The present disclosure relates in general to the fields of photovoltaics, microelectronics, and optoelectronics. And more particularly, methods, architectures, and apparatus relating to uniformly forming a porous semiconductor layer or multilayer on a substrate are disclosed.
Currently, crystalline silicon (including multi- and mono-crystalline silicon) is the most dominant absorber material for commercial photovoltaic applications, with (mono and multi) crystalline silicon modules accounting for over 80% of the photovoltaic market today. The relatively high efficiencies associated with mass-produced crystalline silicon solar cells in conjunction with the abundance of material garner appeal for continued use and advancement. But, the relatively high cost of crystalline silicon material itself (due to its dependency on polysilicon feedstock, silicon ingot growth, or cast brick formation and wafering) limits the widespread use of these solar modules. At present, the cost of “wafering”, or crystallizing silicon and cutting a wafer, accounts for about 40% to 60% of the finished solar module manufacturing cost.
As an alternative to “wafering”, methods of growing monocrystalline semiconductors, such as silicon, and releasing or transferring the grown wafer have been proposed. Yet regardless of the formation methods, a low cost epitaxial semiconductor, such as silicon, deposition process accompanied by a high-volume, production-worthy, uniform and reliable low cost method of forming a release layer or release layers are prerequisites for wider use of solar cells manufactured by semiconductor deposition and release processing.
Porous silicon (PS) formation is a fairly new field with an expanding application landscape. Porous silicon is often created by the electrochemical etching of silicon wafers with appropriate doping in an electrolyte bath. The electrolyte for porous silicon is: HF (49% in H2O typically), isopropyl alcohol (IPA) (and/or acetic acid) or other alcohols, such as ethanol, or combinations thereof, and deionized water (DI H2O). IPA (and/or acetic acid) serves as a surfactant and assists in the uniform creation of PS. Additional additives such as certain salts or acids may be used to enhance the electrical conductivity of the electrolyte, thus reducing its heating and power consumption through ohmic losses.
Porous silicon has been utilized as a sacrificial layer in MEMS and related applications, where there is a much higher tolerance for cost per unit area of the wafer and resulting product than solar PV. Typically, porous silicon is produced using simpler and smaller single-wafer electrochemical process chambers with relatively low throughputs on smaller wafer footprints—a costly and inefficient process. The viability of this technology in solar PV applications hinges on the ability to industrialize the process to large scale (at much lower cost), requiring development of very low cost-of-ownership, high-productivity porous silicon manufacturing equipment. Designing porous silicon equipment and formation methods that allow for a high throughput, cost effective porous silicon manufacturing remains a challenge.
Therefore, a need has arisen for fabrication methods and systems relating to the controlled and uniform formation of porous semiconductor material on a wafer. In accordance with the disclosed subject matter, methods, structures, and apparatus for the high-productivity controlled fabrication of uniform porous semiconductor layers are provided. The present disclosure includes several embodiments for the batch processing of semiconductor (silicon in some embodiments) wafers to produce layers of porous semiconductor. Solutions for minimizing and limiting the effect of byproduct gas formed during anodization, minimizing current leakage, and optimal wafer seals and clamps are provided. These innovations substantially reduce or eliminate disadvantages and problems associated with previously developed porous semiconductor formation methods and systems including cost reductions.
These and other advantages of the disclosed subject matter, as well as additional novel features, will be apparent from the description provided herein. The intent of this summary is not to be a comprehensive description of the subject matter, but rather to provide a short overview of some of the subject matter's functionality. Other systems, methods, features and advantages here provided will become apparent to one with skill in the art upon examination of the following FIGURES and detailed description. It is intended that all such additional systems, methods, features and advantages included within this description be within the scope of the claims.
The features, nature, and advantages of the disclosed subject matter may become more apparent from the detailed description set forth below when taken in conjunction with the drawings in which like reference numerals indicate like features and wherein:
The following description is not to be taken in a limiting sense, but is made for the purpose of describing the general principles of the present disclosure. The scope of the present disclosure should be determined with reference to the claims. Exemplary embodiments of the present disclosure are illustrated in the drawings, like numbers being used to refer to like and corresponding parts of the various drawings.
And although the present disclosure is described with reference to specific embodiments, such as silicon and other fabrication materials as applied to the field of photovoltaics, one skilled in the art could apply the principles discussed herein to other materials, technical areas, and/or embodiments without undue experimentation.
A novel aspect in the porous silicon system designs and processing methods of this disclosure lies in the batch parallel or multi-wafer processing architecture (batch stack architecture), similar to low-cost large batch wet chemical processing in benches or tanks. Presently available porous silicon tools rely on single wafer processing which characteristically burdens each wafer with high capital cost, serial cumulative processing times, relatively high electrical power consumption per wafer, and excessive wafer handling/sealing resulting in potential yield losses. The novel designs of this disclosure may reduce the capital cost by a factor approximately equal to or even greater than the number of wafers in each batch stack or array. Furthermore, the proposed design simplifies and reduces the capital cost of automation, reduces the tool footprint and enables downstream rinsing and drying.
Details about the individual components of the batch reactor are explained below.
The embodiment includes multiple architecture of the electrode assembly. The simple version is a solid electrode plate or a film etc. The inert electrode, such as diamond, graphite, platinum, or any other suitable material, does not corrode or etch during the electrochemical reaction. The second embodiment of the electrode assembly is a compartmentalized electrode chamber as shown in
The process chamber holds the wafers and the electrolyte. The embodiment covers a wide range of process chamber dimensions to be able to create porous silicon on wafers of various geometries such as, but not limited to round, square, pseudo square(square with truncated corners) with rounder corners of varying degrees, as well as rectangular structures. Schematics of a 200 mm round and 165 mm square process chambers are shown in
The substrates involved may be essentially flat with varying degree of roughness or may be structured to form 3-dimensional patterns or structured with films that locally inhibit or enable porous silicon formation.
The process chambers are envisioned to be able to open in multiple sections, like a clam shell as shown in
A key requirement of the porous silicon process is to get substantially uniform porous silicon coverage on the full surface of the wafer, in some embodiments without any edge exclusions. This requires that no areas of wafer edge should be blocked or covered by any material that will prevent uniform electric field distribution and direct contact with the chemistry. One embodiment covers designs of mechanical features that can hold the wafer in place, but with zero to negligible contact points and blocking points on the wafer. As shown in
Another critical item is the choice of sealing material around the inner walls of the process chamber. The chamber walls will be lined with either a single layer of chemically inert (HF and organic resistant) insulating rubber or foam to provide a leak-free seal between the wafer edge and the chamber wall or the wafer holders. This is critical to prevent any chemical leak or electric field leakage in areas where the clam shell chamber walls lock.
The batch chamber design with the compartmentalized electrode chamber allows for electric modulation as well. The parameters such as electrode dimension, gap between electrode and closest wafer, gap from wafer to wafer, etc. may easily be modified to achieve the required uniformity for the electric field. Another key component is the spacers used to hold the membrane discussed above. The shape and patterns on the insulating spacer may also be modified to achieve the best electric field uniformity on the wafer. In circumstances where a varying electric field (thereby varying thickness or porosity of porous silicon) is required for the integrated process flow, the spacer design can be used to control the required electric field without changing the chamber design.
The chamber may be designed with fluid fill and vent ports 208 on the top of the chamber as shown in
One challenge with any porous silicon chamber is handling the hydrogen (H2) gas generated as a result of the anodic etch reaction. Hydrogen evolves from the surface of the wafer and each electrode. Since the bath is integral with electrical current transmission, H2 gas blocks current flow and supply of chemicals to the reaction surface, thus affecting porous silicon formation and continuity/uniformity. It is therefore critical to effectively and rapidly purge or sweep H2 byproducts from the surfaces of the wafer and electrodes. The wafer gap, fluid flow and design of the flow ports determine the effectiveness of the sweep. While sweeping H2 is fairly simple in terms of fluid mechanics, some consideration is warranted to mitigate the current loss from the fluid ports. Since the fluid lines are connected from wafer to wafer, depending on the geometry of the ports, line size and length, current can leak or bypass each wafer. Therefore, isolation of each port is advantageous. Also, for example, reducing the line diameter and increasing the length results in greater electrical resistance which reduces current losses or bypass losses. The current field lines are also influenced by the geometry adjacent to the wafer. So, large flow ports are less desirable compared to multiple small ports.
Typical wet chemical baths and process chambers use direct fluid fill/drain of the process chamber, wherein the chemical is directly pumped in the process chamber. This may require additional fill and drain times before the process can start and results in loss of productivity. This embodiment also covers a design termed as “bath in bath” for the PS production as shown in
There are at least two embodiments of this bath in bath design: a) Prefilled inner chamber that is immersed and lifted out completely into and from the bath; b) Resident bath-in-bath with wafers being handled using auto loader that handles a batch of wafers and that places the batch into the lower holder part of the inside bath, then retreats.
In design a), the process chamber is pre-loaded with wafers and filled with the process chemicals. The entire assembly is then immersed into a larger bath which is pre-filled with the process chemical/electrolyte. The ports/vents on the top of the chamber allow for the electrolyte to fill the process chamber if and when the liquid level drops in the process chamber due to the reaction or other means of loss such as evaporation. Once the process is complete, the process chamber unlocks and is pulled out and the standby process chamber is immediately immersed in the larger bath minimizing loss in productivity due to wafer load/unload and chamber fill and drain. The larger bath is designed with its own pumping and recirculation system to maintain the required concentration and temperature. This methodology allows having multiple process chambers that can be introduced into the main bath without any loss in productivity.
In design b), the chamber is an integral part of the tool or the larger bath and always remains immersed in the main bath, but the chamber can open and close. It is envisioned that loading mechanisms such as robotic handlers can transfer a batch of n wafers into the base of the process chamber. After the wafer handlers have moved away from the process chamber, the outer walls of the process chamber close. This action not only secures the wafers, but also encloses the process chemicals in to the process chamber. The additional vents and ports allow the process chamber to be filled completely to the required level and maintain the same level throughout the process.
In any case, the top of the vent ports may be outside of the liquid, such that an electrically connecting path outside of the inner bath is avoided. This embodiment is shown in
The embodiments of design a) and design b) can be combined into a hybrid utilizing the loading mechanism from design a) and the sealing mechanism from design b). In this hybrid design, the bottom section of the chamber remains in the outer bath. The wafers are pre-loaded into the top (and side) portion of the chamber, which acts both as a handling mechanism and a partial chamber. The preloaded wafers are then immersed in the outer bath until the wafers make contact with the lower portion of the chamber. The chamber walls are then closed tight with an actuator mechanism ensuring a leak-proof chamber.
The batch porous silicon equipment design embodiments described above can be used to form either single-layer or multi-layer porous silicon on one or both sides of the wafers in the batch. Porous silicon can be formed on only one side of the wafers by applying the electrical current flowing in only one direction without a change in the current polarity. On the other hand, porous silicon can be formed on both sides of the wafers by alternating the current flow direction at least once or multiple times. The electrical current density (in conjunction with the HF concentration) controls the layer porosity. Thus, the layer porosity can be increased by increasing the electrical current density and conversely can be reduced by reducing the electrical current density. Multi-layer porous silicon can be formed by modulating or changing the electrical current level in time during the porous silicon formation process. For instance, starting the porous silicon process with a lower current density followed by a higher current density results in formation of a lower porosity layer on top of a higher porosity buried layer. A graded porosity porous silicon layer may be formed by, for instance, linearly modulating or varying the electrical current density in time. One can use this approach to form any porous silicon structure with one to many porous silicon layers with one to many porosity values.
A key factor in the uniform anodization of a wafer surface in a bath reactor is the suppression and minimization of the quantity, density, and impact of gas bubbles formed during the anodization process. As a semiconductor gets anodized in an etching fluid (consisting of HF and typically an additive, such as an alcohol, to reduce the surface tension), a byproduct of the reaction is the liberation of gas bubbles, substantially hydrogen gas bubbles. In a substantially vertical reactor arrangement where the wafers are immersed and held vertically or at least to a substantial angle away from the horizontal direction within the anodization bath, such as that shown in
All or some of the disclosed bubble mitigation systems and methods herein may be combined for an optimized anodization result. One disclosed solution employs the use of sonic energy, such as ultrasonic or megasonic transducers coupled to the fluid bath or to the wafer holder, to effectively dislodge and liberate such bubbles from the surface.
Another disclosed solution utilizes pulsed anodization—the current is pulsed on and off as shown in the graph in FIG. 8—where the formation of the anodization is pulsed to give the hydrogen bubbles that are formed during the reaction sufficient time to travel up and away through a vent hole, or a plurality of vent holes, at the top or along the upper edges of the chamber or wafer tunnel. The bubbles may also be swept away by an upwards chemical flow. It is to be noted that the formation of a high porosity layer is typically accomplished at a substantially higher electrical current than the formation of a low porosity layer. Also, the anodization speed, in terms of anodizing a certain thickness of film, increases with the current (i.e., higher porosity films form faster than lower porosity films). Therefore, high porosity films typically take less time to form but generate more hydrogen per unit time, making the formation of higher porosity films susceptible to hydrogen bubble-induced uniformity problems. Thus, if a plurality of layers with at least one lower porosity layer is formed, the throughput loss from forming the high porosity layer in a pulsed fashion is comparatively minor.
A third disclosed embodiment utilizes fluid transport to dislodge gas bubbles while the anodization is in an off-state. In the current-off state during anodization pulsing, an active fluid transport which sweeps the hydrogen gas away may be turned on—the fluid transport is left turned off during the anodization. This method may be beneficial to the uniformity of the electrical field, as a moving electrolyte tends to carry electrical field lines with it and can cause field distortions.
Alternatively, suitably directed fluid flow during the anodization may also be utilized to create a uniform overall anodization result. To enable fluid flow during the anodization, proper care needs to be taken to avoid contact of fluid volumes from the different fluid compartments between wafers. An exemplary system that may be used to push liquid through the reactor are separate tubes connected to the bottom of each compartment. Through these tubes, the liquid is pushed across the wafers thus displacing other liquid and bubbles through the vent holes on the upper part of the chamber. One skilled in the art may envision various methods and systems for keeping the replenishing fluids as well as the displaced fluids between individual compartments separated.
Yet another alternative method for bubble removal utilizes a suitable vacuum, which may be for instance applied to the vent holes of the anodization chamber.
In order to sweep bubbles away from active surfaces that are to be anodized, it may also be helpful to have a small fluid volume above the wafers, where bubbles will drift to due to an effect of buoyancy.
The electrical power dissipation per wafer in the batch porous silicon tool may be reduced by adding a suitable additive such as a salt or an acid to the anodization bath in order to enhance its electrical conductivity without any detrimental impact on the anodization chemistry and process. An increase in the electrical conductivity of the batch porous silicon bath through a suitable conductivity-enhancement additive such as a chemically-benign salt or acid not only reduces the electrical power dissipation per wafer but also enables an increase in the wafer batch size by reducing the wafer-to-wafer spacing within the bath. The reduction in wafer-to-wafer spacing may be achieved because it is possible within a more conductive electrolyte to equalize the electric field strength across a smaller distance—thus enabling smaller wafer-to-wafer compartments and allowing for an increase in wafer batch size.
For a uniform anodization result, it is important to have a substantially uniform electrical field across each wafer. In a stacked batch array (horizontal, vertical or angled), special importance is given to the wafer closest to the cathode, shown as wafer 314 in
There are however reasons it may be beneficial to have the diameter of the compartment size be slightly larger than the wafer diameter. One consideration is that at least above the wafer, a compartment size 330 larger than the size of the wafer, while connected to the vent hole or holes, allows for bubbles to be temporarily stored during the anodization without being in the direct path of the anodization current. Another consideration is that in the case that the sealing at the wafer edge is asymmetric, as is described below, the additional fluid space between wafers allows the electric field to equalize towards the wafer edge.
When the electrodes have similar size as the wafers to be anodized, such as the configuration shown in
Another important consideration is that all the current between the electrodes, anode 306 and cathode 308, to flow through the stack of wafers to be anodized—thus there should be no parallel connecting fluid path between the electrodes which would divert current from the wafer stack. Such a parallel fluid path would be detrimental to overall power loss, as well as controllability and matching of performance between different baths. Seals 332 in
For the anodization process, it is often necessary to use electrodes that are chemically very inert. Such electrodes tend to be costly which increases substantially with the electrodes size. Therefore, to decrease the overall tool cost it is advantageous to have small electrodes and utilize suitable field shaping to optimally expand the electric field and provide a uniform electric field for the anodization.
To expand the electric field the electrode may be placed at a larger distance from the first and/or last wafer in order to make use of the electrolyte's conductivity to distribute the field evenly—thus increasing/adjusting the distances 320 and 324 in
In
In
Further, as shown in
As shown in the batch anodization arrangements in
However, due to the finite extension of the seal material, which is typically a flexible material, wafer surface areas close to the bevel apex, such as area 404 in
To optimize the performance of the seal, the seal must both eliminate fluid leakage around the edge of the wafer and at the same time minimally affect the anodization of the wafer. Thus, the wraparound of the sealing material around the bevel needs to be minimized, as said wraparound prevents areas contacted by the flexible seal from being anodized, while also performing a fluid tight seal.
A disclosed solution to the optimization problem stated above and depicted in
Another solution allowing for anodization closer to the apex of the bevel is to use a sealing method which keeps the apex of the bevel, or at least the region on the bevel close to the apex, substantially exposed and ready to be anodized.
The ring seal inner diameter is to be minimized, in order to allow the field behind each wafer to reshape into a uniform density prior to reaching the next wafer. Suitable field shaping is required here, both from the seal, as well as from the walls limiting the compartmentalized fluid between wafers.
Another aspect of ring seal 414 positioned at the backside edge of the wafer is to have a sufficiently large extension to accommodate for a change or variation of wafer diameter or thickness as a function of the re-use of the wafer. The ring seal will typically have a shape substantially similar to the wafer itself, such as circular for a round wafer, square for a square wafer and so on. In another embodiment of this asymmetric seal, the small slanted pins may be replaced by a continuous and suitably slanted wedge.
Asymmetric wafer seal 440 is similar to asymmetric wafer seal 410 in
And in a further embodiment, the sponge-like material can extend past the edge area and across the whole wafer or large parts of the wafer. Building on this, in another embodiment the fluid filled compartments between the wafers in a stacked batch anodizing tool, such as that of
All described arrangements and sealing systems and methods may be optimized by suitable choices of material and geometry that accommodate variations in wafer diameter, thickness, warpage, bevel form, and other shape variations in such a way that reliable sealing of the individual fluid compartments on each side of each wafer is achieved.
In all described embodiments, the wafer holders are suitably segmented to allow for wafer loading and unloading which may be accomplished, for example, by segmenting the whole batch into a clamshell-like design with two or more segments for load and unload or by stacking individual wafer holders, similar to arrangements depicted in
In the field of photovoltaics, this disclosure enables low cost, high-throughput fabrication of thin film (or thin crystalline semiconductor foil) substrates to be used for solar cell manufacturing by means of a preferably reusable template which can be used repeatedly to fabricate and release said thin film (or thin foil) crystalline semiconductor substrates. The application fields of this disclosure not only include solar photovoltaics, but also other semiconductor areas including microelectromechanical systems (MEMS) and optoelectronics. The field of the disclosure covers several apparatuses and methods for generating uniform layers or multilayers of porous semiconductor with controlled porosity profile across the porous layer (or multilayer) which then may be used as sacrificial release layers for removing a thin film semiconductor substrate deposited on top of a template with the release layer(s). Other applications of the porous semiconductor layers produced by the methods and apparatus of this invention include non-sacrificial applications such as formation of anti-reflection coatings, optoelectronics, and layers for chemical sensors, etc.
The foregoing description of the exemplary embodiments is provided to enable any person skilled in the art to make or use the claimed subject matter. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without the use of the innovative faculty. Thus, the claimed subject matter is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
It is intended that all such additional systems, methods, features, and advantages that are included within this description be within the scope of the claims.
This application claims is a continuation of U.S. patent application Ser. No. 13/288,721 filed Nov. 3, 2011 which claims priority to U.S. Provisional Pat. App. 61/409,940 filed Nov. 3, 2010, all of which are hereby incorporated by reference in their entirety. This application is also a continuation-in-part of U.S. patent application Ser. No. 12/774,667 filed May 5, 2010 which claims priority to U.S. Provisional Pat. App. 61/175,535 filed May 5, 2009, and is also a continuation-in-part of U.S. patent application Ser. No. 13/244,466 filed Sep. 24, 2010 which claims priority to U.S. Provisional Pat. App. 61/386,318 filed Sep. 24, 2009, U.S. patent application Ser. No. 12/774,667 filed May 5, 2010, and U.S. patent application Ser. No. 12/688,495 filed Jan. 15, 2010 which claims priority to U.S. Provisional Pat. App. 61/145,018 filed Jan. 15, 2009, all of which are hereby incorporated by reference in their entirety.
Number | Date | Country | |
---|---|---|---|
61409940 | Nov 2010 | US | |
61175535 | May 2009 | US | |
61386318 | Sep 2010 | US | |
61175535 | May 2009 | US | |
61145018 | Jan 2009 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 13288721 | Nov 2011 | US |
Child | 14563888 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 12774667 | May 2010 | US |
Child | 13288721 | US | |
Parent | 13244466 | Sep 2011 | US |
Child | 13288721 | US | |
Parent | 12688495 | Jan 2010 | US |
Child | 13244466 | US | |
Parent | 12774667 | May 2010 | US |
Child | 12688495 | US |