DE3135411C2 - Photosensitive device with superimposed pin layers of an amorphous silicon alloy material - Google Patents

Description

The invention relates to a photosensitive arrangement the genus mentioned in the preamble of claim 1. A such an arrangement is already known (IEEE transactions on Electron Devices, Volume ED-27, April 1980, pages 662-670). The known arrangement consists of amorphous a-silicon and is used as a solar cell. Thereby, hydrogen and fluorine into the amorphous material with a photosensitive (intrinsic) area continuously changing amount built. Do it Hydrogen and fluorine both reduce the density of sturgeon cells and change the Band gap, which is thereby between 1.5 and 2.0 eV can be set.

In addition, it is also known (U.S. Patent 4,217,374), amorphous Semiconductor films that contain hydrogen and fluorine to combine different elements. So for the amorphous Semiconductor gas matrix u. a. Germanium and carbon used, of which is known (JP 55-13938A and  JP 55-13939A) that they change the band gap.

The invention has for its object the efficiency a photosensitive arrangement of amorphous structures at the beginning to enlarge the genus mentioned with simple means.

The invention is characterized in claim 1. In subclaims further training of the same is claimed and in the description are particularly preferred exemplary embodiments described.

According to the invention, carbon and / or Germanium in a proportion of significantly more than 20% used; the setting element or the setting elements are within the intrinsic photosensitive area in different amounts across the layer thickness distributed.  

Apart from the continuously changing known per se Quantity distribution gives advantages even if none continuous change in the distribution of the adjusting element takes place, but the distribution in stages and / or with interruptions across the layer thickness of the intrinsically conductive photosensitive Area.

The invention is applicable not only to solar cells, but for example also in xerography and for photo detectors and photodiodes including large-area photodiode array.

The arrangement according to the invention is characterized by improved spectral responsiveness as well as an improved collection the photoelectrically generated carrier.

That or the setting elements germanium and carbon can by vapor deposition, atomization or by glow discharge become. In addition, others can also Adjustment elements such as tin and nitrogen are used.

The photosensitive arrangement can also consist of several superimposed ones There are stacking of layer combinations.

Preferred embodiments of the invention will now be described of the drawings explained. Show:

FIG. 1 shows one type of a PIN solar cell having a graded band gap;

FIG. 2 shows an application in which the precipitation of the amorphous alloy and the application are carried out of the activated fluorine and hydrogen in separate steps and in separate containers;

Fig. 3 is a sectional view of an embodiment of a Schottky barrier solar cell;

Fig. 4 is a sectional view of a layered pin solar cell;

Fig. 5 is a sectional view of a laminated nip solar cell;

Fig. 6 shows a device by means of which amorphous alloys together with one or more of the setting elements by means of a plasma-activated vapor can be deposited; and

Fig. 7, in which the property for various applications available photosensitive standard sunlight wavelengths λ shown in a solar spectral exposure diagram.

In Fig. 1, a PIN solar cell is shown, which has a graduated band gap. The bandgap of the intrinsically conductive area is graded from an n⁺ range to an n⁺ range of 1.2 eV to 1.8 eV, which is achieved by adding one or more adjusting elements that set the bandgap. The intrinsic region i can be formed by the deposition of amorphous Si: F. The smallest band gap of 1.2 eV can be achieved by adding germanium during the precipitation of the intrinsically conductive area. From this value, the bandgap can be increased by reducing the amount of germanium added during the precipitation of the intrinsic layer. The greatest value of the band gap of 1.8 eV can be achieved in that the addition of the adjusting element or elements is terminated towards the end of the precipitation of the intrinsically conductive layer, even before the p⁺ layer is deposited. The bandgap can be further increased by adding additional setting elements, such as nitrogen or carbon, which means that the bandgap can reach a value of up to 1.8 eV and more close to the p⁺ range. The nitrogen can be added in the form of ammonia (NH₃), while the carbon can be added in the form of methane (CH₄). The addition of the band gap adjusting element or elements to achieve a graded band gap can be carried out while maintaining a high quality of the electronic properties and the photoconductivity in the cell as a result of the inclusion of fluorine (and hydrogen) in the alloy.

The graded bandgap improves use the solar energy for the generation of electron-hole pairs reached. Because that or the adjustment elements in any quantity Different forms of gradation can or can be achieved: The gradation can be continuous, with interruptions or be done in stages. Also a gradation the n⁺ and p⁺ layers are of the same type and Possible or by selective doping.

From Fig. 1 it can be seen that the valence band between the n⁺ layer and the p⁺ layer is inclined. This slight inclination allows the photon generated holes to migrate towards the p⁺ layer for collection. Such an inclination can be achieved by selective doping in small amounts during the precipitation of the intrinsic layer. Instead, it can also be achieved by building a space charge. If there is an area with a space charge, the holes created can migrate freely towards the p⁺ layer.

By grading the band gap in the manner shown in FIG. 1, there is a slight reduction in the open circuit voltage compared to an alloy with a uniform bandwidth of 1.8 eV, since the smallest band gap of the graded alloy is essentially decisive for this value. However, the short-circuit current is increased as a result of the more efficient use of solar energy to generate the electron holes and as a result of the more efficient collection of the carriers produced. The operation of the entire arrangement is therefore improved in particular for those applications in which higher currents and lower voltages are required or accepted. Increasing the bandgap to more than 1.8 eV also increases the amount of sunlight available for use in the array.

In the production of the photosensitive arrangement, the further elements can be inserted into the amorphous alloy during their precipitation, but it may also appear advisable to carry out the precipitation of the amorphous alloy and the injection of the further elements into the semiconductor alloy completely separately from one another. This can be advantageous for certain applications because the conditions for the injection of such agents are then completely independent of the conditions for the precipitation of the alloy. Thus, the porosity of the alloy, if a porous alloy is produced during the vapor deposition, can in some cases be reduced much more easily by environmental conditions which differ significantly from those of a normal vapor deposition process. With reference to FIG. 2, it can be stated that it is indicated schematically here that the deposition of the amorphous alloy is carried out completely separately from the diffusion of the other elements.

Fig. 3 shows a Schottky barrier solar cell 142, 144 has a substrate made of a material having a good electrical conductivity and the ability to produce an ohmic contact with an amorphous silicon alloy 146th The substrate 144 is made of a metal of a small work function, such as aluminum, tantalum, stainless steel, or other material that matches the amorphous silicon alloy 146 deposited on the substrate 144 . The alloy has a small density of states in the energy gap of preferably not more than 10¹⁶ per cm³ per eV. It is envisaged that the alloy has a region 148 near the substrate 144 which has an n⁺ conductivity, which is heavily doped and forms a low-resistance boundary layer between the electrode and an undoped region 150 with a relatively high dark resistance, the Area 150 is an intrinsically conductive area with a small n-conductivity and has a graded band gap, as described above.

The upper surface of the amorphous alloy 146 adjoins a metallic region 152 , the boundary layer between this metallic region and the amorphous alloy 146 forming a Schottky barrier layer 154 . The metallic region 152 is transparent or semi-transparent to the solar radiation, has good electrical conductivity and has a high work function of, for example, 4.5 eV and more when using, for example, gold, platinum, palladium and the like. The like compared to that of the amorphous alloy 146 . The metallic region 152 can be a single metal layer or can consist of several layers. The amorphous alloy 146 can have a thickness of approximately 0.5 to 1 µm and the metallic region 152 can have a thickness of approximately 10 nm in order to be semi-transparent to solar radiation.

A grid electrode 156 made of a metal with a good electrical conductivity is placed on the surface of the metallic region 152 . The grid electrode can consist of orthogonally arranged lines of a conductive material and occupy only a small portion of the area of the metallic area, the rest of which is exposed to solar energy. In the exemplary embodiment, the grid electrode 156 takes up only about 5 to 10% of the total area of the metallic region 152 and collects the current uniformly from the metallic region 152 , so as to ensure a good low series resistance for the arrangement.

An anti-reflection layer 158 can be arranged over the grid electrode 156 and the surfaces of the metallic region 152 that are not covered by the grid electrode 156 . The antireflection layer 158 has a surface 160 exposed to solar radiation. This antireflection layer 158 has a thickness on the order of the wavelength of the maximum energy point of the solar radiation spectrum divided by four times the refractive index of the antireflection layer. If the metallic region 152 consists of platinum with a thickness of 10 nm, then zirconium oxide with a thickness of approximately 50 nm and with a refractive index of 2.1 is suitable for the antireflection layer 158 .

The adjusting element or elements serving to adjust the band gap are added to the region 150 producing a luminous flux in varying amounts during its precipitation. The Schottky barrier layer 154 , which is formed in the boundary layer between the regions 150 and 152 , separates the charge carriers in the alloy 146 generated from the photons of the solar radiation, which are collected as current through the grid electrode 156 . An oxide layer can also be inserted between the layers 150 and 152 in order to form an MIS solar cell.

In addition to the Schottky junction or MIS solar cell shown in FIG. 3, it is also possible to create solar cells with a pin layer in the amorphous alloy, with successive precipitation, compensation or change and doping. The solar cells of these designs are shown schematically in FIGS. 4 and 5.

In FIG. 4, a pin solar cell 198 is shown comprising a glass or consisting of a flexible fabric made of stainless steel or aluminum substrate 200. The substrate 200 has a width and a length desired for the particular application and is preferably at least 75 µm (3 mils) thick. An insulating layer 202 is applied to the substrate 200 , which can be done chemically, by vapor deposition or an anodic treatment if aluminum is used for the substrate. The layer 202 has a thickness of 5 μm and can consist of a metal oxide. If the substrate is made of aluminum, then layer 202 should be aluminum oxide (Al₂O₃); if the substrate is made of stainless steel, then the layer 202 can be made of silicon dioxide (SiO₂) or another suitable glass.

An electrode 204 is disposed in one or more layers above layer 202 to form a base electrode for cell 198 . The electrode 204 is evaporated onto the layer 202 , which can be carried out relatively quickly. It is formed from a reflective metal, such as molybdenum, aluminum, chrome, or stainless steel, when the arrangement is used as a solar cell or as a photovoltaic device. Such a reflective electrode is preferred from the point of view that in a solar cell all light not absorbed by the semiconductor alloy is then reflected by such an electrode, so that it is passed through the semiconductor alloy again, which consequently then absorbs further light energy , which improves the efficiency of the solar cell in a known manner.

The substrate 200 then experiences the actual coating with the amorphous alloy. FIGS. 4 and 5 show various pin-laminations. In both cases the alloy has a total thickness between approximately 300 nm and 3 µm. This thickness ensures that the structures have no holes or other physical defects, with which maximum light absorption can also be achieved. A thicker material can absorb even more light, but it can be assumed that no further current is generated from a certain thickness, since the larger thickness allows an increased recombination of the electron-hole pairs generated by the light. (It should be noted that the thickness of the various layers shown in Figs. 3-5 is not drawn to scale.)

In the pin solar cell 198 , a heavily doped alloy layer 206 of the n⁺ type is initially deposited on the electrode 204 . This layer 206 is covered by an intrinsically conductive i-layer 208 , on which in turn a highly doped alloy layer 210 of the p⁺ type is deposited. The individual alloy layers 206, 208 and 210 form the active layers of the pin solar cell 198 .

Although this solar cell is also used for other purposes can be used below for use as a photovoltaic device are described in more detail.

For use as a photovoltaic device, the outer layer 210 has low light absorption and high conductivity. The intrinsic alloy layer 208 has a graded wavelength threshold for solar photosensitivity, high light absorption, low dark conductivity and high photoconductivity, the bandgap of this layer being graded by sufficient and varying amounts of one or more of the adjustment elements. The lower alloy layer 206 has a small light absorption and a high conductivity. The total thickness of these three layers 206, 208 and 210 is, as mentioned above, at least about 300 nm. The thickness of the layer 206 is about 5 to 50 nm. The thickness of the intrinsically conductive layer 208 is about 300 nm to 3 μm. Finally, the thickness of the layer 210 is approximately 5 to 50 nm. As a result of the shorter diffusion length of the holes, the upper layer 210 should be as thin as possible and be of the order of approximately 5 to 15 nm. Such a smallest possible thickness of the outer layer, which can also be an n⁺ layer instead of a p⁺ layer, is also preferred from the point of view of avoiding light absorption, which is why no elements that adjust the band gap are generally inserted into this layer will.

The solar cell 212 according to FIG. 5 is modified in that a p⁺ layer 214 is initially deposited on the electrode 204 , which is covered by an intrinsically conductive layer 216 which, for a gradation of the band gap, changes amounts of one or more of the setting Contains items. An n-alloy layer 218 is arranged over the layer 216 and an outer n⁺-alloy layer 220 is arranged over this. While the intrinsically conductive alloy layers 208 and 216 of the two embodiments are formed from an amorphous alloy, the other layers can also be polycrystalline, in particular layer 214 . In the two embodiments, the order of the individual layers can also be reversed.

As soon as the individual semiconductor alloy layers are deposited in the desired order, further precipitation is preferably carried out in a separate environment. This further precipitation is preferably an evaporation process because this can be carried out the fastest. A TCO layer 222 , ie a transparent conductive oxide layer, is applied, which can be, for example, indium tin oxide (ITO), cadmium stannate (Cd₂SnO₄) or doped tin oxide (SnO₂). This TCO layer is applied after the fluorine and hydrogen have been added, if the films have not been deposited with these elements. The setting elements can also only be added later.

If desired, an electrode grid 224 can also be applied to the TCO layer 222 . If the array has a sufficiently small area, then the TCO layer 222 is generally sufficiently conductive so that the grid 224 is then not needed to maintain good device efficiency. If, on the other hand, the device has a large area or the conductivity of the TCO layer 222 is insufficient, the arrangement of such a grid 224 has the advantage that it shortens the path of the charge carriers and thus improves the efficiency of the arrangement.

FIG. 6 shows a device which allows plasma-activated evaporation within a chamber 226 . A controller 228 is used to control various parameters such as pressure, flow rates, etc. The pressure in chamber 226 is set to about 0.13 Pa or less in this case.

One or more lines, such as lines 230 and 232 , serve to feed the reaction gases, such as silicon tetrafluoride (SiF₄) and hydrogen (H₂), into a plasma area 234 . The plasma region 234 is formed between a coil 236 connected to a direct current source and a plate 238 . The plasma activates the supplied gas or gases to precipitate activated fluorine and hydrogen onto a substrate 240 . The substrate 240 can be heated to the temperature desired for the precipitation by means of a heating device 44 .

The adjustment element or elements and silicon are supplied from evaporation containers 242 and 244 . For example, container 242 may contain germanium and container 244 may contain silicon. The elements can be vaporized by means of an electron beam or other heating device, and the elements are then activated by the plasma.

A closure 246 can be used to change the amounts of the adjustment element (s). This closure 246 is rotatably arranged and its speed can be controlled to change the amount of the bandgap adjusting element (s) for grading the bandgap. It is also possible to deposit the element or elements used to adjust the band gap in varying amounts stepwise or continuously in discrete layers.

In Fig. 7, available to sunlight spectrum is shown. Air mass O (AMO) denotes the sunlight that is available outside the atmosphere, with the sun then being directly above. Air mass AM1 denotes the sunlight that is available after filtering through the earth's atmosphere. Crystalline silicon is an indirect semiconductor with a band gap of approximately 1.1 to 1.2 eV, which corresponds to the wavelength λ of approximately 1.0 µm. For light with a wavelength above the threshold defined by the bandgap, the photon energies are not sufficient to generate a pair of charge carriers, and consequently they do not supply any additional current.

The bandgap referred to in the present specification is defined as the extrapolated section of a curve with the equation (αℏω) ^1/2 , where α is the absorption coefficient and ℏω is the photon energy.

Another photosensitive application arises for the Laser wavelengths, for example an infrared response to obtain. A photosensitive material that with a fast electrophotographic computer output device with a laser, such as a helium-neon laser should have a wavelength threshold of about 0.6 µm  to have. For fiber optic use such as with GaAs lasers, the threshold of the photosensitive Material is about 1 micron or less. The addition one or more setting elements allows the provision of alloys with a graded bandwidth according to the desired application.

Each of the semiconductor alloy layers can be applied to one Base electrode forming substrate by means of the glow discharge be put down using a chamber that is described in US Pat. No. 4,226,898 or DE-PS 31 35 393 or DE-PS 31 35 412. The alloy layers can also be in a continuous process be put down. In these cases, the system the glow discharge is initially evacuated to about 0.13 Pa, to rinse out the chamber and remove any contaminants to remove the atmosphere that the precipitation system forms. The alloy material is preferably silicon tetrafluoride (SiF₄), hydrogen (H₂) and German (GeH₄). The plasma of the glow discharge is preferably from the gas mixture receive. The system according to U.S. Patent 4,226,898 is preferably at a pressure on the order of about 40 to 200 Pa, preferably at a pressure between 80 and operated at 130 Pa and in particular at 80 Pa.

The semiconductor material is preferably on the an infrared device on that for precipitation each alloy layer desired temperature heated substrate knocked out from a self-sustained plasma. The doped layers of the devices are used in different Temperatures in the range between 200 and about Knocked down 1000 ° C, depending on the shape used Material is dependent. The upper limit of the substrate temperature is partly dependent on the type of for Substrate used metal. For aluminum, the top Temperature should not be more than about 600 ° C while for stainless steel they are also above about 1000 ° C can. For an initially compensated with hydrogen  amorphous alloy that is intrinsically conductive Forms layer in an n-i-p or in a p-i-n arrangement, the temperature of the substrate should be less than about 400 ° C and preferably about 300 ° C.

The concentrations of the dopants become dependent chosen what conductivity for each Layers of the p, p⁺, n or n⁺ type during the precipitation of the individual alloy layers are desired. For n- or p-doped layers, this becomes Material with between 5 and 100 ppm of the dopant endowed. For n⁺- or p⁺-doped layers of The doping material is used in quantities between 100 ppm to over 1% used.

The glow discharge precipitation can be a of an alternating current generated plasma, in which the materials are inserted. The plasma is there preferably between a cathode and a through maintain the substrate formed anode, the AC has a frequency of about 1 kHz to 13.6 MHz.  

For one embodiment, a mixture of the gases GeH₄, Ar, SiF₄ and H₂ provided, in which the relative proportion from GeH₄ to SiF₄ in the order of 0.001 to 1% photosensitive alloys with the graduated bandgap generated without the alloys having their desired electronic Lose properties. When the mixture is the ratio of SiF₄ to H₂ between 4 to 1 and 10 to 1. The amount of the setting element (s) in the resulting alloy is much larger than the gas content and is dependent of the application significantly above 20%. Argon is useful as a diluent, but not essential for the implementation of the procedure.

Although that or the setting elements at least for a gradation of the photosensitive Range of arrangements are provided, they can also useful for the other alloy layers of the assemblies his. In addition to the intrinsic area all other alloy layers other than amorphous Layers, such as polycrystalline layers, with the expression "amorphous" an alloy or a material what is meant is a long-range disorder, although it's a short or intermediate order can sometimes have crystalline inclusions.

Claims (6)

1. Photosensitive arrangement with
superimposed pin layers of an amorphous silicon alloy material containing both hydrogen and fluorine, in order to reduce the density of states in the bandgaps; and
at least one adjusting element in different amounts across the layer thickness of the intrinsically conductive photosensitive region in order to adjust the bandwidth and to improve the radiation absorption without changing the impurity state density in the band gap,
characterized in that carbon (C) and / or germanium (Ge) is used in a proportion of significantly more than 20%. 2. Arrangement according to claim 1, characterized in that the distribution of the adjusting element in the photosensitive region ( 208, 216 ) changes continuously over the layer thickness. 3. Arrangement according to one of the preceding claims, characterized in that the band gap within the photosensitive region ( 208, 216 ) is set significantly less than 1.6 eV. 4. Arrangement according to one of the preceding claims, characterized in that the layer thickness of the photosensitive region ( 208, 216 ) is approximately 0.3 to 3 µm. 5. Arrangement according to one of the preceding claims, characterized in that the adjusting element is distributed in stages in the photosensitive region ( 208, 216 ) in different amounts. 6. Arrangement according to one of the preceding claims, characterized in that the adjusting element with interruptions in the photosensitive area ( 208, 216 ) is distributed in different amounts.