MXPA96003156A - Structure and manufacturing process for a self-aligned silicon cell doll, with alumi alloy joint - Google Patents

Abstract

The present invention relates to a solar cell with back contact, comprising: a semiconductor volumetric layer of a first type of conductivity, having a front surface and a back surface, a plurality of semiconductor regions, contaminated with impurities, spaced apart, type of opposite conductivity formed in said volumetric layer near said back surface, and forming a plurality of semiconductor junctions therewith, said front surface being entirely of said first type of conductivity, a first set of spaced-apart contacts connected to said plurality of regions semiconductor, contaminated with impurities, spaced, and located along said back surface, said first set of spaced-apart electrical contacts comprises an alloy of said semiconductor material of the volumetric layer, and a Group III metal comprising the acceptor contaminant for said plurality of semiconductor regions, contaminated with impurities, spaced apart, a second set of ohmic contacts connected to said back surface of said volumetric layer in the spaces between said first set of ohmic contacts, and insulating means for electrically isolating said first set of spaced-apart contacts from said second set of ohmic contacts

Description

STRUCTURE AND MANUFACTURING PROCESS FOR A CELL SOLAR SILICON. OF CONTACT AFTER AUTO-ALIGNED. WITH ALUMINUM ALLOY JOINT FIELD OF THE INVENTION The present invention relates to an improved design and manufacturing method for a subsequent contact solar cell. BACKGROUND OF THE INVENTION The progress in the development of photo-voltaic cells depends on a variety of factors, not all of which are new designs, new materials and new manufacturing techniques. Historically, many of the efforts have been directed directly towards the attempt to increase the efficiency of solar conversion. The progress has been drastic. For AM1 illumination (sunlight through the thickness of a terrestrial atmosphere), a selenium solar cell, in 1914, had an efficiency of 1%, in 1954 an efficiency of 6% was achieved for a single-crystal cell silicon, in the mid-1980s, efficiencies between 22 and 25% in solar cells were presented. With concentration cells, where lenses or mirrors are used to increase sunlight to a considerably higher intensity than normal, efficiencies of 27.5% have been provided, which compare favorably with the thermal efficiency of 38 to 40% in a traditional energy plant from fossil fuels and the efficiency of 32 to 34% of a power plant of a light water nuclear reactor. However, in order to economize a solar cell in large-scale applications, such as for the supply of electric power to residences, other considerations, in addition to high efficiency, are prominent. One factor is the cost of manufacturing a cell. While the majority of independent homes have a sufficient roof area for conventionally designed solar cells, for the supply of 8500 W-hours of electricity annually, which are sufficient for an average household, an obstacle in commercialization is not efficiency , but the decrease in costs per unit area of a solar cell. A promising candidate for this task are the silicon solar cells, especially those cells made of thin silicon substrates ("100 μm), where high quality silicon is effectively used.The current challenge is to reduce the unit costs for these solar cells. , so that they are competitive with traditional fossil fuel energy supplies at the current prices of this energy.

One way to do this is through improved manufacturing techniques. In addition to manufacturing techniques, certain design structures offer advantages over other designs. A superior design is directed to the rear contact solar cells, which employ, in particular, thin silicon substrates. Silicon solar cells from similar splices have a p-n junction to separate the photon-generated electrons from the photo-generated holes. For the solar cell to work properly, the electrons must be directed towards contact by the n-type material and the holes must be directed towards the contact by the p-type material. The intensity of the light in a semiconductor decreases monotonically with the depth, thus, the p-n junction is preferably close to the illuminated surface, to reduce the recombination of the holes and electrons, before being separated by the p-n junction. In thin silicon solar cells, although the thickness of a cell is smaller than in conventional silicon solar cells ("300 μm), and the probability that a photon will become an electron-hole or load-carrier pair , is smaller, the average life time of a couple of photo-generated electron-holes can be such that this pair will survive the sweep of their respective contacts, ie, in a thin silicon solar cell, the diffusion length of the Minority carrier can be relatively large compared to cell thickness, so the performance of the cell is not unduly compromised.In the present invention, the diffusion length of the minority carrier is equal to the thickness of the cell or greater .

In addition, conventional silicon solar cells (front contact) have a structure in which a large p-n junction is formed over the entire substrate on the illuminated side of the cell. This conventional design has the virtue of simplicity, in which no pattern is required for the emitter (typically the p-type layer in a splice cell p-n), since it covers the entire front surface. However, the simultaneous and conflicting requirements are imposed on the front surface and the emitter layer in this type of arrangement. On the one hand, the diffuser of the emitter must be shallow and have a low concentration of impurities (<1 x 1019 cm.-3) / in order to reduce recombination, which occurs with high concentrations of impurities. On the other hand, such a shallow, slightly contaminated emitter will have a high sheet resistance (current flowing laterally through the top layer of a conventional cell, and between any contact grid line and this sheet resistance is inversely proportional to the thickness of the contaminated layer), generally greater than 100 ohms / square, which will require that the grid contact lines be narrowly spaced to avoid excessive ohmic power losses. The closely spaced contact lines in a conventional front contact cell mean reduced power of the cell, due to the shading of the underlying silicon by the contact material. In addition, if the concentration of impurities is low, the interface of the contact-impurities layer will be rectified (as a Schottky diode) rather than ohmic, with a corresponding loss of energy, associated with the ignition voltage of the diodes. However, the higher the concentration of impurities, the greater the recombination of electrons and holes in the emitter layer, which is detrimental and typically occurs at the point closest to the surface, where light rays enter. Finally, the texturization of the frontal surface to increase the contact lines of the elements that trap the light, has to go on a rough surface without loss of continuity, which can be difficult to achieve. In addition, some texturing methods, such as the porous silicon method, will make it more difficult to create a diffusion layer of the emitter of acceptable uniformity.

For this reason and others, for a conventional cell structure, a balance should be sought between the suitability of a highly contaminated surface, to promote the formation of ohmic contact and reduced shading, and the convenience of a slightly contaminated surface, for Reduced recombination of the carrier and effective surface passivation. Restrictions due to texturing and shading are also a problem. An alternative approach is to place the p-n junction on the back (the unlit side) of the cell. In such a posterior contact solar cell, the requirements for the texturization and passivation of the front surface are separated from the requirements for the formation of the p-n junction and for the contact of the emitter and the base. This means that the p-n junction can be deep and the emitter can be highly contaminated without extreme consequences. The shading of the illuminated surface is no longer emitted, since there are no contacts on the front surface and the spacing of the metal contact lines is not a problem either. Since this type of cell generally employs interlocking contacts, almost half of the back surface area is covered with a positive contact metal and the other half is covered with negative contact metal. Because the pn junction is in the back of the cell, however, the diffusion length of the minority carrier in the starting material (base) must exceed the thickness of the cell, in order to obtain satisfactory efficiency of energy conversion. The best results from this point of view are from a group from Stanford University, which has reported efficiencies of 21.3% in a sun illumination (100 mW / cm2) on a silicon cell, after contact, area of flotation, with thickness of 180 μm and 35 cm2 of area; and 22% for AM1 illumination from the sun at 242C (A. Sinton et al., Large-Area 21% Efficient Si Solar Cells, Conf. Record 23rd IEEE Photovoltaic Specialists Conference, p.157 (1933); RA Sinton et al. al., IEEE Electron Device Lett., EDL-7, No. 7, p.567 (1986), both incorporated herein by reference.

A solar cell of Si, of later contact, such as the design of Sinton et al. , requires a relatively complicated and expensive manufacturing, generally associated with the manufacture of integrated circuits. These processes include separate p-type and n-type diffusions (each requiring a masking), the alignment of the negative contact metal with respect to the positive contact metal, the use of photolithography, and the deposit of a metal Multilayer contact by evaporation or electronic, which requires a vacuum system. Thus, although a subsequent contact structure has significant advantages over a conventional frontal contact structure, its realization can be expensive.

BRIEF DESCRIPTION OF THE INVENTION The present invention reduces the manufacturing cost of silicon solar cells, while preserving a relatively high efficiency of solar conversion, by the use of a silicon solar cell using a single material - preferably aluminum ( Al) - as both the p-type contaminated material and the ohmic contact material, in a n-type silicon (Si) volumetric layer. In addition, a novel fabrication is used for the back contact grid lines, which in the preferred embodiment uses a relatively inexpensive self-aligning, printed screen contact system. A novel feature of this contact system is that it is self-aligned by applying anodic oxidation to a set of contacts, to isolate this set from the other set of contacts, thus eliminating any need for the precise alignment of the successive masked assemblies to achieve grid line pattern.

In a preferred embodiment, several other beneficial features are designed in the posterior contact silicon solar cell, according to the present invention, which include, but are not limited to: the surface texture (formed both during crystal growth and chemically) , the mirrors of the minority carrier of the superficial, frontal and posterior field, the passivation of the surfaces that use the layers of silicon oxide, the use of anti-reflective coatings, the use of the ohmic contacts as a surface light reflector Subsequently, the intrinsic protection against the damage of a reverse orientation condition, due to contiguous regions n + and p +, highly contaminated, and improved negative and positive contact busbars, which allow a design of the 'surface mount technology' , when they connect cells in series.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 illustrates a cross section of a dendritic band silicon preform of the preferred embodiment of the invention; Figure 2 illustrates the cross section of the band in the aluminum deposit stage for positive contact; Figure 3 illustrates the cross section of the band in the heat treatment step; Figure 4A illustrates the cross section of the band during the isolation of the first metal stage to form the aluminum oxide; Figure 4B illustrates an approach view of the rear surface portion of the band of Figure 4A; Figure 5A illustrates the cross section of the web during the removal of the silicon oxide from the back surface; Figure 5B illustrates a close-up view of the rear surface of the band of Figure 5A; Figure 6A illustrates the cross section of the band during metal deposit for the negative contact; Figure 6B illustrates an approach view of the rear surface of Figure 6A; Figure 7 is a rear, bottom view of the finished cell; Figure 8 is a plan view of the rear side surface of a substrate having 8 cells; Figure 9 is an enlarged view of the back surface of one of the cells on the substrate of Figure 8; Figure 10 is a sectional view, taken along lines 10-10 of Figure 9; Figure 11 is an enlarged detailed view illustrating a portion of a pair of positive electrodes and the region between them; Y Figure 12 is an enlarged detailed view illustrating a corner portion of the upper part of the cell of Figure 9.

DETAILED DESCRIPTION OF THE PREFERRED MODALITIES Turning now our attention to the figures, a manufacturing technique and preferred designs of the present invention are disclosed. While Figures 1 to 7 show the dendritic band silicon, it will be understood that the invention can be applied to any form of silicon, which includes floating zone silicon, Czochralski silicon, magnetic Czochralski silicon, molded silicon and silicon. sheet silicon, with the condition that the diffusion length of the minority carrier, under the operating conditions of the cell, exceeds the thickness of the cell.

The starting material for the silicon solar cell (Si), after contact, according to the present invention, is any n-type silicon starting material, shown as layer 10. Common n-type impurities for Si include the atoms of Group V of the Periodic Table, and comprise such elements as Li, Sb, P and As. However, it should also be understood that the revealed cell structure will also work for the starting p-type silicon material, or even without impurities, since the layer 10 functions primarily as a light absorbing medium.

In the preferred embodiment, illustrated in the figures, the starting material (for the volumetric layer 10 in Figure 1) is the n-type dendritic band silicon, contaminated with the antimony (Sb). The dendritic band silicon tape is grown using a procedure similar to that used to produce the Czochralski silicon (CZ). However, silicon that grows by other methods can also be used, in addition to dendritic band silicon, such as flotation zone silicon, CZ silicon, molding silicon and sheet silicon. The dendritic band of Si typically grows to a thickness of 100 microns, although other thicknesses can be used. In this thickness, the diffusion length of the minority carrier is usually greater than the thickness of the cell, often two or three times the thickness. On both surfaces, top and bottom, of the Si tape contaminated with Sb, a shallow layer n +, 20, diffused on both surfaces is placed, while the band is still in the growth furnace. If the surface layers n +, 20 are not introduced during the growth of the band, they can be incorporated at the beginning of the process by any proven method known in the art, which includes simultaneous front and back diffusion of an impurity source. - liquid, which use the rapid thermal process. The n + layers create a "surface field" which drives the voids away from the surfaces and reduces surface recombination there, as well as accelerating the holes generated in the surface layers to the pn junction, and other beneficial effects that work to increase the short circuit current and open circuit voltage, to intensify the efficiency of the solar conversion. In addition, the n + back layer promotes ohmic contact to the negative contact metal, as described below.

In addition, the surface texture of both surfaces, top and bottom, is provided in order to trap more incident light. Such a surface texture, shown in the form of a sawtooth pattern, can be grown, introduced by the anodic etch to create a porous layer of Si (as per the method delineated by YS Tsuo et al., Potential Applications of Porous Silicon in Photovoltaics, Conf. Record 23rd IEEE Photovoltaic Specialists Conf. (Louisville, KY) (1933), incorporated herein by reference), or mechanically introduced by sawing or optics, such as by laser engraving. Although textured and contaminated surfaces are shown in the preferred embodiment, their use is optional in the general case. In addition, the texture of the bottom surface of the cell is not shown in Figures 2 - 6 for clarity.

Figures 1 to 6 illustrate important steps in solar cell manufacturing, including the use of aluminum as the contact material, both impurities and ohmic, as well as its masking, using a self-aligning anodic oxidation technique. Aluminum or aluminum material is defined herein as pure Al or an Al-Si alloy, in which the concentration of silicon is lower than the eutectic composition (88.7% Al and 11.3% Si, by weight). This aluminum is deposited on about half the total back surface area in strips spaced from edge to edge by about 100 μm (micras) of separation and every 100 μm of width, as shown in Figure 2, at spacing 40. spacing 40 between strips 50 should be less than a diffusion length of the minority carrier (gap) for efficient collection of the carrier. The line width and spacing can thus be reduced below 100 μm with some advantage. The useful upper limit for the value of the line width is approximately 2000 μm; while the useful range for edge-to-edge spacing 40 is from about 50 μm to 300 μm. Thus the parallel strips of aluminum each form regions separated from the material of the source of impurities by the solder cell of splice pn, and, as explained later here, they are joined (they are contiguous) in their bases to form a region of collector bar for the positive contact of this busbar.

In a preferred method for depositing aluminum, it is deposited by screen printing, a process known per se in the art, in the form of an aluminum paste. The aluminum strips are shown in Figure 2, where the aluminum is deposited, as the strips 50 go in the plane of the paper. However, other methods, in addition to screen printing, for depositing aluminum, are within the scope of the invention, such as evaporation of electron beams or electronic deposit, although these methods may require more expensive patterns. by photolithography and thus are less convenient.

Generally speaking, the Al layer of the p-type material is relatively thin when compared to the n-type volumetric layer, of about 2 to 20 μm in thickness for a volumetric layer with a thickness of approximately 100 μm.

Note that in the preferred embodiment of the present invention, the selection of aluminum serves at least three purposes simultaneously: it acts as the source of p-type impurities, acts as a positive contact metal and acts as a surface light reflector posterior, partial, on the area of the posterior surface that covers, which is approximately 50% of said posterior area.

Turning our attention now to Figure 3, which illustrates the heat treatment, a figure representing the effects of the heat treatment of the deposited layer of Al in an oxygen rich environment, at a temperature of about 850se is shown. In this step, the p-n junction is formed by the Al alloy deposited by screen printing, with Si. It is further believed that maintaining the temperature of 8500C for a prolonged period of time, for example 30 minutes or more, is beneficial to form a satisfactory alloy. The temperature range can be from 577ac, the eutectic temperature of the aluminum-silicon alloy, to 1420SC, the melting point of silicon.

The heating can be provided by a rapid thermal processing unit, a band furnace, a tube furnace, or other means. The ambient atmosphere can be inert, such as argon or nitrogen, or chemically active, such as with oxygen or hydrogen. Mixtures of environmental gases are also possible. Times at high temperatures can vary from 30 seconds to several hours. In the preferred embodiment, it is expected that the use of an oxygen rich environment, at this temperature, will allow the oxide (SiO2) to grow in any exposed Si, which will passivate the surface and decrease the deleterious effects of recombination.

The temperature is then decreased in the Si-Al alloy, and the Si again grows by the epitaxy of liquid phase, until reaching the eutectic temperature (5772C). As a result, the Si that has regrown is now of type p contaminated with Al (about 1018cm ~ 3), as denoted by the composition 60 of the p + layer in Figure 3. The required pn junction is formed according to the concentration of Al exceeds the concentration of the donor in the starting Si, and the eutectic alloy (about 88.7% Al and 11.3% Si, by weight) remains on the surface to serve as a contact to the p-type silicon. . It will be noted that the pn junction can be quite deep (1 to 20 microns from the surface) however, since the splice is at the back of the cell, where very little light is absorbed, the depth of the splice is only of importance secondary, compared to a conventional solar cell frontal contact. The splicing depth of the alloy can be controlled by the use of a mixture of Al-Si as the printed printed screen material, instead of pure Al. This is because as the concentration of the Si increases towards the eutectic composition, the amount of Si that the printed metal can dissolve becomes smaller, and thus the depth of the joint becomes smaller. This depth of the splice can be increased, if desired, by increasing the thickness of the deposited aluminum and increasing the temperature of the alloy, according to the aluminum-silicon phase diagram. Also, the mass life time of the minority carriers will probably increase to the temperature of about 850SC, due to the property of the dendritic band silicon (the preferred type of the Si used in the present invention), will have any defect upon cooling down rapidly , such as Si vacancies and self-interstitial annealing. Cooling at a controlled rate of 10 sec per minute versus faster cooling will also allow rapid chilling defects to be tempered, reducing deleterious recombination sites.

The above heat treatment can be carried out using a web oven process in which the samples are loaded onto a web and this web is pulled slowly through the stable hot zones in a furnace. In the alternative to heating the Si / Al mixture to 8502C for about 30 minutes in such a furnace, a variety of other techniques can be employed to form the eutectic Si / Al, such as the use of a process unit. rapid thermal, which uses, as an example, quartz lamps to heat the Si to 1000SC and maintain that temperature for 30 seconds, which can increase production in a commercial establishment, or by a conventional quartz tube oven.

Having such a p + region (region 60 of Figure 3) immediately adjacent to the n + region in the back surface layer (the back surface layer 20 of Figure 1), also has the unanticipated benefit of protecting the solar cell of overheating, when oriented in reverse, as, for example, by shading in a module. A "module" is a group of interconnected cells that are protected by glass or other roofing material, and that produce a significant amount of power, typically 10 to 100 watts, when illuminated. This design p + n + built in protection against the reverse orientation, eliminates any need to have diodes of protection, commonly designated as "deviation diodes", to protect against reverse orientation. The splice p + n + acts as a Zener diode, which is interrupted under modest inverse orientations with only a small voltage and, consequently, only a small amount of energy dissipates in the cell, thus protecting this cell.

Turning now our attention to Figures 4A and 4B, another step in the preferred embodiment of the present invention is illustrated. This step provides a unique "self-alignment" feature, to align the negative contact (to the region of type n) with respect to the positive contact (to the region of type p). In order to isolate the contact of the p-type layer (positive contact) from the contact of the n-type layer (negative contact), the present invention does not require complicated masking techniques that have been used in the past, but rather, it employs the use of insulation by the formation of an oxidation layer to coat the first set of Al contacts (positive) and electrically isolate this set from the second set of (negative) contacts. As shown in Figures 4A and 4B, this is done by isolating the p-layer composition of Al-Si and the exposed Al-strips 70, which are on the outside of the band preform 10, with an oxide layer 80 ( insulator), which is naturally formed on the exposed materials of Si, Si-Al and Al, in the form of AI2O3, SiO2 or some of its variants, in the presence of oxygen. The oxide layer should grow to cover all of the Al strips 70 to a thickness of about 0.1 to 1 μm. As shown in Figures 4A and 4B, in this step the oxide layer 80 also covers the regions 90 of the surface of the n layer, between the bands of Al, 70. As described more fully below, the oxide layer on the surface regions 90 is subsequently removed (named as the passage illustrated in Figure 5B below) in order to make possible the ohmic contact with the cathode (if of type n) of the solar cell diode.The preferred method of forming the oxidation layer, in Figures 4A and 4B, is by anodic oxidation, in which the surface of the developed cell layer is immersed in a weak electrolyte (such as borates, phosphates or carbonates). ) and is subjected to an applied voltage. The current flows as a result of the voltage applied between an inert electrode and the contact metal (the eutectic Al-Si). The thickness of the anodic oxide can reach 1 miera if the voltage which drives the anodizing current reaches 700 V (14 A / V or 1.4 nm / V). Such oxides must be compact and free of tiny holes. Because the ohmic contact must be made to a positive contact busbar (in region 110, shown in Figure 7), when the solar cell is terminated, in order to make contact with exposed strips 70 of Al, the growth of the anodic oxide must be inhibited in the region of the busbar (and this region of the busbar must be protected throughout the process). One way to do this is to use a still compressive conductive medium, so that the contact of the area is occupied by the bus bar, such as a closed cell sponge, impregnated with carbon. A closed-cell sponge is preferable, since it will not absorb the electrolyte.

In addition to anodic oxidation, any other method for isolating the eutectic layer of aluminum or aluminum-silicon, such as by oxidizing aluminum in an oxygen-containing plasma, is also considered by the present invention.

After the oxidation layer 80 has been added by the anodic oxidation or any other suitable method, the n-type Si surface, covered by the oxidation layer in the regions 90 of the interstitial surface, must be exposed in order to allow the negative contact metal layer of Al to be deposited there. Thus, with reference to Figures 5A and 5B, it is illustrated how the oxidation layer is removed from the Si layer on the back surface, but not the oxidation covering the Al strips 70. In a preferred method of doing this, hydrofluoric acid is used to selectively remove and remove interstitial SiO2 (silicon dioxide) 20, because hydrofluoric acid does not react or remove AI2O3 (aluminum oxide). Consequently, the interstitial Si02 is removed while the AI2O3 insulator layer remains covering the strip contacts 70 (see Figure 5B). Other chemicals that have similar effects can be used, or other oxide removal techniques can be employed, such as by light sand jets on the silicon dioxide layer, which also provides the beneficial effect of lightly scraping the exposed surface of the surface. silicon, which promotes ohmic contact to the n-type base. The use of sandblasting eliminates the need for a diffuse n + coating on the back of the cell, which is usually provided in primary form to promote ohmic contact. The chemical treatment of reactive ions (RÍE) can also be used to remove Si02, while leaving AI2O3 unchanged. Ionic grinding can also be used to lightly scrape the surface to promote ohmic contact in a manner analogous to sandblasting.

Figures 6A and 6B illustrate the next step in the process of manufacturing the solar cell of the present invention, ie the application of a second layer of metal to form the self-aligning negative ohmic contact metal (for the contact of the Si layer of type n). This second metal can be any suitable contact metal, which include aluminum and silver. Again, screen printing is the preferred method for depositing this second metal, but other methods, such as electron beam evaporation or electronic deposit are also acceptable. This second metal layer, designated as the metal layer 100, covers almost the entire back of the cell. This layer is isolated from the first strips 70 of metal contact by the anodic oxide 80, while making the ohmic contact to the 90 n + regions, found between the metal strips 70, formed of the first deposited aluminum layer. The second metal layer also helps to form a light reflector from the back surface, to assist in reflecting any light not absorbed by the silicon material in a first pass, back into the silicon material. Figure 7 shows the finished solar cell, as seen from the back, where the cell is covered with the second metal (aluminum or other ohmic metal) or the aluminum-silicon eutectic. The solar cell of the present invention has an unobstructed front surface, which is a significant advantage over conventional solar cells. On the back, both metal contacts (ohmic metal contacts, 70 and 100) serve as partial reflectors of the light, in addition to being the ohmic contacts. Likewise, the design of the busbar allows the simplified interconnection of the solar cells of the present design in series, as the present model eliminates the need for the manufacture of intercell contacts, external, from the front to the rear, uncomfortable, and uses the design of "surface mount" technology, which dispenses with such contacts.

Referring now to Figure 7, there is shown a bus bar region 110, which is the non-oxidized area, where the contact of the bus bar leading to the first strips 70 of positive contact metal, is located, as already It was explained before. The region 110 of the busbar, as can be seen, is smaller in area than the areas covered by the layer 100, of the negative contact metal, but both metal contacts, positive and negative, lead by themselves precisely to a Modular surface mounting design. The eutectic fingers 70 of Al-Si flow vertically upwards from the bus bar 110 of Figure 7, not visible in that figure, due to the over-position of the second metal layer 100. If desired, other ohmic contact metals other than Al or Ag can be deposited to form the positive and negative ohmic contacts described herein, such as by means of the example using as contacts the titanium / palladium / copper sandwich type. or silver printed on the screen.

With respect to the use of anti-reflective (AR) coatings, one or more coating layers may ordinarily appear on the outermost front illuminated surface, but they have been omitted from Figures 1-7 for clarity. It is possible that an AR coating is optional with the present design, since the texturing or perhaps the texturing in combination with a passivating oxide, as explained above, can be sufficiently effective, so that there is no need for an AR coating. . However, an AR coating, such as silicon nitride, applied by the plasma intensified chemical vapor tank (PECVD) or titanium dioxide applied by the chemical vapor deposition at atmospheric pressure (APCVD) can be used. The implantation of hydrogen ions (to improve the diffusion length of the minority carrier) can also be introduced prior to the deposition of the AR coating, provided that the surface oxide is absent or very thin. REDUCTION TO PRACTICE With reference to Figure 8, solar interleaved back contact 120 contact cells of the aluminum alloy splice (IBC) are fabricated using Czochralski silicon wafers as the starting substrates. These solar cells serve to demonstrate the use of aluminum alloy splices in a subsequent contact configuration. Aluminum was deposited by the evaporation of electron beams, rather than by screen printing. The negative electrodes were not self-aligned using anodic oxidation or some other technique, but rather they were manually aligned with respect to the positive electrodes of aluminum-eutectic silicon, with the help of a contact aligning element, as used for the manufacture of integrated circuits.

The wafers of the silicon substrate were polished on one side, 7.6 cm. in diameter, 279.4 to 431.8 microns thick, with phosphorus impurities, at 3 - 20 O-cm. , and with a surface (111). Two wafers (designated CZ-7 and CZ-8) were processed with the structures and test solar cells on the polished side. Figure 8 shows the back side of a wafer. The overlapping side (hidden from view in Figure 8) has a phosphorus diffusion and an anti-reflective (AR) coating 95 (see Figures 10 and 12). Each of the eight solar cells is a square of 1.00 cm. , ignoring the 2 mm wide bus bar 112 for the negative electrodes. They are referenced below in Tables 2 and 3 by the number of the n + fingers that make up the negative contact (4, 8, 16 and 25) and by their location (inside (I) or peripheral (P)). The four inner cells have only the eutectic alloy in contact with the p + region, while the peripheral cells have the second metal also deposited in the eutectic part. The best results were obtained for the CZ-8 wafers, so only the process and test results for this wafer will be described. The process used in the manufacture of the cells IBC for the CZ-8 wafer, is summarized in the following table. A remarkable feature of this process is that the aluminum alloy back splice and phosphorus-contaminated n + layers, formed through the front surface and the exposed back surface, between the aluminum electrodes, were created in a single step to high temperature. Photolithography was used for the evaporated aluminum standard and to define the second metal that serves as the negative electrode. A view of the rear side of the solar cell of the IBC aluminum alloy splice is shown in Figure 9, while the cross section views are shown in Figures 10 to 12. TABLE 1 TABLE 1 (Continued) Process for the CZ-8 Wafer Some comments regarding the CZ-8 process: The alloy / n + diffusion process includes slow step cooling in the fast thermal processing unit (RTP) (* 50ec / in) from IOOOSC to 825SC).

There is no apparent damage in painting the liquid phosphorus impurities on the aluminum, before the RTP process at 1000SC, which simultaneously forms the p + and n + regions. The depth of the p + region, determined was 5 μm by scanning electron microscopy in cross section, for RTP processes at 1000ac for 30 seconds).

The sheet resistance of the n + frontal surface, measured, was 25 O / D, after the RTP diffusion at 100 ° C for 30 seconds. Mask 2 was skipped, because it is only needed for some test patterns, not for IBC cells. The Ti / Al contact to the n + surface was limited to a thickness of 0.55 μm by the detachment process. Without the chemical treatment of silicon n + between the positive electrode (eutectic) and the negative electrode (Ti / Al), the p-n junction was severely derived. TABLE 2 I-V Illumination Data (AM1.5, loo m / cm2, front illumination) Before AR Coating TABLE 3 I-V Illumination Data (AM1.5, loo m / cm2, front illumination) After AR Coating Note that the operation of the posterior contact solar cells was obtained, thus demonstrating the viability of the revealed structure. Efficiencies have been measured in the conversion of light to electrical energy up to 9.0%. Refinements are expected in the substrate material and process techniques to increase this efficiency to twice that shown up to now.

While the above provides a detailed and complete disclosure of the preferred embodiment of the invention, various modifications, alternative constructions and equivalents may be employed. For example, while the preferred embodiment has been described with reference to aluminum to form p-type diffusion and ohmic contacts, other Group III metals, such as gallium and indium, can be used for this purpose. A suitable element of group III is that which dissolves the silicon and remains as a trace amount, to serve as an impurity when the silicon solidifies. In addition, while the preferred embodiment has been described with reference to a n-type volumetric silicon layer 10, p-type volumetric silicon can be used to fabricate a subsequent contact solar cell. When the p-type volumetric silicon layer is used, a p + layer is formed on the upper surface as the layer 20, but an n + layer is formed on the bottom of the volumetric layer 10. As will be appreciated by those skilled in the art, the Minority carriers are electrons. Therefore, the above description and illustrations should not be construed as limiting the scope of the invention, which is defined by the appended claims.

Claims (32)

1. NOVELTY OF THE INVENTION Having described the present invention, it is considered as a novelty and, therefore, the content of the following is claimed as property: CLAIMS 1. A posterior contact solar cell, which comprises: a semiconductor volumetric layer, of a first type of conductivity, which has a front surface and a back surface; a plurality of spaced regions of semiconductors with impurities, of the opposite conductivity type, located in the volumetric layer closest to the back surface and forming a plurality of semiconductor junctions therewith; a first set of spaced ohmic contacts, connected to the plurality of spaced regions of semiconductor with impurities, and located along the back surface; a second set of ohmic contacts, connected to the posterior surface of the volumetric layer, in the spaces between the first set of the ohmic contacts; and insulating elements, to electrically isolate the first set of ohmic contacts spaced from the second set of ohmic contacts.
2. 2. The solar cell, according to claim 1, in which the first set of ohmic contacts is in the form of conductive strips, substantially parallel in mutual form.
3. 3. The solar cell, according to claim 2, wherein the conductive strips are joined at one end, to form a bus bar contact.
4. 4. The solar cell, according to claim 1, wherein the volumetric semiconductor layer has a thickness no greater than the diffusion length of the minority carriers of the first type of conductivity.
5. 5. The solar cell, according to claim 1, wherein the volumetric layer is formed of n-type silicon.
6. 6. The solar cell, according to claim 5, wherein the n-type silicon is the dendritic band silicon.
7. The solar cell, according to claim 1, in which the first set of spaced ohmic contacts comprises an alloy of the semiconductor material of the volumetric layer and a Group III metal comprising the impurities of the acceptor, for the plurality of regions spaced from the semiconductor with impurities.
8. 8. The solar cell, according to claim 7, in which the Group III metal is selected from the group consisting of aluminum, gallium and indium.
9. 9. The solar cell, according to claim 1, wherein the insulating element comprises an insulating layer, which covers the first set of ohmic contacts.
10. 10. The solar cell, according to claim 1, further including an anti-reflective coating on the front surface.
11. 11. The solar cell, according to claim 1, wherein the volumetric layer is formed of n-type material and the front and back surfaces are n + contaminated initially.
12. 12. The solar cell, according to claim 1, in which the second set of ohmic contacts is comprised of ohmic metals, selected from the group consisting of silver, aluminum, copper, titanium and palladium.
13. 13. The solar cell, according to claim 1, wherein at least one of the surfaces, front and rear, is textured to increase the entrapment of the light in the volumetric layer.
14. 14. A method for manufacturing a solar contact cell posterior, which has self-aligned ohmic contacts, this method comprises the steps of: (a) supplying a volumetric semiconductor layer, of a first type of conductivity, this volumetric layer has a front surface and a back surface; (b) forming a plurality of semiconductor diffusion regions, of opposite conductivity type, in the volumetric layer, near the back surface; (c) forming a first set of ohmic contacts spaced apart by the diffusion regions, on the back surface, using an ohmic contact metal material; (d) electrically isolating the first set of ohmic contacts from the spaces between them; and (e) forming a second set of ohmic contacts on the back surface in the spaces, using an ohmic contact metal material, this second set of ohmic contacts is electrically isolated from the first set of ohmic contacts.
15. 15. The method for manufacturing a rear contact solar cell, according to claim 14, in that the supply stage (a) is carried out with n-type silicon.
16. 16. The method for manufacturing a rear contact solar cell, according to claim 14, wherein the supply step (a) is carried out with the n-type silicon, which has a n + surface diffusion layer on its surface frontal.
17. 17. The method for manufacturing a rear contact solar cell, according to claim 14, wherein the supply step (a) is carried out with the n-type silicon, which has a n + surface diffusion layer, on its front and back surfaces.
18. 18. The method for manufacturing a rear contact solar cell, according to claim 14, wherein steps (b) and (c) are carried out concurrently.
19. 19. The method for manufacturing a posterior contact solar cell, according to claim 14, wherein steps (b) and (c) are carried out concurrently, applying a layer with a pattern, which contains a Group III metal , to the rear surface of the volumetric layer, heating at least the back surface and the adjacent inner regions of the volumetric layer, so that the material of the volumetric layer in the interior regions and the patterned layer forms an alloy, and allowing this alloy to cool, so that the diffusion regions are formed using the Group III metal as an acceptor and the first set of contacts are formed from the cooled alloy that remains on the back surface.
20. 20. The method for manufacturing a rear contact solar cell, according to claim 19, wherein the patterned layer comprises a mixture of Group III metals and the bulk material.
21. 21. The method for manufacturing a rear contact solar cell, according to claim 20, wherein said mixture comprises aluminum and silicon.
22. 22. The method for manufacturing a rear contact solar cell, according to claim 19, wherein the patterned layer comprises a plurality of individual strips.
23. 23. The method for manufacturing a rear contact solar cell, according to claim 22, wherein the strips are substantially mutually parallel.
24. 24. The method for manufacturing a rear contact solar cell, according to claim 19, wherein the patterned layer is applied by the printing of the screen. lla.
25. 25. The method for manufacturing a rear contact solar cell, according to claim 14, wherein the step (d) of the electrical insulation is carried out by forming an insulating layer on the first set of ohmic contacts and the spaces between them, and selectively removing the portions of the insulating layer that overlap the spaces of the back surface of the volumetric layer, so that the insulating layer substantially covers only the first set of ohmic contacts and the spaces are exposed.
26. 26. The method for manufacturing a rear contact solar cell, according to claim 25, wherein the step of selective removal comprises etching the portions of the insulator layer that overlap the spaces.
27. 27. The method for manufacturing a posterior contact solar cell, according to claim 26, wherein the etching step is chemical.
28. 28. The method for manufacturing a posterior contact solar cell, according to claim 26, wherein the etching step is by means of reactive ions.
29. 29. The method for manufacturing a rear contact solar cell, according to claim 25, in that the step of selective removal comprises the treatment by sandblasting of the portions of the insulating layer that overlaps the spaces.
30. 30. The method for manufacturing a rear contact solar cell, according to claim 25, in which the step of selective removal comprises the treatment by ions of the portions of the insulator layer, which overlaps the spaces.
31. 31. The method for manufacturing a rear contact solar cell according to claim 14, further comprising the steps of texturing at least one of the front and rear surfaces of the volumetric layer.
32. 32. The method for manufacturing a rear contact solar cell, according to claim 14, further comprising the step of applying an anti-reflective coating on the front surface.