Article pubs.acs.org/IC
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
HF-Free Synthesis of Li2SiF6:Mn4+: A Red-Emitting Phosphor Christiane Stoll,† Jascha Bandemehr,‡ Florian Kraus,‡ Markus Seibald,§ Dominik Baumann,§ Michael J. Schmidberger,§ and Hubert Huppertz*,† †
Institut für Allgemeine, Anorganische und Theoretische Chemie, Universität Innsbruck, Innrain 80-82, 6020 Innsbruck, Austria Anorganische Chemie, Fluorchemie, Fachbereich Chemie, Philipps-Universität Marburg, Hans-Meerwein-Straße 4, 35032 Marburg, Germany § OSRAM Opto Semiconductors GmbH, Mittelstetter Weg 2, 86830 Schwabmünchen, Germany ‡
Inorg. Chem. Downloaded from pubs.acs.org by IDAHO STATE UNIV on 03/21/19. For personal use only.
S Supporting Information *
ABSTRACT: Li2SiF6:Mn4+ was synthesized via a new HF-free synthesis route by a high-pressure/high-temperature doping experiment at 5.5 GPa and 750 °C. It is proven that the phosphor cannot be synthesized by the common wet-chemical precipitation route in aqueous HF. The sample was characterized by powder X-ray diffraction, EDX, and luminescence spectroscopy. At room temperature, Li2SiF6:Mn4+ exhibits seven emission lines with the strongest line at λmax ≈ 630 nm and a dominant wavelength of λdom ≈ 618 nm. The CIE coordinates are 0.688 and 0.312 for x and y, respectively. The compound shows a luminous efficacy of radiation (LER) of 218 lm Wopt−1, which exceeds the LER of current state-of-the-art red LED phosphor K2SiF6:Mn4+ by 7% due to a blue-shift of the emission. It reveals excellent thermal quenching behavior up to 125 °C. hexafluorides A2MF6:Mn4+ (A = K, Na, Cs; M = Si, Ge, Zr, Ti) was prepared since the discovery of K2SiF6:Mn4+ by Adachi and co-workers.5,8−18 Usually, these kind of phosphors are prepared via wet chemical etching processes in aqueous HF solutions.6,7,9,19 Due to high security measures, which are mandatory during the application of HF in industrial processes, HF-free synthesis methods are in high demand. The first “HFfree” synthesis of K2SiF6:Mn4+ was published by Wang and coworkers in 2006.20 They used a hydrothermal synthesis route with a mixture of H3PO4/KHF2 as a stabilizer for Mn4+ in solution and simultaneously as a fluorinating agent.20 Xie and co-workers optimized this synthetic method to get better internal quantum efficiencies by using a less toxic NH4F/HCl solution.21 Sohn’s group managed to develop a solid state strategy to synthesize K3SiF7:Mn4+ by grinding KHF2, K2SiF6, and the dopant K2MnF6 together and heating of the mixture under an H2/N2 environment.3 In 2018, the first “HF-free” solid-state route to a hexafluoridotitanate phosphor was discovered by Jiao and co-workers.22 They mixed the hexafluoridotitanate with the dopant K2MnF6 and various amounts of KHF2 and heated the sample in closed PTFE-lined vessels.22 In this work, we present a new HF-free synthesis route via a high-pressure/high-temperature experiment for the preparation of the new hexafluoridosilicate phosphor Li2SiF6:Mn4+, which probably cannot be synthesized via a common wetchemical etching process. Not many lithium-containing
1. INTRODUCTION Lighting accounts for a major portion of energy consumption worldwide and thus offers a high potential to reduce costs. With the development and improvement of the light-emitting diode (LED), there was an increase in efficiency of lighting. The development and enhancement of white light-emitting diodes (WLED) provides potential for significant reduction in energy consumption.1 One of the best ways to optimize WLEDs seems to be the improvement of luminescent materials. The red component especially is of great importance regarding the energy-saving potential in WLEDs, due to the fact that light emitted in the deep-red and infrared region is hardly noticeable by the human eye.1 As of today, there are different kinds of luminescent materials that are used as red emitters. For example, there are band-emitters with Eu2+ or Ce3+ as dopants or line-emitters with Mn4+ as dopant.2−4 Sr[LiAl3N4]:Eu2+ is a representative for red band-emitters, which has gained a lot of attention in the past few years.2 K2SiF6:Mn4+ is a state-of-the-art line-emitter in the red region of the visible spectrum.5−7 Both Sr[LiAl3N4]:Eu2+ and K2SiF6:Mn4+ are suited for application in WLEDs. However, even though Sr[LiAl3N4]:Eu2+ has a very small bandwidth (fwhm ≈ 50 nm, 1180 cm−1), its emission maximum of ∼650 nm already lies in the deep-red region,2 where the sensitivity of the human eye decreases rapidly. In comparison to Sr[LiAl3N4]:Eu2+, the main emission maximum of K2SiF6:Mn4+ is blue-shifted (∼ 630 nm)5 with a narrow bandwidth resulting from its line-emitting nature. Due to the small bandwidth and the good spectral position, the class of manganese-doped fluorides gained a lot of attention. A whole series of © XXXX American Chemical Society
Received: December 10, 2018
A
DOI: 10.1021/acs.inorgchem.8b03433 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
Figure 1. (top) Rietveld plot (pseudo-Voigt approximation) of the sample obtained by doping Li2SiF6 with Cs2MnF6. The experimental powder pattern is shown in black (crosses), the calculated pattern is in red (solid line), and the difference plot is shown in blue (solid line). The reflection positions of Li2SiF6 (ICSD 425923)27 and Cs2SiF6 (ICSD 26871)29 are shown in green and orange, respectively. Li2SiF6 is observed as the main component, with Cs2SiF6 as a side product. (bottom) Rietveld plot (pseudo-Voigt approximation) of the sample obtained by doping Li2SiF6 with K2MnF6. The experimental powder pattern is shown in black (crosses), the calculated pattern is in red (solid line), and the difference plot is shown in blue (solid line). The reflection positions of Li2SiF6 (ICSD 425923)27 and LiF (ICSD 18012)30 are shown in green and pink, respectively. Reflections belonging to KLiSiF631 are marked with purple dots. Li2SiF6 is observed as the main component. LiF and KLiSiF6 are present as side products.
hexafluoridosilicates are known so far; LiNa2AlF6:Mn4+23 is one of them. However, the lithium ion is the lightest cation besides H+; therefore, a blueshift of the emission is expected for lithium-rich phases in comparison to compounds of its heavier congeners. Because even a small blueshift can lead to a higher luminous efficacy of radiation (LER), we expect good performance of Li2SiF6:Mn4+ in WLEDs.
Germany). A pressure of 5.5 GPa was applied by a 1000 t multianvil press (Max Voggenreiter GmbH, Mainleus, Germany) equipped with a Walker-type module (Max Voggenreiter GmbH, Mainleus, Germany). The above-named mixture was compressed to 5.5 GPa within 145 min and kept at this pressure during the heating program. The temperature was raised to 750 °C within 10 min, kept for 150 min, and lowered to 350 °C within 180 min. Subsequently, the sample was quenched to room temperature, followed by decompression of the assembly within 430 min. A colorless product, which showed red luminescence, was recovered. Powder X-ray analysis showed Li2SiF6 as the main product with Cs2SiF6 as side product (Figure 1, top). The use of high-pressure conditions for this step is crucial, because at ambient pressure Li2SiF6 decomposes to LiF and SiF4 at temperatures above 250 °C.27 A second experiment employing the same reaction parameters was conducted by using a mixture of Li2SiF6 and K2MnF6 with a molar ratio of 1:0.059. A colorless product, which also showed a red luminescence, was recovered. Powder X-ray analysis exhibited Li2SiF6 as the main product with LiKSiF6 and LiF as side products (Figure 1, bottom). A third experiment was conducted employing the same high-pressure/high-temperature reaction parameters but using only Cs2MnF6 as reactant.
2. EXPERIMENTAL SECTION 2.1. Synthesis. Caution: HF and its derivatives are potentially hazardous materials and must be handled using suitable protective gear with immediate access to proper treatment procedures. Doping of Li2SiF6 with Mn4+ was realized under high-pressure/ high-temperature conditions by employing Cs2MnF6 as dopant. For this experiment, a mixture of Li2SiF6 and Cs2MnF6 with a molar ratio of 1:0.065 was weighed in and ground together in an agate mortar under argon atmosphere in a glovebox. The mixture was filled into a platinum capsule (99.95%, Ö gussa, Vienna, Austria) and then placed in a boron nitride crucible (Henze Boron Nitride Products AG, Lauben, Germany), which was inserted in an 18/11 assembly (details of the assembly and its preparation are described elsewhere)24−26 and compressed by eight tungsten carbide cubes (Hawedia, Marklkofen, B
DOI: 10.1021/acs.inorgchem.8b03433 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
Figure 2. Emission spectra of Li2SiF6:Mn4+ (bottom, 1: sample doped with K2MnF6; 2: sample doped with Cs2MnF6) in comparison to the spectra of K2SiF6:Mn4+ (middle) and Cs2MnF6 (top). Furthermore, it was attempted to synthesize K2SiF6:Mn4+ and Li2SiF6:Mn4+ via a wet-chemical approach in hydrofluoric acid. For the synthesis of K2SiF6:Mn4+, SiO2 (267.1 mg, 4.44 mmol, Alfa, 99.5%) was added to hydrofluoric acid (20 mL, 58−62%, Fisher Chemical). After stirring for 45 min, the SiO2 was completely dissolved, and K2CO3 (1.9640 g, 14.211 mmol, Grüssing GmbH, 99.5%) was carefully added until a precipitate formed. The mixture was cooled to 0 °C and solid KMnO4 (1.2202 g, 7.7208 mmol) was dissolved. Over a period of 1 h, H2O2 (35%, Roth, ROTIPURAN p.a.) was added dropwise under stirring until the violet color disappeared. After filtration, the light yellow K2SiF6:Mn4+ was washed twice with hydrofluoric acid (10 mL, 20%), two times with acetone (10 mL, techn. grade), and dried for 1 h at 60 °C. Powder X-ray analysis showed that phase-pure K2SiF6 was obtained (Figure S1). The successful doping of K2SiF6:Mn4+ was revealed by a red luminescence under UV irradiation (∼395 nm). The emission spectrum (Figure S2, solid line) shows the typical lines of K2SiF6:Mn4+. Synthesis of Li2SiF6:Mn4+ was attempted in the same way as described for K2SiF6:Mn4+. For this experiment, 261.2 mg (4.348 mmol) of SiO2, 1.1240 g (7.114 mmol) of KMnO4, and 341.0 mg (4.614 mmol) of Li2CO3 (Merck, Suprapur) were used. Powder X-ray analysis of the colorless product showed LiF and K2SiF6 as crystalline products (Figure S3). The mixture showed no luminescence under UV irradiation and no signal in luminescence spectroscopy (Figure S2, dotted line). 2.2. Powder X-ray Diffraction. The powder was analyzed by a Stoe Stadi P powder diffractometer (STOE, Darmstadt, Germany) in transmission geometry with Mo Kα1 radiation (λ = 70.93 pm) utilizing a focusing Ge(111) primary beam monochromator and a Mythen 2 DCS4 detector. The measurement was performed in the 2θ range 2− 40.4°. Figure 1 shows the Rietveld analysis of the powder sample concerning the doping experiments with Cs2MnF6 (Figure 1, top) and K2MnF6 (Figure 1, bottom). The analysis was carried out using a LaB6 standard. The program suite TOPAS 4.228 was used for the reflection shape refinements. The powder samples obtained via the wet-chemical synthesis route were prepared between 3M Scotch Magic-tape and subsequently measured using a Stoe Stadi MP powder diffractometer (STOE, Darmstadt, Germany) utilizing Cu Kα1 radiation (λ = 154.06 pm), a Ge(111) primary beam monochromator and a Mythen1K detector. The measurements were performed in the 2θ range of 5.0−93.1°. The powder patterns concerning the experiments via the wet-chemical route are depicted in Figures S1 and S3.
2.3. Luminescence Spectroscopy. The emission spectra of Li2SiF6:Mn4+, K2MnF6, K2SiF6:Mn4+, and Cs2MnF6 were recorded using a setup equipped with a blue laser diode (λ = 448 nm, THORLABS, Newton, USA) and a CCD-Detector (AVA AvaSpec 2048, AVANTES, Apeldoorn, Netherlands). Prior to the experiments, a tungsten-halogen calibration lamp was used for a spectral radiance calibration of the setup. The software AVA AvaSoft full version 7 was used for data handling. The emission data is depicted in Figure 2. Luminescence investigations of the powder material of the synthesis via precipitation were carried out using a QUANTAURUS-QY spectrometer (HAMAMATSU, Japan) equipped with a full integrating sphere (diameter approximately 8.4 cm) and a 150 W xenon excitation light source. In order to prevent fluorescence of the sample container, the sample was measured inside a closed silica-glass cell (outer diameter 17 mm) positioned on a Teflon-covered sample holder. An excitation wavelength of 462 nm was chosen with a maximum spectral full width at half-maximum (fwhm) of 10 nm. A spectrum was measured in the wavelength range between 200 and 960 nm with 0.77 nm step size. The emission data are depicted in Figure S2. Temperature-dependent luminescence data and the excitation spectrum were recorded using a Fluoromax 4 spectrophotometer (HORIBA, Japan). The emission spectra were measured in the wavelength range between 550 and 700 nm (step size 1 nm) using an excitation wavelength of 460 nm. The excitation spectrum was measured in the wavelength range between 250 and 600 nm, monitored at 629 nm. 2.4. Energy-Dispersive X-ray Spectroscopy. The chemical composition was analyzed by energy-dispersive X-ray spectroscopy (EDX) using a SUPRA35 scanning electron microscope (SEM, CARL ZEISS, Oberkochen, Germany, field emission) equipped with a Si/Li EDX detector (OXFORD INSTRUMENTS, Abingdon, Great Britain, model 7426).
3. RESULTS AND DISCUSSION The powder X-ray pattern (Figure 1) of both doping experiments (Li2SiF6, with Cs2MnF6 or K2MnF6 as dopant) shows that Li2SiF6 is obtained as the main component. By doping with Cs2MnF6, a small amount of Cs2SiF6 was detectable as a side product, and by doping with K2MnF6, small amounts of LiKSiF6 and LiF were obtained as byproducts. During these doping experiments, Mn4+ cations C
DOI: 10.1021/acs.inorgchem.8b03433 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry are likely to replace the octahedral coordinated Si4+ cations, due to similar ionic radii of both ions (Si4+ = 54 pm, Mn4+ = 67 pm).32 In addition, less distortion defects are expected because of the same valence of the ions. The emission spectra of the high-pressure/high-temperature samples exhibited identical emission profiles, as can be seen in Figure 2. Since there is no noticeable difference in the emission spectra, it is reasonable to assume that the observed emissions can be attributed solely to Li2SiF6:Mn4+ as it is the only component that is observed as a product in both of the doping experiments. EDX measurements confirmed the silicon to fluorine content to 1:6; therefore, within the margin of error, the nominal composition of Li2SiF6 could be verified. The manganese content is beneath the detection limit of the method. Additionally, it was possible to yield single crystals of Li2SiF6 by this synthesis method (Table S1). To rule out that the emission spectrum stems from the dopant material, which might undergo a phase-change during the high-pressure/hightemperature experiment, another experiment was conducted, where only the dopant Cs2MnF6 was compressed. The emission spectra before and after the high-pressure/hightemperature experiment as well as a comparison to the emission spectrum of Li2SiF6:Mn4+ are shown in Figure S4. The emission spectrum of Cs2MnF6 after the high-pressure/ high-temperature experiment is very slightly blue-shifted in comparison to the spectrum taken before the experiment. Both emission spectra miss the zero-phonon line, which is well visible for Li2SiF6:Mn4+. Therefore, the emission spectrum of Li2SiF6:Mn4+ cannot be the result of a phase-change of the dopant material. By excitation with λexc = 448 nm, Li2SiF6:Mn4+ exhibits seven emission lines at room temperature, whereas the emission line of maximum intensity is recorded at λmax ≈ 630 nm. Similar to K2SiF6, the bulk sample still shows luminescence after several weeks of storage under air. This indicates a comparable level of chemical stability. In comparison to the emission spectrum of K2SiF6:Mn4+ at room temperature, Li2SiF6:Mn4+ exhibits an additional emission line and the emission is slightly blue-shifted. The splitting of the emission into seven instead of six emission lines is likely to be attributed to the different crystal structures of Li2SiF6 and K2SiF6. Li2SiF6 crystallizes in the Na2SiF6 structure type in the trigonal space group P321 (no. 150) with a = 8.219(2) Å and c = 4.5580(9) Å.27 In comparison, K2SiF6 crystallizes in the K2PtCl6 structure type in the cubic space group Fm3̅m (no. 225) with a = 8.134(1) Å.33 A comparison of both crystal structures is depicted in Figure 3. It is noticeable that there is only one silicon position in K2SiF6, whereas there are two different silicon positions in Li2SiF6. In K2SiF6, the fluorine atoms form, in terms of crystallography, a
perfectly symmetric octahedron around the Si atoms (4a, m3̅m). In comparison, the two silicon atom positions in Li2SiF6 (1a, 32., and 2d, 3..) show distorted octahedral coordination spheres due to lower site symmetry (Table S2). Therefore, the emission peaks of the spin-forbidden 2Eg → 4 A2g transition at ≈596, ≈608, ≈611, ≈620, ≈630, ≈632, and ≈646 nm can likely be assigned to the transitions of the vibronic modes ν3(t1u), ν4(t1u), ν6(t2u), ZPL, ν6(t2u), ν4(t1u), and ν3(t1u), respectively. Due to the lower symmetry of the surrounding of Mn4+ in Li2SiF6:Mn4+, the zero-phonon line at λZPL ≈ 620 nm, which is electronic dipole forbidden, can be observed. This is not the case in K2SiF6:Mn4+, where Si4+/ Mn4+ exhibits a higher symmetric surrounding and therefore, the ZPL shows only little to no intensity.34 The comparison of the emission spectra of Li2SiF6:Mn4+, K2SiF6:Mn4+, and Cs2MnF6 reveals a slight blue-shift of the line positions for the lighter alkali metal ions in comparison to the heavier ones. However, even a small blueshift changes the color of emission, which is described by the CIE coordinates (Table 1). Table 1. Dominant and Maximum Emission Wavelengths, x and y Coordinates in the CIE Diagram, and LER Values for Li2SiF6:Mn4+, K2SiF6:Mn4+, and Cs2MnF6a
Li2SiF6:Mn4+ K2SiF6:Mn4+ Cs2MnF6
λdom/ nm
λmax/ nm
xCIE
yCIE
LER /lm Wopt−1
618 621 622
630 631 634
0.688(1) 0.693(1) 0.696(1)
0.312(1) 0.307(1) 0.304(1)
218 204 187
a
Coordinates rounded to significant numerical values regarding the measurement accuracy.
Especially in the red spectral region, where the human eye loses sensitivity rapidly (see Figure S6), this small shift in the emission leads to a strongly increased LER (luminous efficacy of radiation) value. The LER of Li2SiF6:Mn4+ is about 218 lm Wopt−1. The CIE coordinates for Li2SiF6:Mn4+ are x = 0.688 and y = 0.312. The quantum efficiency of the first lab sample is quite low (
