Chimera Diimine Ligands in Emissive [Cu(PˆP)(NˆN)][PF 6 ] Complexes

: The syntheses and characterizations of the chelating ligand 6-chloro-6 (cid:48) -methyl-2,2 (cid:48) -bipyridine (6-Cl-6 (cid:48) -Mebpy) and of the copper(I) compounds [Cu(POP)(6-Cl-6 (cid:48) -Mebpy)][PF 6 ] and [Cu(xantphos) (6-Cl-6 (cid:48) -Mebpy)][PF 6 ] (POP = bis(2-(diphenylphosphanyl)phenyl)ether and xantphos = 4,5-bis (diphenylphosphanyl)-9,9-dimethyl-9 H -xanthene) are described. The single crystal structures of both complexes were determined; the copper(I) ion is in a distorted tetrahedral environment and in [Cu(xantphos)(6-Cl-6 (cid:48) -Mebpy)][PF 6 ], the disorder of the 6-Cl-6 (cid:48) -Mebpy ligand indicates there is no preference of the ‘bowl’-like cavity of the xanthene unit to host either the methyl or chloro-substituent, consistent with comparable steric e ﬀ ects of the two groups. The electrochemical and photophysical properties of [Cu(POP)(6-Cl-6 (cid:48) -Mebpy)][PF 6 ] and [Cu(xantphos)(6-Cl-6 (cid:48) -Mebpy)][PF 6 ] were investigated and are compared with those of the related compounds containing 6,6 (cid:48) -dichloro-2,2 (cid:48) -bipyridine or 6,6 (cid:48) -dimethyl-2,2 (cid:48) -bipyridine ligands. Trends in properties of the [Cu(PˆP)(NˆN)] + complexes were consistent with 6-Cl-6 (cid:48) -Mebpy behaving as a combination of the two parent ligands. (cid:48) ), 128.6 (C C5 ), 127.5 (C A5 ), 127.2 (C B5 ), 126.2 (t, J CP = 2 Hz, C C4 ), 122.6 (t, J CP = 13 Hz, C C2 ), 122.2 (C B3 ), 121.5 (C A3 ), 36.8 (C xantphos-bridge ), 29.5 (C xantphos-Me ), 28.1 (C xantphos-Me ), 26.8 (C bpy-Me ). 31 P{ 1 H} NMR (202 MHz, 298 K, acetone- d 6 ) δ / ppm − 13.5 (broad, xantphos), − 144.3 (heptet, J PF = 707, PF 6 − ). ESI-MS (CH 2 Cl 2 / MeOH, positive mode) m / z 845.16 [M − PF 6 ] + (calc. 845.17), 641.10 [M − PF 6 − (6-Cl-6 (cid:48) -Mebpy)] + (base peak, calc. 641.12). Found: C 58.63, H 4.11, N 2.71; C 50 H 41 ClCuF 6 N 2 OP 3 · 0.5CH 2 Cl 2 requires C 58.65, H 4.09, N 2.71.


Introduction
Light-emitting diode (LED) and organic LED (OLED) solid-state lighting technologies are firmly embedded in our society [1]. Light-emitting electrochemical cells (LECs) represent a less well developed, but nonetheless promising, area of solid-state lighting [2,3]. Among the compounds targeted for emissive components in the active layers of LECs are heteroleptic [Cu(PˆP)(NˆN)] + complexes in which PˆP is a wide bite-angle bisphosphane [4] and NˆN contains a diimine metal-binding domain. Investigations in this field were inspired by the work of McMillin and coworkers who showed that [Cu(PˆP)(NˆN)] + complexes possess low-lying metal-to-ligand charge transfer (MLCT) excited states [5,6]. In the last decade, it has been shown that [Cu(PˆP)(NˆN)] + complexes exhibit thermally activated delayed fluorescence (TADF) [7,8]. This phenomenon involves fast intersystem crossing from the lowest-lying singlet excited state (S 1 ) to the triplet excited state (T 1 ); because the latter is long-lived and shows a relatively slow phosphorescence, thermal repopulation of the singlet excited state via a reverse intersystem crossing occurs with a concomitant fluorescence emission from S 1 . This process significantly increases the photoluminescent quantum yield (PLQY), making [Cu(PˆP)(NˆN)] + complexes of particular interest for use in LECs and allows the harvesting of both triplet and singlet excited states which are generated in a 3:1 ratio [7,8].

Synthesis and Characterization of 6-Cl-6'-Mebpy
The preparation of 6-Cl-6'-Mebpy reported in the patent literature [16] involved the Grignard reaction of MeMgBr with 6,6'-Cl2bpy. We favored a Negishi coupling between 2,6-dichloropyridine and 6-methyl-2-pyridinylzinc bromide under microwave conditions, which led to 6-Cl-6'-Mebpy in a 23.5% yield after purification. The base peak in the electrospray mass spectrum was observed at m/z 205.02 and was assigned to the [M + H] + ion. The isotopomer distribution ( Figure S1 in the supplementary materials) was as predicted. The 1 H and 13 C{ 1 H} NMR spectra were assigned using COSY, NOESY, HMQC and HMBC methods. In the 1 H NMR spectrum, two sets of pyridine ring signals ( Figure S2) and the presence of one methyl signal were consistent with the structure of 6-Cl-6'-Mebpy shown in Scheme 1. The HMQC and HMBC spectra are displayed in Figures S3 and S4.

Synthesis and Characterization of 6-Cl-6 -Mebpy
The preparation of 6-Cl-6 -Mebpy reported in the patent literature [16] involved the Grignard reaction of MeMgBr with 6,6 -Cl 2 bpy. We favored a Negishi coupling between 2,6-dichloropyridine and 6-methyl-2-pyridinylzinc bromide under microwave conditions, which led to 6-Cl-6 -Mebpy in a 23.5% yield after purification. The base peak in the electrospray mass spectrum was observed at m/z 205.02 and was assigned to the [M + H] + ion. The isotopomer distribution ( Figure S1 in the supplementary materials) was as predicted. The 1 H and 13 C{ 1 H} NMR spectra were assigned using COSY, NOESY, HMQC and HMBC methods. In the 1 H NMR spectrum, two sets of pyridine ring signals ( Figure S2) and the presence of one methyl signal were consistent with the structure of 6-Cl-6 -Mebpy shown in Scheme 1. The HMQC and HMBC spectra are displayed in Figures S3 and S4 6 ] were recorded in acetone-d 6 and a comparison of the aromatic regions of the 1 H NMR spectra is given in Figure 1. The atom labeling for the NMR spectroscopic assignments is given in Scheme 1. As in the 1 H and 13 C{ 1 H} NMR spectra of related compounds [11,12,18,19], two sets of signals are observed for the phenyl rings of the PPh 2 groups. In Figure 1, these are labeled D and D'. In the NOESY spectrum of each complex, cross-peaks between signals for the methyl group of 6-Cl-6 -Mebpy and one set of protons-D2 were observed, defining D2 as the ortho-protons on the phenyl rings closest to the NˆN ligand. Combined use of NOESY and HMQC allows the 13 C resonances for C D2 and C D2 to be distinguished ( Figure 2). The distinction between the sets of phenyl rings is more clearly recognizable in the structural discussion that follows later. were recorded in acetone-d6 and a comparison of the aromatic regions of the 1 H NMR spectra is given in Figure 1. The atom labeling for the NMR spectroscopic assignments is given in Scheme 1. As in the 1 H and 13 C{ 1 H} NMR spectra of related compounds [11,12,18,19], two sets of signals are observed for the phenyl rings of the PPh2 groups. In Figure 1, these are labeled D and D'. In the NOESY spectrum of each complex, cross-peaks between signals for the methyl group of 6-Cl-6'-Mebpy and one set of protons-D2 were observed, defining D2' as the ortho-protons on the phenyl rings closest to the N^N ligand. Combined use of NOESY and HMQC allows the 13 C resonances for C D2 and C D2' to be distinguished ( Figure 2). The distinction between the sets of phenyl rings is more clearly recognizable in the structural discussion that follows later.   , diffusion of Et2O into a CH2Cl2 solution of the complex was used to grow crystals. The POP-and xantphos-containing compounds crystallize in the monoclinic P21/c and triclinic P-1 space groups, respectively. The copper atom in each complex cation is in the expected distorted tetrahedral geometry with two chelating ligands ( Figure 3). The structure of the [Cu(POP)(6-Cl-6'-Mebpy)] + cation was severely disordered in both the N^N and POP domains. Within the POP ligand, half of the {(C6H4)2O} unit and two phenyl rings of one PPh2 group were also disordered, and each fragment was modeled with a 50% site occupancy. The Cl and Me substituents in the 6-Cl-6'-Mebpy ligand in both compounds were disordered and were modeled over two sites, each with a 50% occupancy. The disorders do not affect the {CuP2 N2}-coordination sphere and relevant bond lengths and angles are given in Table 1. ORTEP representations of the complex cations and the disorders in the [Cu(POP)(6-Cl-6'-Mebpy)] + cation are depicted in Figures S12-S14 in the supporting information. In the [Cu(POP)(6-Cl-6'-Mebpy)] + cation, the bpy unit is twisted with an angle between the least squares planes through the two pyridine rings of 19.5°. The corresponding angle in the [Cu(xantphos)(6-Cl-6'-Mebpy)] + cation is 5.7°.   6 ], diffusion of Et 2 O into a CH 2 Cl 2 solution of the complex was used to grow crystals. The POP-and xantphos-containing compounds crystallize in the monoclinic P2 1 /c and triclinic P-1 space groups, respectively. The copper atom in each complex cation is in the expected distorted tetrahedral geometry with two chelating ligands ( Figure 3). The structure of the [Cu(POP)(6-Cl-6 -Mebpy)] + cation was severely disordered in both the NˆN and POP domains. Within the POP ligand, half of the {(C 6 H 4 ) 2 O} unit and two phenyl rings of one PPh 2 group were also disordered, and each fragment was modeled with a 50% site occupancy. The Cl and Me substituents in the 6-Cl-6 -Mebpy ligand in both compounds were disordered and were modeled over two sites, each with a 50% occupancy. The disorders do not affect the {CuP 2 N 2 }-coordination sphere and relevant bond lengths and angles are given in Table 1 , diffusion of Et2O into a CH2Cl2 solution of the complex was used to grow crystals. The POP-and xantphos-containing compounds crystallize in the monoclinic P21/c and triclinic P-1 space groups, respectively. The copper atom in each complex cation is in the expected distorted tetrahedral geometry with two chelating ligands ( Figure 3). The structure of the [Cu(POP)(6-Cl-6'-Mebpy)] + cation was severely disordered in both the N^N and POP domains. Within the POP ligand, half of the {(C6H4)2O} unit and two phenyl rings of one PPh2 group were also disordered, and each fragment was modeled with a 50% site occupancy. The Cl and Me substituents in the 6-Cl-6'-Mebpy ligand in both compounds were disordered and were modeled over two sites, each with a 50% occupancy. The disorders do not affect the {CuP2 N2}-coordination sphere and relevant bond lengths and angles are given in Table 1. ORTEP representations of the complex cations and the disorders in the [Cu(POP)(6-Cl-6'-Mebpy)] + cation are depicted in Figures S12-S14 in the supporting information. In the [Cu(POP)(6-Cl-6'-Mebpy)] + cation, the bpy unit is twisted with an angle between the least squares planes through the two pyridine rings of 19.5°. The corresponding angle in the [Cu(xantphos)(6-Cl-6'-Mebpy)] + cation is 5.7°.    (5) 117.08 (11), 105.66 (10), 114.44(10), 113.21 (11) In [Cu(xantphos)(NˆN)] + complexes which contain asymmetrical 6-substituted 2,2 -bipyridines or 2-(pyridin-2-yl)quinolines, the NˆN ligand may favor a conformation in which the 6-substituent or quinoline unit lies over, or is remote from, the xanthene unit [11,12,18,19]. In some cases, NMR spectroscopy has provided evidence for mixtures of conformers in solution [11,18]. Among the 152 hits arising from a search of the Cambridge Structural Database (CSD, v. 5.4.1 [20]) for compounds containing a {Cu(POP/xantphos)(bpy/phen)} or closely related unit, only a few contain asymmetrical NˆN ligands [10][11][12]14,18,[21][22][23]. Within the context of these NˆN ligands, 6-Cl-6 -Mebpy is atypical in being asymmetrical by virtue of having different substituents in the 6-and 6 -positions of the bpy domain. The disorder in [Cu(xantphos)(6-Cl-6 -Mebpy)][PF 6 ], modeled with equal occupancies of Cl and Me groups, indicates that there is no preference for the orientation of the 6-Cl-6 -Mebpy ( Figure 4). This is consistent with the more general observation of similar steric effects for these two substituents [24,25].   (11) In [Cu(xantphos)(N^N)] + complexes which contain asymmetrical 6-substituted 2,2'-bipyridines or 2-(pyridin-2-yl)quinolines, the N^N ligand may favor a conformation in which the 6-substituent or quinoline unit lies over, or is remote from, the xanthene unit [11,12,18,19]. In some cases, NMR spectroscopy has provided evidence for mixtures of conformers in solution [11,18]. Among the 152 hits arising from a search of the Cambridge Structural Database (CSD, v. 5.4.1 [20]) for compounds containing a {Cu(POP/xantphos)(bpy/phen)} or closely related unit, only a few contain asymmetrical N^N ligands [10][11][12]14,18,[21][22][23]. Within the context of these N^N ligands, 6-Cl-6'-Mebpy is atypical in being asymmetrical by virtue of having different substituents in the 6-and 6'-positions of the bpy domain. The disorder in [Cu(xantphos)(6-Cl-6'-Mebpy)][PF6], modeled with equal occupancies of Cl and Me groups, indicates that there is no preference for the orientation of the 6-Cl-6'-Mebpy ( Figure  4). This is consistent with the more general observation of similar steric effects for these two substituents [24,25].

Electrochemical and Photophysical Properties of [Cu(POP)(6-Cl-6'-Mebpy)][PF6] and [Cu(xantphos)(6-Cl-6'-Mebpy)][PF6]
The electrochemical behavior of the copper(I) compounds was investigated using cyclic voltammetry. For the xantphos-containing compound, a reversible oxidation process (Figures 5 and S15) was assigned to the Cu + /Cu 2+ redox process, but for [Cu(POP)(6-Cl-6'-Mebpy)][PF6], the corresponding process was irreversible ( Figure S16). Two ligand-centered reductive processes are   The electrochemical behavior of the copper(I) compounds was investigated using cyclic voltammetry. For the xantphos-containing compound, a reversible oxidation process ( Figure 5 and Figure S15) was assigned to the Cu + /Cu 2+ redox process, but for [Cu(POP)(6-Cl-6 -Mebpy)][PF 6 ], the corresponding process was irreversible ( Figure S16 Table 2 compares values of the Cu + /Cu 2+ oxidation potential of [Cu(POP)(6-Cl-6'-Mebpy)] + and [Cu(xantphos)(6-Cl-6'-Mebpy)] + with those of the analogous complexes containing bpy, 6,6'-Me2bpy and 6,6'-Cl2bpy. Oxidation from copper(I) to copper(II) is accompanied by a change from preferred tetrahedral to square-planar geometry. On going from bpy to 6,6'-Me2bpy, the shift in E1/2 to higher potentials can be rationalized in terms of steric effects of the methyl substituents which hinder the flattening of the coordination sphere. In this case, steric effects appear to dominate over electronic effects since the electron-donating methyl substituents would also stabilize copper(II). On going from bpy or 6,6'-Me2bpy to 6,6'-Cl2bpy, the shift in E1/2 to higher potentials is consistent with a combination of both steric and electronic effects; the electron-withdrawing chlorine substituents stabilize the copper(I) oxidation state. The value of E1/2 for [Cu(xantphos)(6-Cl-6'-Mebpy)][PF6] is in accord with an interplay of the effects of both chlorine and methyl groups, and the value of Epc for [Cu(POP)(6-Cl-6'-Mebpy)][PF6] is also consistent with this trend.  [PF6] exhibit absorption spectra comprising intense high-energy, ligand-centered absorptions and a broad, lower intensity band arising from metal-to-ligand charge transfer (MLCT). The spectra are shown in Figure 6a and data are presented in Table 3. The trend in the MLCT maximum on changing the N^N ligand from bpy, 6,6'-Me2bpy or 6,6'-Cl2bpy to 6-Cl-6'-Mebpy reflects the changing electronic properties of the substituents. Figure 3b compares [14]. The values of λmax for the MLCT absorption are 400, 383, 379 and 420 nm, respectively. The LUMO of a [Cu(P^P)(N^N)] + complex is localized on the N^N ligand, and introducing electron-withdrawing chloro groups into  Table 2 compares values of the Cu + /Cu 2+ oxidation potential of [Cu(POP)(6-Cl-6 -Mebpy)] + and [Cu(xantphos)(6-Cl-6 -Mebpy)] + with those of the analogous complexes containing bpy, 6,6 -Me 2 bpy and 6,6 -Cl 2 bpy. Oxidation from copper(I) to copper(II) is accompanied by a change from preferred tetrahedral to square-planar geometry. On going from bpy to 6,6 -Me 2 bpy, the shift in E 1/2 to higher potentials can be rationalized in terms of steric effects of the methyl substituents which hinder the flattening of the coordination sphere. In this case, steric effects appear to dominate over electronic effects since the electron-donating methyl substituents would also stabilize copper(II). On going from bpy or 6,6 -Me 2 bpy to 6,6 -Cl 2 bpy, the shift in E 1/2 to higher potentials is consistent with a combination of both steric and electronic effects; the electron-withdrawing chlorine substituents stabilize the copper(I) oxidation state. The value of E 1/2 for [Cu(xantphos)(6-Cl-6 -Mebpy)][PF 6 ] is in accord with an interplay of the effects of both chlorine and methyl groups, and the value of E pc for [Cu(POP)(6-Cl-6 -Mebpy)][PF 6 ] is also consistent with this trend. Compound  6 ] exhibit absorption spectra comprising intense high-energy, ligand-centered absorptions and a broad, lower intensity band arising from metal-to-ligand charge transfer (MLCT). The spectra are shown in Figure 6a and data are presented in Table 3. The trend in the MLCT maximum on changing the NˆN ligand from bpy, 6,6 -Me 2 bpy or 6,6 -Cl 2 bpy to 6-Cl-6 -Mebpy reflects the changing electronic properties of the substituents. Figure 3b compares 6 ], leading to a red shift in the MLCT absorption. A smaller red shift is observed when 6,6 -Cl 2 bpy is replaced by 6-Cl-6 -Mebpy (Table 3 and Figure 3b), consistent with the combined characters of the NˆN ligand.  (Table 3 and Figure 3b), consistent with the combined characters of the N^N ligand.    [10,11,14]. As expected, the PLQY values are significantly enhanced from solution to powdered samples (    6 ] compounds [10,11,14]. As expected, the PLQY values are significantly enhanced from solution to powdered samples (  [14]. The excited-state lifetime, τ, (which was determined using a biexponential fit [27] as detailed in Table 4) 7 µs [14]). Similar trends were seen for the xantphos-containing compounds with values of τ being 3.3, 4.0, and 11 µs for NˆN = 6,6 -Cl 2 bpy [14], 6-Cl-6 -Mebpy (Table 4) and 6,6 -Me 2 bpy [10]. In terms of quantum yields, we observed that the PLQY of 16% for powdered [Cu(xantphos) (  a A biexponential fit to the lifetime delay was used because a single exponential gave a poor fit; τ is calculated from the equation A i τ i / A i and A i is the pre-exponential factor for the lifetime and values of τ(1), τ(2), A 1 and A 2 are also given.  [14]. The excited-state lifetime, τ, (which was determined using a biexponential fit [27] as detailed in Table 4) 7 μs [14]). Similar trends were seen for the xantphos-containing compounds with values of τ being 3.3, 4.0, and 11 μs for N^N = 6,6'-Cl2bpy [14], 6-Cl-6'-Mebpy (Table 4) and 6,6'-Me2bpy [10]. In terms of quantum yields, we observed that the PLQY of 16% for powdered [Cu(xantphos) (
Cyclic voltammograms were recorded using a CH Instruments 900B potentiostat (CH Instruments, Bee Cave, TX, USA) with [ n Bu 4 N][PF 6 ] (0.1 mol dm −3 ) as the supporting electrolyte and a scan rate of 0.1 V s −1 ; the solvent was HPLC grade CH 2 Cl 2 and solution concentrations were ca. 1 × 10 −4 mol dm −3 . All solutions were degassed with argon. The working electrode was glassy carbon, the reference electrode was an Ag wire, and the counter-electrode was a Pt wire. Final potentials were internally referenced with respect to the Fc/Fc + couple.

Crystallography
Single crystal data were collected on a STOE StadiVari diffractometer (STOE & Cie GmbH, Darmstadt, Germany) equipped with a Pilatus300K detector and a Metaljet D2 source (Ga Kα radiation). The structure was solved using Superflip [29,30] and CRYSTALS [31]. Structure analysis including the ORTEP representations, used Mercury CSD v. 4.1.1 [32,33]. The cation in [Cu(POP)(6-Cl-6 -Mebpy)][PF 6 ] suffered from severe disorder and all the aromatic rings were refined as rigid bodies; one ring was refined isotropically as it was not possible to identify two distinct orientations, and the C atom of the methyl group was also refined isotropically. In the 6-Cl-6 -Mebpy ligand in [Cu(xantphos)(6-Cl-6 -Mebpy)][PF 6 ], the Cl/Me sites were disordered and were modeled over two sites with equal occupancies.