Positional Isomerism in the N^N Ligand: How Much Difference Does a Methyl Group Make in [Cu(P^P)(N^N)]⁺ Complexes?

Abstract

The synthesis and structural characterization of 5,6′-dimethyl-2,2′-bipyridine (5,6′-Me₂bpy) are reported, along with the preparations and characterizations of [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] (POP = bis(2-(diphenylphosphanyl)phenyl)ether, xantphos = 4,5-bis(diphenylphosphanyl)-9,9-dimethyl-9H-xanthene). Single-crystal X-ray structure determinations of [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] confirmed distorted tetrahedral copper(I) coordination environments with the 5-methylpyridine ring of 5,6′-Me₂bpy directed towards the (C₆H₄)₂O unit of POP or the xanthene unit of xantphos. In the xantphos case, this preference may be attributed to C–H…π interactions involving both the 6-CH unit and the 5-methyl substituent in the 5-methylpyridine ring and the arene rings of the xanthene unit. ¹H NMR spectroscopic data indicate that this ligand orientation is also preferred in solution. In solution and the solid state, [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] are yellow emitters, and, for powdered samples, photoluminescence quantum yields (PLQYs) are 12 and 11%, respectively, and excited-state lifetimes are 5 and 6 μs, respectively. These values are lower than PLQY and τ values for [Cu(POP)(6,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(6,6′-Me₂bpy)][PF₆], and the investigation points to the 6,6′-dimethyl substitution pattern in the bpy ligand being critical for enhancement of the PLQY.

1. Introduction

The development of solid-state lighting technologies has revolutionized modern domestic and commercial lighting, primarily through the development of devices which are cheaper to manufacture and operate [1]. Although light-emitting diodes (LEDs) and organic light-emitting diodes (OLEDs) lead the market, light-emitting electrochemical cells (LECs) offer a promising alternative [2,3]. LECs employ ionic transition metal compounds (iTMCs) as the light-emitting materials, the most commonly encountered of which are cyclometallated iridium(III) complexes [2,4] and heteroleptic [Cu(P^P)(N^N)]⁺ complexes [5] in which P^P is a wide bite-angle bisphosphane [6], such as xantphos and POP (Scheme 1), and N^N is typically a 2,2′-bipyridine (bpy) or 1,10-phenanthroline (phen) chelating ligand. This last class of compound follows from the seminal work of McMillin and coworkers, who observed that [Cu(P^P)(N^N)]⁺ complexes exhibit low-lying metal-to-ligand charge transfer (MLCT) excited states [7,8]. More recently, it has been demonstrated that many [Cu(P^P)(N^N)]⁺ complexes exhibit thermally activated delayed fluorescence (TADF) [9,10], and this has increased interest in this family of copper(I) emitters. Triplet and singlet excited states are statistically present in a 3:1 ratio, and, in TADF, there is a fast intersystem crossing from the lowest-lying singlet excited state (S₁) to the triplet excited state (T₁). The T₁ state is long-lived, with a relatively slow phosphorescence. If the energy gap between the S₁ and T₁ states is small, thermal repopulation of the S₁ state via a reverse intersystem crossing can occur, leading to enhanced fluorescence from the S₁ state and an increased photoluminescence quantum yield (PLQY).

Since the potential for copper(I) iTMCs as light-emitting components in LECs was first demonstrated [11,12], xantphos and POP have been combined with many structurally and electronically diverse diimine ligands [5]. Some of the highest PLQY values and the best device electroluminescence (EL) performances have been achieved using bpy ligands containing simple substituents, including 6-alkyl and 6,6′-dialkyl groups [13,14], 6-alkyloxy and 6-alkylthio substituents [15], 6,6′-bis(alkyloxy) groups [16] and 6,6′-dihalo substituents [17]. Methyl substituents, in particular, lead to high PLQY values [13,14] and LEC performances. For example, a maximum luminance of 53 cd m⁻² was observed in a LEC containing [Cu(POP)(6,6′-Me₂bpy)][PF₆] in the emissive layer [13]. Considering photoluminescence data for powdered samples, there is a significant improvement in quantum yield on going from [Cu(POP)(bpy)][PF₆] (3% [17]), to [Cu(POP)(6-Mebpy)][PF₆] (9.5% [13]) or [Cu(POP)(6,6′-Me₂bpy)][PF₆] (43.2% [13]), and from [Cu(xantphos)(bpy)][PF₆] (1.7% [17]) to [Cu(xantphos)(6-Mebpy)][PF₆] (34% [14]) or [Cu(xantphos)(6,6′-Me₂bpy)][PF₆] (37% [14]) (6-Mebpy = 6-methyl-2,2′-bipyridine, 6,6′-Me₂bpy = 6,6′-dimethyl-2,2′-bipyridine). These data indicate that the introduction of at least one methyl group into the 6-position of bpy is beneficial for enhanced emission. In the case of the POP complex, there is a further enhancement when a second methyl group is introduced, but this is less pronounced in the case of the xantphos-containing analogue. Much of the effort in ligand design has emphasized electronic tuning through substituents attached at the 4- and 6-positions of the pyridine rings. In contrast, substitution at the 5-position has a similar inductive but a slightly different resonance electronic effect (6-Me, σ –0.17, σ_I –0.04, σ_R –0.04; 5-Me, σ –0.07, σ_I –0.04, σ_R –0.14) [18,19] but significantly changes the steric demands of the ligand. We were interested in comparing complexes containing the N^N ligand 5,6′-dimethyl-2,2′-bipyridine (5,6′-Me₂bpy) with those containing 6-Mebpy or (isomeric) 6,6′-Me₂bpy. Here we describe the synthesis, characterization, and photophysical and electrochemical properties of [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] and assess the effects of introducing the second methyl group into the 5- rather than the 6-position.

2. Results and Discussion

2.1. Synthesis and Structural Characterization of 5,6′-Me₂bpy

The ligand 5,6′-Me₂bpy has previously been prepared in 66% overall yield using a Stille coupling [20,21] but we preferred to develop a tin-free approach. The palladium-catalyzed cross-coupling shown in Scheme 2 (in which the intermediate is made in situ and was not isolated) yielded 5,6′-Me₂bpy, after work up, in 48% yield. Although the yield is lower than those reported for the Stille route [20,21], we consider that the procedure described here is advantageous in that no intermediate organometal reagent needs to be isolated. The ¹H NMR spectrum (assigned using COSY and NOESY methods) agreed with that published [20], and the ¹³C{¹H} NMR spectrum (not previously reported) was assigned using HMQC and HMBC methods.

Single crystals of 5,6′-Me₂bpy were obtained by diffusion of cyclohexane into an ethyl acetate solution of the compound. 5,6′-Me₂bpy crystallizes in the monoclinic space group P2₁/n and the structure is shown in Figure 1. The compound adopts the expected trans-conformation and the angle between the least-squares planes through the two pyridine rings is 16.5°. Bond lengths and angles are unexceptional, and selected bond lengths are listed in the caption to Figure 1. The molecules pack in centrosymmetric pairs (Figure 2a, pairs of molecules A/B, B/A’, and A’/B’). The first interaction involves face-to-face π-stacking of pyridine rings between molecules A/B and A’/B’. The distance between the planes of the pyridine rings containing N2 and N2i (symmetry code i = 1 – x, 1 – y, 1 – z) is 3.46 Å, and the intercentroid distance is 3.73 Å. For the second pairing (B/A’ in Figure 2a), a C_Me–H…π interaction is important with a C_Me…centroid distance of 3.51 Å. The shortest C_Me…C_py distances are 3.68 and 3.69 Å, which are similar to the sum of the van der Waals radii for C_Me and C_sp2 (3.70 Å) [22]. We consider the C_Me…C_py rather than the H_Me…C_py distances because the H atoms of the methyl group are in calculated positions. When the lattice is viewed along the a-axis, we observe an assembly into zigzag chains (Figure 2b). The latter features a close interaction between the 5-methyl substituent and the 6-substituted pyridine (py) ring of an adjacent molecule. The closest C_Me…C_py separation is 3.79 Å, which is slightly longer than the sum of the C_Me and C_sp2 van der Waals radii [22]. The relationship between the zigzag chains and the centrosymmetric pairs of molecules can be appreciated by comparing Figure 2a,b.

2.2. Synthesis and Characterization of [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆]

The syntheses of [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] were carried out by either sequential (for POP) or simultaneous (for xantphos) addition of the bisphosphane and 5,6′-Me₂bpy ligands to [Cu(MeCN)₄][PF₆]. This is a standard procedure for the different P^P ligands, and the reasons for the different strategies have previously been detailed [14,15]. The POP and xantphos-containing complexes were isolated in yields of 74.4% and 89.6%, respectively. The electrospray mass spectrum of each compound exhibited a base peak corresponding to the [M–PF₆]⁺ ion (m/z 785.2 for [Cu(POP)(5,6′-Me₂bpy)]⁺ and m/z 825.3 for [Cu(xantphos)(5,6′-Me₂bpy)] (Figures S1 and S2; see the Supplementary Materials).

Single crystals of [Cu(POP)(5,6′-Me₂bpy)][PF₆]·Me₂CO and Cu(xantphos)(5,6′-Me₂bpy)][PF₆]·0.5CH₂Cl₂ 0.75Et₂O were grown by diffusion of Et₂O into an acetone or dichloromethane solution of the POP- and xantphos-containing compounds, respectively. The compounds crystallize in the monoclinic space groups P2₁/c and P2₁/n, respectively, and the structures of the cations are presented in Figure 3. Bond parameters for the copper(I) coordination spheres are listed in

Table 1. Bond lengths and angles in the coordination sphere of each [Cu(P^P)(5,6′-Me₂bpy)]⁺ cation, and values of the τ₄ parameter.

Parameter	[Cu(POP)(5,6′-Me2bpy)]+	[Cu(xantphos)(5,6′-Me2bpy)]+
Cu–N/Å	2.075(2), 2.089(2)	2.053(2), 2.105(2)
Cu–P/Å	2.2967(7), 2.2517(7)	2.2488(8), 2.2555(8)
P–Cu–P/o	113.48(3)	117.36(3)
N–Cu–N/o	79.90(8)	80.02(9)
N–Cu–P/o	131.84(6), 108.61(6), 110.64(6), 103.82(6)	120.61(7), 116.05(7), 103.55(7), 110.47(7)
τ4 1	0.83	0.87

¹ τ₄ parameter, see reference [23].

along with Houser’s τ₄ parameter [23]. The latter may be used to assess the distortion of the copper(I) centre away from T_d symmetry (τ₄ = 1.00) towards C_3v (τ₄ = 0.85), C_2v see-saw geometries (τ₄ ≤ 0.64) and finally square-planar (τ₄ = 0.00). The large P–Cu–N angles of 131.84(6)° in [Cu(POP)(5,6′-Me₂bpy)]⁺ and 120.61(7)° and 116.05(7)° in [Cu(xantphos)(5,6′-Me₂bpy)]⁺ contribute to values of τ₄ (Table 1) that are close to τ₄ = 0.85, calculated for a C_3v (trigonal pyramidal) geometry. In both complex cations, the 5,6′-Me₂bpy ligand is positioned with the 5-methylpyridine ring facing the backbone of the bisphosphane ligand, and, in [Cu(xantphos)(5,6′-Me₂bpy)]⁺, this results in the C51–H51 unit being accommodated within the bowl-shaped cavity of the xanthene unit (Figure 4a). Considering the centroids of the two outer rings of the xanthene unit, the C51–H51…centroid distances are 3.01 and 3.22 Å, consistent with C–H…π interactions [24]. We note that, in many of the crystal structures of [Cu(xantphos)(6-Xbpy)][PF₆], in which 6-Xbpy is a 6-substituted-2,2′-bipyridine, the asymmetric 6-Xbpy ligand is oriented with the X group (X = Me, Et, OMe, OEt, OPh, SMe, SEt, SPh, Br [14,17,25,26]) lying over the xanthene bowl, rather than being remote from it. Examples of this alternative configuration are less commonly encountered in the solid state [14,27,28], and, in several structures, disordering indicates little energy difference between the two possible orientations [25,29]. Thus, it is noteworthy that the 5,6′-Me₂bpy ligand exhibits a preference for an orientation with the 6-methyl group remote from the xanthene unit. A possible explanation is the combined effects of the C–H…π interactions described above and weak interactions between the 5-methyl substituent and the arene rings of the xanthene unit. The closest C_Me…C_arene distance is 4.12 Å which, although greater than the sum of the C_Me and C_sp2 van der Waals radii [22], may indicate a stabilizing influence. As in the earlier discussion of the structure of the free ligand, we choose to use the C_Me…C_arene distance rather than the H_Me…C_arene separation because the H atoms of the methyl group are in calculated positions. Intramolecular π-stacking interactions are considered to be important in [Cu(P^P)(N^N)]⁺ complexes in terms of contributing towards enhanced PLQY values [30], and, in [Cu(xantphos)(N^N)]⁺ cations, face-to-face π-stacking interactions are commonly observed between two phenyl rings of different PPh₂ units [15]. However, in [Cu(xantphos)(5,6′-Me₂bpy)]⁺, the angle between the planes of the phenyl rings containing atoms C7 and C34 (Figure 3b) is too large (39.5°) for a meaningful interaction. In contrast, in [Cu(POP)(N^N)]⁺ cations, π-stacking between a phenyl ring and one arene ring of the O(C₆H₄)₂-unit in the POP ligand is often a feature in the solid state, and [Cu(POP)(5,6′-Me₂bpy)]⁺ is no exception. Figure 4b shows the stacking of the rings containing atoms C3 and C9; the angle between the ring-planes is 14.7°, and the distance between the ring-centroids is 3.80 Å.

The solid-state structures discussed above assist in an interpretation of the solution NMR spectroscopic properties of the copper(I) compounds. ¹H, ¹³C{¹H}, COSY, NOESY, HMBC and HMQC-NMR spectra were recorded, allowing full assignment of the ¹H and ¹³C{¹H} NMR resonances in the complexes. 1D and 2D spectra of the compounds are shown in Figure 5 and Figure 6, and in Figures S3–S10 in the Supplementary Materials. Figure 5 displays the aromatic regions of the ¹H NMR spectra of the two compounds. Figure 5 reveals a substantial shift to lower frequency for the signal for proton A6 on going from [Cu(POP)(5,6′-Me₂bpy)][PF₆] to [Cu(xantphos)(5,6′-Me₂bpy)][PF₆]. This is consistent with the 5,6′-Me₂bpy ligand adopting the same configuration in solution as in the solid state, such that proton A6 is affected by the ring currents of the two arene rings of the xanthene unit. The appearance of two sets of signals for the PPh₂ phenyl D-rings (labelled D and D’ in Figure 5 and Figure S5) is consistent with ¹H and ¹³C-NMR spectroscopic data for related POP and xantphos-containing copper(I) complexes; see, for example, [14,15,28,31,32]. In [Cu(POP)(5,6′-Me₂bpy)][PF₆], exchange (EXSY) peaks are observed between pairs of protons D2/D2’ and D3/D3’ (Figure 6). Corresponding peaks are not observed in the xantphos-containing complex (Figure S10). NOESY crosspeaks are observed in [Cu(POP)(5,6′-Me₂bpy)][PF₆] between protons D2/A6 and D2’/A6, and between D2/Me-B6 and D2’/Me-B6, indicating that the flexible POP backbone undergoes dynamic behavior on the NMR timescale at 298 K, which exchanges the axial and equatorial phenyl rings of each PPh₂ unit [14,28].

2.3. Electrochemical and Photophysical Properties

Both [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] are redox active and their electrochemical behavior was investigated in propylene carbonate solution using cyclic voltammetry (Figure 7). Each of [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] exhibits an irreversible copper-centered oxidation with E_pc values of +0.81 and 0.89 V, respectively, and a reversible bpy-centered reductive process with E_1/2 values of −2.12 V and −2.10 V, respectively (E_pa − E_pc = 100 V in each case). Data for the oxidative process are presented in

Table 2. Copper(I)/(II) oxidation potentials in [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] in propylene carbonate solutions (ca. 10⁻⁴ mol dm^–3); values are referenced to internal Fc/Fc⁺ = 0.0 V; [ⁿBu₄N][PF₆] as supporting electrolyte and a scan rate of 0.1 V s^–1. Values are compared to literature data (versus Fc/Fc⁺) for related compounds; see footnotes for solvents.

Compound	E1/2 /V	Epc–Epa/mV	Epca/V	Reference
[Cu(POP)(5,6′-Me2bpy)][PF6]			+0.81	This work
[Cu(xantphos)(5,6′-Me2bpy)][PF6]			+0.89	This work
[Cu(POP)(6,6′-Me2bpy)][BF4]	+0.82 b	– d		[33]
[Cu(xantphos)(6,6′-Me2bpy)][PF6]	+0.90	150		[29]
[Cu(POP)(6-Mebpy)][PF6]	+0.69	– d		[13]
[Cu(xantphos)(6-Mebpy)][PF6]	+0.85 c	100		[34]
[Cu(POP)(bpy)][PF6]	+0.72	110		[29]
[Cu(POP)(bpy)][BF4]	+0.72 b	– d		[33]
[Cu(xantphos)(bpy)][PF6]	+0.76	110		[29]
[Cu(xantphos)(bpy)][BF4]	+0.67 b	– d		[33]

^a The value of E_pc is stated when the process is irreversible. ^b Recorded in MeCN. ^c Recorded in CH₂Cl₂. ^d No E_pc − E_pa value reported.

and are compared with oxidation potentials for closely related compounds. Note that a common solvent was not used for all the compounds in Table 2. Nonetheless, the data for the xantphos-containing complexes are consistent with a trend of copper(I) oxidation occurring at the highest potential for [Cu(xantphos)(6,6′-Me₂bpy)]⁺ and following a sequence according to the N^N ligand of 6,6′-Me₂bpy > 5,6′-Me₂bpy > 6-Mebpy > bpy. Copper(I)-to-copper(II) oxidation is accompanied by a geometrical change from tetrahedral to square-planar, and the trend in the oxidation potentials is consistent with decreasing steric hindrance in the coordination sphere of the copper center along the series from 6,6′-Me₂bpy to 5,6′-Me₂bpy to 6-Mebpy to bpy. The 5,6′-Me₂bpy also fits into this pattern for the POP-containing compounds, although (from the literature data) the trend from 6-Mebpy to bpy is less well defined than for the xantphos family of compounds (Table 2).

The solution absorption spectra of the heteroleptic copper(I) compounds (Figure 8) show intense absorptions below 330 nm arising from ligand-based π*←π transitions. The broad absorption band with λ_max at 376 nm in [Cu(POP)(5,6′-Me₂bpy)][PF₆] and at 374 nm in [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] arises from MLCT transitions. The MLCT λ_max values for the 5,6′-Me₂bpy complexes are similar to those for [PF₆]⁻ or [BF₄]^– salts of [Cu(POP)(6,6′-Me₂bpy)]⁺ (372 nm [33]) and [Cu(xantphos)(6,6′-Me₂bpy)]⁺ (375 nm [33], 378 nm [14]), but are blue-shifted with respect to those of salts of [Cu(POP)(bpy)]⁺ (389 nm [12], 385 nm [33]), [Cu(xantphos)(bpy)]⁺ (383 nm [29], 382 nm [33]), [Cu(POP)(6-Mebpy)]⁺ (380 nm [13]) and [Cu(xantphos)(6-Mebpy)]⁺ (379 nm [14]). Since the LUMO of a [Cu(P^P)(N^N)]⁺ complex is localized on the N^N ligand [14], the general trends in λ_max are consistent with a destabilization of the LUMO upon introducing electron-donating methyl substituents into the bpy domain.

Both aerated and deaerated solutions of [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] are weakly emissive when excited into the MLCT band at 375 and 374 nm, respectively. Values of λ_em^max were 626 and 622 nm (yellow emitters), respectively, PLQY values were around 1%, and excited state lifetimes (τ) ≤ 1 μs. Low PLQYs in solution, even in non-coordinating solvents, have been attributed to the copper(I) complex undergoing exciplex formation, which favors non-radiative decay [35]. Excitation of powdered samples of the compounds (λexc = 365 nm) leads to emission maxima at higher energies than the solution samples while maintaining a yellow emission color. Such a blue-shift is typical of [Cu(P^P)(N^N)]⁺ complexes [36,37].

Table 3. Solid-state emission data (298 K) for the copper(I) compounds (λexc = 365 nm) compared to data for related compounds.

Compound	λemmax/nm	PLQY/% a	τ/μs a	Reference
[Cu(POP)(5,6′-Me2bpy)][PF6]	553	12	6 b	This work
[Cu(xantphos)(5,6′-Me2bpy)][PF6]	555	11	5 b	This work
[Cu(POP)(6,6′-Me2bpy)][PF6]	535	43.2	10.5 c	[13]
[Cu(xantphos)(6,6′-Me2bpy)][PF6]	539	37	11 c	[14]
[Cu(POP)(6-Mebpy)][PF6]	567	9.5	2.6 b	[13]
[Cu(xantphos)(6-Mebpy)][PF6]	547	34	9.6 b	[14]
[Cu(POP)(bpy)][PF6]	580	3.0	1.5 c	[17]
[Cu(xantphos)(bpy)][PF6]	587	1.7	1.5 c	[17]

^aValues from the literature are quoted as reported, hence the inconsistency in the number of significant figures. ^bA biexponential fit to the decay was used, using the equation: τ = ΣA_iτ_i/ΣA_i where A_i is the pre-exponential factor for the lifetime. Values of A₁, A₂, τ₁ and τ₂ are given in Table S1. ^cA monoexponential fit to the decay was used.

summarizes the solid-state emission data, and the emission spectra are shown in Figure 8. Values of PLQY and τ for [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] are lower than for their 6,6′-Me₂bpy-containing analogues (Table 3). The blue-shift in the emission maximum on going from the heteroleptic complexes containing bpy to those with 6-Mebpy, 5,6′-Me₂bpy or 6,6′-Me₂bpy is consistent with the electron-donating character of the methyl substituents. However, comparing values of λ^em_max for [Cu(P^P)(5,6′-Me₂bpy)][PF₆] and [Cu(P^P)(6,6′-Me₂bpy)][PF₆] indicates that a red-shift in the emission accompanies a change from 6,6′-Me₂bpy to 5,6′-Me₂bpy, and this in turn is consistent with the introduction of an electron-donating methyl group in a position meta to the N-donor stabilizing the LUMO of the complex. The PLQY values in Table 3 underline the importance of a second methyl group adjacent to the coordination site of the bpy ligand.

3. Materials and Methods

3.1. General

¹H, ¹³C{¹H} and ³¹P{¹H} NMR spectra were recorded at 298 K on a Bruker Avance III-500 NMR spectrometer (Bruker BioSpin AG, Fällanden, Switzerland). ¹H and ¹³C-NMR chemical shifts were referenced to residual solvent peaks with respect to δ(TMS) = 0 ppm and ³¹P NMR chemical shifts with respect to δ(85% aqueous H₃PO₄) = 0 ppm. A Shimadzu LCMS-2020 (Shimadzu Schweiz GmbH, 4153 Reinach, Switzerland) was used to record electrospray ionization (ESI) mass spectra. Solution absorption and emission spectra were recorded using Shimadzu UV-2600 and Shimadzu RF-6000 instruments (Shimadzu Schweiz GmbH, 4153 Reinach, Switzerland), respectively. A Hamamatsu absolute photoluminescence quantum yield spectrometer C11347 Quantaurus-QY was used to measure PLQYs, and emission lifetimes and powder emission spectra were measured using a Hamamatsu Compact Fluorescence Lifetime Spectrometer C11367 Quantaurus-Tau with an LED light source (λ_exc = 365 nm).

Electrochemical measurements were carried out using a CH Instruments 900B potentiostat (CH Instruments, Texas 78738, USA) with [ⁿBu₄N][PF₆] (0.1 M) as supporting electrolyte and at a scan rate of 0.1 V s^–1. The working electrode was glassy carbon, the reference electrode was a leakless Ag⁺/AgCl (eDAQ ET069-1) (eDAQ Europe, 01-471 Warszawa, Poland), and the counter-electrode was a platinum wire. Final potentials were internally referenced with respect to the Fc/Fc⁺ couple.

POP and xantphos were purchased from Acros (Fisher Scientific AG, 4153 Reinach Switzerland) and Fluorochem (Chemie Brunschwig AG, 4052 Basel, Switzerland), respectively. [Cu(MeCN)₄][PF₆] was prepared by the literature route [38].

3.2. Synthesis of 5,6′-Me₂bpy

nBuLi (4.66 mL, 11.7 mmol, 2.5 M in n-hexane, 1.1 eq.) was added to a degassed solution of 2-bromo-6-methylpyridine (2.01 g, 11.7 mmol, 1.1 eq.) in dry THF (10 mL) at −78 °C under a nitrogen atmosphere. The mixture was stirred for 30 min during which time it turned deep red. Then, a solution of ZnCl₂ in Et₂O (10.6 mL, 10.6 mmol, 1.0 M in Et₂O, 1.0 eq.) was added dropwise and the reaction mixture was stirred for 1.5 h. It was then allowed to warm to ambient temperature (ca. 22 °C) and was added dropwise to a solution of 2-bromo-5-methylpyridine (20.1 g, 11.7 mmol, 1.1 eq.) in dry THF (30 mL) containing [Pd(PPh₃)₄] (612 mg, 0.53 mmol, 0.05 eq.). This reaction mixture was heated to 80 °C and this temperature was maintained with stirring for 60 h. After cooling to ambient temperature, an aqueous solution of NaOH (ca. 100 mL, 3 M) was added until most of the precipitate had dissolved. The mixture was extracted with CH₂Cl₂ (3 × 50 mL) and the combined organic fractions were washed with aqueous NaOH, and then dried over MgSO₄. The solvent was then removed under reduced pressure. The crude material was purified by column chromatography (neutral alumina, cyclohexane:ethyl acetate 40:1, followed by a second column with silica, cyclohexane:ethyl acetate 20:1) to give 5,6′-Me₂bpy (937 mg, 5.08 mmol, 48%) as a white solid. ¹H NMR (500 MHz, CDCl₃) δ/ppm: 8.52–8.48 (m, 1H, H^A6), 8.29 (d, J = 8.1 Hz, 1H, H^A3), 8.13 (d, J = 7.9 Hz, 1H, H^B3), 7.68 (t, J = 7.7 Hz, 1H, H^B4), 7.61 (ddd, J = 8.1, 2.2, 0.7 Hz, 1H, H^A4), 7.14 (d, J = 7.6 Hz, 1H, H^B5), 2.63 (s, 3H, H^Me-B6), 2.39 (s, 3H, H^Me-A5). ¹³C{¹H} NMR (126 MHz, CDCl₃) δ/ppm: 158.0 (C^B6), 155.8 (C^B2), 154.0 (C^A2), 149.7 (C^A6), 137.6 (C^A4), 137.2 (C^B4), 133.3 (C^A5), 123.1 (C^B5), 120.9 (C^A3), 118.0 (C^B3), 24.8 (C^MeB6), 18.5 (C^MeA5). ESI MS: m/z 184.1 [M + H]⁺ (base peak, calc. 184.1). Found: C 78.01, H 6.75, N 15.40; C₁₂H₁₂N₂ requires C 78.23, H 6.57, N 15.21%.

3.3. [Cu(POP)(5,6′-Me₂bpy)][PF₆]

[Cu(MeCN)₄][PF₆] (93.2 mg, 0.250 mmol, 1.0 eq.) and POP (148 mg, 0.275 mmol, 1.1 eq.) were dissolved in CH₂Cl₂ (30 mL) and the reaction mixture was stirred for 1.5 h at room temperature (ca. 22 °C). 5,6′-Me₂bpy (46.1 mg, 0.250 mmol, 1.0 eq.) was then added, and stirring was continued for another 1.5 h. The yellow solution was filtered, and the solvent volume of the filtrate was reduced (under reduced pressure) and added to n-hexane (ca. 40 mL) to precipitate the product. The precipitate was separated and was washed with n-hexane (4 × 10 mL) using sonication and dried under vacuum. [Cu(POP)(5,6′-Me₂bpy)][PF₆] was isolated as a yellow solid (173 mg, 0.186 mmol, 74.4%). ¹H NMR (500 MHz, acetone-d₆) δ/ppm: 8.39 (d, J = 8.4 Hz, 1H, H^A3), 8.36 (d, J = 8.3 Hz, 1H, H^B3), 8.21 (s, 1H, H^A6), 8.06 (t, J = 7.8 Hz, 1H, H^B4), 7.86 (dd, J = 8.4, 2.1 Hz, 1H, H^A4), 7.48–7.42 (m, 3H, H^B5+C5), 7.42–7.33 (m, 4H, H^D4+D4’), 7.31 (t, J = 7.5 Hz, 4H, H^D3), 7.28–7.18 (m, 10H, H^C6+D2+D3’), 7.15 (td, J = 7.5, 1.1 Hz, 2H, H^C4), 7.05–6.97 (m, 4H, H^D2’), 6.90 (m, 2H, H^C3), 2.48 (s, 3H, H^Me-B6), 2.11 (s, 3H, H^Me-A5). ¹³C{¹H} NMR (126 MHz, acetone-d₆) δ/ppm: 159.6 (C^B6), 158.9 (t, J_PC = 6 Hz, C^C1), 152.8 (t, J_PC = 2 Hz, C^B2), 150.9 (t, J_PC = 2 Hz, C^A2), 150.4 (C^A6), 140.0 (C^B4), 139.9 (C^A4), 137.0 (C^A5) 135.0 (C^C3), 134.2 (t, J_PC = 8 Hz, C^D2), 133.7 (t, J_PC = 8 Hz, C^D2’), 133.2 (C^C5), 132.1 (overlapping C^D1’+A3D1), 131.2 (C^D4), 130.8 (C^D4’), 129.8 (t, J_PC = 4 Hz, C^D3), 129.6 (t, J_PC = 4 Hz, C^D3’), 126.6 (C^B5), 126.1 (t, J_PC = 2 Hz, C^C4), 125.0 (overlapping d, J_PC = 14 Hz, C^C2), 123.0 (C^A3), 121.2 (C^C6), 120.5 (C^B3), 26.8 (C^Me-B6), 18.2 (C^Me-A5). ³¹P{¹H} NMR (202 MHz, acetone-d₆) δ/ppm: –12.5 (broad, FWHM ≈ 450 Hz), –144.2 (septet, J_PF = 707 Hz, [PF₆]^–). ESI-MS: m/z 785.20 [M–PF₆]⁺ (base peak, calc. 785.19), 601.05 [Cu(POP)]⁺ (calc. 601.09). UV-Vis (CH₂Cl₂, 2.5 × 10^–5 mol dm^–3): λ/nm (ε/dm³ mol^–1 cm^–1) 253 (29,600), 293 (23,000), 316 (16,000), 376 (3,400). Found: C 62.03, H 4.87, N 2.68; C₄₈H₄₀CuF₆N₂OP₃ requires C 61.90, H 4.33, N 3.01%.

3.4. [Cu(xantphos)(5,6′-Me₂bpy)][PF₆]

[Cu(MeCN)₄][PF₆] (93.2 mg, 0.250 mmol, 1.0 eq.) was dissolved in CH₂Cl₂ (15 mL). A solution of xantphos (145 mg, 0.250 mmol, 1.0 eq.) and 5,6′-Me₂bpy (46.1 mg, 0.250 mmol, 1.0 eq.) was added and the mixture turned orange then yellow while it was stirred for 1.5 h at room temperature (ca. 22 °C). The yellow solution was filtered, and the solvent volume was reduced under vacuum. Et₂O was then added to precipitate the product, and the solid was collected by filtration, washed with Et₂O (4 × 10 mL) using sonication, and dried under vacuum. [Cu(5,6′-Me₂bpy)(xantphos)][PF₆] was isolated as a yellow solid (218 mg, 0.224 mmol, 89.6%). ¹H NMR (500 MHz, acetone-d₆) δ/ppm: 8.45 (d, J = 8.0 Hz, 1H, H^A3), 8.43 (d, J = 7.5 Hz, 1H, H^B3), 8.13 (t, J = 7.8 Hz, 1H, H^B4), 7.90−7.86 (overlapping m, 3H, H^A4+C5), 7.62 (s, 1H, H^A6), 7.52 (d, J = 7.7 Hz, 1H, H^B5), 7.38 (t, J = 7.6 Hz, 2H, H^D4/D4’), 7.35 (t, J = 7.2 Hz, 2H, H^D4/D4’), 7.29 (t, J = 7.7 Hz, 2H, H^C4), 7.23 (m, 4H, H^D3’), 7.20 (m, 4H, H^D3), 7.08 (m, 4H, H^D2), 7.01 (m, 4H, H^D2’), 6.60 (dtd, J = 7.6, 3.8, 1.4 Hz, 2H, H^C3), 2.40 (s, 3H, H^Me-B6), 2.05 (s, 3H, H^Me-A5), 1.87 (s, 3H, H^Me-xantphos), 1.77 (s, 3H, H^Me-xantphos). ¹³C{¹H} NMR (126 MHz, acetone-d₆) δ/ppm: 159.2 (C^B6), 155.9 (t, J_PC = 6 Hz, C^C1), 152.5 (t, J_PC = 2 Hz, C^B2), 151.0 (t, J_PC = 2 Hz, C^A2), 149.6 (C^A6), 140.3 (C^B4), 140.2 (C^A4), 137.1 (C^A5), 135.1 (t, J_PC = 2 Hz, C^C6), 133.8 (t, J_PC = 8 Hz, C^D2’), 133.6 (t, J_PC = 8 Hz, C^D2), 132.6 (t, J_PC = 17 Hz, C^D1’), 132.3 (t, J_PC = 17 Hz, C^D1), 131.6 (C^C3), 131.2 (C^D4’), 131.0 (C^D4), 129.8 (t, J_PC = 5 Hz, C^D3’), 129.8 (t, J_PC = 5 Hz, C^D3), 128.4 (C^C5), 126.6 (C^B5), 126.4 (t, J_PC = 2.4 Hz, C^C4), 123.3 (C^A3), 121.4 (overlapping d, J_PC = 14 Hz, C^C2), 120.7 (C^B3), 37.0 (C^Cq-xantphos), 28.7 (C^Me-xantphos), 27.8 (C^Me-xantphos), 26.9 (C^Me-B6), 18.1 (C^Me-A5). ³¹P{¹H} NMR (202 MHz, acetone-d₆) δ/ppm: −12.1 (broad, FWHM ≈ 330 Hz), −144.2 (septet, J_PF = 707 Hz, [PF₆]⁻). ESI-MS: m/z 825.25 [M − PF₆]⁺ (base peak, calc. 825.22), 641.10 [Cu(xantphos)]⁺ (calc. 641.12). UV-Vis (CH₂Cl₂, 2.5 × 10^–5 mol dm^–3): λ/nm (ε/dm³ mol^–1 cm^–1) 251 (35,000), 287 (29,700), 315 (14,000), 374 (3900). Found: C 62.82, H 4.54, N 2.97; C₅₁H₄₄CuF₆N₂OP₃ requires C 63.06, H 4.57, N 2.88%

3.5. Crystallography

Single-crystal data for the ligand were collected on a STOE StadiVari diffractometer equipped with a Pilatus300K detector and a Metaljet D2 source (GaKα radiation). Data reduction used STOE software [39] and the structure was solved using Olex2 [40], ShelXT [41], and ShelXL v. 2014/7 [42]. For the copper(I) compounds, single-crystal data were collected on a Bruker APEX-II diffractometer (CuKα radiation) with data reduction, solution and refinement using the programs APEX2 [43], Superflip [44,45] and CRYSTALS [46]. Structure analysis including the ORTEP diagrams employed the program Mercury CSD v. 4.1.1 [47,48]. For [Cu(xantphos)(5,6′-Me₂bpy)][PF₆], SQUEEZE [49] was used to treat part of the solvent region, and an additional 0.75 Et₂O was found that sums with the 0.5 CH₂Cl₂ that could be refined.

3.6. 5,6′-Me₂bpy

C₁₂H₁₂N₂, M_r = 184.24, colorless plate, monoclinic, space group P2₁/n, a = 6.4300(2), b = 19.2147(4), c = 8.0931(2) Å, β = 91.147(2)^o, V = 999.71(4) ų, D_c = 1.224 g cm^–3, T = 130 K, Z = 4, μ(GaKα) = 0.369 mm⁻¹. Total 36,616 reflections, 2030 unique (R_int = 0.0291). Refinement of 1952 reflections (130 parameters) with I > 2σ(I) converged at final R₁ = 0.0564 (R₁ all data = 0.0574), wR₂ = 0.1686 (wR₂ all data = 0.1698), gof = 1.190. CCDC 2005673.

3.7. [Cu(POP)(5,6′-Me₂bpy)][PF₆]^.Me₂CO

C₅₁H₄₆CuF₆N₂O₂P₃, M_r = 989.39, orange needle, monoclinic, space group P2₁/c, a = 9.5613(9), b = 14.8515(13), c = 32.472(3) Å, β = 90.338(4)^o, V = 4611.0(7) ų, D_c = 1.425 g cm^–3, T = 130 K, Z = 4, μ(CuKα) = 2.227 mm^–1. Total 31,246 reflections, 8211 unique (R_int = 0.029). Refinement of 7611 reflections (586 parameters) with I > 2σ(I) converged at final R₁ = 0.0392 (R₁ all data = 0.0420), wR₂ = 0.0529 (wR₂ all data = 0.0536), gof = 0.9967. CCDC 2005674.

3.8. [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] ^.0.5CH₂Cl₂^.0.75Et₂O

C_54.50H_52.50ClCuF₆N₂O_1.75P₃, M_r = 1069.38, yellow block, monoclinic, space group P2₁/n, a = 17.6502(15), b = 16.8504(15), c = 18.4582(16) Å, β = 104.685(3)^o, V = 5310.4(8) ų, D_c = 1.338 g cm^–3, T = 130 K, Z = 4, μ(CuKα) = 2.422 mm^–1. Total 55,278 reflections, 9603 unique (R_int = 0.0319). Refinement of 8839 reflections (608 parameters) with I > 2σ(I) converged at final R₁ = 0.0544 (R₁ all data = 0.0575), wR₂ = 0.1728 (wR₂ all data = 0.1768), gof = 1.025. CCDC 2005675.

4. Conclusions

We have reported the synthesis and structural characterization of the 5,6′-Me₂bpy ligand, and the preparations and solution and solid-state characterizations of the heteroleptic [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] compounds. Crystallographic structure determinations of the complexes confirmed distorted tetrahedral copper(I) coordination environments with the 5,6′-Me₂bpy ligand oriented with the 5-methylpyridine ring directed towards the (C₆H₄)₂O unit of POP or the xanthene unit of xantphos. In the latter, this preference appears to arise from C–H…π interactions involving both the 6-CH unit and the 5-methyl substituent in the 5-methylpyridine ring and the arene rings of the xanthene unit. ¹H NMR spectroscopic data are consistent with this same orientation being preferred in solution. The electrochemical behavior of [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] was investigated and the copper(I) oxidation occurs at a lower potential than in corresponding complexes with 6,6′-Me₂bpy. [Cu(POP)(5,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(5,6′-Me₂bpy)][PF₆] are yellow emitters in both solution and the solid-state, and, for powdered samples, PLQY values of 12 and 11%, respectively, and excited-state lifetimes of 5 and 6 μs, respectively, were observed. These values are lower than the PLQY and τ values for [Cu(POP)(6,6′-Me₂bpy)][PF₆] and [Cu(xantphos)(6,6′-Me₂bpy)][PF₆] [13,14]. Our results underline the importance of the 6,6′-dimethyl substitution pattern in the bpy ligand for enhancement of the PLQY in particular.