Abstract
YAG:Ce, a widely employed phosphor for LED wavelength conversion applications, comes in different forms, including polycrystal (powder), single crystal, and composite single crystal varieties. We investigated Epoch NeoTM, a composite single crystal incorporating Al2O3 as a heat distribution and optical guiding material embedded within its luminescent core. The properties of Epoch Neo were analyzed and compared with those of its polycrystal counterparts in this paper. Multiple characterization techniques like chromaticity points, color temperature, phosphor temperature, and speckle pattern were employed to study the effect of variable laser input power on both polycrystal powder and composite single crystal of YAG:Ce. It was found that under the influence of pumping blue laser high input optical power the composite single crystal phosphor's performance was significantly better compared to that of the polycrystal.
Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
1. Introduction
For solid-state lighting applications, cerium-doped yttrium aluminum garnet (YAG:Ce) stands out as a prominent phosphor for generating white light. This is due to its favorable characteristics, such as wide availability, low cost, broad emission spectrum, appropriate absorption and emission wavelengths, rapid luminescence decay, high thermal conductivity, and minimal thermal interference with luminescence [1, 2]. Furthermore, the presence of Ce3+ ions in YAG results in an exceptionally short luminescence lifetime. This short lifetime results in a high absorption saturation threshold, especially when exposed to incident blue laser light [3].
Despite these advantages, one significant drawback associated with using YAG:Ce is its susceptibility to thermal quenching, where its luminescence efficiency diminishes as the temperature rises. In typical polycrystalline YAG:Ce powders, this thermal quenching effect leads to a noticeable intensity reduction. Substantial efforts have been devoted to mitigating this thermal luminescence quenching phenomenon. A significant reduction in the intensity drop was achieved by using an Al2O3/YAG:Ce composite ceramic [4, 5]. The advantage of having minimal thermal quenching becomes particularly apparent under high power pumping. Besides, another area of improvement is in enhancing the luminescence saturation threshold (LST) of YAG:Ce based phosphors. The findings from various published studies indicate that both composite ceramic containing Al2O3 and YAG:Ce and single-crystalline YAG:Ce exhibit superior LST compared to other types of YAG:Ce phosphors, such as YAG:Ce phosphor embedded in glass (phosphor-in-glass) or phosphor layers [6, 7]. In the case of single crystal YAG:Ce phosphors, it is seen that they possess an exceptionally high LST, with no indication of a decline in luminescence intensity or luminous flux even when the laser pump power density reaches 360 Wmm−2 [8–10].
The primary objective of this paper is to analyze and compare the luminescence behavior of polycrystal YAG:Ce and a single crystal composite of Al2O3 and YAG:Ce for laser based application. The spectrum, chromaticity points, color temperature, phosphor temperature, and speckle were studied to evaluate the performance of the phosphors for high input blue laser power excitation.
2. Phosphors for use with laser diode pumping
2.1. Polycrystal YAG:Ce phosphors
For a few years now, it has been understood that traditional powder phosphors are not suitable for pumping with intense laser beams. Very high local power density generated by laser beams on phosphor surfaces can result in very high local temperatures that can result in both heat-induced luminescence quenching and phosphor material degradation [4, 11, 12]. Thus, a high optical damage threshold is an essential attribute of laser-pumpable phosphors. High structural integrity and thermal conductivity are needed for this. Traditional powder phosphors cannot effectively serve in this role because of the use of organic binders to hold phosphor particles together. This results in low structural strength and very low thermal conductivity. Hence, such phosphor compositions have very low damage thresholds and are not suitable for excitation by even moderate power laser beams. Glass has been used as a binder to circumvent this issue to a certain extent. Glass-bound phosphors are called phosphor-in-glass (PiG) compositions. Glass confers highly increased temperature resistance, compared to organic binders. PiG materials have been used in both high-power LED- and laser diode (LD) based luminaires [3, 13–18]. However, at high LD power levels, glass-based phosphors also prove inadequate, and an even better alternative is needed.
2.2. Single crystal YAG:Ce phosphors
Single crystal (monocrystalline) phosphors have proven to be the best wavelength conversion material where high temperature capability and damage resistance are needed [19]. These are cut from Czochralski-grown boules of Ce-doped YAG garnet. Their spectral characteristics are similar to that of powdered YAG:Ce phosphors that have long been used with blue-emitting GaN/InGaN LEDs. Nanocrystalline varieties of this phosphor are also under development [20]. Being single crystal in form, these plate-like single crystal YAG:Ce phosphors are exceptionally well-suited for pumping with blue-emitting GaN/InGaN LDs. Their structural rigidity, hardness, and thermal conductivity are all much superior to that of YAG:Ce PiG phosphors. Several research groups have reported on the use of single crystal YAG:Ce phosphors with LD pumping [10, 19, 21–23]. But an even better alternative is now available as given below.
2.3. Composite YAG:Ce/Alumina single crystal phosphor
Typical single crystal phosphors designed for LD-pumped light sources are produced through the incorporation of additional material like alumina (Al2O3) into the YAG melt during Czochralski crystal growth. The resulting YAG:Ce/Al2O3 composite phosphors possess exceptional properties, rendering them highly attractive for LD-pumped illumination applications [24, 25]. In their pioneering work, Song et al detailed the design of laser-driven phosphor-converted luminaires tailored for automotive applications, leveraging composite YAG:Ce/Al2O3 single crystal phosphors [26–28].
Referred to as a ceramic phosphor plate, this single crystal phosphor presents two distinct advantages over simpler single crystal YAG:Ce phosphors. Firstly, its enhanced temperature endurance stems from the high thermal conductivity (∼20 W m−1 K−1) of alumina inclusions, approximately twice that of typical powdered YAG:Ce phosphor [26]. This augmentation raises both the luminance quenching temperature and the optical damage threshold, enabling its use with significantly higher incident laser beam intensities. The second advantage arises from the random alumina phase inclusion within the YAG matrix. Through the proper admixture of alumina, random-continuous and quasi-continuous alumina regions form within the YAG:Ce host. These regions act as conduits, facilitating the passage of light and creating a waveguide-like effect. This property allows for efficient extraction of light from the phosphor, thereby enhancing the external quantum efficiency of the composite single crystal phosphor. These combined advantages bestow significant operational benefits upon LD-based lighting systems, establishing this composite phosphor as the preferred conversion medium for high fluence wavelength up-conversion in LD-pumped luminaires.
Hu et al have conducted studies on YAG:Ce composite phosphors as well [29]. Their findings indicate that the inclusion of an optimized number of α-Al2O3 particles, of selected sizes, allows for the simultaneous optimization of light propagation, luminous efficiency, and thermal stability of luminescence in high-power phosphor-converted lighting systems energized by LDs. Cozzan et al have also explored composite YAG:Ce/Al2O3 single crystal phosphors, employing a spark plasma sintering technique [30]. Their investigations reveal similar performance metrics for this category of phosphors, aligning with findings from other researchers. Additionally, Xu et al have reported on this technique for producing composite ceramic phosphors, highlighting its potential as a more straightforward method compared to Czochralski growth [31].
In a more recent development, Song Hu et al have demonstrated that composite crystal phosphors can be powdered, mixed with photocurable resin, and 3D printed into desired shapes, offering a novel approach to crafting specifically structured wavelength up-conversion media [32]. A review article published in 2022 provides a more extensive overview on various composite phosphors used for solid state lighting [33].
3. Experimental details
A 450 nm laser diode module was used as the excitation source in our experiment. The laser was driven by an external constant current driver operated with a 12 V DC source and a pulse-width modulation (PWM) control interface. The laser spot was rectangular in shape and power-modulated using PWM by varying the duty cycle through a function generator. A concave lens, having a focal length of 120 mm, was placed in front of the LD module and it served to expand the beam coming from the module. The laser power was measured using an 818 T-30 thermopile detector. Initially, the power meter was positioned at the sample location indicated in figure 1. Subsequently, the power meter was substituted with the sample. This allowed measurement of the actual laser power density incident on the sample. An Ocean Optics spectrometer was used to capture the irradiance spectrum of the sample, as shown in figure 1. Irradiance is the measure of the amount of energy that is emitted from the sample surface at each wavelength. Some investigations on optical characterization of phosphors use more elaborate experimental set up using an integrating sphere [34, 35], however we opted to use a simpler arrangement aided with a cosine corrector. A cosine corrector CC3-UV-S (inset of figure 1) was used for measuring the irradiance, CIE chromaticity point and correlated color temperature (CCT) of the light emitted by the sample. The cosine corrector provides a diffused uniform light field to ensure accurate and consistent color and CCT measurements that correlate well with visual observations. The cosine corrector was calibrated using a tungsten halogen light source HL-2000 and the collection area used for the cosine corrector was 0.4 cm2 [36–38].
Figure 1. Schematic of the assembly used for measuring YAG:Ce sample parameters.
Download figure:
Standard image High-resolution imageTwo samples were used for measurements. The EPOCH Neo composite single crystal phosphor mounted on an integral heat sink from Oxide Corporation is an Al2O3/YAG:Ce phosphor composite with a 5 × 5 mm active optical area. This has a thickness of 0.2 mm and it is mounted on a copper-tungsten alloy heat sink as depicted in the inset of figure 2(c) [26].
Figure 2. (a) Absorption spectra of YAG:Ce phosphor. Reprinted from [39], with the permission of AIP Publishing. (b) Spectrum of YAG:Ce polycrystal sample along with the image of the sample (c) Spectrum of YAG:Ce/Al2O3 composite single crystal sample along with the image of the sample.
Download figure:
Standard image High-resolution imageThe other sample studied here, shown in figure 2(b) is a YAG:Ce phosphor powder from Sigma Aldrich Corporation, deposited on a 5 × 5 mm area, with a thickness of 0.2 mm on a copper base. A widely used binding agent, KASIL 1624 (Potassium Silicate) sourced from the PQ corporation was used.
4. Photoluminescence mechanism of YAG:Ce phosphor
The Ce3+ ion possesses only one electron in the 4f state. Upon examination of the ground state of Ce3+ ion we observe a bifurcation into two energy states, namely 2F7/2 and 2F5/2. The succeeding energy level originates from the 5d state, and it is worth noting that 4f–5d transitions are permitted in terms of parity. The 5d state is subject to crystal field splitting, leading to the emergence of multiple Ce3+ ion absorption bands within the excitation spectrum as depicted in figure 2(a) within the 300–500 nm range. Specifically, two wide bands are discernible, with their peaks at 340, and 460 nm [39]. Of particular significance is the intense excitation band spanning from 400 to 550 nm. This band forms the foundation for using the phosphor with the blue laser. We opted for the 450 nm wavelength for excitation, which conveniently aligns with the widespread availability of high-power LEDs and LDs emitting at or close to this wavelength. In the emission spectrum as illustrated in figures 2(b) and (c), the broadband spanning from 500 nm to 700 nm assumes the role of a typical yellow light source, complementing the blue light emitted by the laser.
The irradiance spectrum and chromaticity points for both samples as a function of the excitation laser was operating at different power are shown in figures 2(b) and (c). The spectral shape remained the same at different phosphor excitation levels and the intensity increased with an increase in laser power level for both samples. Varying the laser power had minimal effect on changing the chromaticity point of the YAG:Ce single crystal as seen in the inset of figure 2(c) whereas the inset of figure 2(b) shows that the polycrystal phosphor was sensitive to the variation in input pump power to the degree of ∼1% of Δx coordinates and ∼3% of Δy coordinates. The emission spectrum shows a wide wavelength band, with its peak luminous intensity situated at 575 nm. The absorption of incident photons occurs around 450 nm, corresponding to the transition between the 4 f(2F5/2) to 5d1(2D3/2) energy levels. Due to the Stokes shift, the resulting wavelengths correspond to energy levels 5d1 to 2F5/2 and 5d1 to 2F7/2, and the two emitted wavelengths superimpose to form the observed broad emission peak [40–42]. In the case of a composite single crystal, the yellow light generated by YAG:Ce phosphor excited by blue laser light undergoes scattering at the boundaries between YAG:Ce and Al2O3 which is efficiently extracted from the surface. Thus, the effect of light trapping due to total internal reflection within the YAG:Ce is greatly reduced. Meanwhile, the heat produced due to Stokes shift or non-radiative transition is dissipated through Al2O3 [26].
5. Comparative study of single crystal composite and polycrystal YAG:Ce phosphors
5.1. CCT and CIE color coordinates
To assess the performance of composite single crystal and polycrystal phosphors, emission spectra, chromaticity points, and CCT were analyzed. A summary of measured CCT and CIE chromaticity coordinates (x, y) for both samples is presented in table 1, showing a significant deviation of the x to y coordinates as mentioned earlier in section 4.
Table 1. CCT and CIE color coordinates for composite single crystal and polycrystal YAG:Ce phosphors.
| Power density (Wmm−2) | CCT (K) | x | y | CCT (K) | x | y |
|---|---|---|---|---|---|---|
| Polycrystal | Composite single crystal | |||||
| 0.25 | 4103 | 0.3834 (±5%) | 0.4035 (±5%) | 4731 | 0.3746 (±5%) | 0.3875 (±5%) |
| 0.50 | 4129 | 0.3814 (±5%) | 0.3996 (±5%) | 4733 | 0.3745 (±5%) | 0.3873 (±5%) |
| 0.75 | 4150 | 0.3797 (±5%) | 0.3959 (±5%) | 4738 | 0.3744 (±5%) | 0.3874 (±5%) |
| 1.00 | 4172 | 0.3781 (±5%) | 0.3927 (±5%) | 4742 | 0.3740 (±5%) | 0.3869 (±5%) |
| 1.25 | 4185 | 0.3770 (±5%) | 0.3904 (±5%) | 4743 | 0.3743 (±5%) | 0.3877 (±5%) |
The illustration in figure 3(a) shows the normalized irradiance for both samples at 50% duty cycle which corresponds to an incident pump power density of 0.75 Wmm−2. The shapes for both spectra are identical, with slight change in intensity values. The location of both samples in the color diagram with variable power intensity is provided in the inset of figure 3(a). The chromaticity coordinates x and y for both phosphors are affected differently by the excitation power density. There is a linear increase in the x and y chromaticity coordinates in the case of the polycrystal phosphor, whereas in the case of the composite single crystal, the coordinates of the chromaticity points are rather constant. Since both x and y coordinates are increasing, it implies that the chromaticity points are shifting color temperature which is evident in the CCT graph presented in figure 3(b). The graphs in figure 3(b) clearly show that the CCT parameter is more affected by the excitation power density in the case of the polycrystal phosphor, as compared to the single crystal phosphor. The composite single crystal phosphor sample showed a CCT of around 4700 K whereas the polycrystalline phosphor sample's CCT was approximately 4100 K—and both are suitable for general lighting applications. The CCT variation with power is consistent with the shift in x, y coordinates of the chromaticity points. This implies that the yellow-to-blue ratio of the emitted light has a role in determining the CCT. The tuning of color temperature with the ratio of area between the yellow region to the blue region (Y/B) at different power densities is shown in figure 3(b). The composite single crystal does not show much variation in terms of CCT, or Y/B, as compared to the polycrystal phosphor. An increase in yellow intensity results in shifting of the emission toward warmer colors, generating lower CCT values compared to when the Y/B ratio is lower. When blue laser light is incident on yellow-emitting phosphor, the luminescent yellow centers convert the blue light to yellow light. As more and more blue photons get converted to yellow, luminescent centers get saturated and a bleed of blue photons occurs, and thus the amount of blue component increases as input pump power increases. From figure 3(b) it can be inferred that the composite single crystal sample is more color-robust compared to the polycrystalline sample, as it has more stable CCT at different operating powers. A fluctuation in CCT as observed in the polycrystalline sample (see figure 3(b)) is caused by increasing temperature as input power increases. The measurement points on figure 3 correspond to power densities of 0.25, 0.50, 0.75, 1.00, and 1.25 Wmm−2, respectively.
Figure 3. (a) Spectrum comparison of YAG:Ce composite single crystal and YAG:Ce polycrystal phosphor samples (b) Ratio of the area under the curve of blue (B) and yellow (Y) parts of the spectrum (Y/B) and corresponding CCT of both samples with variable power density.
Download figure:
Standard image High-resolution image5.2. Variation of phosphor temperature with pump power level
Earlier in the paper, thermal quenching in YAG:Ce phosphor was addressed as a well-known and significant drawback. Non-radiative heat transfer leads to poorer luminescence efficiency resulting in lower emission intensity and a redshift (e.g. warmer white light). Besides this, when regions in phosphor are excited by a laser beam, local thermal degradation is commonly observed in most YAG:Ce phosphors. This is usually due to the heat dissipation of the thermal energy generated from the Stokes shift and quantum efficiency losses. To study and compare how the temperature of both the composite single crystal and the polycrystalline YAG:Ce changed with excitation time (i.e. time of exposure to the blue laser beam), we recorded the temperature change using surface mount thermocouples at two distinct power levels. As can be seen in figure 4, at a low level of excitation power (0.25 Wmm−2), Δt curves for both samples appear similar. However, when the excitation laser beam intensity is increased to 2.5 Wmm−2 the thermal budget in both the phosphors is significantly higher and differences in heat dissipation of the two samples are more apparent. Since typical epoxy resins (binding agents) have low thermal conductivity ranging ∼0.2 Wm−1 K−1 [43], KASIL used as the binding agent for the polycrystalline sample acts as a thermal insulator and efficient heat dissipation is hindered.
Figure 4. Averaged changes in temperature of the two types of YAG:Ce phosphors with respect to laser excitation time for two different power levels of 0.25 Wmm−2 and 2.5 Wmm−2. The insert shows alumina regions spread within a YAG:Ce matrix of a composite phosphor crystal [33]. Reproduced from [33].
Download figure:
Standard image High-resolution imageThus, dark spots were observed at the site of excitation on the powdered YAG:Ce phosphor from Sigma Aldrich, indicating thermal degradation of the KASIL/YAG:Ce composite when exposed to 4 W power for 50 s. Contrary to this, the damage threshold reported from Oxide Corp. for their composite single crystal phosphor is much higher (∼60 W) [26]. The single crystal phosphor does not limit its thermal conductivity with any resins, furthermore, the inclusion of Al2O3 waveguide-like features (visible from the SEM micrograph, inset in figure 4) in the composite material further accentuates its thermal conductive properties. Furthermore, a specially engineered heat sink base for the single crystal phosphor ensures efficient spread and extraction of the generated heat without straining the phosphor. As seen in the figure 4, heat dispersion and thermal budget management in the composite single crystal phosphor is better than in the polycrystalline phosphor.
5.3. Speckle pattern
When working with laser-based luminaires, a critical consideration is the potential presence of speckles, as noted in [44]. This phenomenon arises from the interaction of coherent laser radiation with uneven surfaces. Various techniques have been developed to mitigate speckle effects stemming from laser sources [45–47]. As pump laser light traverses among phosphor particles, its coherence diminishes, leading to minimal speckle in any radiation that passes out through the phosphor layer. Figure 5(a) depicts the speckle pattern of the pump laser light at the position of the phosphor plate. This pattern was captured using a CCD camera-based laser beam Ophir's BeamGage profilometer. An attenuator was placed in front of the detector to prevent any damage. The intensity is color-coded in correspondence with the scale displayed to the right of the image. Lower intensities are represented at the scale's bottom, while higher intensities are shown at the top. As figures 5(b) and (c) illustrate, the down-converted light and the residual pump light constituting white light do not exhibit pronounced speckle patterns.
Figure 5. Speckle pattern of (a) blue laser light source (b) emission from YAG:Ce composite single crystal (c) emission from YAG:Ce polycrystalline phosphor at various power densities.
Download figure:
Standard image High-resolution image
Speckle patterns in figure 5 offer a direct comparison between the input (i.e. excitation laser light) and output light (i.e. white light), both traveling to and from the phosphor plate. Clearly, the speckle observed in the initial excitation light has been effectively eliminated by either of the phosphor plates. Traditionally, non-composite single crystal YAG:Ce phosphors have employed external surface roughening to scramble phase and reduce laser speckle. One notable advantage of using the composite single crystal phosphor is that it eliminates the necessity for additional external crystal preparations. This is because it naturally reduces speckling through the randomization of trajectories, facilitated by the presence of random alumina inclusions within the matrix [48].
6. Conclusions
In summary, the behavior of a new class of composite YAG:Ce/alumina phosphor with LD pumping was characterized through CIE chromaticity coordinates and temperature, and compared with that of polycrystalline YAG:Ce phosphor. We investigated the effect of incident photon flux on wavelength up-conversion and on the spectral characteristics of wavelength up-converted light. The spectral shapes for both phosphors were identical, but the effect of incident power density causing shifts in luminescence-related parameters was much more noticeable in the case of the polycrystal phosphor. Both the chromaticity points and the CCT plot showed that the composite single crystal was significantly more stable compared to the polycrystal powder for different pump input powers. We further investigated the effect of phosphor temperature; it was inferred that due to the lower thermal conductivity of the binding agent used in powdered phosphor, the polycrystal phosphor mixed with a biding agent is not suitable for high-input power laser excitation as compared with the single crystal phosphor. Another benefit of using the Epoch Neo composite single crystal phosphor is that there is almost no speckle present in the converted light. All of these findings attest to the superiority of composite single crystal phosphors in comparison with conventional powder phosphors contained in a binder matrix.
Acknowledgments
We acknowledge valuable assistance from Tsuneo Kusunoki of Oxide Corporation, Japan, in the performance of this work. We also acknowledge help from the Department of Electrical Engineering and Computer Science at Ohio University through the iSURF grant.
Data availability statement
All data that support the findings of this study are included within the article (and any supplementary files).