2.1 Materials and synthesis

All copper(I) halides, ligands, and solvents available commercially were used as received, without further purification. Unless otherwise stated, all manipulations were conducted in the air. The Cu(I) complexes [CuCl(PPh₃)₃]·CH₃CN [18], [CuBr(PPh₃)₃] [19], [CuI(PPh₃)₃] [20] were prepared according to the reported procedures. [CuX(PPh₃)₂(py)] (1-X, X = Cl⁻, Br⁻, or I⁻) were prepared in the same way and confirmed crystallographically to have the same crystal structures as those reported previously [21]. Synthetic procedures for 3-I are described in our previous report [14]. Although the preparation of [CuI(PPh₃)₂(iq)] (2-I) is also reported in our previous study [14], we found that polymorphic crystals are produced depending on the amount of PPh₃ added (see Section 3.1), which are named as 2-I-α and 2-I-β. The preparation of 2-I-β is described below.

[CuCl(PPh₃)₂(iq)] (2-Cl). To [CuCl(PPh₃)₃]·CH₃CN (90.4 mg, 0.098 mmol) was added iq (1 mL), and the yellow precipitate was formed. CHCl₃ (1 mL) was added to the mixture to dissolve the precipitate. The title compound was obtained as yellow crystals (43.8 mg, 60%) by slow diffusion of diethyl ether vapor into the copper solution for 18 h. Elemental analysis calculated for C₄₅H₃₇CuClNP₂: C 71.80, H 4.95, N 1.86; found C 71.86, H 4.74, N 1.91. ¹H NMR (270 MHz, CDCl₃) δ ppm 7.19–7.44 (m, 30 H) 7.56–7.78 (m, 3 H) 7.82 (d, J = 8.2 Hz, 1 H) 7.94 (d, J = 8.2 Hz, 1 H) 8.53 (s, 1 H) 9.27 (s, 1 H).

[CuBr(PPh₃)₂(iq)] (2-Br). To a suspension of CuBr (37.5 mg, 0.24 mmol) in iq (1 mL) was added PPh₃ (153.8 mg, 0.59 mmol) and the yellow precipitate was formed. CHCl₃ (2 mL) was added to the mixture to dissolve the precipitate, and then filtered to remove the green impurity. The title compound was obtained as yellow crystals (128.7 mg, 69%) by slow diffusion of diethyl ether vapor into the copper solution for 14 h. Elemental analysis calculated for C₄₅H₃₇CuBrNP₂: C 67.80, H 4.68, N 1.76; found C 67.66, H 4.49, N 1.78. ¹H NMR (270 MHz, CDCl₃) δ ppm 7.15–7.45 (m, 30 H) 7.57–7.75 (m, 3 H) 7.83 (d, J = 7.9 Hz, 1 H) 7.94 (d, J = 7.6 Hz, 1 H) 8.52 (d, J = 5.6 Hz, 1 H) 9.27 (s, 1 H).

[CuI(PPh₃)₂(iq)] (2-I-β). To a suspension of CuI (48.9 mg, 0.26 mmol) in iq (1 mL) was added PPh₃ (660 mg, 2.5 mmol) and the yellow precipitate was formed. CHCl₃ (1 mL) was added to the mixture to dissolve the precipitate. The title compound was obtained as yellow crystals (177.1 mg, 85%) by slow diffusion of diethyl ether vapor into the copper solution for 10 h. Elemental analysis calculated for C₄₅H₃₇CuINP₂: C 64.02, H 4.42, N 1.66; found C 64.28, H 4.23, N 1.68. ¹H NMR (270 MHz, CDCl₃) δ ppm 7.06–7.49 (m, 30 H) 7.58–7.76 (m, 3 H) 7.83 (d, J = 8.2 Hz, 1 H) 7.96 (d, J = 8.2 Hz, 1 H) 8.54 (d, J = 5.9 Hz, 1 H) 9.27 (s, 1 H).

[CuCl(PPh₃)₂(nap)] (3-Cl) (nap = 1,6-naphthyridine). [CuCl(PPh₃)₃]·CH₃CN (49.6 mg, 0.054 mmol) was dissolved in a CHCl₃ solution (1 mL) of nap (100 mg, 0.77 mmol). The title compound was obtained as yellow crystals (20.5 mg, 52%) by slow diffusion of diethyl ether vapor into the copper solution for 3 days. Elemental analysis calculated for C₄₄H₃₆CuClN₂P₂: C 70.12, H 4.81, N 3.72; found C 70.17, H 4.67, N 3.76. ¹H NMR (270 MHz, CDCl₃) δ ppm 7.20–7.43 (m, 30 H) 7.56 (dd, J = 8.2, 4.3 Hz, 1 H) 7.93 (d, J = 5.9 Hz, 1 H) 8.26–8.37 (m, 1 H) 8.75 (d, J = 5.9 Hz, 1 H) 9.13 (dd, J = 4.3, 2.0 Hz, 1 H) 9.32 (s, 1 H).

[CuBr(PPh₃)₂(nap)] (3-Br). [CuBr(PPh₃)₃] (56.8 mg, 0.061 mmol) was dissolved in a CHCl₃ solution (1 mL) of nap (100 mg, 0.77 mmol). The title compound was obtained as yellow crystals (43.1 mg, 89%) by slow diffusion of diethyl ether vapor into the copper solution for 38 h. Elemental analysis calculated for C₄₄H₃₆CuBrN₂P₂: C 66.21, H 4.55, N 3.51; found C 66.10, H 4.47, N 3.48. ¹H NMR (270 MHz, CDCl₃) δ ppm 7.13–7.46 (m, 30 H) 7.56 (dd, J = 8.6, 4.3 Hz, 1 H) 7.92 (d, J = 5.9 Hz, 1 H) 8.25–8.36 (m, 1 H) 8.75 (d, J = 5.9 Hz, 1 H) 9.13 (dd, J = 4.3, 2.0 Hz, 1 H) 9.32 (s, 1 H).

2.2 Luminescence measurements

The luminescence spectrum of each sample was measured using a JASCO FR-6600 spectrofluorometer at room temperature. The luminescence quantum yield was recorded on a Hamamatsu Photonics C9920-02 absolute photoluminescence quantum yield measurement system equipped with an integrating sphere apparatus and a 150-W CW xenon light source. An A10095-03 non-luminescent quartz sample holder (Hamamatsu Photonics) was used for the absolute photoluminescence quantum yield measurement in the air. The accuracy of the instrument was confirmed by the measurement of the quantum yield of anthracene in ethanol solution (0.27) [22]. Emission lifetime measurements were conducted by using a Hamamatsu Photonics C4334 system equipped with a streak camera as a photo detector and a nitrogen laser as an excitation light source (λ_ex = 337 nm). A liquid N₂ cryostat (Optistat-DN optical Dewar and ITC-503 temperature controller, Oxford Instruments) was used to control the sample's temperature.

The emission decays were analyzed using two exponentials, i.e., I = A₁ exp(−t/τ₁) + A₂ exp(−t/τ₂), where τ₁ and τ₂ are the lifetimes, and A₁ and A₂ are the pre-exponential factors. For the determination of the radiative and non-radiative rate constants, we used the averaged lifetimes (τ_av), which were estimated based on equation (1) for the two exponential decay components [23]:

2.3 Single-crystal X-ray diffraction measurements

All single-crystal X-ray diffraction measurements were conducted using a Rigaku Mercury CCD diffractometer with graphite monochromated Mo Kα radiation (λ = 0.71069 Å) and a rotating anode generator. Each single crystal was mounted on a MicroMount using paraffin oil. The crystal was then cooled using a N₂-flow-type temperature controller. Diffraction data were collected and processed using the CrystalClear software [24]. Structures were solved by the direct method using SIR-2004 [25]. Structural refinements were conducted by the full-matrix least-squares method using SHELXL-97 [26]. Non-hydrogen atoms were refined anisotropically, and hydrogen atoms were refined using the riding model. All calculations were conducted using the Crystal Structure crystallographic software package [27]. Crystallographic data obtained for each complex are summarized in Table 1.

2.4 Characterization measurements

The ¹H NMR spectrum of each sample was measured using a JEOL EX-270 NMR spectrometer at room temperature. Elemental analysis was conducted at the analysis center at Hokkaido University.

2.5 Theoretical calculations

DFT calculations were performed with the B3LYP functional [28,29] and the LANL2DZ basis [30,31] set using Gaussian 03 [32] for 1-Cl, 1-Br, 2-Cl, 2-Br, 2-I-β, 3-Cl, and 3-Br. As for the calculations of 1-I, 2-I-α, and 3-I, we have already reported the results using the same functional and basis set [14]. The atomic coordinates determined for the individual molecular structures by X-ray crystallographic analysis were used for the DFT calculations. The drawing of molecular orbitals for the complexes were made using Avogadro 1.10 [33].

3.1 Crystal structures

Figs. 1 and 2 show the molecular structures of 2-X and 3-X (X = Cl⁻, Br⁻, I⁻), respectively. Crystal structures of 1-X prepared herein were identical to those reported by Engelhardt et al. [21]. All complexes adopt a tetrahedral coordination geometry occupied by one halide, one N atom of the N-heteroaromatic ligand, and two P atoms of the PPh₃ ligands. Selected bond lengths and angles around the copper(I) atom of 2-X and 3-X are listed in Table 2. For complex 2-I, two polymorphic crystal structures (2-I-α and 2-I-β) were obtained. Structure 2-I-α was crystallized from the isoquinoline/chloroform solution including a 1:2 ratio of CuI and PPh₃ [14], while 2-I-β was obtained from the isoquinoline/chloroform solution including 1:10 ratio of CuI and PPh₃. The important difference between the polymorphic forms of 2-I-α and 2-I-β is the cell volume. The volume of 2-I-α (3864(2) ų) is larger by about 100 ų than that of 2-I-β (3771.2(7) ų) in spite of both polymorphs having the same chemical formula. The larger cell volume of 2-I-α, with a loosely packed structure, is expected to allow structural relaxation in the excited state more effectively than 2-I-β (see 3.2.2.). There can be seen another notable difference between the conformation of the iq ligand for 2-Cl and the other complexes 2-X (X = Br⁻ and I⁻). The C37 carbon atoms at 3-position of iq are oriented in the direction of the halide anion for 2-Br and 2-I, while 2-Cl adopts an inverted orientation, which is probably a steric effect of the larger halides of 2-Br and 2-I. Interestingly, the complexes of 3-X, 2-Br, and 2-I-β were found to be isomorphous to each other. Although 3-Cl has a similar molecular structure to 2-Cl, the crystal structure of 3-Cl is different from that of 2-Cl. The Cu-X bond of 3-Cl is shorter than that of 2-Cl (2.3330(9) Å for 2-Cl, 2.312(1) Å for 3-Cl) and the Cu-N bond of 3-Cl is longer than that of 2-Cl (2.091(3) Å for 2-Cl, 2.125(3) Å for 3-Cl). The shorter Cu-X and longer Cu–N bonds of 3-Cl indicate steric repulsion between the Cl ion and nap ligand. Thus, the steric repulsion between the chloride ion and nap ligand likely causes the differences in the crystal structures between 2-Cl and 3-Cl.

3.2 Emission properties

3.2.1 Complexes 1-X (X = Cl⁻, Br⁻, I⁻)

Fig. 3 shows the emission spectra of 1-X at 298 K and 77 K in the crystals, and the photophysical parameters of 1-X are summarized in Table 3. Complexes 1-X exhibit blue-green light emissions when they are excited with UV light, and the emission maxima at 298 K were observed at 498, 483, and 485 nm for 1-Cl, 1-Br, and 1-I, respectively. All spectra of 1-X at both 298 and 77 K are broad without vibronic progressions, indicating that the emissive excited states have a charge-transfer character in this temperature range. Each emission spectrum measured at 77 K shifted to lower energies by ca. 540–850 cm⁻¹ than the corresponding spectrum measured at 298 K. Interestingly, the luminescence quantum yields of 1-X in crystals were found to be extremely high (0.98, 0.95, and 0.99 for 1-Cl, 1-Br, and 1-I, respectively) although the measurements in solution were difficult because of the ligand-dissociation and/or the dimerization as usual for mononuclear Cu(I) complexes [13]. The emission lifetimes of complexes 1-X in the crystals were estimated to be in the range from several microseconds to several tens of microseconds by least-square fitting of the emission decays (see Experimental). These emission properties of complexes 1-X are similar to those of [CuX(PPh₃)₂(4-Mepy)], whose emission properties were characterized as delayed fluorescence at room temperature [13]. Therefore, the luminescence of complexes 1-X could have a similar origin. The different radiative rate constants (k_r) at 298 K and 77 K, k_r values at 298 K are several times larger than that at 77 K, suggest that luminescence of 1-X may be assigned to delayed fluorescence at room temperature.

Table 3

Luminescence properties of complexes 1-X, 2-X, and 3-X in the crystals at 298 K and 77 K.

Compound	1-Cl		1-Br		1-I
	298 K	77 K	298 K	77 K	298 K	77 K
λmaxa(nm)	498	516	483	496	48g	506
τ (μs (An))b	3.3 (0.13) 12.5 (0.89)	7.5 (0.12) 27.4 (0.88)	21.1 (0.84) 27.0 (0.14)	40.0 (0.43) 53.6 (0.57)	10 (0.21) 20 (0.82)g	28 (0.14) 94 (0.84)
Φc	0.98	0.89	0.95	0.93	0.9g	0.99
τavd(μs)	12	27	22	49	19	91
kre(s−1)	8.0·104	3.3·104	4.3·104	1.9·104	5.2·104	1.1·104
knrf(s−1)	1.6·103	4.1·103	2.2·103	1.4·103	5.3·102	1.1·102
Compound	2-Cl		2-Br		2-I-α		2-I-β
	298 K	77 K	298 K	77 K	298 K	77 K	298 K	77 K
λmaxa(nm)	491, 515, 553	476, 488, 511, 522, 549	554	486, 497, 507, 520, 532, 561	53g	486, 521, 561, 610	487, 520, 558	484, 496, 520, 532, 561, 607
τ (μs (An))b	154 (0.60) 616 (0.48)	227 (0.49) 1436(0.54)	73 (0.58) 165 (0.47)	152 (0.84) 825 (0.23)	27 (0.54) 78 (0.52)g	123 (0.94) 534 (0.18)	79 (0.74) 493 (0.30)	126 (0.86) 777 (0.23)
Φc	0.77	0.56	0.46	0.20	0.6g	0.51	0.65	0.50
τavd(μs)	506	1285	132	553	65	310	376	528
kre(s−1)	1.5·103	4.4·102	3.5·103	3.6·102	9.8·103	1.6·103	1.7·103	9.5·102
knrf(s−1)	4.5·102	3.4·102	4.1·103	1.4·103	5.7·103	1.6·103	9.3·102	9.5·102
Compound	3-Cl		3-Br		3-I
	298 K	77 K	298 K	77 K	298 K	77 K
λmaxa(nm)	636	674	608	634	571g	585
τ (μs (An))b	0.32 (0.12) 1.5 (0.91)	3.8 (0.20) 14.4 (0.84)	1.6 (0.29) 3.6 (1.6)	21.9 (0.033) 61.9 (0.97)	4.7 (0.87)g 6.3 (0.14)	46.2 (0.16) 62.7 (0.84)
Φc	0.16	0.08	0.44	0.25	0.7g	0.63
τavd(μs)	1.5	14	3.5	38	5.0	73
kre(s−1)	1.1·105	5.8·103	1.2·105	6.6·103	1.5·105	8.9·103
knrf(s−1)	5.7·105	6.7·104	1.6·105	2.0·104	5.3·104	5.2·103

^a Emission maximum.

^b Emission lifetime. Emission decays were analyzed with two components: I = A₁ exp(−t/τ₁) + A₂ exp(−t/τ₂).

^c Photoluminescence quantum yields in the solid states.

^d Average emission lifetimes were determined by using Eq. (1).

^e Radiative rate constants k_r were estimated by Φ/τ_av.

^f Nonradiative rate constants k_nr were estimated by k_r(1 − Φ)/Φ.

^g Ref. [14].

To clarify the emission properties in detail, the temperature dependence of the emission lifetimes of complexes 1-X was investigated. As shown in Fig. 4, the lifetimes of complexes 1-X changed gradually in the low-temperature region. However, a sharp decrease in the lifetime was observed with increasing temperature above 170 K. Assuming a two-state model involving the lowest excited singlet state (S₁) and the lowest excited triplet state (T₁), the observed lifetime can be expressed as a Boltzmann average by using equation (2) [11,34–36],

τobs=3+exp−ΔERT3τT1+τs1exp−ΔERT

(2)

where ΔE is the energy difference between the singlet and triplet states, τS1 and τT1 are the lifetimes of S₁ (fluorescence) and T₁ (phosphorescence) states, R is the ideal gas constant, and T is the absolute temperature. In this two-state model analysis, the average lifetime was used instead of τ_obs, i.e. τ_obs = τ_av. The corresponding fitting parameters are listed in Table 4. The obtained ΔE values for 1-X (740–850 cm⁻¹) are roughly consistent with the spectral shifts of ∼1000 cm⁻¹ between 298 and 77 K, indicating that the two-state model is reasonable for this temperature range. Using fitting parameters, singlet components of luminescence (I(S₁)/I_total) can be estimated by using equation (3) [35],

IS1Itotal=1−1+ΦS1τT13ΦT1τS1exp−ΔERT−1

(3)

where, ΦS1 and ΦT1 are luminescence quantum yields of the singlet and triplet states. As noted above, the k_r value at 298 K and 77 K and lifetime analysis of complexes 1-X revealed that the emissive states of complexes 1-X at 298 K are mainly singlet states and those at 77 K are triplet states, therefore we approximate ΦS1 and ΦT1 by Φ at 298 K and 77 K, respectively. As a result, the estimated singlet components at 298 K are 57, 53, and 80% for 1-Cl, 1-Br, and 1-I, respectively (Table 4). While the emission of 1-I is mainly from the singlet state at 298 K, the singlet components of emission from 1-Cl and 1-Br are no more than 60%. The smaller singlet emission components would result in the larger k_r values of 1-Cl and 1-Br for the phosphorescence component (1.9–3.3 × 10⁴ s⁻¹) compared to that of 1-I (1.1 × 10⁴ s⁻¹), because ΔE values are comparable to each other. Taking account of the heavy atom effect, the k_r values of 1-X at 77 K are expected to increase in the following order: 1-Cl < 1-Br < 1-I. However, the obtained results showed the opposite trend. It is noteworthy that the trend can be seen also in the 4-Mepy complexes, [CuX(PPh₃)₂(4-Mepy)] [13] and 3-X (vide infra), which have the ^1,3(M + X)LCT emission state, while the 2-X complexes with the ³ππ* emission state exhibit a normal trend of the halide effect (vide infra). Further investigation concerning the properties of the triplet sublevels would be necessary for the elucidation of the halide effects on the (M + X)LCT emission state.

Table 4

Fitting parameters of 1-X and 3-X based on the two-state model Eq. (2), and estimated luminescent singlet components of 1-X and 3-X at 298 K.

	1-Cl	1-Br	1-I	3-Cl	3-Br	3-I
ΔEa (cm−1)	740	850	810	710	670	550
τ(S1)b (ns)	203	230	155	18	51	119
τ(T1)b (μs)	27	47	91	11	37	71
I(S1)/Itotalc (%)	57	53	80	93	94	94

^a Energy difference between S₁ and T₁.

^b Lifetime of the S₁ and T₁ states determined by fitting the data using Eq. (2) with the averaged emission lifetime.

^c Determined by using Eq. (3).

3.2.2 Complexes 2-X (X = Cl⁻, Br⁻, I⁻)

Fig. 5 shows the emission spectra of 2-X at 298 K and 77 K in the crystals. In contrast to the spectra of 1-X, those of 2-X exhibit vibronic structures except for those of 2-Cl and 2-I-β at 298 K, which suggests ³ππ* character of the luminescence. The longer lifetimes (300–1300 μs at 77 K) for 2-X compared to those found for 1-X support this assignment. The order of k_r at 77 K ([4.4, 3.6, 9.5, and 16] × 10² s⁻¹ for 2-Cl, 2-Br, 2-I-β, 2-I-α, respectively) show a normal trend of the heavy atom effect of the halide ions. Furthermore, complexes 2-Cl and 2-I-β exhibit rather long lifetimes even at room temperature (506 and 376 μs, respectively), which is consistent with the spectral profiles of the vibronic structures. On the other hand, the broad spectra of 2-Br and 2-I-α can be assigned to an emission from the triplet CT state because of their shorter lifetimes (132 and 65 μs, respectively) compared to those of 2-Cl and 2-I-β at 298 K. Thus, the complexes 2-X bearing an isoquinoline are interesting systems with competitive emission states between the triplet CT and ³ππ*. It is noteworthy that the two polymorphic forms 2-I-α and 2-I-β provide different emission spectra at 298 K, although the structured spectra at 77 K are essentially the same. As discussed in the crystal structures section, the cell volume of 2-I-α is ca. 100 ų, which is larger than that of 2-I-β, which suggests that structural relaxation could occur more effectively in the excited state. Structural relaxation in the excited state probably stabilizes the CT excited state more effectively than in the ππ* state, because the MLCT excited state for tetrahedral Cu(I) complexes is liable to flattening distortion resulting in the broad emission spectrum observed for 2-I-α at 298 K. The emission wavelength of 2-Br is longer than that of 2-I, indicating that the emission energy depends on the halide anion as observed for the charge-transfer type emissions of halide-bridged dinuclear complexes [15]. Therefore, the CT excited state of 2-Br is more stabilized than the excited state of 2-I, which results in the broad emission spectrum of 2-Br. In addition, 1-X and 2-X also exhibit relatively high luminescence quantum yields (0.44–0.73) at 298 K; however, the values decreased to 0.20–0.51 at 77 K. The reverse trend in the quantum yields supports the contribution to the emission from different emission states (i.e. ³ππ* and ³(M + X)LCT) at 298 K, although the emission occurs from only ³ππ* state at 77 K.

3.2.3 Complexes 3-X (X = Cl⁻, Br⁻, I⁻)

Fig. 6 shows the emission spectra of 3-X at 298 K and 77 K in the crystals. The spectral profiles of 3-X are similar to those of 1-X, suggesting that the emission occurs from the triplet CT state, although the emission maxima appear at much longer wavelengths (571, 608, and 636 nm for 3-I, 3-Br, and 3-Cl, respectively). In contrast to 1-X and 2-X, the emission wavelengths of 3-X clearly show the halide-ligand dependence. The order of the emission wavelengths is consistent with the increasing ligand field strength of the halide ion in the complexes (I⁻ < Br⁻ < Cl⁻). This trend is similar to that observed for the charge-transfer type emissions of halide-bridged dinuclear complexes [8,17,35]. Lower energy shifts (420–890 cm⁻¹) in the emission spectra at 77 K again suggest the contribution from delayed fluorescence at 298 K.

One additional characteristic found for 3-X is the trend of the emission quantum yield decreases drastically in the order of 3-I, 3-Br, and 3-Cl (0.77, 0.44, and 0.16, respectively), which may be explained by the energy gap law [37,38]. As shown in Fig. 7, the plots of ln(k_nr) vs. wavenumber at the emission maxima roughly exhibit a linear relation. The decrease in emission quantum efficiency with decreasing emission energy is interesting since competing nonradiative processes become more relevant (vide infra). The lower luminescence quantum yields at 77 K (0.08 for 3-Cl, 0.25 for 3-Br, and 0.63 for 3-I) compared with those at 298 K suggest that the emissive states at room temperature and low temperature are different than in the case of 1-X.

As shown in Fig. 8, a two-state analysis of the temperature dependence of averaged emission lifetime for 3-X is presented in the same way as that described for 1-X. The results for 3-X as well as 1-X are summarized in Table 4. The ΔE values (550–710 cm⁻¹) for 3-X are roughly consistent with the spectral shifts of ∼1000 cm⁻¹ between 298 and 77 K, which indicates that the two-state model is reasonable for this temperature region. The luminescent singlet components of 3-X were estimated to be 93–94% at 298 K by using equation (3), which indicates that 3-X emit mainly from the singlet state at room temperature.

3.3 Luminescence mechanism

Scheme 2 shows the schematic energy level diagram of complexes 1-X–3-X. For 1-X, the ¹(M + X)LCT state is easily thermally accessible from the ³(M + X)LCT emissive state at room temperature because the ΔE between these states is small. However, the luminescent singlet components of 1-X are only 53–80%. This is probably due to the large phosphorescent rate constant of 1-X; thus the delayed fluorescent process competes with the phosphorescent process. Therefore, 1-X emits light from both the ¹(M + X)LCT and ³(M + X)LCT states at room temperature. At low temperatures, on the other hand, complexes 1-X emit phosphorescence from the ³(M + X)LCT state because the thermally activation to ¹(M + X)LCT state is suppressed. Complexes 3-X also exhibit similar emission behavior and the luminescent singlet components of 3-X are almost 100%. Therefore, complexes 3-X mainly emit delayed fluorescence from the ¹(M + X)LCT state at room temperature. In contrast to 1-X and 3-X, the origins of the emission of 2-X are different. The emission studies clearly indicate the involvement of the ³ππ* state in the emission states of 2-X, although the TD-DFT calculations corresponding to the light absorption process suggest that the lowest singlet excited states are all the (M + X)LCT state, including those of 2-X (see Supporting Information). It is interesting to note that the ππ* (L) transition energies of 1-I − 3-I (HOMO-20 to LUMO, HOMO-5 to LUMO, and HOMO-13 to LUMO for 1-I, 2-I, and 3-I, respectively) are different from each other (6.512, 4.804, and 4.900 eV for 1-I, 2-I, and 3-I, respectively). Thus the ππ*(L) transition energies are relatively small for 2-X, which are consistent with those of the free ligand L [39–42]. The small energy difference for 2-I suggests that the (M + X)LCT state and the ππ* state of 2-I are more accessible to each other than those of 1-I and 3-I. Experimentally the phosphorescence maximum of the iq ligand is observed at 520 nm [41,42] and the emission maxima of 2-I is 520–530 nm, which proves that the ³(M + X)LCT and ³ππ* states of 2-I should be close in energy and can be thermally accessible to each other, although the ³ππ* levels of 1-I and 3-I are much higher than the ³(M + X)LCT levels. The characteristic properties of 2-I appeared as the different emission spectra between the two morphs of 2-I (α and β) at room temperature. The difference in the flexibility between them in the crystal could give rise to the different spectral features as discussed in section 3.2.2; more flexible 2-I-α can be relaxed structurally forming more stabilized ³(M + X)LCT state.