Ligand Engineering Triggered Efficiency Tunable Emission in Zero-Dimensional Manganese Hybrids for White Light-Emitting Diodes

Zero-dimensional (0D) hybrid manganese halides have emerged as promising platforms for the white light-emitting diodes (w-LEDs) owing to their excellent optical properties. Necessary for researching on the structure-activity relationship of photoluminescence (PL), the novel manganese bromides (C13H14N)2MnBr4 and (C13H26N)2MnBr4 are reported by screening two ligands with similar atomic arrangements but various steric configurations. It is found that (C13H14N)2MnBr4 with planar configuration tends to promote a stronger electron-phonon coupling, crystal filed effect and concentration-quenching effect than (C13H26N)2MnBr4 with chair configuration, resulting in the broadband emission (FWHM = 63 nm) to peak at 539 nm with a large Stokes shift (70 nm) and a relatively low photoluminescence quantum yield (PLQY) (46.23%), which makes for the potential application (LED-1, Ra = 82.1) in solid-state lighting. In contrast, (C13H26N)2MnBr4 exhibits a narrowband emission (FWHM = 44 nm) which peaked at 515 nm with a small Stokes shift (47 nm) and a high PLQY of 64.60%, and the as-fabricated white LED-2 reaches a wide colour gamut of 107.8% National Television Standards Committee (NTSC), thus highlighting the immeasurable application prospects in solid-state display. This work clarifies the significance of the spatial configuration of organic cations in hybrids perovskites and enriches the design ideas for function-oriented low-dimensional emitters.


Introduction
Low-dimensional organic-inorganic hybrid metal halides (OHMHs) have been widely recognized for their outstanding optical properties, including high-efficiency tunable emissions, near-unity photoluminescence quantum yields (PLQYs), and long decay lifetimes, etc., which are attributed to the strong quantum confinement at the molecular levels and the most convenient radiative recombination of photo-generated excitons [1][2][3][4]. However, the absorption peaks of most OHMHs are located in the ultraviolet (UV) region, which limits the luminous efficiency of white light-emitting diodes (w-LEDs) because they cannot be excited by commercial blue chips, thus hindering their industrialization in the fields of solid-state lighting and display. Remarkably, the zero-dimensional (0D) hybrid manganese halides become the preference for blue-light-excited luminescent materials since their photoluminescence excitation (PLE) bands lie in the near-ultraviolet and blue regions, while N-Methyldiphenylamine (C 13 H 13 N, 98%, Aladdin, Shanghai, China), N,N-Dicyclohexylmethylamine (C 13 H 25 N, 98%, Aladdin, Shanghai, China), manganese bromide tetrahydrate (MnBr 2 ·4H 2 O, 98%, Aladdin, Shanghai, China), hydrobromic acid (HBr, 40%, Aladdin, Shanghai, China) and ethanol (C 2 H 5 OH, 99.7%, Guangfu, Tianjin, China). MnBr 2 ·4H 2 O needs to be heated at 120 • C for 6 h to remove the crystal water for later use.
The single-crystals of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 were synthesized by a simple solution-phase crystallization method. First, C 13 H 13 N (0.3665 g, 2 mmol) or C 13 H 25 N (0.3907 g, 2 mmol) was dissolved in 1.5 mL HBr for protonation. Then MnBr 2 (0.2148 g, 1 mmol) and 5 mL C 2 H 5 OH were added to the above protonated solution, and heated and stirred at 75 • C until forming a clear solution. After the solution was naturally cooled to room temperature, yellow (C 13 H 14 N) 2 MnBr 4 and green (C 13 H 26 N) 2 MnBr 4 bulky Nanomaterials 2022, 12, 3142 3 of 11 crystals were precipitated overnight. Finally, the crystals were washed several times with acetone and dried in a vacuum oven.

Characterization
Single-crystal X-ray diffraction (SCXRD) data of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 were collected by using an Agilent Technologies Gemini EOS (Palo Alto, CA, USA) diffractometer at 298 K using Mo Kα radiation (λ = 0.71073 Å). Powder X-ray diffraction (PXRD) data of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 for Rietveld analysis was collected at room temperature with a Bruker D8 ADVANCE powder diffractometer EOS (Karlsruhe, Germany) (Cu-Kα radiation) and linear VANTEC detector. The step size of 2θ was 0.011 • , and the counting time was 2 s per step. Rietveld structure refinements were performed by using TOPAS 4.2. PL, PLE spectra, PL decay curves and PLQYs were measured by an Edinburgh FLS920 fluorescence spectrometer EOS (Edinburgh, UK) with a picosecond pulsed diode laser. Temperature-dependent emission spectra were measured on the same spectrophotometer installed with a heating apparatus as the heating source. Morphology observation and elemental mappings were conducted by a scanning electron microscope (SEM, JEOL JSM-6510, EOS, Peabody, MA, USA). UV-vis absorption curves were recorded on a TU-1901 Ultraviolet spectrometer EOS (Beijing, China) at room temperature, in which BaSO 4 was used as the standard reference. CIE chromaticity coordinates were calculated using the CIE calculator software based on the emission spectra excited at 450 nm. The emission spectra, correlated colour temperature (CCT), luminous efficacy, and CIE coordinates of w-LEDs were performed on the integrating sphere spectroradiometer system (ATA-100, Everfine, EOS, Hangzhou, Cina).

Computational Methods
The electronic band structure and density of state (DOS) were calculated by CASTEP based on plane-wave pseudopotential density functional theory (DFT) [23]. Perdew-Burke-Ernzerhof (PBE) functionals in the form of general gradient approximation (GGA) were used for electronic structure calculations [24]. A kinetic energy cut off value of 450 eV and a Monkhorst-pack k-point mesh spanning less than 0.03 Å −1 in the Brillouin zone were chosen.

Results and Discussion
As shown in Figure 1a,d, two ligands with similar atomic arrangements but different spatial configurations were screened out to synthesize hybrid manganese bromides (C 13 where d 0 represents the average Mn-Br bond length and d i are four individual lengths of Mn-Br bond. where θ i refer to the individual Br-Mn-Br angles. The ∆d values of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 are 1.78 × 10 −4 and 2.39 × 10 −4 , respectively. The σ 2 values are 24.64 and 6.08. It is worth noting that the difference in bond length distortion is small and can be ignored, since the large distinction in the bond angle variance maybe the underlying reason for the disparate optical properties of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 . Nanomaterials 2022, 12, 4 of 12 length distortion (∆d) and bond angle variance (σ 2 ) of individual [MnBr4] 2− are calculated by the following formulas [25,26]: where 0 represents the average Mn-Br bond length and are four individual lengths of Mn-Br bond.  The phase purity and crystallinity of (C13H14N)2MnBr4 and (C13H26N)2MnBr4 powders were monitored by PXRD, and the results are shown in Figure S1. All peaks were indexed by triclinic cell (P1) and monoclinic cell (P21/c) with parameters close to those obtained from a single crystal experiment, respectively. Therefore, these structures were considered as a starting model for Rietveld refinement, as shown in Figure 2a,b, which were performed using TOPAS 4.2. The refinement results were stable and gave low R-factors. The main parameters of processing and refinement of (C13H14N)2MnBr4 and (C13H26N)2MnBr4 were listed in Table S3. The elemental mapping images (Figure 2c,d) indicate that the elements N, Br and Mn are evenly distributed on the above manganese bromides. The phase purity and crystallinity of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 powders were monitored by PXRD, and the results are shown in Figure S1. All peaks were indexed by triclinic cell (P1) and monoclinic cell (P2 1 /c) with parameters close to those obtained from a single crystal experiment, respectively. Therefore, these structures were considered as a starting model for Rietveld refinement, as shown in Figure 2a,b, which were performed using TOPAS 4.2. The refinement results were stable and gave low R-factors. The main parameters of processing and refinement of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 were listed in Table S3. The elemental mapping images (Figure 2c,d) indicate that the elements N, Br and Mn are evenly distributed on the above manganese bromides.
As shown in Figure S2b,c, the manganese bromides of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 with yellow-green body colour exhibit bright yellow-green fluorescence under a 365 nm UV lamp, and the corresponding CIE coordinates are depicted in Figure S2a. To further reveal their PL properties, the PLE and PL spectra of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 are initially investigated in Figure 3a,b. Both manganese bromides exhibit similar excitation bands attributed to the d-d transition of tetrahedrally coordinated Mn 2+ centres and can be excited by blue light, but they possess distinct emission spectra. (C 13 H 14 N) 2 MnBr 4 displays a broadband emission peaked at 539 nm with a full width at half maximum (FWHM) of 63 nm and the Stokes shift of 70 nm, which is wider than most 0D hybrid manganese halides [19,27,28]. In contrast, the emission peak of (C 13 H 26 N) 2 MnBr 4 appears at 515 nm, with a FWHM of 44 nm and the Stokes shift of 47 nm. Meanwhile, we synthesized the powder chlorides of (C 13 H 14 N) 2 MnCl 4 and (C 13 H 26 N) 2 MnCl 4 using the above two ligands and they exhibit the same effect and difference on their PL properties as  Figure S3). We consider that the essence for the PL difference may be caused by the different spatial configuration of organic components in above manganese bromides. The ligand C 13 H 13 N contains two planar benzene rings with little steric hindrance. However, the twelve carbon atoms are not in the same plane for C 13 H 25 N, so that there is a larger steric hindrance. Proceeding from (C 13 H 14 N) 2 MnBr 4 to (C 13 H 26 N) 2 MnBr 4 , the increasing cation volume and the steric hindrance not only shrink the free space for the atom movement or the cation/anion rotation, but also inhibit the lattice distortion [29,30]. Therefore, (C 13 H 14 N) 2 MnBr 4 produce stronger electron-phonon coupling and crystal field strength than (C 13 H 26 N) 2 MnBr 4 , resulting in a broadband emission with large Stokes shift. As shown in Figure S2b,c, the manganese bromides of (C13H14N)2MnBr4 and (C13H26N)2MnBr4 with yellow-green body colour exhibit bright yellow-green fluorescence under a 365 nm UV lamp, and the corresponding CIE coordinates are depicted in Figure  S2a. To further reveal their PL properties, the PLE and PL spectra of (C13H14N)2MnBr4 and (C13H26N)2MnBr4 are initially investigated in Figure 3a,b. Both manganese bromides exhibit similar excitation bands attributed to the d-d transition of tetrahedrally coordinated Mn 2+ centres and can be excited by blue light, but they possess distinct emission spectra. (C13H14N)2MnBr4 displays a broadband emission peaked at 539 nm with a full width at half maximum (FWHM) of 63 nm and the Stokes shift of 70 nm, which is wider than most 0D hybrid manganese halides [19,27,28]. In contrast, the emission peak of (C13H26N)2MnBr4 appears at 515 nm, with a FWHM of 44 nm and the Stokes shift of 47 nm. Meanwhile, we synthesized the powder chlorides of (C13H14N)2MnCl4 and (C13H26N)2MnCl4 using the above two ligands and they exhibit the same effect and difference on their PL properties as (C13H14N)2MnBr4 and (C13H26N)2MnBr4 ( Figure S3). We consider that the essence for the PL difference may be caused by the different spatial configuration of organic components in above manganese bromides. The ligand C13H13N contains two planar benzene rings with little steric hindrance. However, the twelve carbon atoms are not in the same plane for C13H25N, so that there is a larger steric hindrance. Proceeding from (C13H14N)2MnBr4 to (C13H26N)2MnBr4, the increasing cation volume and the steric hindrance not only shrink the free space for the atom movement or the cation/anion rotation, but also inhibit the lattice distortion [29,30]. Therefore, (C13H14N)2MnBr4 produce stronger electron-phonon coupling and crystal field strength than (C13H26N)2MnBr4, resulting in a broadband emission with large Stokes shift. Figure 3c depicts the temperature-dependent PL spectra of (C13H14N)2MnBr4 and (C13H26N)2MnBr4 in the range of 80-300 K. The PL intensity decreases with the increasing temperature, which is consistent with the PL quenching behaviour caused by thermally activated non-radiative recombination. Meanwhile, the emission peak positions have a To further reveal the photo-physical properties of (C13H14N)2MnBr4 and (C13H26N)2MnBr4 composed of organic moieties with different spatial configurations. The theoretical calculations involving electronic band structure and densities of states (DOS) are conducted based on the density functional theory. As shown in Figure 4a,b, the calcu-  Figure 3c depicts the temperature-dependent PL spectra of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 in the range of 80-300 K. The PL intensity decreases with the increasing temperature, which is consistent with the PL quenching behaviour caused by thermally activated non-radiative recombination. Meanwhile, the emission peak positions have a blue-shift. It is speculated that the reason for this phenomenon is that the thermally induced lattice expansion weakens the crystal field strength, or reduces the energy loss due to the spin-spin coupling between localized neighbouring Mn 2+ ions [16]. Furthermore, the FWHM versus temperature (Figure 3d) can reveal the origin of the PL differences in above manganese bromides. The Huang-Rhys factor S defines the degree of electron-phonon coupling; it can be solved by the following equation [18,31,32]: where ω is the phonon frequency,hω is the maximum phonon energy, k is the Boltzmann constant, S is Huang-Rhys factor, and coth(x) = e x +e −x e x −e −x = 1 + 2 e 2x −1 . When¯h ω kT is small enough, e¯h ω kT − 1 ≈¯h ω kT , it can be obtained the following equation: Further written as: where a = 5.57 × S × (hω) 2 and b = 5.57 × S × (hω).
The obtained S factor is 2.89 and phonon energyhω phonon is 44.26 meV for (C 13 H 14 N) 2 MnBr 4 , while for (C 13 H 26 N) 2 MnBr 4 , it corresponds to S = 0.77,hω = 75.85 meV. The higher S factor indicates that there is a stronger electron-phonon coupling in (C 13 H 14 N) 2 MnBr 4 , which is favourable for the formation of broadband emission with large Stokes shift.
In addition, the crystal field strength plays a key role in affecting the PL properties of manganese bromides. Previous references reported that the crystal field strength is related to the polyhedral distortion, and the increasing distortion leads to strong crystal field splitting and low position of the lowest 3d excited energy levels, resulting in the red-shift of the emission band [33][34][35][36]. The σ 2 values of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 are 24.64 and 6.08, respectively, as calculated above, indicating that the larger bond angle distortion contributes to a red-shift of the emission peak from 515 nm to 539 nm.
To verify the electronic transition process, we measured the lifetimes excited at 450 nm and monitored at 539 nm and 515 nm, respectively. As shown in Figure 3e,f, the PL decay curves can be fitted with a single order exponential equation: where I(t) and I 0 are the luminescence intensity at time t and t τ, respectively. A is a constant, and τ is the decay time for an exponential component. At 298 K, the lifetimes are on millisecond scale, and the values are determined to be 0.245 ms and 0.370 ms, respectively, which are close to the previously reported lifetimes of hybrid manganese bromides, demonstrating that these emissions belong to the d-d transition ( 4 T 1 -6 A 1 ) of Mn 2+ [17,37,38]. At 80 K, the lifetimes are prolonged, but the variation tendency of (C 13 H 14 N) 2 MnBr 4 is greater than that of (C 13 H 26 N) 2 MnBr 4 , which may be because that (C 13 H 14 N) 2 MnBr 4 has more vibration options. In addition, PLQY is also affected by the spatial configuration of organic cations in these manganese bromides. Figure S4 shows that the [C 13  To further reveal the photo-physical properties of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 composed of organic moieties with different spatial configurations. The theoretical calculations involving electronic band structure and densities of states (DOS) are conducted based on the density functional theory. As shown in Figure 4a,b, the calculated band gap of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 are 2.54 eV and 2.80 eV, respectively, which are coincide with the observed optical absorption spectra ( Figure S5). It is worth noting that the (C 13 H 14 N) 2 MnBr 4 has a lower conduction band level compared to (C 13 H 26 N) 2 MnBr 4 , resulting from the higher degree of conjugation of organic component, which is also responsible for the red-shift of the emission peak of (C 13 H 14 N) 2 MnBr 4 . Both the frontier orbitals are flat and dispersion-less, indicating that the photoelectrons are localized for above manganese bromides. This is because that the isolated [MnBr 4 ] 2− tetrahedrons separated by bulky organic cations have weak interactions, resulting in spatial confinement and electronic confinement effects. Figure 4c  According to the electron transition process, we employ the T-S energy diagram to visualize the effect of spatial configuration of organic ligands on PL behaviours. It is well known that the Mn 2+ ions possess the outermost electron configuration of 3d 5 , and the original five degenerate d orbitals of free Mn 2+ are split into multiple energy levels under the electrostatic field effect generated by ligands. In the tetrahedral field, the ground state is denoted as 6 A1, the excited states of 4 F, 4 P, 4 D and 4 G correspond to the excited energy levels in the UV (200-400 nm) and blue (400-500 nm) regions, respectively [6,7]. The electrons located at the ground state are transferred to different excited states under excitation, and then transferred to the lowest excited state energy level 4 T1 through non-radia- According to the electron transition process, we employ the T-S energy diagram to visualize the effect of spatial configuration of organic ligands on PL behaviours. It is well known that the Mn 2+ ions possess the outermost electron configuration of 3d 5 , and the original five degenerate d orbitals of free Mn 2+ are split into multiple energy levels under the electrostatic field effect generated by ligands. In the tetrahedral field, the ground state is denoted as 6 A 1 , the excited states of 4 F, 4 P, 4 D and 4 G correspond to the excited energy levels in the UV (200-400 nm) and blue (400-500 nm) regions, respectively [6,7]. The electrons located at the ground state are transferred to different excited states under excitation, and then transferred to the lowest excited state energy level 4 T 1 through nonradiative relaxation, and finally the electrons return to the ground state with the release of photons. In the T-S energy diagram, the strength of the crystal field is represented by the abscissa (∆) and the energy difference between excited states 4 T 2 and 4 T 1 . The stronger the crystal field strength, the more obvious the splitting of 4 T 2 and 4 T 1 , making the larger values of E [ 4 T 2 ]-E[ 4 T 1 ]. In view of this, the luminescent mechanism is depicted in Figure 5. To further evaluate their practical applications, the w-LEDs (LED-1 and LED-2) were encapsulated by coating (C13H14N)2MnBr4/(C13H26N)2MnBr4 and KSF:Mn 4+ red phosphors on a commercial InGaN blue chip (450 nm), respectively. As shown in Figure 6, LED-1 presents the standard white-light emission with a relatively high colour-rendering index (CIR) of Ra = 82.1, and a CIE chromaticity coordinate of (0.3211, 0.3206). LED-2 presents a cool white-light emission with a wide colour gamut of 107.8% NTSC in the CIE 1931 colour space, and a CIE chromaticity coordinate of (0.3037, 0.3297). It is concluded that (C13H14N)2MnBr4 is more suitable for solid-state lighting, while (C13H26N)2MnBr4 is unprecedentedly promising as a narrow-band green emitter for solid-state displays. The emission spectra of LED-1 and LED-2 under different drive current different current (20−120 mA) in Figure S7 suggests that the w-LEDs possess excellent colour stability. Of course, we also evaluated the thermal quenching behaviour of (C13H14N)2MnBr4 and (C13H26N)2MnBr4. As shown in Figure S8, the PL intensity decreases with increasing temperature. The PL intensity of (C13H14N)2MnBr4 and (C13H26N)2MnBr4 at 373 K remain around 79.86% and 84.84% of that at room temperature, respectively, suggesting that the above manganese bromides have relatively good thermal stability. To further evaluate their practical applications, the w-LEDs (LED-1 and LED-2) were encapsulated by coating (C 13 H 14 N) 2 MnBr 4 /(C 13 H 26 N) 2 MnBr 4 and KSF:Mn 4+ red phosphors on a commercial InGaN blue chip (450 nm), respectively. As shown in Figure 6, LED-1 presents the standard white-light emission with a relatively high colour-rendering index (CIR) of Ra = 82.1, and a CIE chromaticity coordinate of (0.3211, 0.3206). LED-2 presents a cool white-light emission with a wide colour gamut of 107.8% NTSC in the CIE 1931 colour space, and a CIE chromaticity coordinate of (0.3037, 0.3297). It is concluded that (C 13 H 14 N) 2 MnBr 4 is more suitable for solid-state lighting, while (C 13 H 26 N) 2 MnBr 4 is unprecedentedly promising as a narrow-band green emitter for solid-state displays. The emission spectra of LED-1 and LED-2 under different drive current different current (20−120 mA) in Figure S7 suggests that the w-LEDs possess excellent colour stability. Of course, we also evaluated the thermal quenching behaviour of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 . As shown in Figure S8, the PL intensity decreases with increasing temperature. The PL intensity of (C 13 H 14 N) 2 MnBr 4 and (C 13 H 26 N) 2 MnBr 4 at 373 K remain around 79.86% and 84.84% of that at room temperature, respectively, suggesting that the above manganese bromides have relatively good thermal stability. sion spectra of LED-1 and LED-2 under different drive current different current (20−120 mA) in Figure S7 suggests that the w-LEDs possess excellent colour stability. Of course, we also evaluated the thermal quenching behaviour of (C13H14N)2MnBr4 and (C13H26N)2MnBr4. As shown in Figure S8, the PL intensity decreases with increasing temperature. The PL intensity of (C13H14N)2MnBr4 and (C13H26N)2MnBr4 at 373 K remain around 79.86% and 84.84% of that at room temperature, respectively, suggesting that the above manganese bromides have relatively good thermal stability.

Conclusions
In summary, the two new 0D manganese bromides were developed by using organic cations with different spatial configuration. (C13H14N)2MnBr4 exhibits a broadband emission peaked at 539 nm with a FWHM of 63 nm and a Stokes shift of 70 nm, while

Conclusions
In summary, the two new 0D manganese bromides were developed by using organic cations with different spatial configuration. (C 13 H 14 N) 2 MnBr 4 exhibits a broadband emission peaked at 539 nm with a FWHM of 63 nm and a Stokes shift of 70 nm, while (C 13 H 26 N) 2 MnBr 4 shows a narrowband emission peaked at 515 nm with a FWHM of 44 nm and a Stokes shift of 47 nm, which is ascribed to that the ligand of C 13 H 13 N with planar configuration induces a stronger electron-phonon coupling (S = 2.89) and crystal field strength (σ 2 = 24.64) than the ligand of C 13 H 25 N with chair configuration. DFT calculations reveal that (C 13 H 14 N) 2 MnBr 4 possesses a narrower bandgap compared to (C 13 H 26 N) 2 MnBr 4 , resulting in the red-shift of the emission peak of (C 13 H 14 N) 2 MnBr 4 . Moreover, the steric effect of organic cations on PL behaviours in manganese halides is comprehensively reflected in T-S energy diagram. The as-fabricated white LED-1 based on (C 13 H 14 N) 2 MnBr 4 with a high CIR of Ra = 82.1 is suitable for solid-state lighting, while the as-fabricated white LED-2 based on (C 13 H 26 N) 2 MnBr 4 with a wide colour gamut of 107.8% NTSC in the CIE 1931 colour space presents an unprecedentedly promising utility for solid-state display.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.