Active solid-state devices (e.g. – transistors – solid-state diode – Incoherent light emitter structure – In combination with or also constituting light responsive...
Reexamination Certificate
2000-08-08
2002-09-03
Elms, Richard (Department: 2824)
Active solid-state devices (e.g., transistors, solid-state diode
Incoherent light emitter structure
In combination with or also constituting light responsive...
C257S012000, C257S013000, C438S962000
Reexamination Certificate
active
06445009
ABSTRACT:
BACKGROUND OF THE INVENTION
This invention relates to stackings of GaN or GaInN quantum dots on a silicon substrate.
More precisely, the invention concerns a silicon substrate provided with a coating comprising at least one stacking constituted by a layer of GaN or GaInN quantum dots emitting visible light at room temperature, in a layer of AlN or GaN respectively.
The invention also relates to the procedure for preparing such substrates, as well as the electroluminescent devices and the lighting devices comprising these substrates and stackings.
The technical domain of the invention can be defined as light production and, more precisely, that of white light production, the white light corresponding to the criteria laid down by the CIE in 1964, that is the coordinates in the chromatic diagram corresponding to x=0.33 and y=0.33, for perfect white light.
The production of white light is at present the subject of much research in order to perfect devices presenting in particular the properties of low cost, low energy consumption, a long lifetime and good energy efficiency.
In fact, for a long time white light has been essentially produced by incandescent lamps whose efficiency and lifetime are very low, since they are respectively about 5% and 1,000 hours. The efficiency of fluorescent tubes and their lifetime are higher, since they are about 25% and 1,000 hours for fluorescent lamps, but they present a certain number of drawbacks: in fact, they consist of vacuum tubes, complicated and costly to manufacture, and which, besides, contain several milligrams of highly toxic mercury, so that their destruction at the end of their lifetime poses a serious environmental problem.
In addition, fluorescent tubes produce a light which is not pleasant, which often limits their utilisation.
In terms of efficiency, the best industrial lamps at present are low pressure sodium lamps, which have an efficiency of about 35%. Just like the fluorescent tubes, these are lamps which are not very pleasant, or are even unacceptable for everyday lighting because of their colour. Research is thus directed towards other devices which are safe, solid, reliable, with a long lifetime, and able to produce white light at low cost with an energy efficiency higher than or equal to that of the devices mentioned above.
As an example, such sources of white light have been envisaged using, among others, phosphors, electroluminescent polymers and semiconductors.
Electroluminescent polymers, such as PPV, are very reasonably priced and the technology for using them, which just consists of inserting the polymer between semi-transparent electrodes, is very simple. The whole range of visible colours can be obtained and the emission of white light is obtained either through combination of the colours, or through using a single spectrum compound which is wide enough. However, these compounds are only used at present for the illumination of liquid crystal screens in orange light. The short lifetime of green and blue emitters of this type rule out production of white light with these electroluminescent polymers.
The production of white light by semiconductors depends essentially on nitrides and, in particular, nitrides of group III elements which, alone, emit green or blue with high efficiency and with a long lifetime. The most commonly used nitride type compound is GaInN which emits blue and red light.
A thin layer of nitrides, GaInN for example, inserted into a material such as Ga(Al)N and whose forbidden band fixes the emission wavelength and thus the colour, thus constitutes the elementary brick of the active zone of extremely bright Electroluminescent Diodes (LEDs).
The thickness of the layer of GaInN is generally lower than or equal to 100 Å and one then talks of LEDs with GaInN/Ga(Al)N quantum wells whose emission is centred on a particular colour, for example blue or green.
It is currently accepted that the composition in indium and/or the thickness of the layer of GaInN fixes the transition energy of the quantum well and, consequently, the wavelength of the LED emission. The GaInN/Ga(Al)N quantum wells, however, have very special optical properties for compositions in indium higher than 10%, among which one can mention the abnormally long radiative lifetime of the excitons and the very low energy variation of the forbidden band as a function of pressure.
The extremely bright blue LEDs, whose yield is already very high (greater than 10%), allows the production of white light through a hybrid technology, in which the blue LED serves for pumping the phosphors or polymers. The combination of the yellow luminescence of these compounds with that of the LED produces white light by colour addition. This technology is at present widely used by NICHIA®, as well as by HEWLETT-PACKARD®, GELCORE® or by SIEMENS-OSRAM® and its application to domestic lighting is very promising.
Nonetheless, the coupling of a LED and another constituent, such as phosphor or polymer, with the aim of obtaining white light is, because of its hybrid character, a costly and complex procedure, comprising several technological stages implying a posteriori, for example, the deposition on a blue LED of a phosphor or polymer type compound before encapsulation.
The balance of colours with a hybrid phosphor/polymer-LED is not simple and one does not easily obtain a white described as “pleasant” for domestic lighting. Besides, the device, unlike the procedure for its preparation, is complex, comprises numerous elements and is therefore less reliable than the basic nitride LED whose intrinsic lifetime is about 100,000 hours.
Finally, inherently, the yield of a hybrid system, in which losses are inevitable, is lower than that of the pumping nitride LED.
Therefore it would be interesting to have devices, in particular LED electroluminescent diodes emitting white light directly—that is to say monolithic devices—to be free from the inconveniences of hybrid devices implying coupling of a blue (or green) LED with a phosphor or polymer.
Moreover, in large-scale production of compounds comprising luminescent nitrides, the substrate used most often for growing nitrides is sapphire and, to a lesser extent, silicon carbide SiC. These two materials, particularly sapphire, have a certain number of inconveniences.
The inconveniences of sapphire are that it is an electric insulator and poor heat conductor, whereas the inconveniences of SiC are that it is expensive and of varying quality, and besides this, these two substrates have a limited size.
It was therefore envisaged to replace them by silicon which has evident advantages, both from the economical and technical points of view, compared with the two materials cited above.
In fact, apart from other interesting properties, silicon is a good heat conductor and can easily be eliminated by chemical means.
Moreover, since there already exists a technological channel for silicon which is perfectly controlled on the industrial scale, since its cost is very much lower than those of sapphire and SiC and since it is available under the form of very large-sized substrates, silicon is the substrate to choose for mass production at low cost.
There is therefore a real need to perfect electroluminescent devices on silicon substrates.
However, the growth of nitrides, such as gallium nitride, on silicon substrates come across problems due to the major differences existing between the mesh parameters and the coefficient of thermal expansion of the substrate and the nitride. In order to grow a nitride layer, such as GaN, of high quality, it is generally admitted that, beforehand, it is necessary to deposit on the silicon substrate a thin layer of, for example, AlAs, SiC or AlN, called a “buffer layer”, of a thickness of several tens of nanometres.
The article by S. GUHA and N. A. BOJARCZUK, in Applied Physics Letters 72, 415 (1998) demonstrated that LEDs could be obtained by epitaxial growth through a molecular beam epitaxy procedure (MBE) on silicon (111). In the same way, the luminescence at room temperature of quantum wel
Damilano Benjamin Gérard Pierre
Grandjean Nicolas Pierre
Leroux Mathieu
Massies Jean
Semond Fabrice
Centre National de la Recherche Scientifique
Elms Richard
Oblon & Spivak, McClelland, Maier & Neustadt P.C.
Wilson Christian D.
LandOfFree
Stacking of GaN or GaInN quantum dots on a silicon... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Stacking of GaN or GaInN quantum dots on a silicon..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Stacking of GaN or GaInN quantum dots on a silicon... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2830478