Description

WAVELENGTH CONVERTER HAVING A POLYS ILOXANE MATERIAL, METHOD OF MAKING, AND SOLID STATE LIGHTING DEVICE CONTAINING SAME

Field of the Invention

This invention relates to solid state lighting devices that contain silicone materials. More particularly, this invention relates to the use of silicone resins as matrix materials for luminescent materials in solid state lighting devices.

Background of the Invention

Silicones are used in solid state lighting devices, such as light emitting diode (LED) packages, primarily as adhesives, phosphor encapsulants , and optics, e.g. lenses for light extraction. In a typical general lighting application, the operating temperature of the LED package may be as high as 90 °C to 110°C. For more demanding applications, such as automotive headlamps, LED packages may experience temperatures on the order of 125°C to 150°C, with some localized regions reaching upwards of 175 °C. In the emerging high illuminance laser- activated remote phosphor (LARP) or high power LED projection applications, the temperatures may be even higher than those of the automotive applications.

Most methyl silicones retain their optical and mechanical properties for thousands of hours at temperatures below about 150°C. On the other hand, they can form cracks in a matter of days at temperatures above about 200°C. The currently

available UL-listed optical silicones have relative temperature indices (RTIs) in the range of 105°C to 150°C. Phenyl silicones are even less stable, becoming brittle and developing a yellow color rapidly at temperatures above 150°C. As a result, standard methyl or phenyl silicones are unlikely to survive the extreme conditions present in high temperature and/or high flux applications such as LARP or newer high power LEDs . Therefore, a replacement material for the currently used silicones is necessary for many of the high temperature/flux applications . Summary of the Invention

In one aspect, there is provided a wavelength converter having a luminescent material dispersed in a highly cross-linked siloxane network that is formed from a low viscosity methoxy methyl siloxane precursor.

In another aspect, there is provided a method for making a wavelength converter having a polysiloxane matrix.

In a further aspect, there is provided a solid-state lighting device, such as a phosphor-converted LED (pc-LED) or laser diode, which contains the wavelength converter.

The polysiloxane matrix material of this invention is more stable than standard methyl silicones in high temperature and high flux applications and may include the following additional advantages :

• A highly cross-linked polysiloxane-based wavelength

converter can be made with cleaner/sharper edges than a screen printed silicone-based converter;

· The wavelength converter can be made using an inexpensive process at room temperature (or slightly elevated temperature if it is desired to speed up the curing process) ;

Because the fabrication process does not require high temperature or solvents, the process is compatible with nearly all phosphors, so different colors from blue to red, including combinations (e.g. cool- and warm-white blends) , are possible;

The technology is compatible with tape-casting and punching, which simplifies the manufacturing steps and reduces cost;

The highly cross-linked polysiloxane matrix material is much less tacky than standard silicones and converter elements can be punched without fouling the tooling;

The method of fabrication can produce wavelength

converters that are more uniform in terms of brightness and color point than other methods;

Because the precursor material is a liquid, it is possible to incorporate different additives into the converter element, such as nanoparticles , metal alkoxy precursors, organic molecules, polymers, etc.; and

The methoxy methyl siloxane precursor lends itself to be used with or without solvent.

Brief Description of the Drawings

Figure 1 is an exemplary structure of a methoxy functionalized methyl polysiloxane and/or methoxy functionalized methyl polysiloxane precursor wherein the methoxy content is ~32 wt% . The number of repeat units, n, can vary. Figure 2 is a schematic illustration of how a nanoparticle with hydroxy (-OH) groups can be incorporated into the polysiloxane network. This is for illustrative purposes only. It is not intended to imply that four polysiloxane units will bond to each nanoparticle, or that each hydroxyl group would be involved in the cross-linking.

Figure 3 shows Attenuated total reflectance-Fourier transform infrared spectroscopy (ATR-FTIR) spectra / curves in ATR Units (ATR U) for a preferred methoxy methyl siloxane in liquid resin (11) and cured (dashed, 12) forms (u - wavenumber in cm ^-1 ) .

Figure 4 is an exemplary structure of a highly cross-linked polysiloxane that could result from the hydrolysis and

condensation of the precursor shown in Figure 1. The dangling bonds could indicate a continuation of the structure or a terminal group.

Figure 5 is a schematic illustration of a solid state lighting device according to this invention for an LED application.

Figure 6 is a schematic illustration of a solid state lighting device according to this invention for a laser diode

application .

Detailed Description

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 wherein like numerals designate like parts. As used herein, the term "wavelength converter" means an individual component that is designed to be used with a light source, preferably a semiconductor light source. The

wavelength converter contains a luminescent material that converts a primary light emitted by the light source into a secondary light having a different wavelength than the primary light. The luminescent material may comprise a phosphor, quantum dots, or other photoluminescent material which is excited by the primary light emitted by the light source, preferably in the ultraviolet (UV) or visible region of the electromagnetic spectrum. The secondary light emitted by the luminescent material typically has a longer wavelength than the primary light (down-conversion.) Unconverted primary light that passes through the converter combines with the emitted secondary light to provide the overall emission from the solid state light source, e.g., a blue-emitting LED may be combined with a wavelength converter having a yellow-emitting phosphor to produce an overall white light emission. In other

applications, the primary light may be completely converted to secondary light (full conversion) so that a single color emission is generated from the solid state lighting device.

As used herein, references to the color of a phosphor, LED or conversion material refer generally to its emission color unless otherwise specified. Thus, a blue LED emits a blue light, a yellow phosphor emits a yellow light and so on.

In a preferred embodiment, the wavelength converter of this invention has a polysiloxane matrix that is preferably formed from a liquid methoxy methyl polysiloxane precursor at room temperature (or slightly elevated temperature if it is desired to speed up the curing process.) A preferred methoxy methyl polysiloxane precursor is shown in Figure 1. The terminal groups of the polysiloxane precursor may comprise one or more chemically reactive groups such as alkoxy, vinyl, hydroxyl, carboxylic acid, ester, or other reactive functional group. In other embodiments, the terminal groups could be a less reactive group, for example alkyls such as methyl and ethyl. While methyl and methoxy side groups are preferred, this does not exclude other functional groups such as ethyl, ethoxy, phenyl, phenoxy, vinyl, trifluoropropyl, etc. Moreover, other

combinations of a polysiloxane backbone with methyl and methoxy side groups are possible. In addition to phosphor particles, many other additives can be accommodated by the liquid precursor and, in turn, the final solid polysiloxane. These additives may include inorganic nanoparticles , metal alkoxy precursors, organic molecules, and other polymers. The different additives serve different purposes, such as controlling the viscosity, providing crack resistance and enhanced mechanical strength, adjusting the refractive index, and increasing thermal conductivity.

An additional potential advantage of the disclosed polysiloxane materials over traditional silicones is how certain

nanoparticles are incorporated. In traditional silicones, nanoparticles are usually present as simple physical mixtures. They are contained within the silicone network, but they are not chemically bonded. In the case of this invention, the precursor material has reactive alkoxy groups that may react with other alkoxy groups or hydroxy groups (among others) . If the nanoparticles are made in such a way that their surfaces contain alkoxy or hydroxy groups, they may chemically bond to the matrix material, making them part of the siloxane network, instead of just being physically trapped in the network.

Figure 2 is a simplified schematic of how this may work. The reactive methoxy groups on the polysiloxane can become

hydrolyzed, leaving behind a silanol group and forming methanol as a byproduct. The hydroxy groups on the nanoparticle and the silanol groups of the polysiloxane could then undergo a condensation reaction, linking them together and forming water as the byproduct. This description is oversimplified since there would also be competing reactions, but the net result of main reactions is the same; a highly cross-linked network of nanoparticles and polysiloxanes .

This has implications for thermal conductivity, transparency, mechanical strength, etc. For example, a phonon will be able to propagate more easily along a continuous network of chemical bonds than through a physical mixture where there are thermal interfaces at every particle-silicone boundary. The

transparency is also influenced by the type of composite. In simple physical mixtures, unless the particles are very small and are very well dispersed, there will be light scattering at every interface between materials of different refractive index, i.e. the particles and the matrix. On the other hand, if the nanoparticles are chemically bonded to the matrix material, it is more like a homogeneous system and scattering will be reduced or even eliminated. A preferred polysiloxane precursor is a low molecular weight methoxy methyl polysiloxane where the methoxy content ranges from 10 to 50 weight percent (wt%) , more preferably from 15 to 45 wt%, and even more preferably from 30 to 40 wt% . During the curing process, each methoxy group that participates in the crosslinking reaction forms methanol (or similar volatile product) , which easily evaporates from the system. Therefore, the more methoxy groups there are to begin with, the more organic content is removed during the curing. Preferably, the cured polysiloxane material loses less than about 20% of its initial weight when heating in air to 1000°C. The precursor is preferably a liquid at room temperature. The molecular weight (and number of siloxane units, n) should be such that the viscosity is in the range of 1-50 mPa ^■ s , and preferably in the range of 2-20 mPa -s . The infrared spectra of a preferred methoxy methyl polysiloxane in liquid and cured forms are shown in Figure 3 and contain absorption bands that are assigned as follows: methyl (-CH ₃ ) 2969-2944 cm ^"1

methoxy (-O-CH3) 2840 cm ^"1

silicon methoxy (S1-O-CH3) 1193, 851 cm ^"1

siloxane (Si-O-Si) 1034-1000 cm ^"1

silicon methyl (-S1-CH3) 1269 cm ^"1

silanol (-Si-OH) 919 cm ^"1

In the spectrum of the cured material, the peaks associated with the methoxy functionality (~2840 cm ^-1 and 1192 cm ^-1 ) have almost disappeared, indicating that most of the methoxy groups were involved in the crosslinking process or were replaced by silanol groups (the peak at 919 cm ^-1 in the cured material is most likely due to the presence of silanol) . The differences between the strongest peaks in the region of ~1000 cm ^-1 to ~1150 cm ^-1 suggest that there was a significant change in the siloxane structure going from liquid to cured forms. The broad peak at ~3400 cm ^-1 indicates that the cured material absorbed some moisture.

There is a sharp peak at 1269 cm ^-1 in both spectra. This peak is associated with S1-CH3 motions. The location of this peak is dependent on the number of oxygen atoms to which the silicon is bonded. For siloxane compounds, a common naming scheme for describing the monomeric unit is shown in Table 1. This nomenclature system is based on the ratio of oxygen to methyl groups for a given silicon atom. The presence of the 1269 cm ^-1 peak and the lack of any peaks at 1250 cm ^-1 and 1260 cm ^-1 suggests that only T-type units are present in either the resin or cured material indicating a structure similar to that shown in Figure 4. This is confirmed by the peaks at ~762 cm ^-1 and 767 cm ^-1 , which are also indicative of T units.

Table 1. Abbreviations for siloxane monomeric and associated IR bands.

IR bands for

Monomeric unit

Abbreviation Si-R

(R = methyl)

(cm ^"1 )

R 12 60

TOUVQ Si 0 ^JUUUUL 8 60 (weak)

8 00

O 12 7 0

TOUVQ Si 0 ^JUUUUL 7 60 - 7 80

As described previously, phosphor powders may be added to the liquid precursor. The concentration of the phosphor (or phosphor blend) should be ≤80 wt% of the final phosphor-in- polysiloxane wavelength converter. For a given embodiment, the concentration of a given phosphor in the cured polysiloxane matrix depends on the activator (e.g. Ce ³⁺ or Eu ²⁺ )

concentration, the phosphor's absorptance and quantum

efficiency, the target color point, and if there are other scattering additives present in the system. The following is a nonexclusive list of phosphors that may be incorporated into the polysiloxane matrix of this invention: (REi-xCe _x ) 3 (Ali-yA' _y ) 5O12 where RE is at least one of Y, Lu, Tb, and Gd, x is in a range 0<x≤0.1, A' is at least one of Sc and Ga, and y is in a range 0≤y≤l) ;

(REi-xCe _x ) 3 (Al5-2yMg _y Si _y ) O12 where RE is at least one of Y, Lu, Tb, and Gd, x is in a range 0<x≤0.1, and y is in a range 0≤y≤2;

(REi-xCex) 3Al5-ySiyOi2-yN _y where RE is at least one of Y, Lu, Tb, and Gd, x is in a range 0<x≤0.1, and y is in a range 0≤y≤0.5;

(REi-xCex) ₂ CaMg ₂ Si30i2 : Ce ³⁺ where RE is at least one of Y, Lu, Tb, and Gd, and x is in a range 0<x≤0.1;

(AEi-xEux) 2S15 8 where AE is at least one of Ca, Sr, and Ba, and x is in a range 0<x≤0.1;

(AEi-xEux) AI S 1N3 where AE is at least one of Ca, Sr, and Ba, and x is in a range 0<x≤0.1;

(AEi-xEux) 2AI 2S 12 6 where AE is at least one of Ca and Sr, and x is in a range 0<x≤0.1;

( Sri-xEux) L1AI3N4 where x is in a range 0<x≤0.1;

(AEi-xEux) 3Ga3 5 where AE is at least one of Ca, Sr, and Ba, and x is in a range 0<x≤0.1;

(AEi-xEux) S12O2N2 where AE is at least one of Ca, Sr, and Ba, and x is in a range 0<x≤0.1;

(AExEuy) Sii2-2x-3yAl2x+3yO _y Ni6-y where AE is at least one of Ca, Sr, and Ba, x is in a range 0.2≤x≤2.2, and y is in a range 0<y≤0.1;

(AEi-xEux) 2S 1O4 where AE is at least one of Ca, Sr, and Ba, and x is in a range 0<x≤0.1; and

(AEi-xEux) 3S1O5 where AE is at least one of Ca, Sr, and Ba, and x is in a range 0<x≤0.1. Other phosphors may also be used including slight modifications of the examples listed above, e.g. incorporation of fluoride or other halide ions. For an automotive forward lighting

application, the phosphor of choice would be a cerium

activated, gadolinium doped garnet, (Yi-x- _y GdxCe _y ) 3AI5O12 , where x is from 0≤x≤0.2 and y is from 0<y≤0.05.

For many embodiments of this invention, additives play an important role in the processing ability of the precursor and/or the final properties of the highly cross-linked

polysiloxane . One class of additives are inorganic

nanoparticles (including crystalline and non-crystalline phases.) The following is a non-exclusive list of example materials :

Oxides: S1O2, Zr0 ₂ , Ti0 ₂ , AI2O3 , and ZnO;

Nitrides: A1N, Si ₃ N ₄ , and BN; and

Carbon-based: carbon nanotubes and graphene .

In some embodiments, the surfaces of the inorganic

nanoparticles may be modified with capping agents to make them miscible with the polysiloxane precursor. The concentration of nanoparticle additives may vary from 0 to 75 wt% . Preferably, the nanoparticles are comprised of Zr0 ₂ and/or Si0 ₂ .

A second class of additives are liquid metal-containing compounds such as zirconium, titanium, and halfnium alkoxides. The concentration of this type of additive may vary from 0 to 75 wt%. A third class of additives are organic molecules that act as adhesion promoters, plasticizers , de-foamers,

thickeners, thinners, etc. The concentration of this type of additive may vary from 0 to 10 wt% . A fourth class of

additives are the organically modified silicas, silicates or ceramics. These additives may be added as-is to the siloxane precursor, but it may be more preferable to add the un-reacted precursors of these materials to the siloxane. A fifth class of additives are polymers, which may be organic (carbon-based chain) or inorganic (non-carbon-based chain) . Some nonexclusive examples include poly (dimethyl siloxane),

poly (methylphenyl siloxane), poly (diphenyl siloxane),

poly ( silphenylene-siloxane) , polyphosphazenes , polysilazane, perhydropolys ilazane . The concentration of this type of additive may vary from 0 to 75 wt% .

The typical combination of catalysts for the curing of the siloxane precursor to cross-linked polysiloxane is humidity and tetra-n-butyl titanate. The tetra-n-butyl titanate catalyst is not necessarily required since the presence of water (liquid or gas) is sufficient for the curing step in a number of

applications. While these catalysts are preferred, other catalysts or combinations of catalysts may be used. When exposed to water, with or without a catalyst, the methoxy methyl polysiloxane precursor undergoes hydrolysis and

condensation reactions, which cross-link the low molecular weight polysiloxane units into a dense polysiloxane network. Figure 4 illustrates a polysiloxane network that would result if all the methoxy groups participated in cross-linking. In practice, not all of the methoxy groups would likely result in cross-linking. Some of them might remain intact and some of them might be replaced by silanol groups. It should be noted that the structure drawn in Figure 4 is for illustrative purposes and should not be construed as limiting the

polysiloxane material in any way. In one embodiment, a phosphor-in-polysiloxane wavelength converter is made by first adding fumed silica to the liquid methoxy methyl polysiloxane precursor to increase the

viscosity. Once the fumed silica is thoroughly incorporated, a desired phosphor powder or blend of phosphor powders is dispersed in the polysiloxane precursor. If desired, other additives, e.g., other oxide nanoparticles , may also be added to the liquid precursor. Typically, a catalyst such as a titanium alkoxide is added in a range of lwt% to 5wt%, but is not required. The relative amounts of phosphor and additives would depend on many criteria, such as the size of the

particles, the desired color point, the volume and thickness of the conversion layer, the wavelength of the exciting radiation, the type of solid state lighting device (e.g. laser or LED), and the amount (s) of other additive materials.

The precursor mixture would then be dispensed by any of a number of techniques, such as spray-coating, dip-coating, spin- coating, drop-casting, tape-casting, doctor-blading, etc. It could be dispensed on a permanent substrate, temporary

substrate, or on the LED itself. In the presence of humidity, or if liquid water was added to the precursor solution, the mixture will begin to cure at room temperature. If desired, curing can be accelerated by applying mild heating (~50 °C to ~150 °C or even higher) . If necessary, the resulting phosphor- in-polysiloxane solid could then be punched or diced to the proper shape and size and incorporated into the LED package or laser-activated module. In a preferred method, the polysiloxane precursor is combined with fumed silica, phosphor powder, and a catalyst. The dispersion is tape-cast or spray-coated onto a non-stick carrier foil. The tape is allowed to cure at room temperature in the presence of humidity for several hours to as long as three days and then removed from the non-stick foil.

Wavelength converters are then formed to the desired size by dicing or punching the tape. The converters are heat treated at a temperature slightly above the expected operating temperature in the end-use application, preferably at least about 5°C higher and more preferably 5°C to 50°C higher. The heat treatment step also could be performed before or after forming the individual converters. Preferably, the wavelengths converters are formed as small flat platelets having dimensions on the order of an LED die.

The target thickness of the final phosphor-in-polysiloxane part is preferably on the order of 25 ym to 200 μπι. The

concentration of the phosphor in the polysiloxane matrix depends on the activator concentration (e.g., Ce ³⁺ in

Y3AI5O12 : Ce ) , the phosphor's absorptance and quantum efficiency, the target color point, and if there are other scattering additives present. A range of applicable phosphor

concentrations is about 15-80 wt% . In a preferred embodiment, wavelength converter contains about 50 wt% phosphor.

The individual converter platelets may be bonded to LEDs or other similar solid state lighting devices with an optical glue which could be a thin layer of polysiloxane that was made from the same precursor as the matrix material, only without the phosphor. The precise configuration of the solid state lighting device is not a limiting factor of this invention. For example, the peak of the exciting radiation could be in the near-UV, violet, indigo, and/or blue region of the

electromagnetic spectrum, i.e., about 365 nm to 490 nm. A more preferred range is from 430 nm to 465 nm. For the LED

embodiments, one or more LED dies could be directly bonded to a circuit board (chip-on-board configuration) or incorporated into a package. There could be some combination of these two configurations within a given module or light engine.

Example

Fumed silica (12.5 wt%) and a (Yi-x- _y GdxCe _y ) 3AI5O12 type phosphor (25 wt%) are added to a liquid methoxy methyl polysiloxane precursor (percentages correspond to total weight including the silica, phosphor, and liquid polysiloxane) . After being

thoroughly mixed, 1% to 5% tetra-n-butyl titanate is added to act as a catalyst. The precursor mixture is then tape-cast on, for example, a non-stick polymer sheet. The target thickness of the final, cured phosphor-in-polysiloxane converter is on the order of 10 μπι to 500 μπι, but preferably between 25 ym and 200 μπι. The tape-cast material is then allowed to cure at room temperature in ordinary ambient conditions. The curing time depends on the thickness of the tape and the relative humidity. Typically the material is dust dry after a few hours, but requires a longer curing time for full mechanical strength.

Once cured, the phosphor-in-polysiloxane wavelength converters are punched out of the tape, using, for example, a numerical control (NC) punch tool. Preferably, the converters are heat treated to a temperature that is slightly above the anticipated application temperature. Depending on that temperature, the heat treatment step could be performed before or after

punching. Once punched and heat treated, the individual converter platelets are ready to be incorporated into a lighting device based on a light emitting diode (LED) or laser diode (LD) .

To test their stability, converter platelets (1 mm x 1 mm with bond notch) made with a (Y,Gd)AG:Ce phosphor in the

polysiloxane matrix were annealed for two hours at different temperatures up to 500 °C. There was no browning or mechanical failure of the phosphor-in-polysiloxane converter even after annealing at 500 °C. Conversely, a converter platelet made with a standard silicone matrix became unstable after annealing at 400°C for two hours and crumbled at the slightest imparted force.

LED applications :

Figure 5 illustrates an embodiment of a chip level conversion application in which solid state lighting device 500 has a separately formed wavelength converter 504 that is adhered to the light emitting surface 507 of LED die 502. The wavelength converter 504 is comprised of a phosphor-in-polysiloxane converter according to this invention. The wavelength

converter 504 is in the form of a flat platelet that is adhered to the light emitting surface 507 of LED die 502 using a thin layer of an adhesive 526. More particularly, the converter 504 may be attached to the LED die 502 with a thin (<10 μπι) adhesive layer which may be any number of optical silicones. It may also be a low-melting glass or a water glass type of material, e.g. A2(SiC>2)nO where A is some combination of Li, Na, or K and n is in the range of approximately 1 to 4, or

monoaluminum phosphate, MALP. It could also be attached to the LED die with the disclosed highly cross-linked polysiloxane, by depositing the liquid siloxane precursor on the LED and/or wavelength converter, bringing the two together so that the precursor is between the LED and wavelength converter, and letting the material cure. It could also be attached to the LED die with the disclosed highly cross-linked polysiloxane that is filled with nanoparticle additives such as S1O2 and/or ZrC>2, by depositing the filled liquid siloxane precursor on the LED and/or wavelength converter, bringing the two together so that the precursor is between the LED and wavelength converter, and letting the material cure. The thickness of the wavelength converter could vary from about 10 ym to about 500 μπι, but is preferably in the range of 25 μπι to 200 μπι, or even more preferably 25 μπι to 100 μπι. In this embodiment, the thickness of the wavelength converter 504 is chosen such that primary light 506 emitted by LED die 502 is only partially converted into secondary light 516. Secondary light 516 and unconverted primary light 506 exit the light emitting surface 520 of the solid state lighting device 500 to produce the desired overall emission from the solid state light device 500 which may be a white light.

Laser Diode applications:

In the case where the solid state light source of the solid state lighting device is a semiconductor laser diode, the phosphor-in-polysiloxane wavelength converter is preferably located remotely from the laser diode, typically separated by additional optical elements. As shown in Figure 6, the solid state lighting device 600 has a wavelength converter 504 that is bonded to a transparent substrate 610, e.g., sapphire, that is at least transmissive to the primary light 606 emitted by laser diode 602. As previously described, an adhesive layer 526 may be used to bond the converter 504 to the substrate 610. The primary light 606 emitted by laser diode 602 impinges upon and is transmitted by substrate 610 whereupon it enters converter 504 to generate secondary light 516. In this transmissive embodiment, it is often preferable to incorporate a dichroic layer on the incident side of the converter, which allows the exciting radiation to pass, but would reflect the down-converted radiation. In an alternative embodiment, the substrate 610 is reflective, e.g., a silver reflector, and the primary light 606 is impinges directly on the light emitting surface 520 of the converter 504.

While there have been shown and described what are at present considered to be 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 patent application claims the priority of US patent application with the application number 62/356,569, the

disclosure content of which is hereby incorporated by

reference . References

11 ATR-FTIR for methoxy methyl siloxane in liquid resin

12 ATR-FTIR for methoxy methyl siloxane in cured form

500 solid state light device

502 LED die

504 wavelength converter

506 primary light

507 light emitting surface

516 secondary light

520 light emitting surface

526 adhesive

600 solid state lighting device

602 laser diode

606 primary light

610 substrate