Structure containing hydrogenated amorphous silicon and process.

0001The present invention is concerned with a structure for providing semiconductor devices and to a process for fabricating such. The present invention is especially applicable to structures that, in turn, are suitable for providing thin-film field effect transistors. The present invention provides a relatively simple and readily reproducible process for obtaining such struc­tures.

0002In obtaining semiconductur devices, it is often ne­cessary to obtain a predetermined pattern of thicker and thinner layers of semiconductor material. Such pattern is generally more useful if the semiconductor material is a deposited thin film (as opposed to bulk grown crystal) material, for example, hydrogenated amorphous silicon, or crystalline silicon formed by the high temperature or laser annealing of the amorphous material. One particular application be­ing thin-film field effect transistors.

0003A typical thin-film field effect transistor includes an electrically conductive gate above an insulating substrate such as glass, a gate insulator such as silicon dioxide above the gate, a hydrogenated amorphous silicon layer above the gate insulator, an electrically conductive source and drain regions above the amorphous silicon. It is desired for such devices to control the current such that when the device is in the "on" state, as high a current as possible exists, but when the device is in the "off" state, the residual current is as small as possible. To achieve such it is desirable to provide doping in the amorphous silicon layer beneath the source and drain regions, but to limit the doping in the amorphous silicon layer in the region between the source and drain regions referred to as the "channel".

0004Presently, such patterns are fabricated by plasma depositing a layer of uniform thickness hydrogenated amorphous silicon, protecting the desired thicker part of the pattern with a resist image, and etching the unprotected area for a predetermined length of time. However, in order to achieve a well-­defined remaining thickness in the etched areas, meticulous control of conditions and of timing is required. Such is relatively expensive and often not practical to implement. Accordingly, these methods are not very well developed and not convenient from a manufacturing or commercial viewpoint.

0005The present invention is concerned with a structure and process capable of providing a pattern multilevel structure that is relatively simple and readily re­producible. In particular, the structure of the pre­sent invention includes a substrate (1) that has at least one major surface area. Above the major surface area of the substrate is a first layer (4) of hydro­genated amorphous silicon. The first layer of hydro­genated amorphous silicon contains a first concentration of hydrogen incorporated therein. Located above the first layer of hydrogenated amorphous silicon is a second layer (5) of hydrogenated amorphous silicon. The second layer of hydrogenated amorphous silicon has a second concentration of hydrogen incorporated there­in that differs from the concentration of hydrogen in­corporated in the first layer.

0006In addition, the present invention is concerned with a process for fabricating a structure. The process in­cludes providing a substrate (1) having at least one major surface area. Above the major surface area of the substrate is provided a first layer (4) of hydro­genated amorphous silicon. The first layer of hydro­genated amorphous silicon has a first concentration of hydrogen incorporated therein. A second layer (5) of hydrogenated amorphous silicon is provided above the first layer. This second layer has a second concentration of hydrogen incorporated therein that differs from the concentration of hydrogen incorporated in the first layer. A vertically differentiated pattern is provided by selectively etching the layer of hydrogenated amorphous silicon having the higher hydrogen concentration of the first and second layers. <ul id="ul0001" list-style="none"><li>Figures 1 to 4 are schematic diagrams showing the structure in various stages of processing in accordance with the present invention.</li><li>Figures 5 to 8 illustrate the application of the inven­tion in fabricating thin film field effect transistors.</li></ul>

0007In Figure 1 is illustrated a substrate (1).

0008The substrate (1) includes any suitable substrate ma­terial such as glass (e.g., quartz), sapphire, sili­con, metal, or metallized substrate.

0009A first layer (4) of hydrogenated amorphous silicon containing a first concentration of hydrogen incor­porated therein is deposited (see Figure 2). In the preferred aspects of the present invention, hydro­genated amorphous silicon layer (4) contains a re­latively low concentration of hydrogen as compared to the subsequently to be applied hydrogenated amor­phous silicon layer (5). The hydrogenated amorphous silicon layer (4) can be provided by well-known plasma deposition techniques such as placing the structure in a plasma reaction chamber using silane as the source of the hydrogenated amorphous silicon. For the lower con­centration of hydrogen, it is preferred to use tempe­ratures of about 225°C to about 325°C and most prefer­ably about 275°C and at a pressure of about 20 µbar (15 millitorr) to about 1.33 mbar (1 torr) and most preferably about 306µbar (230 millitorr). A typical power density is about 5 milliwatts/cm² of combined surface area of the electrodes exposed to the plasma. The power source is typically operated at a radio fre­quency of about 13.6 megahertz.

0010Although the preferred source of the hydrogenated amor­phous silicon is 100% silane, such, if desired, can be diluted with an inert gas such as helium, neon, argon, and krypton or diluted with hydrogen. It is known that the presence of hydrogen in the diluent gas will not significantly effect the amount of hydro­gen deposited along with the silicon in the hydro­genated amorphous silicon layer.

0011Moreover, other methods for depositing the hydrogenated amorphous silicon such as those disclosed in U.S. Patent 4,363,828 can be employed if desired.

0012Next, the second layer (5) of hydrogenated amorphous silicon is deposited (see Figure 3). In the preferred aspects of the present invention the second layer of hydrogenated amorphous silicon (5) contains the greater quantity of hydrogen of the two layers. The increased quantity of incorporated hydrogen can be achieved by the same plasma deposition employed for the first hydro­genated amorphous silicon layer, except by employing a reduced temperature and/or increased pressure. In the preferred aspects of the present invention, the tempe­rature employed for the second hydrogenated amorphous silicon layer is reduced to about room temperature to about 200°C and preferably about 125°C while employing the same pressure as used for depositing the first hydrogenated amorphous silicon layer (4).

0013The thicknesses of the two layers may be chosen at will within the limits of practicality of thin films depositions. Important benefits of the present inven­tion as compared to the prior art will, however, derive when the second layer is thicker or not much thinner than the first layer and when both layers are relative­ly thin. Thus, typical thicknesses of the first hydro­genated amorphous silicon layer (4) are about 10 to about 5,000 nm and preferably about 10 to about 100 nm.

0014Typical thicknesses of hydrogenated amorphous silicon layer (5) are about 5 to about 5,000 nm and preferably about 10 to about 100 nm.

0015The higher concentration of hydrogen in the second hydrogenated amorphous silicon layer results in that layer being more rapidly etched, thereby making it possible to accurately etch down to the interface with first hydrogenated layer (4) allowing for a well-de­fined remaining thickness as well as a flat surface.

0016Accordingly, the second hydrogenated amorphous silicon layer (5) can be patterned (see Figure 4), for instance, by covering the layer (5) with a photoresist material (not shown) exposing the photoresist material to imaging radiation in a predetermined pattern and, thereafter, removing that portion of the photoresist material ex­posed to the imaging radiation in the case of a positive photoresist material and that material not exposed to radiation in the case of a negative photoresist material as is well-known in the art.

0017Next, the portion of the second hydrogenated amorphous silicon layer (5) no longer protected by photoresist material is exposed to a suitable etchant in order to develop the desired pattern in the layer (4). Suitable wet chemical etching compositions include strongly alkaline solutions such as alkali metal hydroxide solutions and, in particular, potassium hydroxide and sodium hydroxide. Such compositions preferably are 1 molar or higher. The etching is preferably carried out at normal room temperatures. When more elevated tempera­tures are employed, the pH of the solution can be some­what less than that obtained from a 1 molar solution, such as down to about 12.5. Moreover, if desired, other methods of etching can be employed, although not preferred, such as plasma etching or reactive ion etching. The most pronounced effects from the present invention are achieved when employing a wet chemical etching.

0018In view of the fact that the second layer (5) has the higher hydrogen content and, therefore, etches more rapidly than the first layer, the etching can be carried out such that the etch rate will slow down as the interface to layer (4) is reached, thereby allow­ing a well-defined remaining thickness and a planar surface.

0019The level of hydrogen in hydrogenated layers (4) and (5) can now be equilibrated once layer (5) has been patterned by merely raising the temperature of the substrate to that at which the first layer was formed, such as at about 225°C to about 325°C and preferably at about 275°C and maintain the substrate there for a sufficient amount of time in order to lower the hydro­gen content in the second layer. This usually takes about 1 to about 30 minutes.

0020Reference to Figures 5 to 8 illustrates the application of the present invention in fabricating thin film effect transistors.

0021In Figure 5 is illustrated a substrate (1) containing an electrically conductive gate (2) thereon and a gate insulator (3).

0022The substrate (1) includes any suitable substrate ma­terial such as glass (e.g., quartz), sapphire, silicon, metal, or metallized substrate.

0023The gate (2) is of a metallic-type high electrical conductivity material, preferably a metal such as chromium, nickel, molybdenum and aluminum, as well as non-metallic materials such as highly doped poly­crystalline silicon or intermetallic silicides which, nevertheless, have electrical conductivities of the magnitude generally possessed by metals. The gate (2) is defined by well-known photolithographical techniques that need not be discussed herein in any detail.

0024Electrically insulating gate (2) is a gate insulator (3) such as silicon dioxide and silicon nitride.

0025A first layer (4) of hydrogenated amorphous silicon containing a first concentration of hydrogen incor­porated therein is deposited (see Figure 6). In the preferred aspects of the present invention, hydroge­nated amorphous silicon layer (4) contains a relatively low concentration of hydrogen as compared to the sub­sequently to be applied hydrogenated amorphous silicon layer (5). The hydrogenated amorphous silicon layer (4) can be provided by well-known plasma deposition tech­niques such as placing the structure in a plasma re­action chamber using silane as the source of the hydrogenated amorphous silicon. For the lower concen­tration of hydrogen, it is preferred to use tempera­tures of about 225°C to about 325°C and most preferably about 275°C and at a pressure of about 20 µbar (15 millitorr) to about 1.33 mbar (1 torr) and most pre­ferably about 306 µbar (230 millitorr). A typical power density is about 5 milliwatts/cm² of combined surface area of the electrodes exposed to the plasma. The power source is typically operated at a radio frequency of about 13.6 megahertz.

0026Although the preferred source of the hydrogenated amor­phous silicon is 100% silane, such, if desired, can be diluted with an inert gas such as helium, neon, argon, and krypton or diluted with hydrogen. It is known that the presence of hydrogen in the diluent gas will not significantly effect the amount of hydrogen deposited along with the silicon in the hydrogenated amorphous silicon layer.

0027Moreover, other methods for depositing the hydrogenated amorphous silicon such as those disclosed in U.S. Patent 4,363,828 can be employed if desired.

0028Next, the second layer (5) of hydrogenated amorphous silicon is deposited (see Figure 3). In the preferred aspects of the present invention the second layer of hydrogenated amorphous silicon (5) contains the greater quantity of hydrogen of the two layers. The increased quantity of incorporated hydrogen can be achieved by the same plasma deposition employed for the first hydro­genated amorphous silicon layer, except by employing a reduced temperature and/or increased pressure. In the preferred aspects of the present invention, the temperature employed for the second hydrogenated amorphous silicon layer is reduced to about room temperature to about 200°C and preferably about 125°C while employing the same pressure as used for deposi­ting the first hydrogenated amorphous silicon layer (4).

0029As discussed above, the thickness of the two layers may be chosen at will within the limits of practicality of thin films depositions. Important benefits of the present invention as compared to the prior art will, however, derive when the second layer is thicker or not much thinner than the first layer and when both layers are relatively thin. Thus, typical thicknesses of the first hydrogenated amorphous silicon layer (4) are about 10 to about 5,000 nm and preferably about 10 to about 100 nm.

0030Typical thicknesses of hydrogenated amorphous silicon layer (5) are about 5 to about 5,000 nm and preferably about 10 to about 100 nm.

0031In addition, for the preferred aspects of the present invention, when the present invention is to be used to provide a thin-film field effect transistor, the se­cond hydrogenated amorphous silicon layer is also doped in order to increase its conductivity. A suitable n-type dopant is phosphorous and a suitable p-type dopant is boron. The phosphorous can be incorporated, for instance, by including in the plasma gas, phosphine (PH₃), such as in amounts of a few ppm to about 1% by volume of the gaseous mixture employed. The boron can be provided by using a gaseous boron-containing compound such as diborane (B₂H₆).

0032The higher concentration of hydrogen in the second hydro­genated amorphous silicon layer results in that layer being more rapidly etched, thereby making it possible to accurately etch down to the interface with first hydrogenated layer (4) allowing for a well-defined remaining thickness as well as a flat surface.

0033Accordingly, the second hydrogenated amorphous silicon layer (5) can be patterned (see Figure 4), for instance, by covering the layer (5) with a photoresist material (not shown) exposing the photoresist material to imaging radiation in a predetermined pattern and, thereafter, removing that portion of the photoresist material exposed to the imaging radiation in the case of a positive photoresist material and that material not exposed to radiation in the case of a negative photoresist material as is well-known in the art.

0034Next, the portion of the second hydrogenated amorphous silicon layer (5) no longer protected by photoresist material is exposed to a suitable etchant in order to develop the desired pattern in the layer (4). Suit­able wet chemical etching compositions include strongly alkaline solutions such as alkali metal hydroxide solu­tions and, in particular, potassium hydroxide and sodium hydroxide. Such compositions preferably are 1 molar or higher. The etching is preferably carried out at normal room temperatures. When more elevated tempera­tures are employed, the pH of the solution can be some­what less than that obtained from a 1 molar solution, such as down to about 12.5. Moreover, if desired, other methods of etching can be employed, although not preferred, such as plasma etching or reactive ion etching. The most pronounced effects from the present invention are achieved when employing a wet chemical etching.

0035In view of the fact that the second layer (5) has the higher hydrogen content and, therefore, etches more rapidly than the first layer, the etching can be carried out such that the etch rate will slow down as the interface to layer (4) is reached, thereby allowing a well-defined remaining thickness and a planar surface.

0036That portion of layer (5) that remains is patterned so as to beneath source and drain regions to be sub­sequently provided, while that portion of layer (5) removed corresponds to the area between the source and drain regions to be provided, referred to as the "channel". Accordingly, by proper doping of layer (5), increased conductivity beneath the source and drain regions can be obtained without concomitantly increas­ing the conductivity between the source and drain regions when the device is in the "off" state. This, in turn, allows for improved ohmic contact. In other words, increased ohmic contact is provided by the doping between the hydrogenated amorphous silicon and the source and drain regions, but the highly doped amor­phous layer is cleanly removed from the channel region between the source and drain regions. Source and drain regions (7) and (8) are provided by well-known techni­ques, for examples, by depositing a blanket metallization (not shown) prior to application of the photoresist on top of layer (5) and patterning it with the mask as used for patterning layer (5).

0037The level of hydrogen in hydrogenated layers (4) and (5) can be equilibrated once layer (5) has been patterned by merely raising the temperature of the substrate to that at which the first layer was formed, such as at about 225°C to about 325°C and preferably at about 275°C and maintain the substrate there for a sufficient amount of time in order to lower the hydro­gen content in the second layer. This usually takes about 1 to about 30 minutes.

0038Depending on application, the etched region can be covered with a passivation or encapsulation layer (6).

0039Although the present invention has been described with respect to only two different layers of hydrogenated amorphous silicon, it is understood that the present invention can be carried out with three or more diffe­rent layers or thicknesses of hydrogenated amorphous silicon, each having a different concentration of hydrogen as compared to the layer juxtaposed it.

0040In order to demonstrate the significant variation in the etching rate of different hydrogenated amorphous silicon layers, the following examples are presented.

Example 1

0041A number of samples are prepared whereby hydrogenated amorphous silicon is deposited using the same apparatus and conditions, except for the substrate temperature and the presence or absence of dopant gas. No signi­ficant variation in the deposition rate of the hydro­genated amorphous silicon layer is observed. The substrate employed is glass, the pressure of the plasma deposition is about 306 µbar (230 millitorr), the power density is about 5 milliwatts/cm² of com­bined surface area of the electrodes exposed to the plasma, and the power is radio frequency of about 13.6 megahertz. The dopant gas, when employed, is phosphine. The hydrogen content and distribution of the samples are obtained from infrared spectra. For etch rate measurements, the samples are protected with groove-­pattern wax and are held in a jig that allows simul­taneous immersion into the etchant of all of the samples. The etchant is a continuously stirred 1 molar solution of potassium hydroxide at 23°C. The depth of grooves etched into the hydrogenated amorphous silicon in three different etch times is measured with a "Tencor Alphastep" instrument. The results obtained are presented hereinbelow in Table I.

0042As apparent from Table I, the lowering of the substrate temperature results in a significant increase of the total hydrogen content, of the relative intensity of the hydrogen bending modes (a measure of the fraction of hydrogen in other than monohydride form), and of the etch rate. Moreover, the addition of the dopant, even without a temperature change, has a similar, although much smaller, effect.

Example 2

0043A set of samples, similar to those in Example 1, are prepared. The samples are etched in a dry etching apparatus, in the plasma etching mode, at various power levels, at a pressure of 40 µbar (30 mtorr), in an 80:20 mixture of carbon tetrafluoride and oxygen. High hydrogen content samples are found to etch at rates approximately 60 percent higher than corresponding identically doped samples with low hydrogen content. <tables id="tabl0001" num="0001"><img file="EP0253201A2_D0001.tif" /></tables>