian spectrophotometer model Cary 5000 in the region of 800 to 200 nm (resolution of 1 nm) using Cary WinUV software. Elemental analyses of C, H, N and S were performed on a Carlo Erba instrument (EA 1110). Luminescence spectra were measured at room temperature using a Jobin-Yvon Model Fluorolog FL3-22 spectrometer equipped with a H10330-75 Hamamatsu detector, TE: cooled NIR-photomultiplier module and a 450 W Xe excitation lamp. Excitation and emission spectra were recorded under CW excitation and were corrected with respect to the Xe lamp intensity and spectrometer response. Fluorescence intensity decays were obtained using the time-correlated single-photon counting technique. The excitation source was a mode-locked Ti:saphire laser (Tsunami 3950 pumped by Millennia X Spectra Physics) producing 5 ps FWHM pulses ranging from 0.5 to 8.0 MHz repetition rate, regulated by the 3980 Spectra Physics pulse picker. The laser was tuned to give output at 892 nm and a second harmonic generator BBO crystal (GWN-23PL Spectra Physics) gave the 448 nm excitation pulses that were directed to an Edinburgh FL900 spectrometer adjusted in L-format configuration. The emission wavelength of 620 and 670 nm were selected by a monochromator and emitted photons were detected by a cooled Hamamatsu R3809U microchannel plate photomultiplier. The whole instrument response function was typically 100 ps. Energy transfer rate constant (k EnT ) and efficiency of energy transfer (η EnT ) were obtained using ruthenium 3 MLCT decay values from ruthenium precursors and the respective heterobimetallic complexes.

Synthesis [Nd(TTA-Si) 3 ] (NdL 3 )
This complex was obtained by the procedure adapted from Duarte et al. 17 6 mmol (2.30 g) of the TTA-Si and 2 mmol (0.50 g) of neodymium chloride anhydrous were added to 20 mL of anhydrous ethanol. The reaction mixture was kept under nitrogen atmosphere and stirred for 18 h. The solvent was evaporated and the powder was then washed with pentane and diethylether, and dried under vacuum. The Nd(TTA-Si) 3  This complex was obtained by the procedure described by Duarte et al. 17

Results and Discussion
Synthesis and structural characterization of silylated ruthenium(II)-neodymium(III) heterobinetallic complexes The synthesis routes of three new mononuclear complexes denoted by (i) RuL 1 , (ii) RuL 2 and (iii) NdL 3 has been achieved in Supplementary Information (SI) section ( Figure S1). Equimolar reaction of ruthenium complex bearing the bridging ligand, RuL 1 , with silylated tris(diketonate) neodymium complex, NdL 3 , gives rise to RuL 1 -NdL 3 heterobimetallic complex (i, Scheme 1). On the other hand, the combination of RuL 2 complex, containing both silyl and bridging groups, with tris(diketonate) neodymium complex, Nd(TTA) 3 denoted by NdL 4 (synthesis route and structure of NdL 4 were done in Figure S2, SI section), leads to heterobimetallic RuL 2 -NdL 4 complex with the trialkoxysilyl functional group on the ruthenium moiety (ii, Scheme 1).
FTIR spectra (Figure 1 Raman spectroscopy allows accurate assignment of CCH stretching vibration bands at 1028, 1040, 1202, 1490 and 1580 cm -1 of heterocycles of the bpy and bpmd ligands of ruthenium units in RuL 1 -NdL 3 and RuL 2 -NdL 4 heterobimetallic complexes. The ruthenium(II) coordination was confirmed by the band observed at 1030 cm -1 (v Ru-N ). 23,24 Both spectroscopies appeared well complement one another. Raman scattering spectra are shown in Figure S3 (SI section).
The steric arrangement of the ligands in the metal coordination sphere in NdL 3 and RuL 1 -NdL 3 complexes have been elucidated by assignment of chemical shifts, J H-C and J H-H coupling constant and using complementary NMR studies such as TOCSY1D, 1 H{ 13 C}-HSQC and 1 H{ 13 C}-HMBC experiments ( Figures S4, S5 for NdL 3 and S6-S9 for RuL 1 -NdL 3 , SI section). Numbering of the hydrogen and carbon atoms used in the further characterization is detailed in Figure S1 (SI section) and Scheme 1 for each complexes. Most probable because of the presence of the silyl group, compounds behave as hygroscopic powders and all attempts to grow single crystals have been unsuccessful. Figure 2 shows the UV-Vis absorption spectra for the two types of heterobimetallic complexes and respective precursors. Figure 2a black line, shows the spectrum obtained for the RuL 1 complex. Strong and broad bands are observed at 245 and 424 nm attributed to d→π* 1 MLCT transitions (RuL 1 to bpy and bpmd ligands, respectively) and a strong and sharp band at 286 nm attributed to p→π* transitions of bpy and bpmd ligands. 25 For NdL 3 complex (Figure 2a, red line), the absorption bands at 269 and 340 nm were assigned to singlet-to-singlet transitions in the TTA-Si ligands. 17 The spectrum obtained for RuL 1 -NdL 3 complex (Figure 2a  The RuL 2 -NdL 4 (Figure 3b, blue line) displayed similar transitions bands at 1066 ( 4 F 3/2 → 4 I 11/2 ) and 1334 nm ( 4 F 3/2 → 4 I 13/2 ), respectively. 9,25,26 Excitation spectra obtained by monitoring the NIR emission (λ em : 1065 and 1066 nm for RuL 1 -NdL 3 and RuL 2 -NdL 4 , respectively) are also shown in Figures 3a-3b, red lines. For both heterobimetallic complexes, three sharp bands were detected in visible range and ascribed to neodymium(III) intraconfigurational f-f transitions at 581, 746 and 800 nm. 9,26 The broad band from 270 to 400 nm were assigned to transitions centered at TTA-Si, bpy and bpmd ligands. 14,17,18,26 Bands ranging from 400 to 600 nm were attributed to the ruthenium(II) 3 MLCT transition. 25 Further information on the energy transfer from the ruthenium counterpart to the lanthanide one is these heterobimetallic compounds can be obtained by analyzing the changes observed for the Ru visible emission properties (Figures S10 and S11, SI section). Table 1 shows results for excited state lifetimes. Considering bpmd as a planar bridging ligand that provides short metal-metal distance and a pathway that favors efficient energy transfer in heterobimetallic complexes, especially for neodymium(III) and ytterbium(III) ions, 10 the lifetime values obtained for RuL 1 and RuL 2 substantially reduced in RuL 1 -NdL 3 and RuL 2 -NdL 4 complexes suggesting energy transfer processes for both heterobimetallic complexes. The k EnT and η EnT may be evaluated from lifetimes as shown in Table 1.

Photophysical properties
A pathway of energy transfer (EnT) is presented in Scheme 2. The proposed energy level diagram suggests that the RuL 2 and RuL 1 3 MLCT energy are suitable for the sensitization of the neodymium(III) luminescence. The RuL 2 complex 3 MLCT energy level is centered at 16129 cm -1 and the energy transfer to neodymium(III) could be evaluated. Considering the RuL 2 -NdL 4 complex,  the k EnT and η EnT were 3.04 × 10 7 s -1 and 84%, respectively. For RuL 1 complex, the 3 MLCT energy level is located at 14925 cm -1 . Values of k EnT and η EnT for the RuL 1 -NdL 3 were 0.61 × 10 7 s -1 and 44%, respectively. It is worth emphasizing that higher EnT rate was observed for RuL 2 -NdL 4 complex. This observation could be explained considering that neodymium(III) presents excited levels better located for energy transfer from RuL 2 3 MLCT states.

Conclusions
New ruthenium(II) complexes, named RuL 1 and RuL 2 , both containing the bridging ligand bpym and the last one containing also a silylated bpy ligand (L 2 ) have been prepared. From these complexes and new NdL 3 diketonate complexes (where L 3 is methoxysilyl modified diketonate), ruthenium(II)-neodymium(III) heteronuclear compounds have been prepared. A structural study was performed by EA, MS, FTIR and FT-Raman and 1D and 2D NMR techniques confirming the presence of the alkoxysilyl function and the fabrication of two new silylated d-f heterobimetallic complexes. UV-Vis spectra depicted characteristic transition bands in ruthenium(II) and neodymium(III) mononuclear precursors and both transition bands contributions in RuL 1 -NdL 3 and RuL 2 -NdL 4 complexes corroborating all structural analysis. The luminescent properties of the silylated d-f heterobimetallic complexes were evaluated with energy transfer processes being established from the ruthenium(II) donor to neodymium(III) acceptor units in the IR region. The decrease of lifetime and ruthenium quantum yield resulting from ruthenium(II) units compared to ruthenium(II)neodymium(III) confirm effective energy transfer in these systems. For RuL 1 -NdL 3 complex, k EnT and η EnT values of 0.61 × 10 7 s -1 and 44% were obtained whereas RuL 2 -NdL 4 complex showed k EnT and η EnT values of 3.04 × 10 7 s -1 and 84%, respectively. Our results emphasize that new silylated heterobimetallic ruthenium(II)-neodymium(III) complexes can be excited in the visible range via ruthenium(II) MLCT transitions promoting energy transfer to neodymium(III) units emitting in the IR region. The successful preparation of the silylated ruthenium(II)-neodymium(III) complexes allow grafting onto any silicated matrix suggesting their application as new luminescent NIR-emitting biosensors or biomarkers.