The present invention relates to the deposition of thin film dielectrics for semiconductor device structures, and more particularly relates to nano-layer dielectrics for light emitting device structures and methods of chemical vapour deposition thereof. This invention has particular application to engineered structures for light emitters comprising rare earth luminescent centers in a dielectric host matrix, e.g. for high brightness, solid-state lighting.
Applicant's related copending United States patent publications nos. 2008/0093608 entitled “Engineered Structure for Solid State Light Emitters” published Apr. 24, 2008 in the name of Chik et al, 2007/0181898 entitled “Pixel Structure for a Solid State Light Emitting Device” published Aug. 9, 2007 in the name of Chik et al, and U.S. patent application Ser. No. 12/508,033 entitled “Engineered structure for high brightness solid-state light emitters” filed Jul. 23, 2009 in the name of Calder et al, which are incorporated herein by reference, disclose light emitting device structures and methods of fabrication of light emitting device structures from silicon or other group IV materials. In particular, high brightness, light emitting device structures are disclosed that comprise thin active layers, i.e. light emitting layers, comprising luminescent centres which may be electrically excited, such as rare earth ions in a rare earth oxide or other suitable host matrix. The host matrix may comprise a dielectric, such as silicon dioxide, silicon nitride, silicon oxynitride, silicon aluminum oxynitride (SiAlON), aluminum oxide, or other materials, such as disclosed the above referenced copending applications. For improved excitation efficiency, a drift layer, i.e. an electron acceleration layer, comprising, for example, undoped silicon dioxide or silicon nitride, is provided adjacent each active layer. Preferably, for high brightness light emitters, engineered structures are provided comprising a plurality of active layer and drift layer pairs, in which each drift layer has a thickness, dependent on the applied electric field, related to a required excitation energy of an adjacent active layer. Multilayer light emitting structures may be provided, for example, comprising one or more active layers comprising rare earth doped silicon dioxide or silicon nitride, and corresponding drift layers comprising undoped silicon dioxide or silicon nitride. Active layers are preferably thin, and may be in the range from one monolayer (1 atomic layer) to 10 nm thick Drift layers are typically 2 to 10 nm thick, depending on the applied electric field, to provide a sufficient thickness for acceleration of electrons for appropriate energy matching. Light emitting structures of this type may be fabricated as large area emitters or as pixel structures. Structures may comprise one active layer/drift layer pair or many, e.g. twenty-four, layer pairs emitting one or more colours, to provide light emission of a desired Colour Rendering Index (CRI).
With reference to
For light emitting structures 1, as disclosed in the above mentioned references, which operate at relatively high electric fields, i.e. 5 MV/cm or more, high quality dielectrics with low trap density are required for reliability and extended lifetime. For high brightness and luminous efficiency, careful control of layer thicknesses is desirable. Such structures have been fabricated by plasma enhanced chemical vapour deposition (PECVD), for example.
Literature reports indicate that suitable rare earth containing oxides, or rare earth doped oxides or other dielectrics, can be formed by any of a number of techniques, such as ion implantation, chemical vapour deposition (CVD), physical vapour deposition, i.e. sputtering, spin-on (sol gel) techniques, beam deposition, laser deposition, or any of a large number of similar chemical or physical deposition techniques that are generally well known in the thin film or semiconductor technology fields. Although epitaxial processes are well known for growing nano-structured single crystal semiconductor layers, e.g. for multiquantum well (MQW) semiconductor lasers with QW thicknesses on the order of 3 to 5 nm, to the inventors' knowledge, these processes are typically applied for deposition of crystalline materials and are not applicable for amorphous, i.e. non-crystalline, dielectric layers. Methods, such as chemical vapour deposition (CVD), are well established for achieving high quality, uniform single layers of thin dielectrics, or a few thin film dielectric layers, i.e. the number of layers N=2 or 3, for such structures as transistor gate stacks. However, achieving high quality multilayer stacks (N>3) of thin films of amorphous or non-crystalline materials poses some challenges to known commercial chemical vapour deposition processes for volume production at reasonable cost.
Although light emitting device structures, such as shown in
For commercial production of high quality, thin dielectrics such as silicon dioxide used for gate oxides for electronic devices, industry standard processes use chemical vapour deposition (CVD). That is, silicon dioxide and silicon nitride are typically produced by low pressure LPCVD from a mixture of reactant gases comprising a silicon source gas, typically silane (SiH4), dichlorosilane (SiH2Cl2), or TEOS (tetra-ethyl ortho silicate (Si(OC2H5)4) with an oxidation source gas, e.g. oxygen (O2) or nitrous oxide (N2O), or a nitridation source gas, e.g. ammonia (NH3) or nitrogen (N2), as appropriate. The reaction is thermally driven, at a relatively high temperature, e.g. 400° C. to 900° C. Examples of known systems and processes, and examples of reactant gases and process conditions, for chemical vapour deposition of silicon dioxide, silicon nitride and silicon oxynitride are disclosed, for example, in U.S. Pat. Nos. 6,713,127, 6,884,464, and 7,465,669 to Iyer et al. (Applied Materials). To improve uniformity and flatness of each layer in a multi-layer structure, an optional post deposition treatment or passivation with the oxidation or nitridation source gas is also disclosed. That is, after deposition of a desired layer thickness, the flow of silicon source gas is turned off (or diverted) and the surface of the deposited layer is then exposed to a flow of nitridation or oxidation source gas only, i.e. ammonia or nitrous oxide. This post treatment of the deposited layer with oxidation or nitridation source gas is explained in U.S. Pat. Nos. 6,713,127 and 6,884,464 as terminating unreacted silicon sites, to maximize uniformity and minimize surface roughness of the deposited layers.
However, attempts to fabricate larger area, thin film dielectric layers for a multilayer light emitting structures, as described above, using available commercial thermal CVD processes and equipment, and preferred reactant gas mixtures and process conditions, such as those described in the aforementioned patents, does not provide sufficient control of the layer thicknesses, uniformity, and layer-to-layer reproducibility for such light emitting structures. In fabricating a light emitting device structure wherein active layers comprise rare earth doped silicon nitride, and drift layers comprise silicon dioxide, there is breakdown of the layer structure after deposition of the first few layers on a silicon substrate. For example,
The photomicrograph in
However, as will be seen from
Consequently, improved processes are required for deposition of multilayer light emitting structures comprising nano-layer dielectrics, particularly for layers less than 10 nm thick, where careful control of layer uniformity, thickness, surface roughness or flatness, and layer-to-layer reproducibility is required.
It is desirable to overcome the above-mentioned limitations and provide for improved uniformity and flatness of dielectric layers for light emitting structures and other semiconductor devices.
Thus, there is a need for alternative or improved materials, structures and/or methods of fabrication for solid-state light emitters, particularly for applications requiring higher brightness, luminous efficacy and reliability, such as solid-state lighting.
The present invention seeks to overcome or mitigate the above-mentioned problems relating to deposition of thin film dielectrics and deposition of nano-layer structures for solid-state light emitters, or at least provide an alternative.
One aspect of the invention provides a method of fabricating a thin film dielectric structure for a semiconductor device comprising:
a) depositing a first layer, comprising a first dielectric, on a substrate; and
b) surface treating the first layer with a surface activating reactant gas.
When the thin film dielectric comprises a plurality of layer pairs of first and second dielectric layers, steps of deposition of a first or second dielectric layer and post deposition surface treatment thereof are repeated sequentially to provided a desired number of layer pairs.
Another aspect of the invention provides a method of depositing, by chemical vapour deposition (CVD) in a CVD reaction chamber, a thin film dielectric stack comprising:
a) providing a substrate;
b) depositing a first dielectric layer of a first desired thickness by exposure of the substrate to a first reactant gas mixture comprising a silicon source gas and at least one of one of a nitridation source gas and an oxidation source gas;
c) surface treating the first dielectric layer by exposure to the silicon source gas;
d) depositing a second dielectric layer of a second desired thickness by exposure to a second reactant gas mixture comprising a mixture of a silicon source gas and at least one of one of a nitridation source gas and an oxidation source gas; and
e) surface treating the second dielectric layer by exposure to the silicon source gas. a method
In particular, dielectric layers such as oxides, nitrides or oxynitrides of silicon or other silicon compatible dielectrics, with or without dopants, such as rare earth luminescent centres, may be deposited with significantly improved uniformity, and reduced surface roughness, approaching atomically flat layers. For example, multilayer thin film dielectrics may be deposited by thermal CVD, characterized by a surface roughness of less than 0.21 nm, when the deposited dielectric layer is surface treated with a silicon source gas, and silane in particular, before depositing a subsequent layer thereon.
For improved layer-to-layer uniformity, multilayer dielectric structures suitable for light emitting structures may be deposited by surface treating each individual dielectric layer with silane prior to deposition of a subsequent layer.
Further aspects of the invention provide semiconductor devices structures and light emitting device structures comprising thin film dielectric layers deposited by these methods. For light emitting device structures, active (light emitting) layers may be doped with luminescent rare earth species or other luminescent centres.
Thus, in a preferred embodiment, a light emitting device structure may be provided comprising a plurality of active layer and drift layer pairs, each of the active and drift layers comprising a substantially atomically flat dielectric layer deposited by chemical vapour deposition. In preferred embodiments of this structure, each dielectric layer is selected from the group consisting of doped and undoped oxides and nitrides of silicon. Each active layer comprises luminescent centres, preferably a rare earth, such as cerium, and has a thickness from 1 monolayer to 10 nm. Each drift layer has a thickness from 2 nm to 10 nm. Preferably each layer has a surface roughness of less than 0.21 nm Light emitting structures may comprise a layer stack comprising one or more active layer/drift layer pairs, wherein the drift layers are configured with electrode structures as appropriate, for DC or for AC operation.
For example, for fabrication of a multilayer dielectric stack for in a light emitting structure, the first dielectric layer may comprise rare earth doped silicon nitride (active layer) deposited from a reactant mixture of silane and ammonia (NH3). The second dielectric layer may comprise silicon dioxide (drift layer). To deposit silicon nitride, the first reactant gas mixture comprising silane and ammonia, together with a suitable source of rare earth dopant if required, is introduced into the chamber to deposit a desired thickness of the first dielectric layer, i.e. the rare earth doped silicon nitride. Then the exposure of the substrate to the flow of all reactant gases is stopped, with the exception of the silicon source gas, which in this embodiment is silane. Silane is flowed into the chamber, with an inert carrier gas to maintain pressure and flow dynamics, for a sufficient time to surface treat the deposited layer. After purging the chamber a second mixture of reactant gases consisting of silane and nitrous oxide are introduced into the chamber to deposit a required thickness of silicon dioxide, and then the oxygen containing reactant gas flow is stopped, to expose the deposited surface to silane. Thus after deposition of each dielectric layer, the surface of the deposited dielectric layer is treated by briefly by exposure to the silicon source gas, plus additional non-reactive carrier gases to maintain chamber pressure and flow dynamics, and then the chamber is purged with the carrier gas or gases, before deposition of the next layer. Silane treatment of the deposited dielectric layer is preferably long enough to improve flatness, e.g. by providing active silicon or a monolayer of silicon on the surface, while avoiding formation of a layer of amorphous silicon or polycrystalline silicon.
While this method is particularly suitable for deposition of engineered light emitting structures as described in the above reference copending patent applications, thin well defined doped or undoped dielectric layers that are substantially atomically flat are also desirable for many other electronic devices such as gate oxides for transistor gate stacks, and gate oxide or tunnel oxides for memory or flash devices.
While methods according to preferred embodiments are particularly suitable for deposition by thermal CVD of commonly used silicon containing dielectrics such as nitrides, oxides and oxynitrides of silicon, it is believed that similar surface activating treatment may be applicable for deposition of other dielectrics, e.g. SiAlON or aluminum oxide, or other dielectrics of similar structure or surface chemistry. with improved layer-to-layer reproducibility and reduced surface roughness. More generally, during chemical vapour deposition of dielectrics comprising oxides, nitrides or oxynitrides of metal or semiconductor species M, deposited layers are surface treated with an appropriate M source gas, before deposition of a subsequent dielectric layer.
Methods and structures according to embodiments of the invention therefore provide for improved flatness, reduced surface roughness, and improved layer-to-layer uniformity in multilayer semiconductor device structures, and particularly for light emitting device structures.
Embodiments of the invention will now be described, by way of example, with reference to the following drawings.
With reference to
Typical materials for thin film dielectric structures comprising multilayer dielectric stacks, e.g. for light emitting devices or in other semiconductor device structures, may comprise, for example, suitable oxides, nitrides or oxynitrides of a metal or a semiconductor. Silicon dioxide, silicon nitride and silicon oxynitride are widely used as dielectrics for silicon process technology. Other examples include silicon aluminum oxide (SiAlON) and aluminum oxide. Metal oxides, such as hafnium or zirconium oxide, or metal silicates may be used for high k gate dielectrics. Dielectrics comprising luminescent centers, such as rare earth doped silicon nitride or silicon dioxide, or rare earth oxides may be used for light emitting devices.
When the multilayer dielectric stack 20 of
Thus, ideally, each layer 12 and 13 is deposited by a method that provides a smooth surface, with minimal surface roughness, and where there is an abrupt, and preferably atomically flat interface 16, between layers 12 and 13, as represented schematically in
With reference to
Relative to the structure fabricated by a conventional known process and shown in the TEM image of
With reference to
The multi-layer device structure 20 is provided in which the surface of each of the first and second dielectric layers 12 and 13 is characterized by a surface roughness of 0.21 nm or less, e.g. as measured by AFM (atomic force microscopy).
The multi-layer device structures 10 and 20, shown in
For deposition of the subsequent dielectric layer, i.e. a second dielectric layer 13 comprising silicon dioxide, the chamber was preheated again (104-1), and the surface of the first dielectric layer 12 (silicon nitride) was pre-treated with an oxidation source gas N2O, as at 104-2. Then, a second reactant gas mixture comprising a silicon source gas, e.g. silane, and oxidation source gas, e.g. nitrous oxide, together with the carrier gas, is introduced into the chamber for deposition of the second dielectric layer 13 comprising silicon dioxide. The substrate 11 and the first dielectric layer 12 were exposed to the second reactant gas mixture for sufficient time, i.e. 8 seconds, to deposit the required thickness of the layer 13, e.g. 4 nm of silicon dioxide, as at 104-3. The flow of the oxidation source gas, i.e. nitrous oxide reactant gas, was then turned off, but the flow of the silicon source gas, i.e. silane, was continued for surface treatment of the deposited silicon dioxide dielectric layer 13, as at 105-1, and then the chamber was purged with the carrier gas (105-2) and cooled (105-3). The sequence was repeated until the required number of periods of first and second dielectric layer pairs 12/13 of the multi-layer device structure 10 or 20 were provided, as at 106.
After preparing the multilayer device structure 10 or 20 by process steps as summarized in
Preferably, the duration of the surface treatment with the surface activating reactant gas, e.g. silane, after deposition of each dielectric layer 12 and 13 is long enough to improve flatness, but sufficiently short to avoid formation of a layer of amorphous silicon or polycrystalline silicon on the dielectric surface. For example: for the process conditions summarized in Tables 1 and 2, surface treatment with the silicon source gas for 1 second after turning off the oxidation or nitridation source gases was sufficient to provide improved flatness, although exposure to the silicon source gas for surface treatment or surface activation may be in the range from about 0.5 seconds up to about 10 seconds.
Preferably, for the step 102 of deposition of each of the first and second dielectric layers 12 and 13, to stabilize the chamber, the chamber is preheated (102-1) with a flow of inert or non-reactive carrier gas, i.e. nitrogen, and the flow of the oxidation or nitridation source gas is initiated first before diverting flow of the silicon source gas into the chamber to initiate the reaction for deposition of the required thickness of the respective dielectric layers 12 or 13. After stopping flow of the oxidation or nitration source gas, the deposited layer 12 or 13 is surface treated by exposure to the surface activating reactant gas, e.g. silane. The chamber is purged (103-2) with the carrier gas after the step 103-1 of surface treatment of the deposited dielectric layer 12 or 13 with the silicon source gas. The process according to the embodiment was carried out in a conventional thermal CVD deposition chamber, which provides for the introduction of the carrier gas and/or reactant gas mixture in the upper region for exposure of the substrate for deposition, and a second input for a bottom flow of carrier/purge gas. The carrier gas flow maintains the chamber pressure and flow dynamics, and is continued during surface treatment of deposited layers 12 and 13 with the surface activating reactant gas, e.g. the silicon source gas: silane. Exemplary gas flows and process conditions for deposition of dielectric nano-layer pairs 12 and 13 comprising 5 nm of silicon nitride and 4 nm of silicon dioxide are summarized below, in Tables 1 and 2, respectively.
Similar process steps can also be used for the deposition of a multi-layer light emitting device structure, similar to the multi-layer device 20 shown in
Thus, as illustrated by the TEM photomicrographs shown in
Notably, the remarkable and unexpectedly large improvement in flatness achieved with the process described above, could not be achieved using conventional known post-deposition treatment processes with oxidation or nitridation source gases. Indeed surface treatment of the deposited silicon nitride or silicon dioxide dielectric layers 12 and 13, which results in the improved flatness and uniformity, appears to be specific to treatment with the silicon source gas: e.g. silane.
By comparison, conventional post-treatment of silicon dioxide layers with the oxidation source gas (nitrous oxide) or post-treatment of a silicon nitride layers with the nitridation gas (ammonia) which has been reported to passivate each layer and terminate active silicon sites for reduced surface roughness, resulted in the rough surface structures, as shown in
As exemplified by the gas flows listed by way of example in Table 1 and Table 2, during deposition of the silicon nitride or silicon dioxide layers, there is a relatively small flow of silicon source gas relative to the flow of oxidation or nitridation source gas. During surface treatment with the silicon source gas, the flows of nitridation and/or oxidation source gases are turned off, and the flow of silicon source gas is continued, so that the surface of the deposited dielectric is exposed to the silicon source gas only, or at least, a very high ratio of the silicon source gas relative to any residual reactant source gases, i.e. residual oxidation and or nitridation or source gas, that may remain in the chamber during the step of surface treatment.
Surface treatment of the deposited layers 12 and 13 with another silicon source gas, i.e. disilane, did not result in as dramatic an improvement in flatness compared with use of silane. The difference may be attributable to different surface chemistry effects, or may, in part, be due to a higher impurity level of the available disilane source compared with the silane source.
Without wishing to be limited by any particular theory, the present inventors believe that the significant improvement in surface flatness using silane chemistry for deposition, and silane surface treatment of the deposited dielectric layer is primarily due to surface chemistry effects on the dielectric surface, i.e. surface activation of the dielectric surface to create active silicon sites. Thus, it is believed that surface treatment, with silane or other silicon source gases, provides active silicon on the surface, e.g. a monolayer of silicon or a silicon rich surface, that facilitates deposition of subsequent layers with improved layer-to-layer uniformity and reduced surface roughness in multilayer structures. Surface treatment with silane is particularly effective.
While specific embodiments of the method are described above, it will be appreciated that, in other embodiments, similar doped and undoped dielectric layers, such as metal or semiconductor oxides or nitrides, e.g. silicon nitride, silicon dioxide silicon oxynitride, hafnium oxide, zirconium oxide, silicon aluminum oxynitride (SiAlON), and aluminum oxide, may be deposited from other reactant gas mixtures comprising a semiconductor or metal, e.g. silicon, source gas, and one or more reactant source gases, e.g. a nitridation and/or an oxidation source gas, using a range of process conditions, for example, systems and processes as described in the above references United States patents to Applied Materials, or other CVD apparatus and processes. However, instead of a conventional post-deposition treatment by passivation of the surface of each deposited layer with either of the reactant source gases, i.e. nitridation or oxidation sources gases, the deposited layers in embodiments of the invention are surface treated after deposition with a brief exposure to a surface activating gas, i.e. the metal or semiconductor source gas, preferably silane, together with appropriate non-reactive carrier gas(es) to maintain chamber pressure and flow dynamics, to provide active metal or semiconductor on the exposed surface before subsequent processing.
Optionally, dopants may be added to one or more of the first and second dielectric layers 12 and 13, by addition of a suitable dopant precursor to the respective reactant gas mixture. Thus, for light emitting structures in which the first dielectric layers 12 are active layers containing rare earth luminescent centres, for example, a suitable rare earth precursor is added the reactant gas mixture during deposition of the active layers.
Thus, for example, in alternative embodiments for deposition of oxides, nitrides and oxynitrides of silicon by thermal CVD, pressure in the reactor or deposition chamber may be maintained in the range from about 50 Torr to 350 Torr, temperatures in the range from about 400° C. to 900° C., and appropriate flow ratios of silicon source gas and nitridation and/or oxidation source gases in the range 1:50 to 1:10000 to provide a suitable deposition rate, e.g. approximately 30 nm/min to enable sufficient control of the thickness of the deposited layers 12 and 13 of less than 10 nm in deposition times of approximately 10 seconds per layer. The silicon source gas may be silane, disilane, methylsilane, or halogenated silanes, or other suitable silicon source gases, although silane is preferred for the step of surface treatment and surface activation of the deposited dielectric layers 12 and 13. Nitrous oxide, ozone or TEOS may be used as an oxidation source gas, for example Ammonia or hydrazine may be used as a nitridation source gas, for example. The carrier gas may be nitrogen, argon, xenon or helium, for example.
The surface activation treatment with silicon source gas, and silane in particular, is applicable to fabrication other similarly structured dielectric thin films comprising other dielectrics which are compatible with silicon technology, such as, silicon aluminium oxy nitride (SiAlON). For other doped or undoped dielectrics, which have similar surface chemistry, such as other semiconductor or metal oxides, nitrides or oxynitrides, surface treatment by exposure of the deposited dielectric layer to a respective semiconductor or metal source gas to activate the surface before deposition of subsequent dielectric layers may similarly reduce surface roughness and improve layer-to-layer uniformity in multilayer structures.
More generally, in deposition of other thin film dielectric layers, for semiconductor devices comprising oxides, nitrides, or oxynitrides of semiconductors and/or metals, a similar surface activation treatment would be beneficial in improving flatness and layer-to-layer uniformity in deposition of multilayer structures. Typically, a semiconductor or metal containing dielectric is deposited from a reactant gas mixture comprising a first reactant, which is metal/semiconductor (M) source gas, and a second reactant, e.g. at least one of an oxidation and nitridation source gas. Instead of surface passivation with the second reactant, i.e. the oxidation or nitridation source gas, as is conventional, deposition of multilayer structures, according to embodiments of the present invention, includes reducing or stopping the flow of the second reactant, during surface treatment, and to exposing of the deposited dielectric layer to the reactive gas source of the appropriate semiconductor or metal (i.e. cationic) species (M) to provide active species of M, or a monolayer of M, on the surface of a deposited dielectric layer. Thus, when the dielectric layer is deposited from a reactant gas mixture comprising a semiconductor/metal M source gas and one or more of a nitridation and an oxidation source gas, after deposition of a dielectric layer of the required thickness, the flow of the oxidation and nitridation source gases are preferably terminated, and the surface of the deposited layer is treated by continued exposure to the flow of the reactant metal/semiconductor source gas alone, together with appropriate carrier gases to maintain the chamber pressure and flow dynamics. One example would be surface treatment of aluminum oxide dielectric layers (Al2O3) with an aluminum source gas.
An engineered structure for a light emitting device 100, as described in the above-referenced copending patent applications by Chik et al, is illustrated in
Deposition of each rare earth doped active layer 12 and 14 and each drift layer 13 and 15 by the chemical vapour deposition methods according to the above described embodiments, including surface treatment of the deposited silicon nitride 12 and 14 and silicon dioxide 13 and 15 layers with silane, provides for improved flatness, layer to layer continuity and lateral continuity of deposited layers.
In variations of the light emitting structure 200 described above, the active layers 12 and 14 may comprise, for example, rare earth doped silicon dioxide, rare earth oxides, or other materials as disclosed in the above referenced related copending applications. The drift layers 13 and 15 may comprise e.g. silicon dioxide or silicon nitride. Selected first and second active layers 12 and 14, may be doped with different rare earth luminescent species for emission of different colours, or active layers 12 and 14 may be co-doped with two or more different rare earth luminescent species, for emission of a light of a specific CRI, e.g. white light. Active layers for these and other light emitting device structures may comprise a concentration of rare earth luminescent centres from 0.1 at. % to about 50 at %. Although this embodiment shows a light emitting structure comprising two layer stacks 12/13 and 14/15 each with a plurality of layer pairs, alternative embodiments may comprise additional stacks of active layer/drift layer pairs comprising rare earth luminescent centres for producing light of different colors and wavelengths or for light emission of a particular CRI.
Preferred rare earth luminescent centres comprise one or more of Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm or Yb. One or more selected rare earth luminescent centre is incorporated into the reactant gas mixture for deposition of the active layers, 12 or 14, by including a suitable rare earth precursor, e.g. a metal organic precursor compound which is sufficiently volatile at the required deposition temperature and pressure.
Examples of rare earth precursors are disclosed, for example, in United States published patent application no. US2007/0181906 published Aug. 9, 2007 in the name of Chik et al, which is incorporated herein by reference. Suitable ligands for the rare earth element may include acetate functions, for example 2,2,6,6-tetramethyl-3,5-heptanedione, acetylacetonate, fluorolacetonate, 6,6,7,7,8,8,8-heptafluoro-2,2-dimethyl-3,5-octanedione, i-propylcyclopentadienyl, cyclopentadienyl, and n-butylcyclopentadienyl.
Typically, each active layer 12 and 14 has a thickness from 1 nm to 10 nm, although active layers 12 and 14 may be as thin as one monolayer (one atomic layer), and each drift layer 13 and 15 typically has a thickness from 2 nm to 10 nm to provide an appropriate electron excitation energy, as explained above. For electroluminescent light emitting device structures, electrodes are provided for applying a suitable electric field, which may be about 5 MV/cm to 8 MV/cm or more, for excitation of luminescent centers. Active layer/drift layer pairs 12/13 or 14/15 and electrodes may be configured for operation with an AC or DC voltage.
The thickness of the drift layers 13 and 15 are determined as a function of the wavelength, and therefore of the required excitation energy of the corresponding active layers 12 and 14 For an electric field applied perpendicular to the plane of the active layers 12 to 14, an electron must gain sufficient energy from the applied electrical field to excite the luminescent centers to the correct energy, i.e. in the ballistic regime, the energy gained in the drift layers 13 and 15 (measured in eV) is equal to the electric field multiplied by the thickness of the drift layer 12 or 14. Accordingly, the drift layers 13 adjacent active layers 12 may have a different thickness than the drift layers 15 adjacent the active layers 14. For example, for an applied electrical field of 5 MV/cm, the thickness of the drift layer must be 3.8 nm or thicker to excite a luminescent center to 1.9 eV (1.9 eV/0.5 eV/nm=3.8 nm), 4.6 nm or thicker to excite a luminescent center to 2.3 eV, or 5.6 nm or thicker to excite a luminescent center to 2.8 eV. For layered light emitting film structures 100 powered by AC electrical power, in which neighboring active layers, e.g. 12 and 14, emit at different wavelengths, the intervening drift layer, e.g. 13, must be thick enough for excitation of the luminescent centers in the higher energy layer.
Chemical vapour deposition is a preferred for commercial production of such multi-layer light emitting structures 200 at reasonable cost. Methods of fabrication of the emitter layer structure 200 comprising multiple active layer/drift layer pairs 12/13, 14/15 according to the embodiments of the present invention comprising surface treatment or activation of deposited dielectric layers with the metal or semiconductor source gas, e.g. silane, overcome limitations mentioned above of known commercial thermal CVD processes and apparatus. Improved flatness of deposited dielectric layers 12 to 15 enables scalability of multilayer nano-layer dielectric structures to a larger number of layers, e.g. 24 layer pairs or more, without breakdown of the uniformity of layer structure and thickness.
As disclosed in copending provisional patent application No. 61/187,424 entitled “Cerium doped gate oxide layers”, filed Jun. 16, 2009, high quality, thin, rare earth doped dielectric layers, particularly oxide layers, may also have applications for gate oxides and tunnel oxides other electronic devices such as transistors, memory cells and flash devices. Preparation of such oxides by use of silane chemistry and silane surface treatment to provide nano-layers, which are substantially atomically flat, would also be beneficial for such structures.
Other device structures where a dielectric layer such as a layer of doped or undoped silicon dioxide, nitride, or oxynitride, or other dielectric provides a gate oxide, tunnel oxide, capacitor dielectric, may also benefit from chemical vapour deposition with surface activation treatment as disclosed above, to provide dielectric thin films or nano-layers with improved flatness and uniformity.
While the specific embodiments described in detail above are directed to light emitting devices, emitter structures and electroluminescent active layers comprising rare earth doped silicon dioxide or silicon nitride, it will be apparent that atomically flat, doped or undoped thin dielectric layers of high quality potentially have a number of other uses, e.g. as gate oxides and gate layer stacks for other electronic device structures, or for gate oxides and tunnel oxides for memory and flash devices. Thin dielectric layers fabricated by methods according to embodiments of the present invention may comprise luminescent species, such as rare earth ions or other light emitting species, that may also find application for light emitting diodes, optically excited light emitting devices, lasers, optical modulators and amplifiers, for example.
Improved chemical vapour deposition processes are provided for deposition of high quality nano-layers of dielectric materials, and particularly oxides and nitrides of silicon for light emitting device structures and other electronic devices. Low surface roughness and substantially atomically flat and abrupt layer structures provide for improved control of layer thicknesses and uniformity, particularly for multilayer stacks of thin film dielectrics. For light emitting device structures, improved quality and flatness of doped and undoped thin dielectric layers potentially offer improved performance, such as higher luminous efficacy, brightness, and reliability for solid-state lighting applications.
Although embodiments of the invention have been described and illustrated in detail, it is to be clearly understood that the same is by way of illustration and example only and not to be taken by way of limitation, the scope of the present invention being limited only by the appended claims.
This application is related to U.S. patent application Ser. No. 11/642,788 filed Dec. 12, 2006, entitled “Engineered structure for solid state light emitters” and U.S. patent application Ser. No. 12/508,033 filed Jul. 23, 2009, entitled “Engineered structure for high brightness solid-state light emitters”, which are incorporated herein by reference.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/CA10/00287 | 3/1/2010 | WO | 00 | 8/24/2012 |