Jiaxiang
Liu‡
a,
Baowen
Wang‡
a,
Zhiming
Zhang
a,
Jiamiao
Yu
a,
Xingqiang
Lü
*a,
Guorui
Fu
*a,
Wentao
Li
*ab and
Wai-Yeung
Wong
*b
aSchool of Chemical Engineering, Northwest University, Xi'an 710069, Shaanxi, China. E-mail: lvxq@nwu.edu.cn; fuguorui@nwu.edu.cn
bDepartment of Applied Biology and Chemical Technology, Research Institute for Smart Energy, The Hong Kong Polytechnic University, Hung Hom, Hong Kong, China. E-mail: wentaowh.li@polyu.edu.hk; wai-yeung.wong@polyu.edu.hk
First published on 30th December 2022
Using the designed chiral [Sm(tta)3(D-phen)] as an emitter, the first example of a chiral organo-Ln3+-based CP-WOLED with both attractive white-light efficiencies (ηMaxEQE = 1.55% and ηMaxCE = 1.61 cd A−1) and high dissymmetry factor (|gEL|Max = 0.011) is reported.
As a matter of fact, CPL activity quantified by the dissymmetry factor glum (corresponding to photo-excited gPL or electric-driven gEL) via the following equation11 originates from the intensity (IL or IR) difference of the emissive left and right CP lights. Meanwhile, from the viewpoint of quantum mechanics, glum can further be simplified to be relative to the electric (μ) and magnetic (m) transition dipole moments and the θ angle between.
Encouraged by the progressive advance of organo-Eu3+/Tb3+/Sm3+-colour-primary OLEDs19 or white-light emitters,20 it is of significance to explore the state-of-the-art chiroptical activity of chiral organo-Eu3+/Tb3+/Sm3+ complexes toward desirable CP-WOLEDs while not monochromatic CP-OLEDs.14–16 Considering the simplification of white-light modulation21 from a dichromatic while not typical trichromatic (RGB) strategy, herein, one efficient chiral organo-Sm3+ complex [Sm(tta)3(D-phen)] (see Scheme 1) is molecularly designed. The conception points include: (i) besides the lower triplet state (1T) energy of Htta than that of D-phen, the compatibility between the 1T of Htta and the first excited state (4G5/2) of the Sm3+ ion should engender efficient and high-purity orange-light for [Sm(tta)3(D-phen)]; (ii) motivated by the higher |glum| while shorter lifetime values of organo-Sm3+ complexes than those of the corresponding organo-Eu3+/Tb3+ counterparts in some cases,13 strong CPL-activity and relatively low efficiency-roll-off from [Sm(tta)3(D-phen)] can be expected; (iii) most importantly, keeping Sm3+-centred stable and high-purity orange-light at hand and further using the popular blue-emitting PVK (poly(N-vinylcarbazole) as the colour-compensated host for low-cost solution-processed device fabrication, the targeted [Sm(tta)3(D-phen)]-doped CP-WOLED ruled by the blue-orange dichromatic principle, could be realized. And thus, this present research, to the best of our knowledge, is the first example of chiral organo-Ln3+-based CP-WOLEDs.
The chiral Phen (phenanthroline) derivative D-phen was synthesized in a higher yield (72% versus 65%) according to the improved synthetic procedure (Scheme S1, ESI†) for 2-L-Phen by the stronger Lewis-base KH instead of NaH as in the literature.22 Following the full characterization by EA, FT-IR and 1H NMR (ESI†), its chiroptical activity was further checked to afford [α]25D = 56.0 ± 0.1°, which almost resembled that ([α]25D = 50°) of (+)-menthol, showing the retention of the original absolute configuration. Also as shown in Scheme S1 (ESI†), further through the self-assembly of the chiral ancillary D-phen with the complex precursor [Ln(tta)3(H2O)] (Ln = Sm or La),23 the corresponding chiral organo-Ln(III) complex [Sm(tta)3(D-phen)] or [La(tta)3(D-phen)] was obtained as the iso-structural product, respectively. The two chiral complexes [Sm(tta)3(D-phen)] and [La(tta)3(D-phen)] soluble in common organic solvents except water, were well characterized by EA, FT-IR, 1H NMR and ESI-MS (see ESI†). In particular, based on the 1H NMR spectrum (Fig. 1) of anti-magnetic24 counterpart [La(tta)3(D-phen)], the signals (δ = 9.57–0.75 ppm) of both (tta)− and D-phen proton resonances were identified, respectively, which together with a stipulated 3:1 molar ratio, well confirms its desirable binary tris-(β-diketonate)-La3+ component. Meanwhile, as compared with the free Htta (δ = 7.83–6.45 ppm), the high-field shifts of all the proton signals (δ = 7.54–6.14 ppm) to the coordinated (tta)− ligands were observed. However, in contrast to the free D-phen (δ = 9.17–0.78 ppm), the aromatic proton resonances (δ = 9.57–6.93 ppm) of the coordinated D-phen within were significantly low-field shifted, which, also upon La3+ coordination, should be the reason for the broadened proton resonances (δ = 9.57–0.75 ppm) of [La(tta)3(D-phen)]. The thermogravimetric analysis (TGA) result (see Fig. S1, ESI†) shows that [Sm(tta)3(D-phen)] has similar thermal stability to [La(tta)3(D-phen)], and the decomposition (5 wt% loss) temperature of about 300 °C is sufficient for the following device fabrication.
Fig. 1 1H NMR spectra of the free ligands Htta and D-phen, and their chiral complex [La(tta)3(D-phen)] in CDCl3 at room temperature, respectively. |
The optical properties of the chiral complex [Sm(tta)3(D-phen)] in solution were explored at room temperature, and the results are summarized in Fig. 2. In contrast to the UV-visible absorptions of the free ligands (229, 244, 271 and 314 nm for D-phen, 262 and 329 nm for Htta; Fig. S2, ESI†), the ligand-centred while significantly red-shifted (229, 246, 284 and 330 nm; see Fig. 2(a)) absorption bands of the chiral complex [Sm(tta)3(D-phen)] are observed, due to the Sm3+-coordination. Upon photo-excitation, λex within the 230–420 nm range just renders the Sm3+-centred emissions for the chiral complex [Sm(tta)3(D-phen)], as shown in Fig. 2(b), where the splitting λem = 565, 595, 643 and 711 nm, corresponding to 4G5/2 → 6HJ/2 (J = 5, 7, 9, 11) transitions of Sm3+ ion, are assigned, correspondingly. Noticeably, the highest intensity (λem = 643 nm) is at the hypersensitive 4G5/2 → 6H9/2 transition and the lowest one (λem= 711 nm) from the 4G5/2 → 6H11/2 transition, meaning that the central Sm3+ ion is located in a site without inversion symmetry.25 Meanwhile, apart from the absence of the ligand-centred emissions, the integration of the two electric dipole (μ) transitions above and the other two 4G5/2 → 6H7/2 and 4G5/2 → 6H5/2 magnetic dipole (m) transitions (λem = 565 and 595 nm), engenders a bright colour-purity orange-light with the CIE (Commission International De L’Eclairage) coordinates of x = 0.604, y = 0.371. Moreover, its outstanding photoluminescence is further reflected from the attractive ΦPL of ca. 10% in the solid-state, which is almost comparable to that26 of non-chiral complex [Sm(tta)3(Phen)]. Furthermore, the decay lifetime of the Sm3+-centred (λem = 643 nm) transition for the chiral complex [Sm(tta)3(D-phen)] is 72 μs, distinctively shorter than those (102 μs grade) of typical tris-(β-diketonate)-Eu3+/Tb3+ complexes,27 and thus, it renders an additional opportunity for relatively weak efficiency-roll-off during device application.
To deeply understand the photo-physical behaviour of the chiral complex [Sm(tta)3(D-phen)], TD-DFT (time-dependent density functional theory) calculations of the anti-magnetic chiral complex [La(tta)3(D-phen)] counterpart for simplification were carried out, and the details are summarized in Tables S1, S2 (ESI†) and Fig. 3. As shown in Fig. 3, similar domination from three (tta)− ligands to each of the HOMOs (12.16/32.47/53.96% for the HOMO; 50.03/12.54/36.06% for the HOMO−1; 36.65/53.02/7.88% for the HOMO−2), and the most contribution (85.05%) also from three (tta)− ligands in non-equivalent mode (33.60/45.47/5.96%) to the LUMO+1 is observed. However, a substantial proportion (12.67%) from the D-phen ligand to the LUMO+1 cannot be neglected, which is different from the minor ones to all the HOMOs. As to the LUMO or the LUMO+2, it is mostly localized at the chiral D-phen unit (94.64% versus 86.37%). Through the calculated HOMO → LUMO transition (388 nm) dominated for the S0 → S1 excitation, the experimentally determined low-energy (over 330 nm; also see Fig. 2(a)) should mainly be attributed to the 1LLCT (LLCT = ligand-to-ligand charge transfer; (tta)− to D-phen) transition. Accordingly, besides the HOMO–LUMO energy of 3.69 eV, a theoretical first excited state level (1T; 3π–π*) of 2.453 eV (19802 cm−1) can further be calculated. By checking the energy level match between the 3π–π* and the 4G5/2 (17064 cm−1) of the Sm3+ ion, a suitable energy gap ΔE (2738 cm−1) within the ideal 2500–4000 cm−1 range according to the so-called Latva's empirical rule,28 reasonably confirms the effective sensitization (see Fig. S3, ESI†) of Sm3+ ions. Therefore, not relevant to the chirality but beneficial to the strengthened absorption of D-phen, the efficient and colour-purity Sm3+-centred orange-light for its chiral complex [Sm(tta)3(D-phen)] is understandable.
Fig. 3 The HOMO and LUMO patterns for the chiral complex [Sm(tta)(D-phen)] based on its optimized S0 geometry, respectively. |
The chiroptical properties including CD (circular dichroism; also see Fig. 2(c)) and CPL (see Fig. 2(d)) spectra of the chiral complex [Sm(tta)3(D-phen)] in solution were studied. As shown in Fig. 2(c), the evident CD signals are exhibited, indicative of the chirality retention29 upon complex formation. And in contrast to the Cotton-effect bands (Fig. S4 (ESI†); (+)-231, (−)-246, (+)-267 and (+)-312 nm) of D-phen, the significantly red-shifted Cotton-effect ones at (+)-245, (−)-291, (+)-327 and (−)-365 nm arising from the Sm3+-coordination are observed for the chiral complex [Sm(tta)3(D-phen)]. In particular, the low-energy Cotton-effect absorption can be assigned to the exciton-coupled CT transition. Upon photo-excitation, as shown in Table S3 (ESI†) and Fig. 2(d), the typically Sm3+-centred CPL spectrum of the chiral complex [Sm(tta)3(D-phen)] was exhibited, where based on the corresponding 4G5/2 → 6H7/2 and 4G5/2 → 6H5/2 magnetic dipole (m) allowed transitions, the largest gPL of +0.009 is detected at λem = 560 nm ascribed to the 4G5/2 → 6H5/2 transition. It is worth noting that the D-phen-ancillary |gPL|Max value (0.009) is relatively lower than those (0.10–1.15; |gPL|) of the previous chiral organo-Sm3+ complexes30 with chirality from the N^O/O^O main ligands, whilst their 3π–π* energies are too high to effectively sensitize the Sm3+-centred orange-light. On the other hand, besides the top level of the |gPL| = 0.009 among those of the reported chiral organo-Sm3+complexes with chirality from the ancillary ligands,31 the (tta)−-incorporation leads to the attractive ΦPL (ca. 10%) for its high-purity orange-light. Importantly, the uniquely CPL-active property is unreachable from chiral non-organo-Ln3+ sources (|gPL| values within the 10−5∼10−4-grade),32 despite the more efficient while non-high-purity orange-light.
Considering the strong CPL activity and high-purity orange-light arising from the chiral complex [Sm(tta)3(D-phen)], it is of special interest for low-cost solution-processed CP-WOLEDs. Through its doping into the commercial electron-transporting PBD (2-(4-tert-butylphenyl)-5-(4-biphenyl)-1,3,4-oxadiazole) and the hole-transporting PVK mixture (5:30:65, weight ratio) as the EML (emitting layer), the colour-integration with blue-light (also see Fig. 2(a)) from the bipolar PVK-PBD host and the orange-light from [Sm(tta)3(D-phen)] should be compensated forward to a desirable CP white-light. In particular, besides an effective Förster energy transfer33 confirmed from the significant spectral overlap (also see Fig. 2(a)) between the emission of PVK-PBD and the CT absorption of [Sm(tta)3(D-phen)], its experimentally obtained HOMO/LUMO energies (−5.40/−2.17 eV; CV as Fig. S5, ESI†) fall well within those (−5.80/−2.16 eV) of PVK-PBD, making PVK-PBD a suitable bipolar host. Profiting from a peculiar stepwise alignment of the HOMO/LUMO levels from the EML to TPBi (−6.20/−2.70 eV; further facilitating electron-transport) and to BCP (−6.70/−3.20 eV; hole-blocking), the carrier-balancing device with ITO/PEDOT:PSS (40 nm)/EML (80 nm)/TPBi (30 nm)/BCP (10 nm)/LiF (1 nm)/Al (100 nm) was configured (see Fig. 4(a and b)).
As expected, upon illumination with the turn-on voltage (Von at 1 cd m−2) of 7.0 V, the normalized electroluminescent spectra (Fig. 4(c)) in the applied bias voltage range of 7.0–13.0 V exhibit the simultaneous emissions of the host-based blue-light (λem = 424 nm) and the Sm3+-centred orange-light (4G5/2 → 6HJ/2 (J = 5, 7, 9, 11) transitions). Noticeably, although the dichromatic colour-integration with CIE coordinates x = 0.268–0.297, y = 0.203–0.224 is highly dependent on the applied bias voltage (see Table S4, ESI†), all the emissive colours fall well within the desirable white-light regime. Moreover, all the warm-white-lights, endowing the CRIs (colour render indices) of 96–97 and the CCTs (correlated colour temperatures) of 2625–2631 K, mostly integrated from Sm3+-characteristic orange-light, are relatively stable. Among them, the minor colour-coordinate shifts (|Δx| = 0.029 and |Δy| = ≤ 0.021) with different blue-to-orange relative intensity ratios should probably be due to doping-induced phase-separation of the EML. As shown in Fig. 4(d and e), in contrast to the monotonous increase of the current density (J, mA cm−2) or the luminance (L, cd m−2) with increasing the applied bias voltage, all the efficiencies (ηC (current efficiency; cd A−1)), ηP (power efficiency; lm W−1) and ηEQE (external quantum efficiency)) increase instantly and then decrease steadily throughout the whole illumination with the ηMaxC = 1.61 cd A−1, the ηMaxP = 0.59 lm W−1 and the ηMaxEQE = 1.55% at 8.5 V (L = 3.86 cd m−2), respectively. Even at a practical luminance of 100 cd m−2, the considerable efficiencies (ηMaxC = 1.09 cd A−1, ηMaxP = 0.36 lm W−1 and ηMaxEQE = 1.03%) can be maintained, from which, the efficiency-roll-offs of 32–39% are acceptable. Intriguingly, as shown in Table S5 (ESI†) and Fig. 4(f), the sizable dissymmetric factor |gEL|max of 0.011 at λem = 560 nm (4G5/2 → 6H5/2 transition of Sm3+ ion) is kept.
As compared with those (ηMaxC, 2.0–50.0 cd A−1; |gEL| ∼ 10−3) of the reported CP-WOLEDs from chiral bis-benzoxanethone8a or spiro-type TADFs,8b the ηMaxC = 1.61 cd A−1 of the [Sm(tta)3(D-phen)]-doped CP-WOLED is relatively lower while satisfactory enough to that (ηMaxC > 0.5 cd A−1) for portable full-colour 3D displays.9 Undoubtedly, one of the merits for the [Sm(tta)3(D-phen)]-doped CP-WOLED rests with the |gEL| size increased by one order of magnitude. More importantly, another evident advantage lies in the high-quality and stable white-lights mainly integrated with the Sm3+-centred orange-light. On the other hand, the device performance of the [Sm(tta)3(D-phen)]-doped CP-WOLED is also at the top-level among those (ηMaxC = 0.65–4.90 cd A−1) of the previously reported WOLEDs34 from non-chiral organo-Sm3+ dyes (ΦPL= 5.6–8.1%). Saliently, the higher-efficiency (ηMaxC = 1.61 cd A−1 and ηMaxEQE = 1.55%) should inherently benefit from the improved ΦPL up to 10% from [Sm(tta)3(D-phen)]. In the meantime, relying on the PBD/TPBi-facilitated electron-transport, more effective carrier confinement and recombination in the EML should be in a subordinate position. Furthermore, upon a stepwise alignment35 of the HOMO/LUMO levels combined with the BCP-interface blocking, the desirable carrier balance might take into effect and especially benefit the relatively weak efficiency-roll-offs.
In conclusion, through the incorporation of the enantiopure D-phen as the ancillary, its chiral complex [Sm(tta)3(D-phen)] displaying both efficient (ΦPL = 10%) Sm3+-centred colour-purity orange-light and strong CPL activity (gPL = 0.009; 4G5/2 → 6H5/2 transition) was molecularly designed. Moreover, by using the chiral complex [Sm(tta)3(D-phen)] as the dopant, the resulting CP-WOLED (ηMaxCE = 1.61 cd A−1) and |gEL|Max = 0.011) was successfully developed. Noticeably, this research work, as the first example of chiral organo-Ln3+-based CP-WOLEDs, suggests that chiral organo-Sm3+/Eu3+/Tb3+complexes should be promising candidates for portable 3D full-colour displays.
Footnotes |
† Electronic supplementary information (ESI) available: Starting materials and characterization; UV; PL. See DOI: https://doi.org/10.1039/d2tc04802g |
‡ These authors contributed equally and should be considered co-first authors. |
This journal is © The Royal Society of Chemistry 2023 |