SHG-active luminescent thermometers based on chiral cyclometalated dicyanidoiridate(iii) complexes

Multifunctional optical materials can be realized by combining stimuli-responsive photoluminescence (PL), e.g., optical thermometry, with non-linear optical (NLO) effects, such as second-harmonic generation (SHG). We report a novel approach towards SHG-active luminescent thermometers achieved by constructing unique iridium(iii) complexes, cis-[IrIII(CN)2(R,R-pinppy)2]− (R,R-pinppy = (R,R)-2-phenyl-4,5-pinenopyridine), bearing both a chiral 2-phenylpyridine derivative and cyanido ligands, the latter enabling the formation of a series of molecular materials: (TBA)[IrIII(CN)2(R,R-pinppy)2]·2MeCN (1) (TBA+ = tetrabutylammonium) and (nBu-DABCO)2[IrIII(CN)2(R,R-pinppy)2](i)·MeCN (2) (nBu-DABCO+ = 1-(n-butyl)-1,4-diazabicyclo-[2.2.2]octan-1-ium) hybrid salts, (TBA)2{[LaIII(NO3)3(H2O)0.5]2[IrIII(CN)2(R,R-pinppy)2]2} (3) square molecules, and {[LaIII(NO3)2(dmf)3][IrIII(CN)2(R,R-pinppy)2]}·MeCN (4) coordination chains. Thanks to the chiral pinene group, 1–4 crystallize in non-centrosymmetric space groups leading to SHG activity, while the N,C-coordination of ppy-type ligands to Ir(iii) centers generates visible charge-transfer (CT) photoluminescence. The PL characteristics are distinctly temperature-dependent which was utilized in achieving ratiometric optical thermometry below 220 K. The PL phenomena were rationalized by DFT/TD-DFT calculations indicating an MLCT-type of the emission in obtained Ir(iii) complexes with the rich vibronic structure providing a few emission bands that variously depend on temperature due to the role of thermally activated vibrations. As these crucial vibrational modes depend on the crystal lattice, the thermometry performance differs within 1–4 being the most efficient in 4 while the SHG is by far the best also for 4. This proves that pinene-functionalized cyclometalated dicyanidoiridates(iii) are great prerequisites for tunable PL-NLO conjunction with the most effective multifunctionality ensured by the insertion of these anions into bimetallic frameworks.


Synthesis of 1
The three-step synthesis of 1 started by mixing iridium(III) chloride hydrate (200 mg, 0.67 mmol, 1.0 eq) and (R,R)-pinppy (417 mg, 1.68 mmol, 2.5eq) with 15 ml of ethoxyethanol and 5 mL of water.After 24 h refluxing this mixture, the resulting solid product was precipitated with water, filtrated on a sintered glass funnel, washed with water and pentane, and dried.The yellow solid was dissolved in dcm, dried over anhydrous magnesium sulfate, and evaporated to dryness.(Yield, 436 mg, ca.90%).Secondly, the as-prepared [Ir III 2(μ2-Cl)2(R,R- pinppy)4] (436 mg, 0.30 mmol, 1 eq) was dissolved in the mixture of methanol (100 ml) and dcm (50 mL).Then, KCN (80 mg, 1.23 mmol, 4.1 eq) was added and the reaction mixture was refluxed for 24 h.After being cooled, the reaction mixture was filtered, and the resulting filtrate was evaporated.The resulting potassium salt, K[Ir III (CN)2(R,R-pinppy)2] was dried and used directly in the last step.Therefore, the entire product was dissolved in 100 ml of methanol, and 215 mg of tetrabutylammonium (TBA) perchlorate (0.63 mmol, ca.two-time excess) was added.The mixture was refluxed for 24 h and the resulting precipitate was filtered.The filtrate was evaporated to dryness, dissolved in a small amount of dichloromethane (dcm), filtered, and the product was precipitated by diethyl ether.The two last steps resulted in 560 mg (95% yield) of the yellow powder of the objective compound 1.A single crystal of 1 was obtained by slow diffusion of diethyl ether into an acetonitrile (MeCN) solution of the compound.Its composition, TBA[Ir III (CN)2(R,R-pinppy)2]•2MeCN (MW = 1065.47g⋅mol −1 , 1), was determined by single-crystal X-ray diffraction (SC-XRD) analysis, while the phase purity and its airstability were proven by the powder X-ray diffraction (P-XRD) method (Fig. S8).The crystals of 1 are quite hygroscopic and, while left in the air, they adsorb water molecules (generating the hydrated form of 1⋅4H2O), which does not cause any distinct structural transformation, as evidenced by the results of the P-XRD and CHN elemental analysis.These water molecules can be easily removed by an inert gas purge, which was confirmed by the results of the TG experiment (see Fig. S2 with the related comment).Elemental analysis calculated for the hydrated form, 1⋅4H2O (C58H86Ir1N7O4; MW = 1137.55g mol −1 ): C, 61.24%; H, 7.62%; N, 8.62%.Found: C, 61.08%; H, 7.54%; N, 8.34%.TG (Fig. S2 with comments): loss of 2 MeCN molecules together with the loss of 1 remaining H2O molecule per formula unit, calculated: 9.24 %; found: 9.35 % (as the other water molecules are removed before the start of the TG measurement).IR spectrum (see Fig. S1 with the related comment): the bands located at 2104 and 2096 cm −1 can be assigned to the stretching vibrations of terminal cyanido ligands, proving the presence of these ligands in the structure.S5,S6 In addition to the set of physical methods presented above, compound 1 was also investigated by the 1 H NMR method in the dissolved form.

Synthesis of 2
The 25 mg portion of freshly filtrated crystals of 1 (0.025 mmol, 1.0 eq) were dissolved in a mixture of methanol (1 ml) and acetonitrile (1 mL), and the 15 mg portion of n-BuDABCO + I -(0.05 mmol, 2 eq) was added directly to this solution.After 5 min of stirring, the mixture was filtrated and diethyl ether was layered on top of the solution.After 2 weeks, a large number of yellow plate crystals of 2 appeared.Yield, 15 mg (50%).Their composition, (nBu-DABCO)2[Ir III (CN)2(R,R-pinppy)2](I)•MeCN (MW = 1247.43g⋅mol −1 , 2⋅MeCN), was determined by SC-XRD analysis, Exposure of the crystals of 2 to the air causes the loss of acetonitrile molecules and adsorption of water molecules (generating the hydrated form, 2⋅2H2O).It does not cause any structural transformation which was confirmed by P-XRD results (Fig. S8), CHN elemental analysis, and TG experiment (see Fig. S2 with the related comment).

Synthesis of 3
The equimolar solution of freshly filtrated crystals of 1 (31 mg, 0.0315 mmol) and lanthanum nitrate hexahydrate (14 mg, 0.0315 mmol) in 2 mL of acetonitrile and 0.1 ml of MeOH was slowly diffused by diethyl ether.After a few days, well-shaped yellow crystals of 3 appeared.Yield, 26 mg (63%).Their composition, (TBA)2{[La III (NO3)3(H2O)0.5]2[IrIII (CN)2(R,R-pinppy)2]2} (MW = 2632.60g⋅mol −1 , 3), was determined by SC-XRD analysis, while the purity of the phase together with its air stability were proven by the P-XRD method (Fig. S8).Similarly to 1, the exposition of 3 in the air causes the surface adsorption of water molecules (generating the hydrated form 3⋅4.5H2O), occurring without any structural transformation, which was confirmed by P-XRD, CHN elemental analysis, and TG results.Note that one coordinated water molecule per the formula unit is present in 3 while another 4.5 water molecules per the formula unit appear as the solvent of crystallization in the hydrated form, 3⋅4.5H2O.Thus, in the TG experiment, a loss of the total number of 5.5 water molecules per the formula unit is observed (for details regarding the TG, see Fig. S2 with the related comment).

Synthesis of 4
The yellow crystals of 4 were obtained by slow diffusion of diethyl ether into the equimolar solution of freshly filtrated crystals of 1 (31 mg, 0.0315 mmol) and lanthanum nitrate hexahydrate (14 mg, 0.0315 mmol) in the solvents mixture containing 1 ml of acetonitrile, 0.1 ml of methanol and 0.1 mL of dimethylformamide.Yield, 22 mg (55%).Their composition, {[La III (NO3)2(dmf)3][Ir III (CN)2(R,R-pinppy)2]}•MeCN (MW = 1264.18g⋅mol −1 , 4), was determined by SC-XRD analysis, while the phase purity together with its air stability were proven by the P-XRD method (Fig. S8).In the air, the crystals of 4 adsorb water molecules (generating the hydrated form of 4⋅2H2O), which can be removed by an inert gas purge, as confirmed by P-XRD, CHN elemental analysis, and TG results (for details, see Fig. S1 with the related comment).
Elemental analysis calculated for the hydrated form, 4⋅2H2O (C49H64Ir1La1N10O11; MW = 1300.22g mol −1 ): C, 45.26%; H, 4.96%; N, 10.77%.Found: C, 45.54%; H, 4.97%; N, 10.51% TG (Fig. S2 with the comment): loss of 3 dmf molecules together 1 MeCN molecule per formula unit, calculated: 20.59%; found: 20.30%.IR spectrum (Fig. S1 with the related comment): the bands situated at 2119 and 2092 cm −1 can be assigned to cyanido ligands; they are noticeably shifted towards higher energy than, e.g., in the ionic salt of 2 which indicates their bridging character.S5 Additional comment on the description of syntheses of compounds 1, 3, and 4: All obtained materials are generally air-stable; however, the prolonged exposition to the air leads to the adsorption of small amounts of water molecules (as determined by the CHN elemental analysis).As confirmed by structural studies, this process does not cause structural changes.We suppose that this adsorption takes place only on the surface of fine crystalline samples of 1, 3, and 4 because most of these water molecules can be easily and quickly removed by an inert gas purge (the detailed analysis of this effect is presented as the comment to Fig. S2, see below).Therefore, all physical measurements were carried out on freshly filtrated compounds which eliminated the adsorption of moisture as much as possible.

Structural studies
Single crystal X-ray diffraction (SC-XRD) analysis for all compounds was performed at 100(2) K, using a Bruker D8 Quest Eco Photo50 CMOS diffractometer, equipped with the Mo Kα (0.71073 Å) radiation source and a graphite monochromator.The single crystals of (R,R)-pinppy ligand and compounds 1-4 selected for the SC-XRD experiments were taken directly from the respective mother solutions, covered by Apiezon N grease, and mounted on the dedicated Micro Mounts TM holders.The SAINT and SADABS programs were used for data reduction and cell refinement processes.The absorption correction was performed using a multi-scan procedure with the help of the TWINABS program.The crystal structures were solved by an intrinsic phasing method using a SHELXT program implemented within the Apex3 package.S7 The further refinements were carried out by a weighted full-matrix least squares method on F 2 of SHELX-2014/7 within the WinGX (ver.2018.3)software.S8 All non-hydrogen atoms were anisotropically refined, while hydrogen atoms for (R,R)-pinppy ligands, aliphatic nBu-DABCO + , and TBA + cations, as well as MeCN solvent molecules were calculated in their idealized positions and refined using a riding model.The methine hydrogen atoms in dmf molecules were found directly from an electron density map.A reasonable number of restraints of the DFIX, ISOR, and DELU types were applied for the selected non-hydrogen atoms of (R,R)-pinppy ligands, counter-cations, coordinated (but highly disordered) nitrate(V) anions, and the part of solvent molecules.It was done to ensure the proper geometries and the convergence of the respective refinement procedures.Due to the significant disorder, two DFIX restraints were applied to the La−O bond distances in 3 to ensure its proper geometry and keep the convergence of the refinement.For similar reasons, the FLAT restraint was applied for one of the coordinated NO3 -anions in the crystal structure of 3. Some of the reflections with intensities endowed with especially large errors (affected by the beamstop) were removed from the final refinement using the OMIT commands.Using all of these procedures, the sets of satisfactory refinement parameters were achieved.The reference CCDC numbers for the crystal data of (R,R)-pinppy and compounds 1, 2, 3, and 4 are 2297257, 2297255, 2297258, 2297256, and  2297254, respectively.Details of the crystal data and refinement of the structure are summarized in Table S1, while representative structural parameters are gathered in Tables S2-S6.The structural figures (Fig. 2, 3, and S3-S7) were prepared using Mercury 3.8 software.Powder XRD data were collected using a Bruker D8 Advance Eco powder diffractometer equipped with a Cu Kα (1.5419 Å) radiation source.The P-XRD measurements were conducted at room temperature for the dried polycrystalline samples inserted into the 0.5 mm glass capillaries using the appropriate commercial experimental setup for rotating capillaries (Fig. S8).

Physical techniques
The 1 H and 13 C{ 1 H} NMR spectra were recorded for the proper solutions of investigated organic compounds at room temperature using a Bruker AVANCE 600 MHz spectrometer, while the 1 H NMR spectrum for 1 (in solution) was recorded at room temperature using a Jeol 400 MHz ECZR spectrometer (see the details in the Synthesis section above).All solid-state physical measurements were performed on freshly prepared and filtrated samples (see the comment to the description of the syntheses above).CHN elemental analyses were performed with standard microanalysis procedures using the Elementar Vario Micro Cube CHN analyzer.The infrared (IR) absorption spectra were collected on a Nicolet iN10 MX FT-IR microscope in transmission mode.Measurements were made in the range of 3800−670 cm −1 for selected single crystals placed on CaF2 windows.Thermogravimetric (TG) measurements were performed on a NETZSCH TG 209 F1 Libra apparatus under inert gas at a heating rate of 1 °C‧min -1 in the temperature range of 20-400 °C.The second harmonic generation (SHG) experiment was carried out on a homemade optical setup using a 1040 nm femtosecond laser as an excitation light source.This setup is described in our previous work.S9 To verify the SHG effect, for all samples 1-4, the power and wavelength dependencies of the output light were measured.To quantify the SH intensities measured in this setup, a potassium dihydrogen phosphate (KDP) was used as a reference sample.Solid-state UV-vis absorption spectra were measured in the range of 220−750 nm on a Shimadzu UV-3600i plus spectrometer using the thin films of power samples inserted between quartz plates.Solid-state photoluminescent properties were measured using an FS5 spectrofluorometer equipped with an Xe (150 W) arc lamp as an excitation source and a Hamamatsu photomultiplier of the R928P type as a detector.Emission lifetime measurements were conducted on the FS5 spectrofluorometer using a time-correlated single photon counting method with EPLED-380 picosecond pulsed light emitting diode (374.4 nm).The temperature-variable emission and excitation spectra were collected on the same spectrofluorometer using a CS204SI-FMX-1SS cooling power optical helium cryostat which is equipped with a DE-204SI closed cycle cryo-cooler (cold head), water-cooled He compressor (ARS-4HW model), and a model 335 cryogenic temperature controller.For all types of photoluminescent measurements, freshly prepared powder samples were placed between two quartz plates.Absolute luminescence quantum yields (QY) were determined by a direct excitation method using an integrating sphere module for the FS5 spectrofluorometer and barium sulfate as the reference material, following the method described in our previous work.S10 Luminescent background corrections were performed within the Fluoracle software.

Comment on Figure 2:
The results of the CHN elemental analysis indicate that the samples adsorb small amounts of water from the air.Moreover, its absorption/removal does not result in a structural transformation, and the above-presented TG results show that most of this solvent is removed in an inert gas purge (even before the start of the TG measurement).Thus, the weight loss in the TG curve of 1 corresponds to two crystallizing acetonitrile molecules and some residual water (about 1 molecule per formula unit).The exposition of the crystals of 2 to air results in the loss of acetonitrile molecules and the adsorption of water molecules.As a result, the thermogravimetric curve of this compound shows the loss of two water molecules.The mass change in the TG curve in 3 corresponds to the weight of 5.5 water molecules per unit of formula, which is consistent with the results of CHN elemental analysis (so, in this case, water molecules are not that easily removed as for the previous cases).Note that among these 5.5 water molecules per the formula unit, one molecule is coordinated to the La(III) center while the remaining 4.5 molecules serve as the solvent of crystallization; however, they are removed under heating in a single, rather featureless, broad step.The sample of 4, similar to 1 and 2, when exposed to air, adsorbs about two water molecules per formula unit, which can be removed by a gas purge.As the temperature increased, the removal of the coordinated dmf molecules and the co-crystallizing acetonitrile molecules can be observed.All compounds, 1−4 are losing the solvents up to ca. 250 o C. Further heating of the samples causes a rapid decrease in mass, which is related to the decomposition of aliphatic cations and ligands and/or the removal of cyanido ligands, which causes the decomposition of the sample.Table S2.Detailed structure parameters of 1.  Table S3.Detailed structure parameter of 2.    Comment on Table S6: Continuous Shape Measure (CShM) analysis for Ir(III) and Ln(III) complexes was performed using SHAPE software ver.2.1.21.S11 The Continuous Shape Measure (CShM) parameter represents the distortion from an ideal geometry.It equals 0 for an ideal polyhedron and increases with increasing distortion.Due to the significant disorder in the crystal structure of 3 (see Experimental Details), some DFIX restraints were applied for the La−O bond distances to ensure the convergence of the refinement procedure.Therefore, the CShM analysis is burdened with a large experimental error.However, for the record, the best geometries for the nine-coordinated lanthanum(III) complexes in 3 are also found as MFF-9 (Cs), while the tencoordinated complexes are best characterized by the geometry of JSPC-10 (sphenocorona, C2v).Comment on the SHG effect shown in Figures S9 and S10: According to the Kleinman symmetry in nondispersive materials, the SHG effect should not be observed in the 422, 432, and 622 point groups.S14,S15 Thus, we suspected that the weak but non-negligible signal obtained for the powder sample of 3 could be connected with the contamination with the precursor or other molecular material; however, the P-XRD analysis (Fig. S8) showed that this sample is homogenous and does not contain any crystalline impurities.After further literature research, we were able to confirm that Kleinman symmetry is often violated for molecular materials apart from the quality of the crystal sample.S16-S18 Therefore, we suppose that the results obtained for 3 are reliable.S7.The emission spectra in the (b) part were normalized to the third (going from lower to higher wavelengths) emission peak.

Selected bond distances in cis-
Table S7.The positions of well-distinguished maxima of the solid-state emission patterns of 1 detected at various indicated temperatures (the position of the main maximum at each temperature was underlined), shown together with the chromaticity parameters of the CIE 1931 scale (Fig. S13).S8.The emission spectra in the (b) part were normalized to the third (going from lower to higher wavelengths) emission peak.
Table S8.The positions of well-distinguished maxima of the solid-state emission patterns of 2 detected at various indicated temperatures (the position of the main maximum at each temperature was underlined), shown together with the chromaticity parameters of the CIE 1931 scale (Fig. S14).S9.The emission spectra in the (b) part were normalized to the third (going from lower to higher wavelengths) emission peak.
Table S9.The positions of well-distinguished maxima of the solid-state emission patterns of 3 detected at various indicated temperatures (the position of the main maximum at each temperature was underlined), shown together with the chromaticity parameters of the CIE 1931 scale (Fig. S15).S10 and S11.See the comment below for details.The emission spectra in the (b) part were normalized to the third (going from lower to higher wavelengths) emission peak.
Table S10.The positions of well-distinguished maxima of the solid-state emission patterns of 4 detected at various indicated temperatures (the position of the main maximum at each temperature was underlined), shown together with the chromaticity parameters of the CIE 1931 scale (Fig. S16).
T     The mono-exponential fitting was applied for each temperature.The best-fit parameters are roughly presented on the graphs while the detailed values are gathered in Table S12.
Table S12.Best-fit parameters for the room temperature emission decay profiles of 1 and 2 at indicated emission wavelengths.Comment on the determination of emission lifetimes: It was possible to measure the room-temperature emission lifetimes for three different emission components only for 1 and 2 (Figure S18 and Table S12).For two other compounds, 3 and 4, only the general emission lifetime for the main emission component was investigated; however, for 3 and 4, it was possible to gather also the temperature dependences of these emission lifetimes (see Figures S19, S20, and S21, and Table S13).

Figure S20.
Temperature-variable emission decay profiles for 3 under λexc = 374 nm and λem = 500 nm, gathered in the 10-290 K temperature range.The mono-exponential fitting was applied for each temperature.The bestfit parameters are roughly presented on the graphs while the detailed values are gathered in Table S13.S13.
Table S13.Best-fit parameters for the emission decay profiles of 3 and 4 to the mono-exponential decay function for the indicated temperature from the 10-290 K range (Figures S18 and S19).

Computational Details
The theoretical calculations for 1 were carried out using the ORCA 5.0.1 quantum chemistry program package.S20 In the first step, the geometry of an anionic complex [Ir III (CN)2(R,R-pinppy)2] -, consisting of Ir(III) center together with the whole surrounding (R,R)-pinppy ligands and cyanido ligands, taken from the SC-XRD experiment, was optimized using DFT methods.In our treatment, we completely omitted counter ions present in the crystal structure of 1.The structure was optimized using the B3LYP hybrid exchange-correlation energy functional which presented its reasonable performance for predicting the geometry parameters of ground and excited states, S21,S22 as well as, excitation energies for various organic and metal transition compounds.S23,S24 The def2-TZVP basis set was used together with the charge-dependent atom-pairwise dispersion correction using D4(EEQ)-ATM model.S25,S26 For the calculations, the LR-CPCM solvation model was used with chloroform as a solvent which is consistent with the solvent used for registering the experimental spectra.S27 The comparison of the SC-XRD structure with the optimized ground state is presented in Figure S23 together with the chosen geometry parameters compared to the experimental ones in Table S14.They show overall good agreement within a deviation of around 0.02 A for the bond lengths and an order of one degree for the angles.The restricted KS determinant of a ground state served then as a reference one for the SOC TD-DFT calculations in the next step.To simulate UV-vis spectra, singlet excited states were optimized using TD-DFT and then mixed with calculated triplet excited states for the optimized ground geometry from the previous step.Scalar relativistic effects were included using the 0 th order regular approximation (ZORA) S28,S29 with a compatible segmented all-electron relativistically contracted basis set SARC-ZORA-TZVP with SARC/J option (generalpurpose Coulomb fitting basis set for all-electron calculations which reduces to def2/J for atoms up to Kr and specially implemented auxiliary basis set for atoms beyond Kr, that is Ir in this case).S30,S31 To accelerate the computation of two-electron integrals, in addition to the resolution of identity approximation for the Coulomb part (RIJ), the chain of spheres algorithm for the exchange part (COSX) was used.S32,S33 The spin-orbit integrals were calculated using the RI-SOMF(1X) approximation that is: using mean-field potential with the inclusion of 1electron terms together with Coulomb term computed with RI approximation and exchange terms evaluated via one-center exact integrals including the spin-other orbit interaction omitting DFT local correlation terms.S34 The maximum number of centers to include in the integrals was set to 4. The list of the first 30 excited singlet states (for the geometry of the ground state optimized in the previous step) is presented in Table S16 together with SOC states obtained by mixing singlets and triplets with the calculated SO-coupling.The theoretical UV-vis spectra from Figure 7 were simulated using the orca_mapspc tool with a broadening of 1800 cm -1 for singlets only (TD-DFT) and spin-orbit states (SOC-corrected) compared to the experimental one.The relevant molecular orbitals with the highest contribution to the first five singlet and triplet states (Table S15) are presented in Figure S24.To better understand the mechanism of absorption (and later the emission from those levels) difference density maps for the first three excited SO-states were plotted in Figure S25.After the inspection, it can be seen that the transitions are mainly of metal-to-ligand charge transfer (MLCT) character with a slight admixture of CN -to ligand CT mechanism.In the last step, to better elucidate the observed luminescence, we simply performed the geometry optimization of the first excited SO-state taking advantage of the possibility to calculate gradients for mixed states in ORCA software.After the optimization, we presented a few first relevant SOC corrected states for the new geometry in Tables S17 and S18 together with the MOs in Figure S26 and geometries compared in Figure S23b analogously to the UV-vis step.One can infer from the density difference maps that the main mechanism of the observed phosphorescence is the MLCT involving (R,R)-pinppy ligands and Ir(III) centers with a slight contribution from the cyanido units and the admixture of intraligand electronic transitions (Figure S27, see the main manuscript for details).The obtained energies of SO-states were compared with emission spectra of 1 (Figures 7 and S28).We did not present relative intensities and therefore lifetimes of the simulated emission bands based on calculated dipole-transition moments, because of the significant impact of vibronic coupling and intersystem crossing rates whose simulations are beyond the scope of this work.Table S17.The energies of the first seven singlet and triplet states for the optimized geometry of the first excited SO-state of cis-[Ir III (CN)2(R,R-pinppy)2] -complexes, together with the weights of molecular orbitals contributing to each excitation.For each state factors greater than 0.1 were bolded.S18 for comparison).

State
The red color represents a positive (build-up) change in electron density, the green represents a negative change (outflow) of electron density.Hydrogen atoms were omitted for clarity.The densities were plotted with an isosurface level of 0.0016.

Figure S2 .
Figure S2.Thermogravimetric (TG) curves, collected in the temperature range of 20-400 °C, for the crystalline samples of (a) 1, (b) 2, (c) 3, and (d) 4. The steps related to the loss of solvent molecules s are depicted.

Figure S3 .
Figure S3.The views of the crystal structure of (R,R)-pinppy along the main a, b, and c crystallographic axes (a−c), and the asymmetric unit with the labeling scheme for symmetrically independent atoms (d).In (b), the C−H••••• aromatic ring interactions between molecules are highlighted.Thermal ellipsoids were presented at a 40% probability level.Hydrogen atoms were omitted for clarity.

Figure S4 .
Figure S4.The views of the crystal structure of 1 along the main a, b, and c crystallographic axes (a−c), the interactions scheme between ions in the crystal structure (d), and the asymmetric unit with the labeling scheme for selected symmetrically independent atoms (e).Thermal ellipsoids were presented at a 40% probability level.Hydrogen atoms were omitted for clarity.Colors: yellow with various hues = Ir centers with cyanido and (R,R)pinppy ligands attached to them, green = TBA + cations, grey = MeCN molecules.

Figure S5 .
Figure S5.The views of the crystal structure of 2 along the main a, b, and c crystallographic axes (a−c), the interactions scheme between ions in the crystal structure (d), and the asymmetric unit with the labeling scheme for = symmetrically independent atoms (e).Thermal ellipsoids were presented at a 40% probability level.Hydrogen atoms were omitted for clarity.Colors: yellow with various hues = Ir centers with cyanido and (R,R)pinppy ligands attached to them, dark green = iodide anions, light green = n-BuDABCO + cations, grey = MeCN.

Figure S6 .
Figure S6.The views of the crystal structure of 3 along the main a, b, and c crystallographic axes (a−c), the interaction scheme between ions in the crystal structure (d), the asymmetric unit with the labeling scheme for symmetrically independent atoms (e), and the arrangement of cyanido-bridged {La III 2Ir III 2} clusters, together with the schematic presentation of metal complexes forming the clusters (f).Thermal ellipsoids were presented at a 40% probability level.Hydrogen atoms were omitted for clarity.Colors: yellow with various hues = Ir centers with cyanido and (R,R)-pinppy ligands attached to them, blue with various hues = La centers with nitrate anions and water molecules attached to them, green = TBA + cations.

Figure S7 .
Figure S7.The view of the crystal structure of 4 along the main a, b, and c crystallographic axes (a−c), the asymmetric unit with the labeling scheme for symmetrically independent atoms (d), the interactions scheme between ions in the crystal structure (e,f), the detailed insight into the intermetallic distances between neighboring metal centers within cyanido-bridged {La III Ir III } chains together with their arrangement in the crystal structure (g).Thermal ellipsoids were presented at a 40% probability.Hydrogen atoms were omitted for clarity.Colors: yellow with various hues = Ir centers with cyanido and (R,R)-pinppy ligands attached to them, blue with various hues = La centers with nitrate anions and dmf molecules attached to them, grey = MeCN molecules.

Figure S9 .
Figure S9.Wavelength dependences of the SH signal for the powder samples 1 (a), 2 (b), 3 (c), and 4 (d), and potassium dihydrogen phosphate (KDP), used as reference material (e), shown together with the comparison of the SH signals for all samples and the KDP reference (f).Insets in (a−e): the photos of respective samples, mounted within the setup of the SHG experiment under the 1040 nm laser light irradiation.

Figure S10 .
Figure S10.Dependences of the SH effect on the excitation intensity for the powder samples of 1 (a), 2 (b), 3 (c), and 4 (d), as well as potassium dihydrogen phosphate (KDP), used as a reference material (e), shown together with the comparison of all measured samples (f).Points represent measured data with intensity uncertainties.Lines correspond to best-fit curves to the quadratic equation pointing to the SH nature of emitted light.

Figure S12 .
Figure S12.Solid-state excitation (a) and emission (b) of (R,R)-pinppy ligand gathered for the indicated emission and excitation wavelengths, respectively, at 77 K.

Figure S13 .
Figure S13.Temperature-variable solid-state excitation (a) and emission (b, normalized) spectra of 1.The inset in (b) represents the enlargement of the 465−580 nm region of the emission spectrum.The related spectroscopic parameters of the emission patterns at each temperature are gathered in TableS7.The emission spectra in the (b) part were normalized to the third (going from lower to higher wavelengths) emission peak.

Figure S14 .
Figure S14.Temperature-variable solid-state excitation (a) and emission (b, normalized) spectra of 2. The inset in (b) represents the enlargement of the 465−550 nm region of the emission spectrum.The related spectroscopic parameters of the emission patterns at each temperature are gathered in TableS8.The emission spectra in the (b) part were normalized to the third (going from lower to higher wavelengths) emission peak.

Figure S15 .
Figure S15.Temperature-variable solid-state excitation (a) and emission (b, normalized) spectra of 3. The inset in (b) represents the enlargement of the 460−560 nm region of the emission spectrum.The related spectroscopic parameters of the emission patterns at each temperature are gathered in TableS9.The emission spectra in the (b) part were normalized to the third (going from lower to higher wavelengths) emission peak.

Figure S16 .
Figure S16.Temperature-variable solid-state excitation (a) and emission (b, normalized) spectra of 4, shown together with the comparison of a few different indicated thermometric parameters (c), and the corresponding relative thermal sensitivity curves (d).The inset in (b) represents the enlargement of the 465−560 nm region of the emission spectrum.The related spectroscopic parameters of the emission patterns at each temperature are gathered in TablesS10 and S11.See the comment below for details.The emission spectra in the (b) part were normalized to the third (going from lower to higher wavelengths) emission peak.

Figure S17 .
Figure S17.Selected optical thermometry characteristics for compounds 1-4.The (a) part contains the best thermometric parameters obtained for all the compounds; in the inset of (a), the repeatability of the optical parameters is presented.Panel (b) contains the relative thermal sensitivity curves calculated based on experimental thermometric parameters shown in (a).The last panel, (c) contains the temperature uncertainty at indicated temperature ranges; this part was also calculated using the best thermometric parameter characteristics shown in (a).See the comment below for details.

Figure S19 .
Figure S19.Wavelength-variable emission decay profiles for 1 and 2 under λexc = 374 nm at room temperature.The mono-exponential fitting was applied for each temperature.The best-fit parameters are roughly presented on the graphs while the detailed values are gathered in TableS12.

Figure S21 .
Figure S21.Temperature-variable emission decay profiles for 4 under λexc = 374 nm and λem = 510 nm, gathered in the 10-290 K temperature range.The mono-exponential fitting was applied for each temperature.The bestfit parameters are roughly presented on the graphs while the detailed values are gathered in TableS13.

Figure S22 .
Figure S22.Temperature dependences of emission lifetime for 3 (a) and 4 (b) with the indicated experimental uncertainties.

Figure S23 .
Figure S23.Comparison of the experimental crystal structure (gold, from the SC-XRD experiment, compound 1) with the optimized ground state (blue) (a) and the comparison of the calculated ground state (blue) with the optimized first excited state (red) (b).Hydrogen atoms were omitted for clarity.

Figure S24 .
Figure S24.Selected molecular orbitals of the optimized ground state for cis-[Ir III (CN)2(R,R-pinppy)2] -complexes.Hydrogen atoms were omitted for clarity.Orbitals are plotted with an isosurface level of 0.15.

Figure S25 .
Figure S25.Difference density maps of three lowest SO-states (SOC corrected TD-DFT states, see TableS16) for the ground state geometry of cis-[Ir III (CN)2(R,R-pinppy)2] -complexes.Red color represents the positive (build-up) change in electron density, green represents the negative change (outflow) of electron density.Hydrogen atoms were omitted for clarity.The densities were plotted with an isosurface level of 0.0016.
Figure S25.Difference density maps of three lowest SO-states (SOC corrected TD-DFT states, see TableS16) for the ground state geometry of cis-[Ir III (CN)2(R,R-pinppy)2] -complexes.Red color represents the positive (build-up) change in electron density, green represents the negative change (outflow) of electron density.Hydrogen atoms were omitted for clarity.The densities were plotted with an isosurface level of 0.0016.

Figure S26 .
Figure S26.Selected molecular orbitals for the optimized geometry of the first excited SO-state of cis-[Ir III (CN)2(R,R-pinppy)2] -complexes.Hydrogen atoms were omitted for clarity.Orbitals are plotted with an isosurface level of 0.15.

Figure S27 .
Figure S27.Density difference maps for the four lowest energy SOC corrected TD-DFT states in cis-[Ir III (CN)2(R,Rpinppy)2] -complexes (for the optimized geometry of the first excited SO-state, see TableS18for comparison).The red color represents a positive (build-up) change in electron density, the green represents a negative change (outflow) of electron density.Hydrogen atoms were omitted for clarity.The densities were plotted with an isosurface level of 0.0016.

Table S4 .
Detailed structure parameters of 3.

Table S5 .
Detailed structure parameters of compound 4.

Table S11 .
Comparison of best-fit parameters to the thermometric calibration curves for 1-4 (see Fig.S16and S17, and related comment for details).

Table S15 .
The energies of the first five singlet and triplet states of cis-[Ir III (CN)2(R,R-pinppy)2] -complexes, shown together with the weights of molecular orbitals that contribute to each excitation.For each state factors greater than 0.1 were bolded.

Table S16 .
The energies of singlet states and SO-states for the ground geometry of cis-[Ir III (CN)2(R,R-pinppy)2] - complexes, and the composition of SO-states in terms of singlet and triplet states for the first thirty states.

Table S18 .
The energies of the first five singlet states and SO-states for the optimized geometry of the first excited SO-state together with the composition of SO-states in terms of singlet and triplet states.