A HF-Free Synthesis Method for High-Luminescent Efficiency Narrow-Bandgap Red Phosphor K₃AlF₆: Mn⁴⁺ with NH₄HF₂ as the Molten Salt

Abstract

Mn⁴⁺-doped fluoride red phosphors are widely used in white LED lighting and display applications due to their excellent luminescent properties. However, their synthesis relies heavily on highly toxic aqueous hydrofluoric acid, which not only causes severe environmental and soil/water pollution but also makes it difficult to control the microstructure of the products due to the rapid reaction rate. In this study, low-melting-point NH₄HF₂ was used as the molten salt, with KMnO₄ and MnF₂ as manganese sources, to synthesize the red phosphor K₃AlF₆: Mn⁴⁺ via the molten salt method. After the reaction, impurities such as NH₄HF₂ were removed by washing with a dilute H₂O₂ solution. The microstructure, photoluminescence properties, thermal quenching behavior, and application in warm white light-emitting diodes (W-LEDs) of the K₃AlF₆: Mn⁴⁺ phosphors were investigated. The results indicate that the phosphors prepared by this method consist of a single pure phase. By adjusting the molten salt content, the morphology of the product can be transformed from nanoparticle-like to nanorod-like structures. All products exhibit the characteristic red emission of Mn⁴⁺ under blue and violet light excitation, with the optimally doped sample achieving an internal quantum efficiency (IQE) of 69% under blue light excitation. The combination of the obtained K₃AlF₆: Mn⁴⁺ with the yellow phosphor YAG enabled the fabrication of W-LEDs. These W-LEDs achieved a color rendering index (Ra) of 86.8, a luminous efficacy (LE) of 77 lm/W, and a correlated color temperature (CCT) of 3690 K, along with excellent color stability under operating conditions.

1. Introduction

Due to the spin-allowed ⁴A_2g→⁴T_2g transition and the electric dipole-forbidden ²E_g→⁴A_2g transition of Mn⁴⁺ ions, the Mn⁴⁺-doped phosphors exhibit broad excitation bands and narrow red emission bands [1,2,3]. Among them, Mn⁴⁺-doped fluoride red phosphors possess narrow emission bands, high quantum efficiency, and excellent thermal stability has attracted much attention [4,5,6]. These characteristics make them particularly attractive for effectively addressing the red deficiency in commercial W-LEDs, garnering significant research interest for their potential application as red-emitting components in blue-light-excited W-LEDs [7,8,9,10].

The ionic radii of Mn⁴⁺ and Al³⁺ are 0.53 Å and 0.54 Å, respectively. This similarity in ionic radius enables Mn⁴⁺ ions to readily substitute for Al³⁺ ions in fluoride matrices, facilitating the preparation of Mn⁴⁺-doped hexafluoroaluminates (with a general formula of M₃AlF₆: Mn⁴⁺, where M = Li, Na, K) [11]. Compared to K₂SiF₆: Mn⁴⁺ and other Mn⁴⁺-doped fluorides, M₃AlF₆: Mn⁴⁺ phosphors exhibit several potential advantages [7,12,13]. Firstly, M₃AlF₆ possesses a melting point exceeding 1000 °C, whereas K₂SiF₆ begins to decompose at 350 °C, resulting in superior thermal stability for M₃AlF₆. Secondly, M₃AlF₆ has lower water solubility compared to K₂SiF₆ and K₂TiF₆, leading to higher chemical stability for M₃AlF₆: Mn⁴⁺ under high-humidity conditions. Thirdly, M₃AlF₆ has long been used as a common chemical solvent for the industrial extraction of aluminum from bauxite and is already produced on a large scale worldwide, which facilitates the low-cost, mass production of M₃AlF₆: Mn⁴⁺. Consequently, the synthesis and luminescent properties of M₃AlF₆: Mn⁴⁺ phosphors have attracted extensive research interest in recent years, such as Na₃AlF₆: Mn⁴⁺ [14], K₃AlF₆: Mn⁴⁺ [12], K₂NaAlF₆: Mn⁴⁺ [15], K₂LiAlF₆: Mn⁴⁺ [16], Cs₃AlF₆: Mn⁴⁺ [17], and A₂BAlF₆: Mn⁴⁺ (where A = Rb, Cs; B = K, Rb) [18].

Several synthetic methods have been established for the preparation of M₃AlF₆: Mn⁴⁺ luminescent materials. For instance, Song et al. successfully synthesized Na₃AlF₆: Mn⁴⁺ phosphors via a co-precipitation route [19], while other researchers have produced K₃AlF₆: Mn⁴⁺ and K₂NaAlF₆: Mn⁴⁺ using techniques such as ion exchange and high-temperature solid-state reactions [20,21]. Nevertheless, the synthesis of these fluoride materials still predominantly depends on highly toxic hydrofluoric acid (HF) solutions, underscoring the pressing demand for greener and safer synthetic strategies [14]. Additionally, in concentrated HF aqueous media, ionic fluorides are prone to rapid nucleation and particle agglomeration, which complicates the control over phosphor particle size and dispersion [22]. To circumvent the use of hazardous HF liquids and F₂ gas, various HF-free approaches have been explored for synthesizing Mn⁴⁺-doped fluoride red phosphors. These include hydrothermal, microwave-assisted, ionothermal reaction, co-precipitation, solid-state reaction, and organic-solvent-assisted methods [23,24,25,26,27,28,29]. Such strategies typically employ less toxic acids and precursors and offer benefits such as shortened reaction duration, high product yields, and the possibility of extending the synthesis to oxyfluoride compositions. However, these alternative techniques also present certain drawbacks, including the frequent need for high-temperature and high-pressure conditions, limited quantum efficiency, relatively small particle size, difficulties in regulating size distribution, the necessity of specific atmospheres for preparation, and challenges in large-scale production [30]. Consequently, the pursuit of more efficient and controllable HF-free synthesis routes continues to be a significant focus in this research field.

In our earlier research, we successfully applied a molten salt method using solid NH₄HF₂ as both the fluorine source and reaction medium to synthesize particulate K₂SiF₆: Mn⁴⁺ red phosphors with particle sizes around 15 μm [31]. This approach entirely eliminates the need for HF solutions or gaseous F₂, which are commonly used in conventional syntheses, thereby mitigating safety hazards and equipment corrosion and aligning with green chemistry principles. To further explore the versatility of the molten salt method in preparing Mn⁴⁺-doped fluorides and the corresponding material characteristics, this work extends the strategy to synthesize K₃AlF₆: Mn⁴⁺ phosphors using NH₄HF₂ as the molten salt medium along with precursors such as K₃AlF₆, KMnO₄ and MnF₂. Upon heating, NH₄HF₂ melts and participates in fluorination reactions with the starting materials at solid–liquid interfaces, yielding an intermediate [MnF₆]²⁻ species. This intermediate subsequently incorporates into the host lattice by partially replacing Al³⁺ in [AlF₆]²⁻ units, leading to the formation of phase-pure K₃AlF₆: Mn⁴⁺. Additionally, the K₃AlF₆ present in the system undergoes dissolution and recrystallization within the molten salt, promoting improved crystallinity and morphological uniformity in the final product. Experimental evidence shows that this technique directly yields rod-like microcrystalline K₃AlF₆: Mn⁴⁺. By tuning synthesis parameters such as the molten salt ratio, reaction temperature, and holding time, the product morphology can be effectively modulated. The overall procedure, based on straightforward solid-state mixing and annealing, constitutes a typical “one-pot” synthesis. It offers notable benefits including operational simplicity, high reproducibility, and minimal equipment demands, rendering it well-suited for large-scale continuous production. This methodology also opens a new avenue for the green and efficient synthesis of other Mn⁴⁺-doped fluoride phosphors.

2. Materials and Methods

2.1. Sample Preparation

The raw materials KF (99.9%, Aladdin^®, Shanghai, China), K₃AlF₆ (99.99%, Aladdin^®, Shanghai), KMnO₄ (AR, Xilong^®, Ganzhou, China), MnF₂ (99.99%, Aladdin^®, Shanghai, China), and NH₄HF₂ (99%, Aladdin^®, Shanghai, China) were all purchased and used directly without further purification. Since KMnO₄ and MnF₂ undergo the following reaction in the NH₄HF₂ molten salt:

2KMnO₄ + 3MnF₂ + 16H⁺ + 24F⁻ → 5MnF₆²⁻ + 2K⁺ + 8H₂O

(1)

The molar ratio of KMnO₄ to MnF₂ to KF was maintained at 2:3:8 during the batching process. The specific experimental procedure is as follows: 10 mmol of K₃AlF₆, x mmol of NH₄HF₂ (x = 10, 20, 30, 40, 50, 70), y mmol of KMnO₄, 1.5y mmol of MnF₂ and 4y mmol of KF (y = 0.25, 0.5, 1, 2, 3) were weighed. The weighed raw materials were thoroughly ground and homogenized in an agate mortar, then transferred into a lidded corundum crucible. The crucible was subsequently placed in a drying oven and reacted at a specified temperature for 12 h. After the reaction was complete, the system was allowed to cool naturally to room temperature. The product was then collected and washed three times with a 20 wt% aqueous H₂O₂ solution. Finally, the product was dried at 60 °C to obtain the K₃AlF₆: Mn⁴⁺ phosphor. A schematic diagram of the sample synthesis process is shown in Figure 1. To facilitate sample identification, the sample codes and their corresponding experimental parameters are listed in

Table 1. The number and corresponding preparation process of samples.

No.	K3AlF6 (mmol)	KMnO4 (mmol)	MnF2 (mmol)	NH4HF2 (mmol)	Reaction Temperature (°C)
KAMF-1	10	0	0	40	140
KAMF-2	10	0.25	0.375	40	140
KAMF-3	10	0.5	0.75	40	140
KAMF-4	10	1	1.5	40	140
KAMF-5	10	2	3	40	140
KAMF-6	10	3	4.5	40	140
KAMF-7	10	1	1.5	10	140
KAMF-8	10	1	1.5	20	140
KAMF-9	10	1	1.5	30	140
KAMF-10	10	1	1.5	50	140
KAMF-11	10	1	1.5	70	140
KAMF-12	10	1	1.5	40	120
KAMF-13	10	1	1.5	40	130
KAMF-14	10	1	1.5	40	160
KAMF-15	10	1	1.5	40	180

.

2.2. Properties Characterization

The phase identification of all samples was carried out using an X-ray powder diffractometer (XRD, Bruker D8 Advanced, Karlsruhe, Germany) with Cu Kα radiation (λ = 1.54056 Å). The morphology and elemental composition of the synthesized samples were examined by scanning electron microscopy (SEM, MLA650F, Hillsboro, OR, USA) equipped with an energy-dispersive spectrometer. Surface elemental information of the samples was detected by X-ray photoelectron spectroscopy (XPS, PHI 5000 VersaProbe III, Chigasaki, Japan). The Mn⁴⁺ ion concentration in the K₃AlF₆: Mn⁴⁺ phosphors was determined using inductively coupled plasma optical emission spectrometry (ICP, Agilent Varian 720, Wilmington, DE, USA). The photoluminescence properties and IQE of the phosphors at room temperature were measured with a fluorescence spectrometer (Edinburgh Instruments^®, FLS 980, Edinburgh, UK) equipped with a 450 W xenon lamp as the excitation source. The temperature-dependent luminescence characteristics (25–200 °C) of the optimal sample were tested using an EX-1000 thermal quenching analysis system (Everfine^®, EX-1000, Hangzhou, China).

2.3. Device Fabrication and Performance Measurements

The W-LED was fabricated using a commercial sapphire-based InGaN chip (1 W power, ~455 nm, supplied by San’an Optoelectronics Co., Ltd.^®, Xiamen, China), commercial yellow phosphor Y₃Al₅O₁₂: Ce³⁺ (YAG, provided by Yantai Shield^®, Yantai, China), and the K₃AlF₆: Mn⁴⁺ red phosphor synthesized in this study. The mass ratio of red phosphor to yellow phosphor was maintained at 6:1, while the mass ratio of the phosphor mixture to UV-curable resin was 1:1. The phosphor blend and resin were thoroughly mixed and then coated onto the surface of the LED chip. After UV irradiation curing, the W-LED device was completed. The photo-electronic performance of the fabricated device was measured using a HAAS-2000 high-accuracy array spectroradiometer (Everfine^®, Hangzhou, China).

3. Results and Discussion

3.1. Structural and Morphological Properties

Figure 2a shows the XRD patterns of samples KAMF-1 and KAMF-4 along with their corresponding standard diffraction cards. All diffraction peaks in Figure 2a match the standard data of pure cubic K₃AlF₆ (JCPDS 03-0615, a = b = c = 8.45676 Å, α = β = γ = 90.0°, V = 604.8 ų). Furthermore, no additional diffraction peaks related to Mn compounds were observed in Figure 2a, indicating that Mn⁴⁺ ions (coordination number CN = 6, ionic radius r = 0.54 Å) either occupy the octahedral sites of Al³⁺ (CN = 6, r = 0.53 Å) or were removed during the sample washing process. The crystal structure corresponding to KAMF-4 is illustrated in Figure 2b. It can be observed that the crystal contains octahedra centered on Al³⁺ ions, where each Al³⁺ coordinates with six F⁻ ions to form regular AlF₆³⁻ octahedra. Due to the nearly identical ionic radii of Mn⁴⁺ (0.54 Å) and Al³⁺ (0.53 Å), despite their different valence states, Mn⁴⁺ can readily substitute Al³⁺ at the octahedral centers and coordinate with F⁻ ions to form stable MnF₆²⁻ octahedra. This structural compatibility facilitates narrow-band red emission under blue light excitation.

Figure 3 presents the XRD patterns of the samples synthesized under varying reaction temperatures and NH₄HF₂ molten salt contents. When the reaction was carried out at 140 °C with 10 mmol of NH₄HF₂, the XRD pattern of sample KAMF-7 displayed impurity peaks at 2θ = 14.6°, 39.7°, 40.3°, and 45.8°, indicating an incomplete reaction between KMnO₄ and MnF₂. As the NH₄HF₂ content increased, the impurity peaks progressively diminished. With 40 mmol of NH₄HF₂, all diffraction peaks could be indexed to a pure K₃AlF₆ phase (Figure 3a), suggesting complete consumption of the precursors. At this stage, Mn ions were either incorporated into the crystal lattice by substituting Al³⁺ or removed during the washing step. Further increasing the NH₄HF₂ amount did not induce any structural changes. In Figure 3b, with fixed amounts of NH₄HF₂ (40 mmol), KMnO₄ (1 mmol), and MnF₂ (1.5 mmol), the reaction temperature was varied from 120 to 180 °C. At 120 °C, the XRD pattern of the sample KAMF-12 showed weak impurity peaks at 2θ = 21°, 22°, and 31°, which can be attributed to the insufficient melting of NH₄HF₂ (melting point ≈125 °C) at this temperature, hindering the complete reaction. When the temperature exceeded the melting point of NH4HF2, all the resulting products formed a pure K₃AlF₆ phase.

Figure 4a–c show the SEM images of the K₃AlF₆ raw material, KAMF-1, and KAMF-4, respectively, while Figure 4d–f present the corresponding energy-dispersive spectrometer (EDS) spectra of the regions marked by circles in the SEM images. The images reveal that the K₃AlF₆ raw material consists of bulk particles larger than 10 μm (Figure 4a). After being maintained in the NH₄HF₂ molten salt for a period, the microstructure of the K₃AlF₆ raw material transformed from large particles into aggregates of numerous small particles with the formation of multiple rod-like structures (Figure 4b). This indicates that the K₃AlF₆ raw material underwent dissolution and recrystallization during the experimental process, which is favorable for Mn⁴⁺ substitution into the K₃AlF₆ lattice by replacing Al³⁺ [19]. As shown in Figure 4c, the addition of KMnO₄ and MnF₂ did not significantly alter the morphology of the sample, suggesting that these additives have limited influence on the morphological evolution of K₃AlF₆. The EDS data in Figure 4f demonstrate that the short rod-shaped samples contain a small amount of Mn atoms, indicating that Mn⁴⁺ has most likely substituted for part of the Al³⁺ and incorporated into the K₃AlF₆ crystal lattice.

In order to further investigate whether the obtained products contain Mn⁴⁺, the XPS spectrum of sample KAMF-4 was measured as shown in Figure 5. Figure 5a clearly demonstrates the presence of K, F, Al and Mn elements in the sample, where the C element is adventitious carbon introduced by the instrument itself for binding energy calibration. Figure 5b and Figure 5c show the high-resolution XPS spectra of Al 2p and Mn 2p, respectively. In Figure 5b, the peak at 76.5 eV corresponds to Al 2p in the trivalent oxidation state. The two peaks at 644.5 eV and 656 eV in Figure 5c are attributed to the paired Mn 2p_1/2 and Mn 2p_3/2 signals from the Mn element [32,33]. The binding energy difference of 11.5 eV between these two peaks is consistent with reported XPS values for Mn⁴⁺ [11].

Based on the aforementioned test results, it was determined that the molten salt NH₄HF₂ is the primary factor influencing the microstructure of the K₃AlF₆ raw material. To further analyze the regulatory effect of the molten salt on the sample morphology, SEM images of samples KAMF-7, KAMF-8, KAMF-9, KAMF-4, KAMF-10, and KAMF-11 were examined, as shown in Figure 6. When the amount of NH₄HF₂ added was 10 mmol (molar ratio to K₃AlF₆ of 1:1), the sample transformed from the large particles of the K₃AlF₆ raw material into finer particles (Figure 4a and Figure 6a), with a small number of rod-like structures observed. Further increasing the molten salt content to 20 mmol resulted in the agglomeration of fine particles and the formation of more rod-like samples (Figure 6b). The inset clearly shows that the rod-like structures are composed of aggregated fine particles. An increase in molten salt content further raised the proportion of rod-like structures (Figure 6c). When the molten salt content reached 40 mmol, the microstructure of the sample was predominantly composed of rod-like structures (Figure 6d). The inset clearly reveals that the nanorods are independent of each other, with their surfaces covered by fine particles. Continuing to increase the NH₄HF₂ content to 50 mmol and 70 mmol did not result in significant changes in the sample morphology, aside from smoother surfaces of the rods (Figure 6e,f). These results indicate that the addition of the molten salt plays a significant regulatory role in the microstructure of K₃AlF₆.

The concentration of luminescent centers (Mn⁴⁺) plays a critical role in determining the luminescent performance of phosphor materials. To quantify the actual Mn⁴⁺ content in the synthesized samples, inductively coupled plasma (ICP) analysis was conducted on samples KAMF-2 to KAMF-6, with the results presented in

Table 2. ICP results of Mn⁴⁺ actual doped concentration of samples.

No.	Mole Ratio of KMnO4 to K3AlF6	Mn4+ Concentration (at%)
KAMF-2	2.5:100	0.92
KAMF-3	5:100	1.59
KAMF-4	10:100	2.53
KAMF-5	20:100	2.11
KAMF-6	30:100	1.47

. Since the molar ratio of KMnO₄ to MnF₂ was fixed at 2:3, only the ratio of KMnO₄ to K₃AlF₃ is provided. The data reveal that as the molar ratio of KMnO₄ to K₃AlF₆ increases from 2.5:100 to 10:100, the Mn⁴⁺ concentration rises from 0.92 at% to 2.53 at%. This suggests that only a limited portion of the manganese species generated during the molten salt reaction is successfully incorporated into the K₃AlF₆ host lattice, while the majority—present as Mn²⁺, Mn³⁺, or unreacted KMnO₄—is removed during washing. Moreover, merely increasing the KMnO₄ content does not effectively enhance the Mn⁴⁺ doping level; in fact, excess KMnO₄ may even suppress Mn⁴⁺ incorporation. It is also important to note that not only the amount of KMnO₄ but also the NH₄HF₂ molten salt content and the reaction temperature significantly influence the resulting Mn⁴⁺ concentration. As illustrated in Table S1, both insufficient NH₄HF₂ and excessively low reaction temperatures hinder the precursor reaction, consistent with the XRD trends in Figure 3. Conversely, overly high amounts of NH₄HF₂ or excessively elevated temperatures also lead to a reduction in Mn⁴⁺ incorporation. The underlying mechanisms for these phenomena warrant further study. In summary, the experimental results confirm that the optimal KMnO₄ concentration is 10 mol% relative to the K₃AlF₆ precursor.

3.2. Optical Properties

Figure 7a,b present the excitation and emission spectra of samples KAMF-2 to KAMF-6 at room temperature. As shown in Figure 7a, monitoring the red emission at 634 nm, all samples exhibit two broad excitation bands within the ultraviolet (UV) to blue spectral region, centered at 370 nm and 467 nm, respectively. These excitation bands correspond to the ⁴A_2g→⁴T_1g and ⁴A_2g→⁴T_2g transitions of the Mn⁴⁺ ions. Under 467 nm blue light excitation, all samples display narrow red emission lines around 634 nm (Figure 7b). These emission lines lie within the spectral range where the eye’s sensitivity remains relatively high, which is beneficial for practical applications. These red emission lines are attributed to the spin- and parity-forbidden ²E_g→⁴A_2g transition of the Mn⁴⁺ ions. The inset in Figure 7b shows digital photographs of the samples under natural light and 365 nm UV lamp irradiation, demonstrating that all samples emit bright red light under blue-violet excitation. Figure 7c shows the relative intensities of the different samples at 634 nm. It can be observed that sample KAMF-4 exhibits the strongest emission intensity due to its highest Mn⁴⁺ doping concentration. Figure 7d depicts the Tanabe–Sugano energy level diagram for the Mn⁴⁺ ion in an octahedral crystal field. In the K₃AlF₆ host with a strong crystal field, the ²E energy level of Mn⁴⁺ is the most stable excited state. Upon UV or blue light excitation, electrons are excited from the ⁴A_2g ground state to the ⁴T_1g and ⁴T_2g excited states, subsequently relax non-radiatively to the ²E_g level, and finally return to the ground state, emitting red light centered near 634 nm. To evaluate the luminescence efficiency, the internal quantum efficiency (IQE) of the KAMF-4 phosphor was measured using an integrating sphere. The results show that under 467 nm excitation, the KAMF-4 phosphor achieves an IQE value of 69%.

Figure 8a shows the PL spectra of samples synthesized with different NH₄HF₂ molten salt contents at 140 °C, demonstrating the influence of molten salt content on the spectral characteristics. Similarly to the results shown in Figure 7b, all samples exhibit narrow-band emission spectra with the main peak centered at 634 nm. As the molten salt content increases, all samples maintain identical spectral profiles while showing variations in emission intensity. The inset displays digital photographs of the samples under ambient light and a 365 nm UV lamp. Figure 8b illustrates the correlation between relative intensity at 634 nm and molten salt content. The luminescence intensity initially increases and then decreases with rising molten salt content, which is attributed to varying concentrations of Mn⁴⁺ ions incorporated into the K₃AlF₆ host under different molten salt conditions. A similar trend is observed in the luminescence properties of samples prepared at different reaction temperatures. As shown in Figure 8c, when the reaction temperature varies from 120 °C to 180 °C, all samples maintain similar spectral shapes but exhibit significant differences in emission intensity, resulting from the thermal instability of K₂MnF₆ at elevated temperatures. In summary, the optimal synthesis parameters are determined to be an NH₄HF₂ to K₃AlF₆ molar ratio of 4:1 and a reaction temperature of 140 °C.

The thermal effects during the LED packaging and usage processes typically lead to reduced luminescence of the phosphors coated on the surface, making it essential to evaluate the thermal stability of the samples. Figure 9 presents the thermal quenching behavior of sample KAMF-4 within the temperature range of 25 °C to 200 °C. The results indicate that the red emission intensity of the sample initially experiences a brief enhancement before progressively decreasing as the temperature rises (Figure 9a). The normalized integrated intensity of Mn⁴⁺ luminescence, shown in Figure 9b, demonstrates that the sample retains 72% of its initial intensity (measured at 25 °C) when heated to 150 °C. Both thermal quenching and thermal degradation can cause shifts in color coordinates and reduce luminous efficiency. Figure 9c illustrates the variation trend of CIE coordinates as the sample temperature increases from 25 °C to 200 °C. The thermal quenching of emission intensity can be explained by the configuration coordinate diagram depicted in Figure 9d. Under blue light excitation, electrons of Mn⁴⁺ transition from the ground state curve (⁴A_2g) to the excited state curve (⁴T_2g), following the process indicated by the arrows in the diagram. At lower temperatures, most electrons return to the ground state, generating red emission. As the temperature rises, the electrons in the excited state will gain more energy and become more active. When this energy exceeds the activation energy (ΔE_a), electrons pass through the crossing point, dissipate energy via lattice vibrations, and return to the ground state non-radiatively. The probability of this thermally activated non-radiative transition is highly dependent on temperature. The higher the temperature, the more excited electrons are activated, consequently leading to reduced emission intensity.

3.3. Device Performance

To evaluate the potential application of the sample in W-LEDs, we fabricated a warm white W-LED by integrating KAMF-4, YAG, and a blue LED chip. The current-dependent device performance was investigated under driving currents ranging from 20 mA to 320 mA, as shown in Figure 10. The results demonstrate that the electroluminescence (EL) spectral shape remains stable without saturation as the driving current increases (Figure 10a). The inset shows digital photographs of the device under ambient light and when powered by 20 mA current. The luminous efficacy (LE) gradually decreased from 77 lm/W at 20 mA to 43 lm/W at 320 mA, primarily due to the efficiency drop of the blue LED chip. Meanwhile, the correlated color temperature (CCT) increases marginally from 3703 K to 3742 K, representing only a 1% change (Figure 10b). As the current rises from 20 mA to 320 mA, the color rendering index (Ra) shows a slight decrease from 86.8 to 84.3 (Figure 10c). Consequently, the chromaticity coordinates shift by Δx = 0.0075 and Δy = 0.017, yet the device continues to emit high-quality warm white light close to the black-body radiation locus (Figure 10d). The key photometric parameters of the device at 20 mA driving current are as follows: CCT is 3690 K, Ra is 86.8, and LE is 77 lm/W. The chromaticity coordinates of this warm white light are (0.3981, 0.3951), positioned on the black-body curve. The relatively low luminous efficacy is attributed to the limited Mn⁴⁺ doping concentration (2.53 at%). Further optimization of experimental parameters to enhance the Mn⁴⁺ doping level is required to meet commercial application standards.

4. Conclusions

In summary, this study demonstrates for the first time the HF-free synthesis of K₃AlF₆: Mn⁴⁺ red phosphor by employing NH₄HF₂ as the molten salt, with KMnO₄ and MnF₂ serving as manganese sources. The NH₄HF₂ molten salt not only provided an acidic environment, facilitating the reaction between KMnO₄ and MnF₂ to form Mn⁴⁺ but also promoted the dissolution and recrystallization of K₃AlF₆. The content of the NH₄HF₂ was identified as a key factor governing the resulting microstructure, leading to either granular or nanorod-like morphology. The presence of Mn in the samples was confirmed by EDS, XPS, and ICP analyses, while the characteristic emission spectrum of Mn⁴⁺ further verified its successful incorporation. The optimal process conditions for the product were determined as follows: NH₄HF₂ and KMnO₄ additions at 400 mol% and 10 mol% relative to the K₃AlF₆ raw material, respectively, with a reaction temperature of 140 °C. Under blue-violet excitation, the resulting samples exhibited a narrow-band red emission centered at 634 nm, with CIE color coordinates of (0.6871, 0.3128). The K₃AlF₆: Mn⁴⁺ sample with optimal doping concentration (2.53 at%) exhibits a room-temperature quantum efficiency of 69%, while retaining 72% of its initial luminescence intensity at 150 °C. The W-LED fabricated using K₃AlF₆: Mn⁴⁺ (2.53 at%) as the red light component achieves a color rendering index of 86.8, a luminous efficacy of 77 lm/W, and a correlated color temperature of 3690 K, while demonstrating excellent color stability during operation. Despite its suitability for large-scale production, the NH₄HF₂-based molten salt synthesis of K₃AlF₆: Mn⁴⁺ still requires further optimization to achieve finer morphological control and higher Mn⁴⁺ doping levels, both essential for advancing the luminescence performance of the material.