The invention generally relates to sheet wafers and, more particularly, the invention relates to doping sheet wafers.
Crystalline sheet wafers can form the basis of a variety of electronic devices. For example, Evergreen Solar, Inc. of Marlborough, Mass. forms solar cells from sheet wafers, which Evergreen Solar designates STRING RIBBON™ wafers or crystals.
Continuous growth of silicon sheets eliminates the need for slicing bulk produced silicon to form wafers. For example, in one implementation, two filaments of high temperature material are introduced up through the bottom of a crucible, which includes a shallow layer of molten silicon, known as a “melt.” A seed connected to two filaments is lowered into the melt, and then pulled vertically upward from the melt.
A meniscus forms at the interface between the bottom end of the seed and the melt, and the molten silicon freezes into a solid sheet just above the melt. The filaments stabilizes the edges of the growing sheet. U.S. Pat. No. 7,507,291, among other documents, describes a method for growing multiple filament-stabilized crystalline sheets simultaneously in a single crucible. Each sheet grows in a growth region, which is referred to in the art as a “lane” in the multi-lane furnace. This multi-lane wafer fabrication process thus reduces the cost of fabricating wafers when compared to crystalline sheet fabrication in a single-lane furnace.
To convert light to electricity, the wafers must be doped. Doping in a multi-lane furnace, however, presents a number of problems. Among them is uneven and inconsistent doping concentrations across the different lanes.
In accordance with one embodiment of the invention, a method and apparatus for forming a sheet wafer adds material to a crucible having a feed area and a remaining area. Specifically, the material is added to the feed area-not the remaining area. The method and apparatus melt the material to form a first growth area and a second growth area, both of which are part of the remaining area. First and second sheet wafers also are drawn (at about the same time) from the first and second growth areas, respectively, and dopant is directly applied to the material at the remaining area. The dopant thus bypasses the feed area to dope at least a portion of the remaining area. In some embodiments, the dopant may diffuse to the feed area.
Various embodiments also apply dopant to material in the feed area of the crucible. Accordingly, this additional dopant bypasses the remaining area to dope the feed area. The method and apparatus may directly apply dopant to the second growth area and not to the first growth area. In this latter case, the directly applied dopant may diffuse from the second growth area and into the first growth area, which is between the feed area and the second growth area. Alternatively, the dopant may be applied to both the first and second growth areas.
Any of a variety of different techniques may directly apply the dopant to the remaining area. In a first implementation, the method and apparatus may directly contact a doped apparatus into the material in the remaining area. For example, the doped apparatus may include a filament that substantially disintegrates after contacting the material, thus releasing the dopant.
Another embodiment releases doped particles from an inkjet apparatus into one or more prespecified portions of the remaining area. Yet another embodiment coats one or more of the filaments of a sheet wafer with dopant. Still other embodiments pass a member (having a dopant) through the material in the remaining area.
Logic may control the amount of dopant applied to the material. To that end, the method and apparatus may measure a quality of at least one of the first and second sheet wafers. Thus, the dopant may be applied as a function of the measured quality. Among other things, the quality may include the resistivity of the at least one of the first and second sheet wafers. The method and apparatus then may change/apply the volume of dopant directly applied as a function of the resistivity.
The method and apparatus also applies to sheet wafer growth systems growing more than two sheet wafers. Moreover, the first and second wafers may be positioned in any of a number of manners. For example, they may be positioned in a side-by-side manner, or face each other.
In accordance with another embodiment, an apparatus for forming a plurality of sheet wafers has a crucible with a feed area and a remaining area, and material inlet for receiving material to be added to the feed area of the crucible. The apparatus also has a wafer puller for drawing a plurality of sheet wafers from the remaining area, and a doping apparatus operably coupled with the crucible. The doping apparatus is configured to directly add dopant to the remaining area, thus bypassing the feed area.
In accordance with other embodiments, a method and apparatus for forming a sheet wafer add material to a crucible having a feed area and a dump area, and melt the material to form a wafer growth area between the feed area and the dump area. The material is added to the feed area and removed through the dump area. The method and apparatus substantially simultaneously draw a plurality of sheet wafers from the growth area, and directly apply dopant to the melted material at the growth area. The dopant thus bypasses the feed area to dope at least a portion of the growth area.
Those skilled in the art should more fully appreciate advantages of various embodiments of the invention from the following “Description of Illustrative Embodiments,” discussed with reference to the drawings summarized immediately below.
In illustrative embodiments, a multi-wafer growth furnace precisely controls its doping processes to produce more consistently doped sheet wafers. Consequently, the wafers should have an optimized efficiency. To that end, rather than simply adding dopant at one remote location within the furnace, various embodiments apply some or all of the dopant directly to the lanes in which the wafers grow. This concept may be referred to as “doping by lane.” Details of a number of different embodiments are discussed below.
The sheet wafers 16 growing in
It should be noted that discussion of silicon sheet wafers 16 is illustrative. For example, the sheet wafers 16 may be formed from a material other than silicon, or a combination of silicon and some other material. As another example, illustrative embodiments may form doped non-crystalline sheet wafers 16. Further, while illustrative embodiments of the invention are described with respect to a furnace 10 with four sub-growth regions (“lanes”) with the sheets generally parallel to each other in a single line, other embodiments may employ more growth lanes or fewer growth lanes, and/or the disposition of the growth lanes with respect to each other may differ.
Any number of different devices can serve the functions of resistance detector 24 and controller 26. For example, one or more eddy current detectors could serve as resistance detectors 24, while a logic element, such as a microprocessor, digital signal processor, application specific integrated circuit, circuit module, or combination of components can serve as the controller 26. Each lane may have an individual resistance detector 24 and controller 26. Alternatively, the various lanes may share resistance detectors 24 and/or controllers 26. The feedback system 22 could be positioned a short distance above the furnace 10 (i.e., outside of the housing 12), or inside the housing 12 if the equipment is robust enough to withstand the high temperatures within the furnace.
The crucible 14 may be considered as having three separate but contiguous regions; namely,
The growth region 32 may be considered as forming four separate crystal sub-regions that each grows a single crystalline sheet 16. To that end, each crystal sub-region has a pair of filament holes 38 for respectively receiving two high temperature filaments that ultimately form the edge area of a growing silicon crystalline sheet wafer 16. Moreover, each sub-region also may be considered as being defined by a pair of optional flow control ridges 40. Accordingly, each sub-region has a pair of ridges 40 that forms its boundary, and a pair of filament holes 38 for receiving filament. These sub-regions may be referred to herein as “lanes.” As shown in the
The crucible 14 should be formed from a material that can withstand high temperatures (e.g., on the order of 1400-1500 degrees C.). To that end, the crucible 14 may be formed from graphite and resistively heated to a temperature capable of maintaining silicon above its melting point. To improve unidirectional liquid flow, the crucible 14 has a length that is much greater than its width. For example, the length of the crucible 14 may be three or more times greater than its width. Of course, in other instances, the crucible 14 is not elongated in this manner. For example, the crucible 14 may have a somewhat square or rectangular shape (e.g.,
As known by those skilled in the art and shown in
As shown in
The crucible 14 of this embodiment is configured to cause the molten silicon to flow at a very low rate from the introduction region 30 toward the removal region 34. If this flow rate were too high, the melt region underneath the growing ribbon would be subject to high mixing forces. It is this low flow that causes a portion of the impurities within the molten silicon, including those rejected by the growing crystal wafers 16, to flow from the growth region 32 toward the removal region 34.
Several factors contribute to the flow rate of the molten silicon toward the removal region 34. Each of these factors relates to adding or removing silicon to and from the crucible 14. Specifically, a first of these factors simply is the removal of silicon caused by the physical upward movement of the filaments through the melt. For example, removal of four sheets wafers 16 at a rate of one inch per minute, where each sheet wafer 16 has a width of about three inches and a thickness ranging between about 190 microns to about 300 microns, removes about three grams of molten silicon per minute in certain sized crucibles 14. A second of these factors affecting flow rate is the selective removal/dumping of molten silicon from the removal region 34.
Consequently, to maintain a substantially constant melt height, the system adds new silicon feedstock as a function of the desired melt height in the crucible 14. To that end, among other ways, the system may detect changes in the electrical resistance of the crucible 14, which is a function of the melt it contains. Accordingly, the system may add new silicon feedstock to the crucible 14, as necessary, based upon the resistance of the crucible 14 and melt level. For example, in some implementations, the melt height may be generally maintained by adding one generally spherical silicon slug having a diameter of about a few millimeters about every one second. See, for example, the following United States patents (the disclosures of which are incorporated herein, in their entireties, by reference) for additional information relating to the addition of silicon feedstock to the crucible 14 and maintenance of a melt height: U.S. Pat. Nos. 6,090,199, 6,200,383, and 6,217,649.
The flow rate of the molten silicon within the crucible 14 therefore is caused by this generally continuous/intermittent addition and removal of silicon to and from the crucible 14. It is anticipated that at appropriately low flow rates, the geometry and shape of various forms of the crucible 14 should cause the molten silicon to flow toward the removal region 34 by means of a generally one-directional flow. By having this generally one directional flow, the substantial majority of the molten silicon (substantially all molten silicon) flows directly toward the removal region 34.
In accordance with illustrative embodiments of the invention, the furnace 10 and accompanying components shown in
The number of printheads for each lane can vary depending upon the requirements and function of the furnace 10. For example, each lane could have two printheads if the melt is to be co-doped. In that case, each line could have one printhead with an n-type dopant (e.g., phosphorus) and another printhead with a p-type dopant (e.g., boron). Other co-doping embodiments may have a single printhead for each lane that has the opposite doping type to that of the melt. Alternatively, a single printhead could have a long length that extends beyond its lane. For example, this long printhead could extend across two to four lanes, and have an outlet orifice at each lane.
More complex furnaces 10 can have a single printhead that moves along a track that is generally parallel with the crucible 14. Such a printhead thus can move between the different lanes in a manner that is similar to the movement of a printhead within a conventional inkjet printer.
The inkjet printheads 42 preferably store the dopant as doped particles within the solvent, such as alcohol. For example, the dopant could comprise boron or phosphorus particles within an alcohol solution. When ejected from the outlet, the high temperature dissolves the alcohol solvent before the solvent reaches the surface of the melt. The particles, however, continue toward and into the surface of the melt, which absorbs the particles after contact.
When exposed to the high temperatures within the furnace 10, however, the solution may evaporate while within their printheads 42, consequently clogging the outlets of the printheads. This could cause catastrophic failure of the entire system. The printheads 42 thus may be spaced a sufficient distance above the crucible 14, where the temperatures are much lower. When spaced far from the crucible 14, however, the dopant may not precisely fall into the melt.
To mitigate this problem, illustrative embodiments may include steerable dopant particles within the solvent. Specifically, the particles could be loaded into the inkjet printheads 42 a the known polarization, i.e., they have a charge. The furnace 10 thus may have electronics and electrodes that generate a controllable electric field near the crucible 14 to steer the doped particles into the melt. The strength and extent of the electric field can be selected based upon a number of parameters, including the position of the printheads 42, the charge of the particles, and the anticipated convective currents within the furnace 10.
In addition to using charged particles, or instead of using charged particles, some embodiments form a heat insulating shield (not shown) in front of the printheads. Of course, the shield should have an opening for each printhead opening. In some embodiments, the shield opening opens only when ejecting dopant, thus further controlling the heat profile behind it. Alternatively, the shields could be integrated into the printheads.
Doping of each lane can be independently controlled on the fly, or through programming, depending upon its requirements. For example, if some of the lanes are co-doped, those lanes near the dump outlet 34 may receive more co-dopant than those near the inlet 18. In any case, as an example, known inkjet printheads 42 should be able to deliver droplets at greater than 1000 hertz in very repeatable, precise droplet sizes. Of course, inkjet printheads 42 can deliver droplets at different rates.
As noted above with respect to
For example, in a furnace 10 implementing co-doping with a primarily p-type doped melt, if the wafer 16 in the lane nearest the outlet port 36 has a resistance indicating too much p-type doping, then the controller 26 may signal its corresponding printhead 42 to deposit more n-type dopant into the melt in that lane. Alternatively, or in addition, the controller 26 may signal its corresponding printhead to deposit less p-type dopant into the melt. In a similar manner, for a furnace 10 that does not co-dope the melt, the controller 26 simply may signal its corresponding printhead 42 to deposit less p-type dopant into the melt.
Rather than using printheads 42, or in addition to using printheads 42, some embodiments use the filaments forming the outside edges of the wafer 16 to dope the melt directly in some or all of the lanes. To that end, the outside surface of some or all of the filaments may be coated with prescribed dopants.
Those skilled in the art can use any combination of the coated filaments to dope the melt. For example, certain lanes may have filaments doped with one type of dopant (e.g., boron), while other lanes may have filaments doped with the other kind of dopant (e.g., phosphorous). In fact, some lanes could use filaments with opposite doping characteristics, i.e., one filament could be doped with an n-type dopant, while the other filament could be doped with a p-type dopant. Moreover, different filaments can have different dopant concentrations to further fine tune the dopant level of the melt in each lane. For example, in a p-type doped melt, in a given lane, the filament nearer the introduction region 30 could have a p-type doping that is greater than that of the filament farther downstream.
Another embodiment passes a filament with dopant on its outside surface directly through prescribe locations of the melt.
Unlike the filaments used to form the edges of the sheet wafers 16, the controller 26 or other control apparatus can pass these filaments through the melt at varying rates. For example, if the sheet wafer resistance is within the prescribed range for a melt having these filaments as their only source of dopant, then the controller 26 may pass these filaments at prescribed rates. These filaments can pass through the melt at an increased or decreased rates, however, if the resistance is outside of the prescribed range.
Other embodiments, such as that shown in
Among other things, the doped apparatus 44 may include a doped/coated filament, wire, plug, highly doped piece of silicon, or other apparatus that can be retracted or extended into the melt. Moreover, the doped apparatus 44 of this embodiment preferably partially or completely dissolves after contact with the melt, although some embodiments do not dissolve. After contact with the melt, some embodiments may remove or retract the doped apparatus 44 from the crucible 14 after substantially all of its dopant diffuses into the melt, or if no further doping is necessary. If the doped apparatus 44 still has dopant, it can be reintroduced into the melt at a later time.
Accordingly, in a manner similar to the embodiments discussed above with respect to
The process begins at step 700, which adds material to the crucible 14. As noted above, silicon or other material may be added to the crucible 14 in a prescribed manner through its introduction region 30. The silicon may be doped or undoped, depending on the doping techniques used downstream. The high temperatures of the crucible 14 and internal environment melt the material into a liquid/molten form.
Next, step 702 draws the four sheet wafers 16 from the melt at substantially the same time. To that end, several pairs of filaments are passed through the crucible 14, which contains the molten silicon. In illustrative embodiments, the filaments are spaced more than about 145 millimeters apart. For example, the filaments may be spaced about 155 or about 156 millimeters apart. Alternative embodiments can space the filaments closer together or farther apart. In any event, drawing the filaments from the melt in this manner causes the filament sheet wafers 16 to grow out of the housing 12, as shown in
At the same time, or at a later time, the process directly applies dopant to one or more lanes of the growth region 32 (step 704). Specifically, illustrative embodiments do not begin drawing wafers 16 as required by step 702 until the melt is appropriately doped. Accordingly, in embodiments using pellets coated with dopant, the pullers of the puller system 20 can begin drawing the wafers 16 as soon as the melt reaches an appropriate volume within the crucible 14. The embodiments using pellets not coated with dopant, however, should not begin drawing the wafers 16 from the melt until it is appropriately doped.
Accordingly, various embodiments directly dope specific lanes, thus bypassing the introduction region 30 (i.e., this dopant is not directly added to the introduction region 30—other dopant can be added to the introduction region 30, but this dopant bypasses that region 30 as it is added to the specific lanes of the crucible 14). For example, the embodiment of
Of course, illustrative embodiments ensure that the dopant levels remain within tight constraints, which produces the most efficient sheet wafers 16. To that end, the resistance detectors 24 then determine if the resistivity of each of the wafers 16 is within the prescribed limits noted above (step 706). If not at the appropriate levels, then step 708 adjusts the doping levels accordingly.
For example, if the melt at the lane in question has too much P-doping, then the controller 26 may stop applying p-dopant into that lane. Moreover, as known by those in the art, the dopant in the melt diffuses to other lanes (or even into the introduction region 30), which then impacts the doping level in other lanes. Accordingly, if the dopant level at a given lane is too high, then the process may reduce the dopant in an upstream lane. Those skilled in the art should calibrate the system to compensate for the impact not only on the lane in question, but the impact on other lanes of the crucible 14.
This process thus enables a number of permutations to the process of doping the melt. Among others, the process can dope the melt with coated silicon pellets added to the introduction region 30 (i.e., this dopant material bypassing the growth region 32) and with dopant applied directly to one or more of the lanes in the growth region 32 (this other dopant material bypassing the introduction region 30). The process also can dope the melt by doping specific lanes in the growth region 32 only.
As noted above, various embodiments apply to other furnace configurations. For example,
Illustrative embodiments therefore permit fine tuned doping levels, thus producing better quality sheet wafers 16. Moreover, various embodiments facilitate co-doping. In either case, such embodiments should produce fewer rejected wafers 16, thus improving wafer yields—and reducing costs.
Although the above discussion discloses various exemplary embodiments of the invention, it should be apparent that those skilled in the art can make various modifications that will achieve some of the advantages of the invention without departing from the true scope of the invention.
This application is related to co-pending U.S. patent application Ser. No. 12/952,288, filed Nov. 23, 2010, attorney docket number 3253/192, naming Brian D Kernan, Gary J Tarnowski, Weidong Huang, Scott Reitsma, and Christine Richardson as inventors, and entitled, “Method of Reducing the Range in Resistivities in Semiconductor Crystalline Sheets Grown in a Multi-lane Furnace,” the disclosure of which is incorporated herein, in its entirety, by reference.