Photovoltaic Cell

This invention relates to solar cells which convert sunlight to electricity, and particularly to solar cells which include layers of tin oxide and silicon in their construction. Such cells have been described in my U.S. Pat. No. 4,146,657, which is incorporated by reference in the present application. In general, this patent disclosed a process for forming fluorine-doped tin oxide layers with such high electrical conductivity that they could be used advantageously as transparent electrodes covering a light-absorbing base material for solar cells made, for example, from silicon. This structure eliminated the need for an expensive and fragile grid of fine metallic lines which serve as an upper electrode in conventional solar cells.

The electrical resistance of these fluorine-doped tin oxide transparent electrodes is satisfactorily and consistently low, when measured with current flowing in the plane of the film. However, current flow between the tin oxide film and the silicon underneath, has been found to be impeded sometimes by an "interfacial resistance" between the two layers. Such interfacial resistance, when present, reduces the overall efficiency of a solar cell.

Another advantageous aspect of the tin oxide layer in these cells is its function as an optical antireflection coating. Bare flat silicon surfaces reflect over one-third of incident sunlight, while a silicon cell coated with tin oxide typically reflects about one-fifth of the incident light. It would clearly be desirable to reduce this 20% reflection loss still further.

It is a principal object of the present invention to reduce the electrical resistance of a tin oxide-silicon solar cell, and thereby to improve its efficiency.

Another object of the invention is to improve the thermal stability of such solar cells.

Another object of the invention is to reduce further the light reflected from the cell, thereby also improving its efficiency.

Other objects of the invention will be discerned by those skilled in the art on their reading of the present disclosure.

The present invention consists of an improved solar cell design in which a thin layer of titanium dioxide, TiO.sub.2, is placed between the tin oxide and the silicon layers. I have discovered that the resulting composite SnO.sub.2 /TiO.sub.2 /Si structure has significantly improved efficiency and also better thermal stability, compared to the previous SnO.sub.2 /Si structures.

The advantages of this new structure arise from a particularly appropriate combination of properties of the materials used. The new structure is more stable, and resists the formation of an insulating layer noted in the previous SnO.sub.2 /Si structures. A simpler structure of TiO.sub.2 /Si would not be as advantageous as the SnO.sub.2 /TiO.sub.2 /Si structure, since TiO.sub.2 alone does not display sufficiently low electrical sheet resistance, to serve alone as the transparent electrode. However, titanium oxide and tin oxide normally have the same crystal structure, and can bond to each other to form an excellent mechanical and electrical contact. Thus, the composite layer SnO.sub.2 /TiO.sub.2 combines the best features of each component (high electrical conductivity of SnO.sub.2 and superior chemical stability of TiO.sub.2) to provide better performance than either compound could individually provide.

Pure titanium dioxide films have a rather high electrical resistance, and would not serve advantageously in the present structures. However, it is known in the art (see, for example, G. H. Johnson, "Influence of Impurities on Electrical Conductivity of Rutile," Journal of the American Ceramic Society, Vol. 36, pages 97-101, 1953) that certain elements such as niobium, tantalum, tungsten, phosphorous, arsenic, antimony and vanadium, can act as dopants serving to decrease the electrical resistance, when they are present as impurities in titanium dioxide. I have also discovered that fluorine can also serve as a dopant increasing the conductivity of titanium dioxide. Thus, the SnO.sub.2 /TiO.sub.2 /Si structures will be understood to contain sufficient amounts of these or other dopants to provide for the required electrical conductivity.

The new SnO.sub.2 /TiO.sub.2 /Si structure also reduces the reflection loss from the cell, because the refractive index of TiO.sub.2 is intermediate between that of SnO.sub.2 and Si. Typically, the addition of a TiO.sub.2 layer between the SnO.sub.2 and Si layers increases the amount of light entering a solar cell by about four percent. Also, the low electrical resistance of the composite structure reduces the area normally blocked by metallic front electrodes, which typically reflect between 5 and 10% of the light incident on the cell. In addition, a final, top layer of a material of lower refractive index, such as silica, SiO.sub.2, (with refractive index optimally equal to the square root of the refractive index of SnO.sub.2, or about 1.42) on top of the SnO.sub.2 can reduce the reflection loss even further.

FIG. 1 illustrates, schematically and in section, a solar cell constructed according to the invention.

FIG. 2 shows, schematically and in section, an apparatus for forming solar cells according to the invention.

Referring to FIG. 1, the base photovoltaic layer 4 consists of a crystalline layer of silicon about 200 microns thick. It is doped with boron to be p-type through most of its thickness, except for a thin layer 6 which doped n-type with phosphorus through a thickness of about 0.6 micron. The surface of this n-type layer is heavily doped with phosphorous in order to promote good electrical contact with the titanium dioxide layer 8, which has a thickness about 0.1 micron. This is covered with a fluorinedoped tin oxide layer 10 of a thickness about 1 micron. Metal front contacts 12 of nickel and back electrode 2 of aluminum complete the essential cell structure. An optional antireflection layer 14 of SiO.sub.2 about 0.1 micron thick is deposited on top. In an alternative form of the cell, the base layer of silicon 4 is doped n-type with phosphorous, while the top layer of silicon is doped p-type with boron, by methods well-known in the art. In other variations, the silicon layer may be replaced by other suitable semiconductor materials, such as gallium arsenide or indium phosphide.

FIG. 2 shows one convenient way to form solar cells with this structure. The silicon layer is first formed and doped by methods which are well-known in the art. Then the titanium dioxide and tin oxide and, optionally silicon dioxide layers are formed by chemical vapor deposition using an apparatus shown schematically in FIG. 2. The bare, doped silicon wafer 22 is placed on a heater 20, above which a gas distributor plate 24 is held. Suitable reactive vapors are introduced through holes 26, and the gases spread across the space 28 over the silicon, and deposit, in turn, the desired sequence of metal oxide layers of the silicon.

Suitable metal oxide layers may also be deposited by other methods, such as vacuum evaporation, sputtering, or glow discharges, as are well known in the art.

In this application and accompanying drawings there is shown and described a preferred embodiment of the invention and suggested various alternatives and modifications thereof, but it is to be understood that these are not intended to be exhaustive and that other changes and modifications can be made within the scope of the invention. These suggestions herein are selected and included for the purposes of illustration in order that others skilled in the art will more fully understand the invention and the principles thereof and will be able to modify it and embody it in a variety of forms, each as may be best suited to the condition of a particular case.

A silicon solar cell blank (n.sup.+ /p type) is placed on the heater 20 at a temperature of about 520.degree. C. A gas mixture of 4% hydrogen, 2% HF in balance nitrogen carrier gas is passed over the wafer for a few minutes, until the thin silicon dioxide layer normally present on silicon is etched away. The gas mixture is then immediately switched to contain 0.2% titanium (IV) ethoxide Ti(OC.sub.2 H.sub.5).sub.4, 0.02% niobium (V) ethoxide Nb(OC.sub.2 H.sub.5).sub.5, 4% hydrogen, balance nitrogen, for about one minute. The gas mixture is formed by injecting a liquid mixture of the titanium (IV) ethoxide and the niobium (V) ethoxide into the hydrogen-nitrogen mixture at about 200.degree. C., using a syringe pump. The gas flow rate is adjusted so that each volume of the reactive gas mixture spends about 0.2 sec in the heated reaction zone. The distributor plate above the gases is held at a temperature of about 250.degree. C. This deposition produced a layer about 0.16 micron thick of titanium dioxide doped with about 10% niobium.

Next, a layer of fluorine-doped tin oxide is formed, by switching the gas mixture to 2.0% tetramethyltin (CH.sub.3).sub.4 Sn, 3.0% bromotrifluoromethane CF.sub.3 Br, balance dry air, for about one minute, which forms a layer about 1 micron thick.

The resulting coated cell has a sheet resistance of about 1 ohm per square on the upper surface. There is negligible interfacial resistance (less than about one ohm per square centimeter) between the metal oxide layer and the silicon, as is demonstrated by the high fill factor when the whole structure is tested as a solar cell. The reflectivity of the coated cell is lower, by about 5 percent, than that of a similar cell without the TiO.sub.2 layer.

Example 1 is repeated, except that tantalum (V) ethoxide is used instead of niobium (V) ethoxide, as the dopant source for the titanium dioxide layer. Similar results are obtained.

Example 1 is repeated, except that titanium (IV) fluoride is used instead of niobium (V) ethoxide, as the dopant source for the titanium dioxide layer. Similar results are obtained.

Examples 1-3 are repeated with solar cells made of GaAs instead of Si.

Examples 1-3 are repeated with solar cells made of InP instead of Si.

Volatile compounds of vanadium, tungsten, phosphorous, arsenic and antimony also can be used as dopants for the titanium dioxide layer. Interfacial resistances achieved thus far with these dopants have been higher than with fluorine, niobium, and tantalum, which are therefore preferred.

A further reduction in reflection from these cells is obtained by a final coat of silica, according to procedures known in the semiconductor industry, or described in my U.S. Pat. Nos. 4,187,336 and 4,206,252.

It is also to be understood that the following claims are intended to cover all of the generic and specific features of the invention herein described and all statements of the scope of the invention which might be said to fall therebetween.