High voltage, high current Schottky barrier solar cell

(54) Name: A Schottky barrier solar cell

TECHNICAL FIELD OF THE INVENTION

The present invention relates to a device for converting energy, more specifically a solar cell, ie. a cell for converting the sunlight's energy into electrical energy.

State of the art 'It is well known about the present difficulties in providing high-efficiency solar cells or converters for the sunlight's energy to electrical energy. Although extensive experiments and experiments have been carried out in this field, and although these have incurred very high costs, it turns out that all proposals for solar cells to date, in which words have been termed such converters of solar energy to electrical energy, have been little effective. Most of the solar cells proposed to date have been photovoltaic, containing a p / n junction. These cells contain a first semiconductor in the form of a layer doped into one conduit type, e.g., n-type, and a second one. semiconductor in the form of a layer doped to the opposite conductor type, thus in the selected example for p-type, so as to obtain a p / n connection. Most of the experiments carried out for the said purpose, especially in connection with space exploration, have used silicon as doped material.

•5

7508320-4

- 2 Sunlight contains, as is well known, a spectrum εν different wavelengths, and thus does not consist of light of a single wavelength. Sunlight therefore also contains more photons of low energy (long-wave light) than high-energy photons (short-wave light). The spectrum of the sunlight is so distributed that an optimal spectral hand is obtained with essentially 1.5 electron volts. However, since the spectral hand affects a silicon cell, it remains in the order of about 1.1 electron volts. Due to the limited wavelength range of silicon, the output voltage of a silicon cell is also very low. It is largely derived only from the half-silicon spectral band and thus gives substantially only 0.55 electron volts. It is otherwise these circumstances that have led to silicon cells being very little efficient in converting the sunlight's energy into electrical energy. Also, the efficiency of conventional silicon solar cells is only in the order of about 11 calculated on existing solar light energy. provided that the silicon solar cells are at room temperature.

In order to increase the efficiency of the conversion, it is envisaged to use p / n cells in which the active semiconductor element is made up of gallium arsenide (GaAs) instead of silicon (Si)._Ok. gallium solar cells.

It is known that GaAs has a spectral range of about 1.4 electron volts, which value comes much closer to the present spectral value of the sunlight. - It should therefore also be expected of gallium solar cells a higher output voltage than that which could be obtained using silicon solar cells, especially in view of the stronger curvature of the spectral hand. In these experiments, however, several secondary problems have been found to be solved before more efficient solar cells can be found.

They have now reached a very advanced material handling technique, with which you can produce GaAs of very high quality and purity, as well as ternary compounds thereof. However, it appears that all of these materials are unstable and that they tend to dissociate at higher temperatures, especially in the order of magnitude required to diffuse and cure doped p / n precursors. Such dissociation can adversely affect the diffusion life of the cell, and this will thereby act on the generation of the charge sensors that give rise to the intended electrical current. Before one can use GaAs or its ternary compounds in photovoltaic cells in the preparation of satisfactory 'p / n * compounds, further development in the field of material processing technology is required.

However, even if a solar cell with GaAs as an active medium could be produced at the present technical stage, it would suffer from severe disadvantages. Namely, in a p / n cell, <jLet is the upper or outer layer, ie,!

the layers, which are doped to the opposite type, are quite thick, and amount to | in practice to several Ångström, usually to several thousand Ångström. such a thick layer of doped material would absorb a considerable g dol of sunlight, especially within part of the spectrum lying at ji and below. In addition, the surface of the counter-doped layer will form so-called recombination centers for the cavities and electrons that. formed in area | the¹ closest to it, thereby further reducing the number of ooh power of the rechargeable barriers which convey the desired electric current. It is therefore clear that, although one can make p / n-cells H [using GaAs or any comparable material such as semiconductors, this material has a more extensive range! spectrum than silicon, however, for the reasons described above, one would only achieve a very limited current.

The object and meaning of the invention

The present invention is primarily intended to produce an improved solar cell. This is to be done according to the invention without the use of any p / n connection, but the solar cell should nevertheless give good efficiency. Thus, the solar cell should be of such a nature that it provides good efficiency in converting sunlight energy into electrical energy, and it must give both high voltage and high current. It should be able to be made on the basis of the technical methods, which are currently available in developed technology!

form.

The above-mentioned requirements are satisfied according to the present invention by arranging a cell in multiple layers or layers, a first such layer of semiconducting material having a fairly wide spectral band, a second layer consisting of a thin film of semi-transparent material deposited on the upper surface of the film. the first layer, so that these two layers together form a Schottky barrier cell. On the opposite side of the one on which the semi-transparent film is arranged, which should be made of metal, the first layer is covered by a semiconducting material with smaller spectral bands. The broad spectral band of the first layer then leads to increased output voltage. In addition, it is found that the former layer contributes greatly to increasing the efficiency of conversion to the desired electric current of the energy of energy-rich photons in the sunlight, ie. the photons whose energy is higher than that of the spectral band. Photons with lower energy than that of the spectral band of the first-mentioned layer pass through this first layer and are absorbed in the second layer

7505320-4 with narrower spectral bands. The charge carriers generated therein move through the first layer to the Schottky barrier and thus participate in the generation of the electric current.

The invention will be described in more detail below in connection with some selected embodiments, but it is understood that the invention is not limited to these particular embodiments, but that various modifications may occur within the scope of the invention. The description will take place with reference to the accompanying drawings.

Attached drawings

In the accompanying drawings, Fig. 1 shows in a highly schematic and enlarged form a solar cell according to the present invention, in cross section, while Fig. 2 shows a band area diagram for the solar cell shown in Fig. 1.

DETAILED DESCRIPTION OF THE EMBODIMENT OF THE INVENTION As mentioned above, FIG. 1 is a cross-section through a solar cell of the present invention, more specifically, an embodiment of the invention. The solar cell is denoted in its entirety by the reference numeral 10. It contains a layer 12 of a doped, semiconducting, wide band material. The material may be an aluminum-gallium arsenide of the general formula:

Al Ga. If x 1 -x

It is noted that gallium is a very complicated metal with variable valence. Therefore, the factor x does not in this formula denote valence but instead the proportion of weight in relation to. weights As. Namely, in the indicated compound, molecular stability is felt, although to a certain extent Al is replaced by As ooh vioe verea. This is a major molecule formation for generating doping action. Thus, it may be the case that with x = 0.8 the doped molecule contains 0.8 atoms Al and 0.2 atoms Ga adjacent to each atom As. Such a major molecule could thus have the chemical formula n-Al 2 GaAs 2, where n is an arbitrary number value.

0m in this formula, for example, the factor x is equal to 0.8, so the layer 12 holds a bandwidth of about 2.1 electron volts. As will be further elucidated in the following, the layer 12 preferably has a thickness of the order of 1 µm or less. On the upper side of this broadband layer 12 is a thin layer 14 of semi-transparent metallic material which will hereinafter be referred to as the metal film 14. This layer is very thin and can, for example, have a thickness of 100 Angstroms or less. Together with the broadband layer 12, the latter layer forms a Schottky

7508320-4

-5 | Barrier.

The broadband layer 12 is on the upper surface of a layer 16 of a narrower band with a narrower bandwidth, e.g. pure GaAs. This compound is affected

Ϊ therefore not by the fairly optional factor x. One contact to the cell consists of -a grid structure 18 which is deposited on the top surface of the metal file, but 14. The second contact member 19 may be continuous and folded on the underside of the layer 16. . D et is particularly preferred that the contact layer 19 j is applied to a further layer 20 of highly doped material with halvlei border properties, e.g. GaAs, which serves as a support for the narrowband layer

16th Thus, both layers 16 and 20 may consist of GaAs, but in that case the layer 20 dries, which may be thicker than all the other layers, mentioned above, being strongly doped. Namely, this layer is intended to support the thinner, overlying layers 16, 12 and 14 and to reduce the serial resistance in the cell. It is also easier to apply the contact layer 19 to the more powerful dimension! layered, highly doped layer 20 than on the, t weaker dimensioned and doped: amaleband layer 16. Thus, it should be pointed out that the layer 20 does not in itself contribute to the generation of the electric current, and it is therefore not necessary, in the following, to further describe the nature of this layer, as in Figure 1, the arrows 25 represent the sunlight which directed to cell 10.

With some advantage, a reflection-preventing agent in the form of a layer 26 can cover those parts of the free surface of the semi-transparent metal film 14; which is not covered by the lattice structure 18, As noted above, the semi-transparent metal film 14 is extremely thin, preferably in the order of a few tens of Angstroms.

f *

Because it is semi-transparent and so extremely thin, a very small proportion of the sunlight will be absorbed in this film. This can also be expressed such that the majority of the photons pass through the semi-transparent metal film 14 and penetrate the broadband layer 12.

One can designate the object of the invention as a solar cell with Schottky barrier. The advantages of such a solar cell may perhaps be best explained by means of an example: Suppose that the bandwidth of the broadband layer 12 is about 2.1 'electron volts. With such a layer, the output voltage should be fairly high, probably in the order of about 1.0 electron volts. That such a high output voltage is actually achieved despite the initial complications, can be attributed to the combination effect of the Schottky barrier and the broadband layer 12. Also, the photons in the sunlight, which have unit energy above 2.1 electron volts, penetrate into the broadband layer 12, they are absorbed , and they contribute to the generation of the charge carriers. Although the number of such photons in the sunlight ioke is large, namely only about 1/6 of x> «a

7508320-4 ^r-> I^:

OF COURSE

Ά

JU - * /:

fairy# _Γ^·

namely, the number of photons in the sunlight, namely, the energy of the existing states of that electron voltage will generate, under highly efficient conditions, charge carriers instead of being converted to heat, which is attributed to the fact that the bandwidth is narrower than the energy of the said high efficiency photons.

It is already known that in semiconductor material with wide bandwidth, the photons absorbed do not penetrate very deeply into the material. In other words, this means that they are absorbed fairly close to the top surface of the material. In the particular example selected, the majority of high-energy photo booths with an electron voltage greater than 2.1 are absorbed within a depth of 1 µl or less from upper surfaces. on the broadband layer 12, or in other words very close to the thin, semi-transparent film 14. Broadband layer 12 should therefore be made very thin, preferably in the order of 1 µl or less, e.g. a few thousands of angstroms. This results in the charge carriers generated in the doped narrowband layer

1- ', in a manner which will be clarified below, is more readily collected by the semi-transparent metal film 14 to produce an output current. Thus, due to the presence of the metal film 14, an electric current will be present in the region commonly referred to as the space charge region within the subject circuit. As a result, the majority of the charge carriers generated in the layer 12 are displaced to the barrier, thereby effectively contributing to the generation of the electric current.

However, the pooled currents emanating from the photons and generated in the broadband layer 12 are not very high, since the number of high-energy photons in the sunlight is rather small. It is for this reason that one introduced the doped narrowband layer 16. As already pointed out above, its bandwidth is narrower than the bandwidth of the broadband layer 12. In the exemplary embodiment, the narrowband layer 16 is made up of GaAs. Bet therefore has a bandwidth of about 1.4 electron volts. In this case, all photons with an energy below 2.1 electron volts pass through the broadband layer 12 and penetrate the sieve band layer 16, and this is also the case with some of the photons with an energy higher than 2.1 electron volts, the quantity of these photons become dependent on the absorption in and thus also on the thickness of the broadband layer 12. These photons are absorbed and generate charge carriers. In the narrowband layer 16, these photons are absorbed in the vicinity of its upper surface, and this layer can therefore be made rather thin, e.g. a number of μ.

The layers 16 and 12 are so constructed that no barrier should arise between them. Therefore, only a small proportion of the charge carriers generated in the narrowband layer 16, Gig moves through the broadband layer 12 to Cchottky7508320-4 ί I j ι! oarriären. It is therefore possible that, since the broadband layer 12 is so small, the space charge area does not extend all the way to the narrow hand layer j 'However, if so, a great deal of the charge | carriers generated in this layer, not to move through diffusion but similar; well to spread to the barrier under the influence of the electric field, in This will increase the electric current. In any case, a whole portion of the charge carriers generated in the narrow hand layer 16 will pass through in the broadband layer 12 and reach the barrier, thereby participating in current; generation. It is striking that higher output power can be expected from a Sehottky barrier solar cell of the present invention, since through the introduction of layers 12 and 16, the energy of multiple photons with energy amounts over 1.4 electron volts is efficiently converted to generate output current of higher voltages than which would otherwise have been the case in a solar cell with only one layer.

In this context, it should be pointed out in particular that the cell of the present invention is manufacturable on the basis of known technology only.

Therefore, it is not necessary to develop any additional technology, which is the case with p / n cells with widespread materials. In the particular example of the invention described above in connection with Fig. 1 of the drawing, and in which the three active layers 12, 16 and 20 are used, it is first necessary to precipitate the strongly doped GaAs layer 20. Then, sequentially after each other layers 16 and 12. The narrow bandwidth layer 16 has a thickness of the order of several yes, while the broadband layer 12 is considerably much thinner, preferably in the order of about 1 yes or something similar. This precipitation should take place in a vacuum at high temperature.

After the three layers have thus been precipitated, all additional manufacturing steps can be carried out at room temperature and normal pressure, and the properties of the layers 12 and 16 are not adversely affected by them. Therefore, the deposition of the thin metal film 14 as well as the contact surfaces 18 and 19 and the reflection preventive cover 26 can be carried out at room temperature using traditional techniques.

This represents a very important advantage of the present invention. Most of the known semiconductor materials with a large bandwidth are of a very complicated nature, and they are, as mentioned above, usually made up of compound chemical compounds. These tend to have a tendency towards dissociations, at higher temperatures, and this adversely affects their properties, especially the activity length of the diffusion. In order to be able to effectively use so 7508320-4

Therefore, it is necessary that, after they have been manufactured, all the following manufacturing regions be conducted at room temperature or in any case at a fairly low temperature. This becomes possible in the present invention. In contrast, for all known semiconductor materials which can be used in a p / n cell, a fairly high temperature is required for diffusing the material onto this substrate.

You already know how to provide for the epitaxial growth of layers of a semiconductor, when the question is about GaAs. Therefore, it is quite possible to prepare the layers 20, 16 and 12 on the basis of known technology. However, due to the difference in position between the bands of the material in the layers 12 and 16, it is important to allow the layer 12 to develop on top of the layer 16. This reduces the risk of a barrier¹ should be educated between them.

The vertical axis of Figure 2 indicates the existing energy in the electron volts of equal depth in the device of Figure 1, which depth is in turn indicated by the horizontal axis, also indicating the boundaries between the three active layers 14, 1 and 16. The curve E indicates the limit value at which the bar transitions to L »conductive state. It is recalled that, for example, copper is a full conductor, which means that copper has the ability to charge free electrons, which provide current conduction through shock ionization. However, this is not the case with a semiconductor, where no free electrons are present. The curve Εθ therefore indicates the boundary curve for conductive properties of the stack, whereas the area between the curves Εθ and indicates the range within which the stack is semi-conductive. The curve Ey is usually referred to as the valence curve. Thus, between these two curves there is an energy gap E E, and preferably there is worked approximately in the middle of this energy gap, as indicated by the mean curve E ^.

However, the temperature also constitutes a quantity of significance to that shown in Fig. 2^- feature. Namely, at high temperature, the electrons gain the ability to pass past the energy gap E, so that the stack obtains conductive G properties. The semiconductor will then become a full leader. However, it should be noted that the values ​​of E_ and E vary depending on it

The VV investigated the distance of the site from the boundary plane between layers 14 and 12, and that this variation occurs after a complicated curve. The curve becomes even more complicated if, in the manner described above, a barrier is built up in that the curve Ey is subjected to a hump 3θ in this barrier, while the curve obtains an approximately asymptotic tip 31. These conditions apply in principle to all semiconductors. , and they are described in more detail in The Energy-Band Model of a Semiconductor, pages 81-86,

The latter conditions find their explanation in that the material in the layer 12 flows to some extent into the material in the layer 16, so that a stable

7508320-4 transition exists, so the curves do not become completely continuous.

The mean value curve is usually considered very closely representative of the properties of an intricate semiconductor as opposed to the properties of an extricate semiconductor. It is seen that this mean curve is very near free from the above-mentioned discontinuity.

A theoretical analysis now provides that if the narrow band 16 of GaAs, preferably made, has a bandwidth of about 1.4 electron volts, and the broadband layer 12 consists of a ternary compound of GaAs with a bandwidth of about 2.1 electron volts, it will not to create any barrier in the boundary layer of the kind shown in Fig. 2 at 30. On the other hand, a discontinuity or an abrupt interruption in the conductive properties band may occur in the manner shown in Fig. 2 at 31. However, this phenomenon can be avoided by gradually passing the narrow band layer 16 into the ternary chemical compound of GaAs. a thickness of a few hundred Angstroms or the order of magnitude thereof, after which you continue with a⁴-1 build up the broadband layer 12 so that it 'maintains the required bandwidth', which can easily be achieved by regulating the temperature and by simultaneously increasing the amount of the additive added to l.

GaAs for formation of the ternary compound. The latter additive may, for example, consist of aluminum (Al).

It is now evident that even if, in the above-described material, the layer is of large bandwidth as if it had the general formula AlGaAs, other chemical compounds, e.g. GaAs ^_χΡ_χ, which ternary compound of GaAs with phosphorus has a bandwidth which is quite large. For example, it becomes equal to 2.0 electron volts, if you choose the factor x to 0.5. Semi-conductive materials other than GaAs and its ternary chemical compounds can also be used for the broadband layer 12 as well as for the narrowband layer 16, if only a barrier layer is formed between the two layers. When choosing materials in both layers, consideration should be given to the environment in which the solar cell will work in the future. For example, aluminum is used to generate the ternary compound (Al), and the factor x is chosen very large. Such a solar cell is also resistant to working in a highly humid atmosphere.

The thin metal film 14 can be made of many different known metals. Examples of suitable such metals include gold, palladium and platinum, which have been found to be very suitable for forming the metal film 14 when the two layers 16 and 12 are formed in the form of an n-type ternary compound.

in

7508320-4

-10. iaisa: -Ά &

<img file="SE405185B_D0001.tif" />