Converter-cooling element assembly with metallic solder connection

Abstract

An assembly is provided that includes a ceramic converter for converting light having a first wavelength into light having a second wavelength, a metal-containing reflective coating, and a cooling element. The surface of the ceramic converter is at least partly coated with the metal-containing reflective coating. The coating dissipates the heat from the converter into the cooling element. The cooling element and the metal-containing reflective coating are connected to one another by a metallic solder connection.

Claims

1. A converter-cooling element assembly, comprising: a ceramic converter for at least partial conversion of light having a first wavelength into light having a second wavelength; a reflective coating comprising metal and glass, wherein the metal is selected from the group consisting of silver, gold, platinum, and alloys thereof, at least portions of a surface of the ceramic converter being directly coated with the reflective coating; and a heat sink is connected with the reflective coating via a metallic solder connection so that the reflective coating dissipates heat from the ceramic converter into the heat sink, wherein the metallic solder connection comprises tin-containing lead-free solder.

2. The assembly as claimed in claim 1, further comprising a thermal heat transfer coefficient of at least 25,000 W/m.sup.2K for at a converter thickness of 200 m.

3. The assembly as claimed in claim 2, wherein the thermal resistance is less than 1.5 K/W.

4. The assembly as claimed in claim 1, wherein the reflective coating comprises silver having a silver content of at least 90 wt %.

5. The assembly as claimed in claim 1, wherein the reflective coating has a layer thickness from 50 nm to 30 m.

6. The assembly as claimed in claim 1, wherein the reflective coating has a glass content from 0.05 to 10 wt %.

7. The assembly as claimed in claim 6, wherein the glass has a glass transition temperature in a range from 300 to 600 C.

8. The assembly as claimed in claim 6, wherein the glass has a refractive index n.sub.D20 in a range from 1.4 to 2.0.

9. The assembly as claimed in claim 6, wherein the glass is selected from the group consisting of PbO glass, Bi.sub.2O.sub.3 glass, ZnO glass, SO.sub.3 glass, and silicate-based glass.

10. The assembly as claimed in claim 6, wherein the glass is a silicate-based glass having a SiO.sub.2 content of more than 25 wt %.

11. The assembly as claimed in claim 1, wherein the heat sink exhibits a thermal conductivity of more than 300 W/mK.

12. The assembly as claimed in claim 1, wherein the heat sink is a heat absorber.

13. The assembly as claimed in claim 1, wherein the ceramic converter is configured as a transmission arrangement with at least a portion of the reflective coating being positioned on a lateral surface of the ceramic converter.

14. The assembly as claimed in claim 1, wherein the ceramic converter is configured as a remission arrangement with at least a surface of the ceramic converter that is positionable away from an excitation light source being coated with the reflective coating.

15. The assembly as claimed in claim 1, further comprising a quality of reflection FOM.sub.CIE-cx defined as: ${\mathrm{FOM}}_{\mathrm{CIE}-\mathrm{cx}}=\frac{c_x\left(\mathrm{measured}\mathrm{sample}\right)-c_x\left({\mathrm{Ref}}_{0R}\right)}{c_x\left({\mathrm{Ref}}_{\mathrm{HR}}\right)-c_x\left({\mathrm{Ref}}_{0R}\right)}.$ is at least 40%; wherein c.sub.x(measured sample) is a chromaticity coordinate of the ceramic converter provided with the reflective coating as determined in remission for a CIE 1931 standard color system; c.sub.x(Ref.sub.HR) is a chromaticity coordinate of the ceramic converter as determined while applied on an ALANOD mirror having a reflectance of 98%; and c.sub.x(Ref.sub.HR) is a chromaticity coordinate of the ceramic converter as determined while applied on a dark background or a light trap.

16. The assembly as claimed in claim 1, wherein the heat sink comprises a copper-containing core and a coating applied thereto.

17. The assembly as claimed in claim 16, wherein the coating comprises a nickel-containing coating and/or a gold-containing coating.

18. The assembly as claimed in claim 1, wherein the reflective coating is a sintered coating.

19. A method for producing a converter-cooling element assembly, comprising the steps of: providing a ceramic converter with at least one polished surface; providing paste comprising a metal powder and a glass powder in an organic pasting medium, wherein the metal powder comprises a metal selected from the group consisting of silver, gold, platinum, and alloys thereof; applying the paste onto at least a portion of the polished surface; drying the paste; firing the ceramic converter and having the pasted dried thereon at a firing temperature above 450 C. to form a metal-containing and glass-containing reflective coating; and soldering a cooling element to the reflective coating with a metallic solder comprising a tin-containing lead-free solder.

20. The method as claimed in claim 19, wherein the paste comprises 70 to 90 wt % silver powder.

21. The method as claimed in claim 19, wherein the glass powder has a D50 value in a range from 1 to 5 m.

22. The method as claimed in claim 19, wherein the glass powder comprises glass having a glass transition temperature in a range from 300 to 600 C.

23. The method as claimed in claim 19, wherein the step of applying the paste onto the polished surface comprises printing the paste onto at least the portion of the polished surface.

24. The method as claimed in claim 19, wherein the step of drying the paste comprises drying at a drying temperature from 150 to 400 C.

25. The method as claimed in claim 19, wherein the firing temperature is in a range from 700 C. to 1000 C.

26. The method as claimed in claim 19, wherein the firing step comprises sintering the metal-containing and glass-containing reflective coating.

27. The method as claimed in claim 19, wherein the metallic solder comprises silver.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention will now be described by way of exemplary embodiments and with reference to FIGS. 1 through 14, wherein:

(2) FIG. 1 is a schematic view of a glued converter-cooling element assembly in a remission arrangement;

(3) FIG. 2 is a schematic view of a first embodiment of the converter-cooling element assembly according to the invention in a remission arrangement;

(4) FIG. 3 is a schematic view of a second embodiment of the converter-cooling element assembly according to the invention in a transmission arrangement;

(5) FIG. 4 is a schematic view of a third embodiment of the converter-cooling element assembly according to the invention in a transmission arrangement;

(6) FIG. 5a shows a graphical comparison of temperature stability and power stability of a glued converter-cooling element assembly;

(7) FIG. 5b shows a graphical comparison of temperature stability and power stability of a converter-cooling element assembly according to the invention;

(8) FIG. 6a show a graphical comparison of chromaticity coordinates of a converter coated according to the invention with similar non-coated converters with and without ALANOD mirror;

(9) FIG. 6b show a graphical comparison of secondary luminous flux of a converter coated according to the invention with similar non-coated converters with and without ALANOD mirror;

(10) FIG. 7a is a cross-sectional SEM image of an embodiment of the invention having a first glass content of the metal-containing coating;

(11) FIG. 7b is a cross-sectional SEM image of another embodiment of the invention having a second glass content of the metal-containing coating;

(12) FIG. 8a is a cross-sectional SEM image of another embodiment of the invention having a third glass content of the metal-containing coating;

(13) FIG. 8b is a cross-sectional SEM image of another embodiment of the invention having a fourth glass content of the metal-containing coating;

(14) FIG. 8c is a cross-sectional SEM image of another embodiment of the invention having a fifth glass content of the metal-containing coating;

(15) FIG. 9 shows graphs of reflectance measurements of different embodiments of the metal-containing coatings on transparent glass ceramic substrates;

(16) FIG. 10 is a graph showing the refractive indices and glass transition temperatures of various glasses;

(17) FIG. 11 schematically illustrates the experimental setup for determining the heat transfer coefficient HTC;

(18) FIG. 12 is a schematic arrangement of the measuring resistors during determination of the heat transfer coefficient HTC; and

(19) FIGS. 13 and 14 illustrate the quality of different embodiments of converters with metallic coating as evaluated based on chromaticity coordinates.

DETAILED DESCRIPTION OF THE INVENTION

(20) FIG. 1 schematically shows a converter-cooling element assembly known from prior art, in a remission arrangement. In this case, the converter 2 is applied on a mirror 4 with the opposite side or side facing away from the primary light source 1, by means of a glue layer 3, and the mirror is connected to a cooling element 5. Mirror 4 ensures that the secondary light 6 produced within the converter 2 and non-absorbed portions of the primary light 1 are reflected.

(21) FIG. 2 schematically illustrates an embodiment of the converter-cooling element assembly, also referred to as converter-cooling element assembly below, according to the invention, in a remission arrangement. Here, the converter 2 is provided with a metal-containing coating 7 on the surface facing away from primary light source 1. Metal-containing coating 7 and cooling element 5 are connected to each other via a solder connection 8. Metal-containing coating 7 is reflective and therefore replaces the mirror 4 shown in FIG. 1.

(22) FIGS. 3 and 4 show the structure of two embodiments of the converter-cooling element assembly according to the invention in transmission arrangements. In this case, the metal-containing coating 7 on the converter surface is configured laterally, in particular the metal-containing coating 7 is not applied in the beam path.

(23) In the embodiment shown in FIG. 3, the metal-containing coating 7 applied on converter 2 has portions, which are not connected to the cooling element 5 via a solder connection 8, for example on the side of the converter, which faces the primary light source. In this case, the metal-containing coating 7 prevents secondary radiation 6 from being emitted from the lateral surfaces of the converter 2 by reflecting the radiation. In addition, the lateral configuration of the metal-containing coating 7 as shown in FIG. 3 achieves high and uniform heat dissipation.

(24) FIG. 4 schematically illustrates an embodiment in a transmission arrangement, in which the converter 2 is cone-shaped. The lateral surfaces of the cone are provided with the metal-containing coating, which in turn is connected to the cooling element via solder connection 8.

(25) FIGS. 5a and 5b illustrate the shift of the chromaticity coordinate of the converter as a function of irradiated laser power for different temperatures. FIG. 5a shows the temperature/power characteristic of a conventional converter-cooling element assembly as illustrated in FIG. 1. FIG. 5b shows the temperature/power characteristic of an embodiment of the converter-cooling element assembly according to the invention as shown in FIG. 2. In each case, the employed converters have the same composition and thickness.

(26) The shift of the chromaticity coordinate as a function of the laser power irradiated onto a small excitation spot can be used to evaluate thermal conductivity. In case of low thermal conductivity, the converter will heat up already at relatively low irradiated laser power to such an extent that conversion efficiency decreases and that the measured chromaticity coordinate decreases. In case of improved thermal conductivity, the chromaticity coordinate will remain at a high level, even for higher laser powers.

(27) This difference can be seen in FIGS. 5a and 5b illustrating a comparison of a glued converter (on a mirror which in turn is connected to the cooling element by means of thermal grease) with the soldered embodiment according to the invention: when comparing the behavior at the same temperature of the cooling element (85 C. or 120 C.), the drop of the color coordinate (and hence of conversion efficiency) of the embodiment according to the invention only occurs at significantly higher laser powers than with the glued variant. At a temperature of the cooling element of 85 C., a shift of the color coordinate Dcy of more than 0.02 occurs at a laser current of about 1000 mA in case of the glued variant, while with the inventive solution the shift of color coordinate Dcy is still less than 0.02 even with the maximum feasible laser current of 1400 mA. At a temperature of the cooling element of 120 C., a shift of the color coordinate Dcy of more than 0.02 is already resulting at a laser current of approximately 820 mA for the glued variant, while for the embodiment according to the invention this is only the case at 1200 mA.

(28) Table 1 shows an approximation of the thermal resistance of a conventional converter-cooling element assembly according to FIG. 1 and of an embodiment of the converter-cooling element assembly according to the invention according to FIG. 2, with homogeneous heat input.

(29) TABLE-US-00001 TABLE 1 Approximation of thermal resistance for a prior art converter assembly and for the solution according to the invention Thermal Cross- resistance Thermal sectional R_th = conductivity area Thickness I/( * A) A I Component [K/W] [W/mK] [mm.sup.2] [m] Prior Art Converter 3.33333 6 10 200 (d = 200 mm) Glue connection 3.33333 0.3 10 10 (d = 10 m) Heat sink 0.00033 300 10 1 (d = 1 mm) Entire assembly 6.7 Metallic connection Converter 3.33333 6 10 200 (d = 200 mm) Metallization 0.02500 40 10 10 (d = 10 m) Solder connection 0.01667 60 10 10 (d = 100 m) Heat sink 0.00033 300 10 1 (d = 1 mm) Entire assembly 3.4

(30) FIG. 6 shows a comparison of chromaticity coordinates (FIG. 6a) and secondary luminous flux (FIG. 6b) of converter-cooling element assemblies with and without reflector and with metal-containing coating, respectively.

(31) Evaluation of reflection properties of a paste reflector at the internal ceramic/reflector interface is not trivial, since the ceramic is a translucent medium having a high refractive index and a slightly porous surface. Thus, it cannot be assumed that the evaluation on a transparent substrate of a similar refractive index (e.g. sapphire or CLEARTRANS glass ceramic) is representative.

(32) FIG. 6a shows an evaluation on the basis of the shift of chromaticity coordinates caused by the reflector: when the converter is irradiated with blue light (e.g. wavelength of 450 nm), this light will be absorbed completely or partially and will be converted into yellow secondary light, for example. This light is emitted isotropically within the converter. Since the converter does not or hardly absorb the secondary radiation, a significant part thereof reaches the back side of the converter. If a reflector is provided there, this light will be directed back towards the emission direction and will, possibly after multiple scattering and reflection events, contribute to the useful luminous flux which has a certain color location. If this back side reflector has a reflectance of less than 100% or is not provided at all, the proportion of the yellow light in the useful light is reduced and the color location will shift towards the blue light. Therefore, the shift of color coordinates in the color space chromaticity diagram (e.g. CIE1931/2 observer) is a quality measure of the reflector. For reliable evaluation, the converter should have a sufficient thickness and should be sufficiently doped with fluorescent active sites so that all blue light is absorbed across the converter thickness. But even if the reflector still reflects components of the blue light, the shift of color coordinates is a suitable measure. The advantage of evaluation on the basis of color location shift is that the color location measured at low power is a measure that is independent of the excitation power and is easily accessible in terms of measurement technology. However, it is only suitable for converters that remit a sufficient proportion of the excitation light.

(33) FIG. 6a shows the chromaticity coordinates of a converter of 200 m thickness irradiated with blue laser light of 450 nm wavelength while lying on a very highly reflective metal mirror (ALANOD Miro Silver) on the one hand, and on a black pad on the other, measured in remission in each case. These chromaticity coordinates represent the reference chromaticity coordinates in the sense of FOM.sub.CIE-cx for the case of a highly reflective mirror and for the case that a mirror is not provided. The chromaticity coordinates measured on an identical converter that has a metallic coating are located between the two reference values. FOM.sub.CIE-cx as calculated from the chromaticity coordinate data is 66%.

(34) Thus, the metal-containing coating exhibits significantly increased reflectivity when compared to the black background, but it is not as highly reflective as the ALANOD reference mirror. However, the converter-cooling element assembly has a substantially better thermal connection.

(35) FIG. 6b shows an alternative possibility for evaluation of the metallic coating based on the luminous flux of the secondary light. In the example, the measurement was carried out with a luminance camera in which, by means of camera optics, a spectrophotometer captures the luminous flux emitted from a measured spot on the converter surface in a certain spatial angle. From the captured spectrum, the secondary light component and the excitation light are then separated by calculation, so that the input variables for calculating FOM.sub.secondary luminous flux can be determined. In the present example, FOM.sub.secondary luminous flux is 59%. When measuring the input variables, particular attention has to be paid to have the same excitation power. Furthermore, measurement setups are conceivable in which the separation of secondary light component and excitation light is accomplished by means of filters, or in which the luminous flux or part of the luminous flux is captured in other measurement configurations.

(36) FIG. 7 shows cross-sectional SEM images (FIB sections) of ceramic converters 9 with sintered silver-containing coatings 10. The thickness of the coating 10 is 9 m (FIG. 7a) and 11 m (FIG. 7b), respectively. The coatings of FIGS. 7a and 7b differ in the glass content in the coating. Coating 10 as shown in FIG. 7a does not include glass, while the proportion of glass in the paste of FIG. 7b is 0.5 wt %. From the orientation contrast, the original grain structure of the paste can be recognized. The coating has a sintered structure, in which the metal particles that existed before firing were sintered together to a large extent so that the coating exhibits a relatively high degree of homogeneity. The cavities 12 or so-called voids, predominantly located at the interface 11 between silver-containing coating and ceramic converter are process-related.

(37) In FIG. 7a, a SiO.sub.2-tungsten layer 16 can be seen, which is applied prior to the cross section preparation by FIB in order to improve the quality of the cross-sectional image, but which is not part of the metallic reflector.

(38) FIGS. 8a to 8c also show cross-sectional SEM images (FIB sections) of a ceramic converter 9 having a silver-containing coating 10, wherein the coatings in FIGS. 8a to 8c differ in the glass content of the coating. The coating 10 as shown in FIG. 8a does not include glass, while the glass content of the paste in FIGS. 8b and 8c is 0.5 wt % and 1.5 wt %, respectively.

(39) The glass-containing coatings (FIG. 7b, 8b, 8c) exhibit a better surface contact at the interface between coating and converter surface than the glass-free coatings (FIGS. 7a and 8a). Thus, the glass content provides for improved adhesion of the coating to the converter surface. This is also apparent from glass gussets 13 which form in case of the glass-containing coatings. As a preparation artifact (due to cross section preparation by ion beam etching), an additional layer is deposited in the voids, which appears brighter than the glass due to material contrast (see, e.g. FIG. 8c on the right edge of the image where this layer is disposed on a glass gusset, or in FIG. 8a where a thin layer can be seen on the inner walls of the voids of the silver-containing layer).

(40) Moreover, the glass content leads to better sintering of the metal particles to one another.

(41) FIG. 9 shows normalized reflectance spectra of different embodiments of the metal-containing coatings according to the invention on transparent glass ceramic substrates, and of corresponding reference samples.

(42) Evaluation of the paste reflector on a ceramic converter is difficult, because the ceramic converter is not transparent. Therefore, different silver-containing coatings which differ with regard to their content of glass or the glass composition used, were applied onto a transparent glass ceramic substrate (CLEARTRANS) and were then examined for their reflection properties through the substrate. To this end, remission of the samples was measured in spectrophotometer Lambda 950. As reference measurements, a sample in which a highly reflective ALANOD silver mirror was placed behind a non-printed CLEARTRANS substrate was measured as a 100% reference (Ref.sub.HR) on the one hand, and on the other a blank non-printed CLEARTRANS substrate as a zero reference (Ref.sub.OR).

(43) Normalization of the spectra was performed for each wavelength according to the rule

(44) $R_{\mathrm{normalized}}=\frac{R\left(\mathrm{measurement}\mathrm{sample}\right)-R\left({\mathrm{Ref}}_{0R}\right)}{R\left({\mathrm{Ref}}_{\mathrm{HR}}\right)-R\left({\mathrm{Ref}}_{0R}\right)}.$

(45) The measurements show that with the employed exemplary embodiments reflectances of more than 83% (based on the reflection of a silver mirror) can be achieved. The reflectance values depend on the glass content and on the composition and hence the refractive index of the employed glass.

(46) Exemplary embodiment 1 does not contain glass, while the coatings of exemplary embodiments 2 and 3 include 0.5 wt % and 1.5 wt % of a silicate glass, respectively, (based on the paste provided in step b), i.e. prior to firing). Exemplary embodiment 4 contains Bi.sub.2O.sub.3-based glass and exhibits a substantially lower reflectance, due to the refractive index of the glass and to possibly occurring redox reactions at the interface.

(47) FIG. 10 is a graph showing the relationship between refractive index and glass transition temperature of various types of glass used in the following exemplary embodiments.

(48) Table 2 shows different exemplary embodiments A to I, which differ with regard to the employed type of glass and the glass content. The proportions by weight as indicated refer to the paste provided in step b). The rest of the composition which is not listed is the organic pasting medium.

(49) TABLE-US-00002 TABLE 2 Details of exemplary embodiments, listing the glass powder component and composition of the paste Glass Glass content Silver content No. component Tg ( C.) (wt %) (wt %) A non 0 85% B SiO.sub.2A 549 0.2% 85% SiO.sub.2-rich C SiO.sub.2A 549 0.5% 84% SiO.sub.2-rich D SiO.sub.2A 549 1.5% 82% SiO.sub.2-rich E SiO.sub.2B 433 0.5% 84% SiO.sub.2-rich F Bi.sub.2O.sub.3A 365 5.0% 84% Bi.sub.2O.sub.3-rich G ZnOPA 455 2.6% 83% ZnO/P.sub.2O.sub.5-rich H ZnOBA 476 1.9% 83% ZnO:B.sub.2O.sub.3-rich I SO.sub.3A 343 1.7% 83% SO.sub.3-rich

(50) Below, the glass compositions of embodiments A to I are listed (in percent by weight):

(51) SiO.sub.2A, SiO.sub.2-rich

(52) TABLE-US-00003 SiO.sub.2 31.1 Al.sub.2O.sub.3 8.8 B.sub.2O.sub.3 23.0 Na.sub.2O 8.2 CaO 17.6 ZnO 11.3

(53) SiO.sub.2B, SiO.sub.2-rich

(54) TABLE-US-00004 SiO.sub.2 56.01 Al.sub.2O.sub.3 5.28 B.sub.2O.sub.3 3.96 Li.sub.2O 18.89 BaO 11.89 ZnO 3.96

(55) Bi.sub.2O.sub.3A, BiO.sub.2O.sub.3-rich

(56) TABLE-US-00005 Bi.sub.2O.sub.3 SiO.sub.2 Al.sub.2O.sub.3 B.sub.2O.sub.3 ZnO 80.7 1.2 0.8 6.2 11.1

(57) ZnOPA, ZnO/P.sub.2O.sub.5-rich

(58) TABLE-US-00006 P.sub.2O.sub.5 51.1 Al.sub.2O.sub.3 1.9 MgO 1.8 CaO 2.5 SrO 4.7 BaO 13.8 ZnO 24.2

(59) ZnOBA, ZnO:B.sub.2O.sub.3-rich

(60) TABLE-US-00007 SiO.sub.2 7.0 Al.sub.2O.sub.3 6.0 B.sub.2O.sub.3 27.0 Na.sub.2O 5.0 K.sub.2O 1.0 MnO.sub.2 6.0 ZnO 48.0

(61) SO.sub.3A, SO.sub.3-rich

(62) TABLE-US-00008 P.sub.2O.sub.5 33.45 SO.sub.3 15.08 Na.sub.2O 14.6 CaO 3.3 ZnO 33.56

(63) FIG. 11 schematically illustrates the structure of a complex converter-cooling element assembly with localized heat input 1 (e.g. via a laser spot) into the converter 2.

(64) In this assembly, a converter of 200 m thickness is coated with a metal-containing coating 7 on the surface facing away from the primary light source 1, and the metal-containing coating 7 is connected to a spatially limited cooling element 5 via a metallic solder connection 8 (not illustrated in FIG. 11). The cooling element 5 is connected to a heat sink 14. The thermal flux is represented by arrows 15.

(65) Arrows 15 illustrate the expansion of thermal flux in the converter-cooling element assembly. Due to the use of a laser as the primary light source 1, heat input is locally limited by the beam spot (radius of about 200 m). Expansion of thermal flux already occurs within the ceramic, so that thermal flux density is already decreased at the interface to the cooling element. In the cooling element, the thermal flux is further expanding so that the heat transfer coefficient (HTC) between the cooling element and the actual heat sink may then be acceptable even in case of a bad HTC of this connection, because of the large contact area.

(66) Accordingly, the thermal resistance determined from such an assembly only represents a figure of merit for the real thermal performance of the overall assembly in case of localized heat input, which thermal resistance will strongly depend on the thickness of the converter and the geometry of the laser spot. However, the thermal resistance as determined from such an assembly is not suitable to evaluate a converter-cooling element assembly independently of its application.

(67) FIG. 12 schematically shows a setup by means of which the thermal resistance of a converter-cooling element assembly as shown in FIG. 2 can be evaluated independently of the optical assembly in which it is operated. In order to determine the thermal resistance of a measured device (e.g. a soldered converter), this device is soldered to a Cu carrier and is contacted to a heat source on one end and to a heat sink on the other end. Thermal flux is determined by means of a thermal measuring resistor.

(68) In the measurement setup, temperature measuring points T1/T2 are available for determining the thermal flux in an upper measuring resistor. T5/T6 are the measuring points of a lower measuring resistor.

(69) Therebetween, the device to be measured is arranged between two Cu carriers with temperature measuring points T3 and T4, where a temperature difference T3T4 is determined. This is illustrated in FIG. 12.

(70) The quotient of temperature difference and thermal flux is the thermal resistance of the measured device. The reciprocal of thermal resistance divided by the surface area of the measured device gives the HTC.

(71) The following devices to be measured were used: Measurement 1: ceramic converters of size 5.2 mm5.2 mm, soldered to Cu carriers (T3/T4), which converters in turn were interconnected by two-component glue. Measurement 2: the Cu carriers (T3/T4) directly connected by two-component glue.

(72) With this experimental setup, the thermal resistance of a converter-cooling element assembly is given by 0.5*(thermal resistance of measurement 1thermal resistance of measurement 2).

(73) The thermal measuring resistor for measuring the thermal flux is made of steel 1.4841 (material name15 CrNiSi 25 20). It has a diameter of 10 mm, a length of 10 mm, and two bores for thermocouples at a spacing of 7 mm. Across these thermocouples with a spacing of 7 mm a temperature difference is measured which can then be converted into a thermal flux using the thermal resistance calculated from the material data of the steel. The thermal resistance of this measuring resistor is between 7.5 and 10 K/W, depending on the temperature of the resistor in a range of up to 100 C. Table 4 shows the dependency of the thermal resistance of the measuring resistor from temperature.

(74) TABLE-US-00009 TABLE 3 Approximation of the expected thermal resistances of the measuring resistor Thermal Thermal conductivity Length Diameter Area resistance HTC [W/m * K] [mm] [mm] [mm.sup.2] [K/W] [W/m.sup.2K] 11.9 7 10 78.54 7.49 1700 13.3 7 10 78.54 9.57 1330

(75) TABLE-US-00010 TABLE 4 Thermal conductivities of measuring resistor Thermal conductivity Temperature [ C.] [W/m * K] 20 11.9 100 13.3 200 15.1 300 16.7 400 18.3 500 19.8 600 21.3 700 22.8 800 24.3 900 25.7 1000 27.1

(76) TABLE-US-00011 TABLE 5 Thermal resistances and HTC of the evaluated converter-cooling element assemblies (OC = optoceramic) Length Therm. or Therm. conductivity thickness Diameter Area resistance HTC Assembly Material [W/m*K] [mm] [mm] [mm.sup.2] [K/W] [W/m.sup.2K] OC OC 6 0.2 5.2 5.2 27.04 1.23 30,000 (200 m) glued silicone 0.3 0.01 5.2 5.2 27.04 1.23 30,000 glue OC with 5.2 5.2 27.04 2.47 15,000 silicone glue OC OC 6 0.2 5.2 5.2 27.04 1.23 30,000 (200 m) soldered Ag paste 430 0.005 5.2 5.2 27.04 0.00 86,000,000 reflector solder 67 0.08 5.2 5.2 27.04 0.04 837,500 layer OC 5.2 5.2 27.04 1.28 28,963 soldered OC OC 6 0.05 5.2 5.2 27.04 0.31 120,000 (50 m) glued silicone 0.3 0.01 5.2 5.2 27.04 1.23 30,000 glue OC with 5.2 5.2 27.04 1.54 24,000 silicone glue OC OC 6 0.05 5.2 5.2 27.04 0.31 120,000 (50 m) soldered Ag paste 430 0.005 5.2 5.2 27.04 0.00 86,000,000 reflector solder 67 0.08 5.2 5.2 27.04 0.04 837,500 layer OC 5.2 5.2 27.04 0.35 104,961 soldered

(77) Table 5 shows that for the conventional system used heretofore (ceramic converter of 200 m thickness, glued with silicone), the thermal resistance of the converter and the resistance of the silicone glue connection are approximately the same (1.23 K/W).

(78) By contrast, in the converter-cooling element assembly according to the invention with a metallic solder connection, the thermal resistance is determined by the resistance of the converter so that the thermal resistance of the converter assembly is almost halved due to the soldering as compared to the glued implementation.

(79) The thinner the converter, the more determining becomes the difference in thermal conductivity of the connecting layer itself (silicone or solder), since in this case the resistance of the converter is less significant. In case of a converter having a thickness of only 50 m, the thermal resistance of the soldered embodiment is already more than four times smaller than that of the glued variation.

(80) In case of highly localized heat introduction (e.g. via a laser spot) it has to be observed that it is no longer the total surface area of the converter that contributes to heat conduction, but a section which is roughly given by the size of the laser spot. In this case, the relative contribution of the ceramic to the thermal resistance is the greater, the smaller the beam spot is. This is illustrated in Table 6.

(81) TABLE-US-00012 TABLE 6 Dependence of thermal resistance from the beam spot size Therm. Length Therm. conductivity or thickness Diameter Area resistance HTC Material [W/m*K] [mm] [mm] [mm.sup.2] [K/W] [W/m.sup.2K] Optoceramic 6 0.2 5.2 5.2 27.04 1.2 30,000 Optoceramic 6 0.2 0.4 0.13 265.3 30,000 (Beam spot D = 0.4) Optoceramic 6 0.2 0.08 416.7 30,000 (Beam spot F = 0.08)

(82) FIG. 13 illustrates the shift of chromaticity coordinates in the CIE 1931 color space as caused by the metallic reflector, for pastes including different glass components. Measurement data of the samples measured on a highly reflective ALANOD mirror (R=98%) prior to metallization are located within the upper right ellipse (HR reference measurement). Measurement data of the same samples but with metallic coating are located within the ellipse more to the left and below. Chromaticity coordinate shift D.sub.cx is exemplified for one sample. Since the chromaticity coordinates for the dark reference measurements of the converters on a dark background or on a beam trap are not available in this case, FOM.sub.CIE-cx (see FIG. 6) cannot be calculated in this case. However, since all samples comprise metalized converters of identical converter material and identical thickness, D.sub.cx is appropriate for evaluating the reflectors in this case.

(83) Furthermore, it is clearly apparent that the color locations of the converters prior to and following the metallic coating are not located on a straight line. That means, the metallic reflector not only has an effect on the ratio between excitation light and secondary light, but in the present example additionally leads to a relative increase in red color components and thus causes an offset of the color location line of the converters provided with the metallic coating to the right and downwards in the CIE 1931 chromaticity diagram. This is caused by a stronger reflection of the longer wavelength spectral components compared to the shorter wavelength spectral components of the secondary spectrum, which can also be seen in FIG. 9. This effect might even be more pronounced in case of other metal-containing coatings, which contain for instance gold, so that it can be used for selectively controlling the chromaticity coordinate of the converter-cooling element assembly.

(84) FIG. 14 shows the determined color location shifts D.sub.cx of FIG. 13 as a function of the glass system used in the metallic reflector. A small color location shift is characteristic for a good reflector. In addition to the Ag coating without glass content, the reflectors based on silicate glass exhibit the best reflective properties.

(85) Exemplary embodiment: Connecting of the metal-containing coating and the cooling element by soldering

(86) The copper cooling elements or copper cooling elements protected against corrosion by a nickel-gold layer are placed in a sample holder so that the surface to be soldered is aligned horizontally and is virtually flush with the sample holder. Then a solder mask is placed thereupon, which has a thickness of 100 m, and is aligned using adjustment pins. Solder paste (Heraeus F169 SA4005-86 D30) is applied onto the solder mask and is spread over the apertures with a doctor knife.

(87) After lifting the solder mask, the ceramic converters are placed on the so formed solder paste fields. Then, a fixing aid may be placed on the sample holder, and the entire assembly is placed on a heating plate. After the solder has softened (at 215 C.), the entire assembly is removed from the heating plate after a holding time of 20 seconds. After cooling, the samples need to be freed of residual flux agents. This is accomplished in an ultrasonic bath in ethanol at 40 C. and an exposure duration of 10 minutes.

(88) When soldering converters having a metallic reflector, it becomes apparent that solderability is highly dependent on the amount of glass used in the paste. Ag pastes with a silicate glass content, for example, exhibit poor wetting during the soldering process in case of a relatively high glass content of 1.5 wt %. With decreasing glass content the wetting improves. Even a glass-free Ag paste exhibits excellent wetting. In case of a glass-free Ag paste, however, reflection is impaired by the soldering process, which is already visually recognizable by a darker appearance of the soldered converter with silver-containing coating compared to the not yet soldered converter with silver-containing coating. For the soldering process described in the exemplary embodiment, a silver-containing coating with a silicate glass content of 0.5 wt % is particularly advantageous, since in this manner good solder wetting is achieved without impairing reflection.

(89) It will be apparent from the exemplary embodiment that the metallic coating, the solder that is used, and the soldering process have to be adapted to one another. When using another solder or another soldering process, other glass components are conceivable which still allow for good solder wetting without impairing reflection.

(90) Converter-cooling element assemblies produced by this soldering process were subjected to a thermal cycle test from 40 C. to +160 C. with two cycles per day for 10 days, with not a single case among 20 tested converter-cooling element assemblies, in which the converter would have become detached from the cooling element.

(91) According to one variation of the exemplary embodiment, a solder furnace is used for the soldering process instead of the hot plate, which permits to better achieve the heating curves required for the solder.

LIST OF REFERENCE NUMERALS

(92) 1 Primary light 2 Converter 3 Glue 4 Mirror 5 Cooling element 6 Secondary light 7 Metal-containing coating 8 Solder connection 9 Ceramic converter 10 Silver-containing coating 11 Interface 12 Pores 13 Glass gusset 14 Heat sink 15 Thermal flux 16 SiO.sub.2-tungsten layer stack