For a better understanding of the present invention, together with other and further objects, advantages and capabilities thereof, reference is made to the following disclosure and appended claims taken in conjunction with the above-described drawings.
The emission spectrum of an indium chloride (InCl) discharge is presented in
There is in principle more than one way of achieving white emission from such a lamp, the main differences being in the number of phosphor components used in the phosphor blend. The simplest of these is to convert part of the blue emission from the discharge into the yellow-orange spectral range by means of only one phosphor component, e.g., a blue-absorbing, yellow-emitting Y3Al5O12:Ce (YAG:Ce) phosphor. However, the result is a relatively low-grade white light. For a high-grade white light, more than one phosphor component would be needed for a good CRI and CCT, preferably including sufficient amounts of a red-emitting phosphor.
The next option would therefore be a two-phosphor blend of green- and red-emitting phosphors that utilizes blue light from the discharge for both excitation and as a color component. More preferably, the phosphor blend would have three or more phosphors, including a blue-emitting phosphor, in order to have a better control over emission parameters.
It is clear from
Phosphors suitable for excitation by the blue radiation emitted by an indium halide discharge include, but are not limited to, red-emitting Ca2Si5N8:Eu (Ca—SiN), green-emitting SrSi2N2O2:Eu (Sr—SiON) and blue-green-emitting Ca8Mg(SiO4)4Cl2:Eu (CAM-Si). In all of the above-mentioned phosphors, the emission is based on Eu2+ activation, exhibiting broad bands peaking at 620 nm, 547 nm and 513 nm, respectively (see
The excitation spectra of the CAM-Si, Ca—SiN, Sr—SiON and BAM phosphors are presented in
Using data from the emission measurements on the InCl discharge and the emission curves of each phosphor, an area-weighed combination of the violet/blue discharge lines and CAM-Si, Sr—SiON and Ca—SiN emissions was obtained. This result was further optimized for the highest CRI and appropriate CCT achievable (and maximum possible lumens thereby) by calculating these parameters and the corresponding color coordinates for a number of different, systematically varied combinations. The results are presented in Table 1 and the 88.1/4867K CRI/CCT spectrum (last line in Table 1) is shown in
With reference to Table 1, a strong blue contribution from the discharge emission seems to benefit both CCT and CRI output parameters. Also, it is important for good color rendering to have a noticeable fraction of the blend contain the red component.
A BAM phosphor (a plasma display panel (PDP) type) was used as a reference for QE at 250 nm and YAG:Ce (Type 251, OSRAM SYLVANIA Products Inc.) for 450 nm excitation. The strongest blue-absorbing phosphor is the Ca—SiN phosphor whose absorption extends far into the green range. None of the spectra has an abrupt, step-like onset of the absorption since the low-energy tail of these curves is a smoothly decaying function. This means that both 411 nm and 451 nm InCl discharge lines will pass through unless the lamp coating is optimized to stop the 411 nm radiation completely. The difference in transmission at these wavelengths may be crudely approximated by e−μ/β where the exponent is the value of the remission function at the wavelength of interest. This yields only a difference of about 2.3 to 4.2 times in the transmission of the blue 451 nm line vs. the violet 411 nm line. In other words, in order to make complete use of the discharge, the coating has to be optimized for zero transmission at 411 nm, which would also reduce the blue radiation below the level required for good color rendering. As the 411 nm and 451 nm lines have an approximate 40%-60% integrated total emission ratio in pure discharge measurements, reducing the former to about a 1% intensity level leaves only 1.5% worth of intensity in the latter. A small modification to this caused by the absorption of phosphor layer will be demonstrated below.
Adding the blue-emitting phosphor component (e.g. BAM) would be necessary for correcting this issue. It is clear, however, from
Maximum Expected LPWUV Values
Subsequently, it was attempted to estimate the lumen per watt (LPW) values for three phosphor components and two of the blend compositions of choice. Spectral distributions were normalized to 1 W of total power in the visible range (see Table 2). “Ideal” in this case means a blend of desired parameters (CRI, CCT) that, depending on the number of components (four or three), either does or does not contain BAM, respectively. The LPW451 and LPW411 for each column have been calculated from the corresponding emission spectrum assuming a certain quantum efficiency (QE) for generating the visible photons when excited by the 451 nm or 411 nm emission line of InCl.
The green-emitting Sr—SiON phosphor produces the highest visible lumens with 518.6 lm per each visible watt generated (2.82×1018 photons in total). With the assumed QE of 0.9, it takes about 10% more blue or violet photons to generate this green photon flux. For this, 1.38 W and 1.51 W of optical power at 451 nm and 411 nm, respectively, is required, yielding LPW451=376.3 and LPW411=343, respectively (i.e. all incident photons assumed to be concentrated at 451 nm or 411 nm wavelength). With the actual mix of excitation lines as 40-60%, the highest possible LPW value for this phosphor has been calculated as LPW40-60=360. A proper blending with two other phosphor components reduces the value to 226.3 LPW40-60. If the actual discharge plus blend emission spectrum is considered (
Other Factors Influencing Blending
One of the disadvantages of the InCl discharge is the high operating temperature required for InCl emission. The wall temperature of the bulb may reach 200° C. or more. However, an infrared reflecting jacket around the lamp, and separated from it, will probably not exceed 150° C. This is the preferred surface for phosphor coating. Phosphors that have been coated on this jacket will have to tolerate this high operating temperature without a significant decrease in conversion efficiency. It is known for most of the phosphors used in various applications that the quantum efficiency will decrease at elevated temperatures due to an increase in non-radiative decay probability. Furthermore, the phosphors have to maintain their chemical (e.g. composition) and physical (e.g. structure) properties while heated to such temperatures in order to prevent the deterioration of their output. The temperature dependence of CAM-Si, Sr—SiON and Ca—SiN was measured under steady-state conditions of 365 nm excitation (Hg—Xe lamp with interference filter). The corresponding weight correction factors due to increased nonradiative processes at elevated temperatures have been incorporated into the blend recipes. One skilled in the art can readily determine these correction factors by empirical measurements.
Experimental Coating of Slides
Physical testing of phosphor blends was first attempted with three components (CAM-Si, Sr—SiON and Ca—SiN) only. Small slides of about 0.8″×1.0″ (20×25 mm2) were cut from regular microscope slides made of quartz and Pyrex. Some of these slides were sandblasted that increased the surface area for the coating but also caused strong scattering of the transmitted light. A preferred method of coating the phosphor blends on a glass substrate uses a slurry of the blend and a polyisobutyl-methacrylate (PIBMA) binder (Elvacite 2045). A vehicle of 13 wt. % PIBMA and 87 wt. % xylene was prepared. Dibutylphthalate and a surfactant (Armeen CD) were added in equal amounts of 1.5 wt. %. A 43 gram amount of the vehicle was mixed with 0.7 grams of a high surface area aluminum oxide powder (Aluminum Oxide C) and rolled for 24 hours. Slurries of the phosphor blends were made by mixing about 4 grams of the phosphor blends with 4-6 ml of the vehicle. The slurry is applied to the glass surface, dried and the binder removed by baking in a nitrogen atmosphere at about 350° C.
The values from Table 1 for the three mixed emission components (10% CAM-Si, 16% Sr—SiON and 41% Ca—SiN) after re-normalization (excluding the discharge) yield 15, 24 and 61%, respectively (Table 4). The next step would be to correct these values for the product of each phosphor's quantum efficiency with discharge intensity, integrated over the spectrum. It is a useful exercise but unfortunately limited to an approximation only due to the fact that excitation intensity is spread over UV and blue spectral regions where the phosphor response is not uniform. It is evident from Table 4 for example that the “match” between CAM-Si and the InCl discharge is relatively less optimal than for other two phosphors. Although not indicated in Table 4, YAG:Ce has a useful overlap of its excitation spectrum and the discharge of only about 53% compared to Sr—SiON.
The second correction comes from the different temperature dependence of each component as demonstrated earlier. Among the three, Sr—SiON is the least affected by temperature quenching. The cumulative values are reflected in the rightmost column of Table 4 and will be used as a starting point for the physical blending of powders.
In addition to the blend shown in Table 4 (designated as blend #1), additional combinations of CAM-Si/Sr—SiON/Ca—SiN were used having the proportions 20/20/60 wt. % (blend #2) and 15/13/72 wt. % (blend #3).
The blends were coated onto slides and after drying (but before baking), the optical density of the slides was checked by using a 450 nm LED and a fiber optic probe (Ocean Optics USB2000). The amount of blue light passed through slides (in peak intensity) was found to be a function of coating density. The dependence of transmitted blue light on the coating thickness is demonstrated in
Testing the Slides
Optical characteristics were measured using InCl lamp excitation and a fiber optic probe. An outer hemispherical glass jacket contains a “shelf”, or circular ring which supports the phosphor test slides in close proximity to the InCl discharge (˜1 cm). The smaller spherical glass discharge bulb is concentric within this outer jacket, supported by thin glass tubes. In this way, the discharge can operate within an insulated, or jacketed, environment, and subject the slides to the UV/Blue discharge emission. An optical fiber protrudes into the jacket from outside via a hole in the glass, and thereby views the phosphor slide emission from the side opposite the discharge, as would be the case in an actual lamp environment. The slides exhibited a strongly varying ratio of transmitted blue discharge and phosphor emission intensities; the other parameters changed relatively less significantly, as evident from Table 6. An example spectrum recorded for the slide with 2.08 mg/cm2 coating weight is presented in
Optical parameters calculated from InCl excited spectra of phosphor-coated slides as functions of coating density are shown in Table 6. The intensity ratios are for peak values; I/I0 is measured for unbaked slides, with 450 nm LED excitation as an indication of the blue transmitted. Relative lumen values have been obtained by normalizing to the maximum measured value in Table 5 (assuming the same experimental conditions).
Some expected trends are evident from the above Tables 5 and 6, particularly for the thinner coating weights (Table 6). When coatings become thinner, the ratio of I451/Iphosphor increases as seen in Table 6 but not in Table 5. Respective integrated areas of 411 nm and 451 nm emissions for the coating densities of 2.50 and 15.8 mg/cm2 as examples are 23%-77% and 35%-65%, a modification expected from 40%-60% ratio measured for the pure discharge. With more blue light being included in the emission from slides, the CRI improves and the color temperature rises. Relative lumens calculated on the basis of emission spectra show an increase with decreasing coating thickness. For collecting the data that are presented in both tables, the same experimental conditions were used and therefore all the relative lumen values are normalized to the same number (corresponding to 10.2 mg/cm2 in Table 5). The trend in lumen values is most likely caused by re-absorption of visible light generated in the phosphor layer itself as mentioned above.
While there have been shown and described what are presently considered to be the preferred embodiments of the invention, it will be apparent to those skilled in the art that various changes and modifications can be made herein without departing from the scope of the invention as defined by the appended claims.
This application claims the benefit of U.S. Provisional Application No. 60/747,617, filed May 18, 2006.
Number | Date | Country | |
---|---|---|---|
60747617 | May 2006 | US |