JP4610515B2 - Peeling method - Google Patents

Description

The present invention relates to a method for peeling a functional thin film, and particularly to a method for peeling a film or layer having various elements. In addition, the present invention provides a transfer method for attaching a peeled film to a film substrate, and a thin film transistor (hereinafter referred to as TFT) formed using the transfer method.
The present invention relates to a semiconductor device having the above and a manufacturing method thereof.

  2. Description of the Related Art In recent years, attention has been focused on a technique for forming a TFT using a semiconductor thin film (having a thickness of about several to several hundred nm) formed on a substrate having an insulating surface. TFTs are widely applied to electronic devices such as ICs and electro-optical devices, and are particularly developed as switching elements and driver circuits for display devices.

  In such a display device, a glass substrate or a quartz substrate is often used, but there is a disadvantage that it is easily broken and heavy. For this reason, it is difficult to increase the size of a glass substrate or a quartz substrate in mass production. Therefore, attempts have been made to form TFT elements on a flexible substrate, typically a flexible plastic film.

  However, when a high-performance polycrystalline silicon film is used for the active layer of a TFT, a high-temperature process of several hundred degrees Celsius is required in the manufacturing process, and it cannot be directly formed on a plastic film.

Therefore, a method has been proposed in which a layer to be peeled existing on a substrate via a separation layer is peeled from the substrate. For example, a separation layer made of amorphous silicon, a semiconductor, nitride ceramics, or an organic polymer is provided, and the substrate is irradiated with a laser beam to cause in-layer separation or the like in the separation layer. They are separated (see Patent Document 1). In addition, there is a description of using this technique to complete a liquid crystal display device by attaching a layer to be peeled (referred to as a layer to be transferred in the publication) to a plastic film (see Patent Document 2). Also, looking at articles on flexible displays, the technology of each company is introduced (see Non-Patent Document 1).
Japanese Patent Laid-Open No. 10-125929 Japanese Patent Laid-Open No. 10-125930 Nikkei Microdevice, Nikkei Business Publications, July 1, 2002, July 1, 2002, p. 71-72

  However, in the method described in the above publication, it is essential to use a highly light-transmitting substrate, and in order to give sufficient energy to pass through the substrate and further release hydrogen contained in amorphous silicon, There is a problem that irradiation with a relatively large laser beam is required and the layer to be peeled is damaged. In addition, the above publication includes a description of providing a light shielding layer or a reflective layer in order to prevent damage to the peeled layer, but in that case, it is difficult to manufacture a transmissive liquid crystal display device or a light emitting device that emits light below. It is. Furthermore, in the above method, it is difficult to peel off a layer to be peeled having a large area.

  In view of the above problems, the present invention provides a substrate having a metal film provided over the substrate and a layer to be peeled that includes the metal-containing oxide film and the silicon-containing film provided over the metal film. And the layer to be peeled off by physical means or mechanical means. Specifically, an oxide layer containing the metal is formed over the metal film, and the oxide layer is crystallized by performing a heat treatment, and from within the oxide layer or from both interfaces of the oxide layer. A TFT obtained by peeling is formed.

  The TFT formed according to the present invention can be employed in any of top emission and bottom emission light emitting devices, transmissive, reflective, and transflective liquid crystal display devices.

  By using the peeling method of the present invention, it can be peeled over the entire surface, so that the yield can be improved and a TFT or the like can be formed on a flexible film substrate. Further, the present invention does not apply a load to the TFT or the like by a laser or the like. A light-emitting device, a liquid crystal display device, and other semiconductor devices each having the TFT are thin, are not easily broken even when dropped, and are lightweight. In addition, display with a curved surface or an irregular shape is possible.

  The TFT on the film substrate formed according to the present invention can achieve an increase in size of a display device in mass production. In the present invention, the substrate on which TFTs and the like are formed can be reused before transfer, and the cost of the semiconductor device can be reduced because an inexpensive film substrate is used.

  Embodiments of the present invention will be described below with reference to the drawings.

(Embodiment 1)
First, as shown in FIG. 1A, a metal film 11 is formed over a first substrate 10. Note that the first substrate only needs to have rigidity enough to withstand a subsequent peeling step. For example, a glass substrate, a quartz substrate, a ceramic substrate, a silicon substrate, a metal substrate, or a stainless steel substrate can be used. As the metal film, an element selected from W, Ti, Ta, Mo, Nd, Ni, Co, Zr, Zn, Ru, Rh, Pd, Os, and Ir, or an alloy material or a compound material containing the element as a main component A single layer made of the above or a laminate of these can be used. A sputtering method may be used as a method for forming the metal film, and the metal may be targeted and formed over the first substrate. The thickness of the metal film is 10 nm to 200 nm, preferably 50 nm to 75 nm.

  In place of the metal film, a nitrided metal film (metal nitride film) may be used. Furthermore, nitrogen or oxygen may be added to the metal film. For example, nitrogen or oxygen may be ion-implanted into the metal film, the film formation chamber may be nitrogen or oxygen atmosphere, the metal film may be formed by a sputtering method, and metal nitride may be used as a target.

At this time, when an alloy of the above metal (for example, an alloy of W and Mo: W _x Mo _1-X ) is used for the metal film, the first metal (W) and the second metal (Mo) are formed in the film formation chamber. A plurality of targets, or a target of an alloy of the first metal (W) and the second metal (Mo) may be disposed and formed by a sputtering method.

  In addition, when a metal film is formed by using a sputtering method, the film thickness at the peripheral edge of the substrate may be nonuniform. For this reason, it is preferable to remove the peripheral film by dry etching. However, since the first substrate is not etched at that time, a SiON film, a SiNO film, or the like is provided between the first substrate 10 and the metal film 11. An insulating film may be formed with a thickness of about 100 nm.

  Thus, by appropriately setting the formation of the metal film, the peeling process can be controlled and the process margin is widened. For example, when a metal alloy is used, the temperature of the heat treatment and further the necessity of the heat treatment can be controlled by controlling the composition ratio of each metal of the alloy.

  Thereafter, a layer to be peeled 12 is formed on the metal film 11. This layer to be peeled has an oxide film and a semiconductor film for forming an oxide layer containing the metal on the metal film 11. Note that the semiconductor film of the layer to be peeled may be a TFT, an organic TFT, a thin film diode, a photoelectric conversion element made of a silicon PIN junction, a silicon resistance element, or a sensor element (typically a pressure-sensitive type using polysilicon). A fingerprint sensor) or the like may be formed.

  As the oxide film, silicon oxide, silicon oxynitride, or the like may be formed by a sputtering method or a CVD method. Note that the thickness of the oxide film is preferably about twice or more that of the metal film 11. Here, the silicon oxide film is formed to a thickness of 150 to 200 nm by a sputtering method using a silicon target.

  In the present invention, when an oxide film is formed, an oxide layer containing the metal is formed on the metal film (not shown). The thickness of the oxide layer may be 0.1 nm to 1 μm, preferably 0.1 nm to 100 nm, and more preferably 0.1 nm to 5 nm.

  In addition, the oxide layer other than the above is prepared by a thin oxide formed by treatment with an aqueous solution containing sulfuric acid, hydrochloric acid or nitric acid, an aqueous solution in which sulfuric acid, hydrochloric acid or nitric acid is mixed with hydrogen peroxide, or ozone water. A membrane can be used. Further, as another method, plasma treatment in an oxygen atmosphere or oxidation treatment may be performed by generating ozone by irradiating ultraviolet rays in an oxygen-containing atmosphere, and heating to about 200 to 350 ° C. using a clean oven. A thin oxide film may be formed.

  In the layer to be peeled 12, an insulating film containing nitrogen such as SiN, SiON, or the like is preferably provided as a base film in order to prevent intrusion of impurities and dust from the metal film or the substrate, particularly on the lower surface of the semiconductor film.

  Thereafter, heat treatment is performed at a temperature higher than 400 ° C. By this heat treatment, the oxide layer is crystallized, and hydrogen contained in the layer to be peeled 12, especially hydrogen in the semiconductor film is diffused. The heat treatment may be combined with manufacture of a semiconductor device to reduce the number of steps. For example, when an amorphous semiconductor film is formed and a crystalline semiconductor film is formed using a heating furnace or laser irradiation, heat treatment at 500 ° C. or higher is performed at the same time as the crystalline semiconductor film is formed for crystallization. Hydrogen diffusion can be performed.

  Next, as shown in FIG. 1B, a second substrate 13 for fixing the layer to be peeled 12 is attached with a first adhesive material (adhesive) 14. Note that the second substrate 13 is preferably a substrate having higher rigidity than the first substrate 10. As the first adhesive 14, it is preferable to use a peelable adhesive, for example, an ultraviolet peelable adhesive that is peeled off by ultraviolet rays, a heat peelable adhesive that is peeled off by heat, a water-soluble adhesive, a double-sided tape, or the like.

  Next, the first substrate 10 provided with the metal film 11 is peeled off using physical means (FIG. 1C). Although the drawing is a schematic diagram, it is not described, but at this time the crystallized oxide layer or the interface on both sides of the oxide layer, that is, the interface between the oxide layer and the metal film or the oxide layer Peel off at the interface with the peeled layer. In this way, the layer to be peeled 12 can be peeled from the first substrate 10.

  Next, as shown in FIG. 1D, the peeled layer 12 to be peeled is attached to a third substrate 16 serving as a transfer body with a second adhesive 15. As the second adhesive 15, an ultraviolet curable resin, specifically, an adhesive such as an epoxy resin adhesive or a resin additive, a double-sided tape, or the like may be used. Note that when the surface of the third substrate has an adhesive function, the second adhesive may not be used. Further, the third substrate may be covered up to the side surface of the layer to be peeled 12. The third substrate 16 may be a plastic substrate such as polycarbonate, polyarylate, or polyethersulfone, a thin substrate such as a polytetrafluoroethylene substrate or a ceramic substrate, or a flexible (flexible) substrate. (Hereinafter, such a substrate is referred to as a film substrate).

  Next, the first adhesive material 14 is removed, and the second substrate 13 is peeled off (FIG. 1E). Specifically, in order to peel off the first adhesive, irradiation with ultraviolet light, heating, or washing with water may be performed. Furthermore, it is preferable to perform plasma cleaning or Bergrin cleaning using argon gas and oxygen gas.

  In addition, a plurality of layers to be peeled provided with TFTs corresponding to each application may be transferred to a third substrate serving as a transfer body. For example, a layer to be peeled of the TFT for the pixel portion and the TFT for the driver circuit may be formed and transferred to a predetermined region of the third substrate.

  A TFT or the like formed on the film substrate obtained as described above can be used as a semiconductor element of a light emitting device or a liquid crystal display device.

  For example, the light emitting device is formed by forming a light emitting element on the layer to be peeled 12 and forming a protective film serving as a sealing material. When the light emitting element is formed on the layer to be peeled 12, the film substrate on which the TFT is formed is flexible. Therefore, each light emitting layer may be formed by vacuum deposition by fixing to a glass substrate with an adhesive, for example, tape. Note that it is preferable to continuously form a light emitting layer, an electrode, a protective film, and the like without being exposed to the atmosphere.

  The order in which the light-emitting device is manufactured is not particularly limited, and after the light-emitting element is formed on the layer to be peeled, the second substrate is bonded, the layer to be peeled having the light-emitting element is peeled, and then the third substrate You may affix on the film substrate which is. Alternatively, after the light emitting element is formed, a film substrate which is a third substrate may be designed large, and the entire apparatus may be wrapped with the film substrate.

  In the case of manufacturing a liquid crystal display device, after peeling the second substrate, the counter substrate may be bonded with a sealant and a liquid crystal material may be injected. The order in which the liquid crystal display device is manufactured is not particularly limited, and a liquid crystal may be injected after the second substrate is bonded as a counter substrate and the third substrate is bonded.

  Also, when manufacturing a liquid crystal display device, spacers are formed or dispersed to keep the substrate interval, but in order to keep the flexible substrate interval, about 3 times more spacers are formed than usual. It is good to spray. The spacer is preferably made softer than that used for a normal glass substrate. Furthermore, since the film substrate has flexibility, it is necessary to fix the spacer so that the spacer does not move.

  By using such a peeling method, a TFT or the like can be formed over a flexible film substrate which can be peeled over the entire surface with high yield. Further, the present invention does not apply a load to the TFT or the like by a laser or the like. A light-emitting device, a liquid crystal display device, and other display devices having the TFT are thin, are not easily broken even when dropped, and are lightweight. In addition, display with a curved surface or an irregular shape is possible. Further, the TFT on the film substrate formed according to the present invention can achieve an increase in the size of the display device in mass production. Further, in the present invention, the first substrate and the like can be reused, and the cost of the display device can be reduced because an inexpensive film substrate is used.

  Hereinafter, experimental results of the present invention, and a light-emitting device, a liquid crystal display device, and other electronic devices manufactured using the present invention will be described.

    In this example, the results of a peeling experiment and the observation results of a transmission electron microscope (TEM) will be described.

First, in the sample shown in FIG. 2, an AN100 glass substrate (126 × 126 mm) is laminated on the substrate 200, a SiON film is formed on the insulating film 201 by a CVD method, and a tungsten (W) film formed on the metal film 202 by a sputtering method. did. Next, a SiO ₂ film formed by a sputtering method is used for the protective film 203 constituting the layer to be peeled, a SiON film formed by a CVD method is used for the base film 204, and an amorphous silicon film formed by a CVD method is used for the semiconductor film 205, respectively. It was.

  Sample A without heat treatment, Sample A with heat treatment at 220 ° C./1 hour, Sample B with heat treatment at 550 ° C./4 hours after 500 ° C./1 hour Samples C were used, and each was observed by TEM. The results are shown in FIGS. 3A to 5A, and schematic diagrams corresponding to TEM photographs (TEM images) are shown in FIGS. 3B to 5B, respectively.

  3 to 5, it can be seen that a layer is formed at the interface between the W film that is the metal film 202 and the protective film 203. In addition, a certain layer was not a complete layer and was sometimes scattered.

  Then, EDX measurement was performed in order to specify the composition of a certain layer. 8 to 10 show spectra and quantitative results of EDX measurement on samples A to C. FIG. The peaks of Al and Mo in the EDX spectrum are due to the sample fixing holder at the time of measurement. From the results of FIGS. 8 to 10, it can be seen that a certain layer contains tungsten and oxygen (hereinafter, a certain layer is referred to as an oxide layer).

  3A to 5A, it can be seen that the oxide layer of Sample C has a crystal lattice arranged in a specific direction. In addition, it can be seen that the thickness of the oxide layer of Samples A and B is about 3 nm, whereas the thickness of the oxide layer of Sample C is slightly thinner (3 nm or less).

  As a result of the peeling experiment in such samples A to C, only the sample C in which the oxide layer has a crystal lattice could be peeled off.

  Further, FIG. 6 and FIG. 7 show TEM photographs (A) of Sample D that was subjected to heat treatment at 400 ° C./1 hour and Sample E that was subjected to heat treatment at 430 ° C./1 hour for the sample shown in FIG. ) And a schematic diagram (B) corresponding to each TEM photograph. Note that the heating temperature 400 ° C. of the sample D is a boundary temperature at which crystallization can be performed and a predicted temperature and a boundary temperature at which separation can occur.

  6 that a crystal lattice is formed in part of the oxide layer of sample D, and a crystal lattice is formed as a whole in the oxide layer of sample E. FIG.

  As a result of the peeling experiment of the samples D and E, only the sample E could be peeled.

  From the results of the above peeling experiment and TEM photograph, it can be seen that an oxide layer is formed at the interface between the metal film and the protective film, and the crystallization of the oxide layer starts to occur from about 400 ° C. And when an oxide layer has crystallinity, it will be in the state which can peel. That is, it can be seen that an oxide layer on the metal film, specifically, an oxide layer having W on the W film needs to be formed in order to peel off.

That is, since the oxide layer can be peeled off in a crystallized sample, crystal distortion or lattice defects (point defects, line defects, surface defects (for example, oxygen vacancies) during crystallization of the oxide layer by heat treatment) Surface defects caused by crystallographic shear planes
, Extended defects), and is considered to peel from their interface.

  Next, in order to obtain information on the formation state of the oxide layer on the metal film, a peeling experiment was performed by changing the presence or absence of the protective film on the W film and the production conditions of the protective film.

As shown in FIG. 11, a sample A formed by sequentially laminating a SiON film 301 formed by a CVD method on a substrate 300 and a W film 302 formed by a sputtering method, and a protective film on the W film. Sample B in which Si film 303 is formed by sputtering using argon gas, Sample C in which SiO ₂ film 304 is formed by sputtering using argon gas and oxygen gas instead of Si film, Silane gas, and nitrogen A sample D in which a SiO ₂ film 305 was formed by a CVD method using a gas was prepared.

  FIGS. 12A to 15A show photographs obtained by observing the cross sections of the samples A to D with TEM, and schematic diagrams corresponding to the TEM photographs are shown in FIGS. 12B to 15B, respectively. Shown in

First, from FIGS. 12A to 14A, it can be seen that the oxide layer is formed on the W film in the sample C, but the oxide layer is not formed in the other samples. .
Although a natural oxide film is formed on the W film of Sample A, it is not clearly seen in the TEM photograph because of the thin film thickness.

  This is considered that the oxide layer was formed on the W film by the oxygen gas used when forming the sample C. On the other hand, when forming the protective film in sample B, only argon gas was used, and it seems that the oxide layer was not formed on the W film. From the viewpoint of the film thickness, the oxide layer formed on the sample C is considered to be different from the natural oxide film formed on the sample A. It is also considered that the oxide layer is formed when the protective film is formed.

In the sample D, the SiO ₂ film is formed on the W film by the CVD method capable of forming an oxide layer. As can be seen from FIG. 15A, the oxide layer is confirmed in the TEM photograph. I could not.

Here, considering the sample C on which the oxide layer is formed and the sample D, the silane gas used for the CVD method of the SiO ₂ film of the sample D is the source gas used in the production process of the SiO ₂ film of the sample C. It can be seen that it has hydrogen. That is, it is predicted that no oxide layer was formed in Sample D due to the presence of hydrogen. That is, in Sample D, it can be considered that the state has changed even if an oxide layer is formed on the W film due to hydrogen.

  As a result, it is conceivable that an oxide layer different from the natural oxide film is formed when the protective film is formed on the metal film. In the case of the W film, it is considered that the thickness of the oxide layer is preferably about 3 nm. And in order to form an oxide layer reliably, it is preferable to form so that a protective film may not have hydrogen.

  From the above results, it is considered that the conditions under which peeling can be performed are that an oxide layer (metal oxide layer) having the metal is formed on the metal layer.

  In particular, when W is used for the metal film, it is found that it is necessary to perform heat treatment at 400 ° C. or higher to crystallize the oxide layer of about 3 nm. From the results of this experiment, it can be seen that it is preferable to perform the heat treatment at 430 ° C. or higher because the crystallization of the oxide layer is performed throughout.

  Further, it can be seen that the metal oxide layer on the metal layer is formed when the protective film is formed, and the protective film does not contain hydrogen or is formed in a low hydrogen concentration state. Specifically, in the case of a W film, it can be seen that it is preferable to form a protective film by a sputtering method using a source gas containing oxygen gas.

  In this example, the result of observing the oxide layer on the substrate side after peeling and the semiconductor film side by TEM is shown.

  On a glass substrate, a W film is formed by sputtering to a thickness of 50 nm, a silicon oxide film is formed to 200 nm by sputtering as a protective film, a silicon oxynitride film is formed to 100 nm by plasma CVD as a base film, and a plasma CVD method is also used as a semiconductor film. A 50 nm thick amorphous silicon film was formed. Thereafter, heat treatment was performed at 500 ° C. for 1 hour and 550 ° C. for 4 hours, and the film was peeled off by physical means such as a polytetrafluoroethylene tape. FIG. 19 shows a TEM photograph of the W film and oxide layer on the substrate side, and FIG. 20 shows a TEM photograph of the oxide layer and silicon oxide film on the semiconductor film side.

  In FIG. 19, the oxide layer remains non-uniformly in contact with the metal film. Similarly, also in FIG. 20, the oxide layer remains non-uniformly in contact with the silicon oxide film. Both TEM photographs demonstrate that peeling was performed in the oxide layer and at both interfaces, and that the oxide layer remained in close contact with the metal film and the silicon oxide film.

  In this example, the result of examining the composition of the oxide layer using XPS (X-ray photoelectron spectroscopy) is shown.

  16A shows the result of sample A, FIG. 16B shows the result of sample B, and FIG. 16 (A) to (C), the horizontal axis indicates the depth direction (the inside of the oxide layer is exposed by ion sputtering, and when tungsten is detected 1 (atomic%), pos. 1 and tungsten is 2 ( atomic%) is detected as pos.2, and tungsten is detected as 3 (atomic%) as pos.3), and the vertical axis represents the bond occupancy ratio (%).

  When comparing FIGS. 16A to 16C, sample C has a larger relative ratio of tungsten (W) indicated by a circle than samples A and B. That is, sample C has a higher ratio of tungsten and lower ratio of tungsten oxide than samples A and B.

Next, FIG. 17 shows the result of normalizing the data of FIG. FIGS. 17A and 17D show the results for sample A, FIGS. 17B and 17E show the results for sample B, and FIGS. 17C and 17F show the results for sample C. FIGS. FIGS. 17A to 17C are graphs in which WO _{3 is set} to 1 and the bond occupancy ratio of other compositions is normalized, and FIGS. 17D to 17F are WO ₂ to 1 and other compositions are combined. It is the graph which normalized the occupation ratio.

First, when FIGS. 17A to 17C are compared, sample C has a larger relative ratio of WO ₂ indicated by crosses than samples A and B. That is, sample C has a higher ratio of WO ₂ than samples A and B, and pos. 1 to pos. As the depth increases to 3, the ratio of WO ₂ increases. Sample C has a small ratio of WO _X and pos. 1 to pos. It can be seen that as the depth increases to 3, the ratio of WO _X decreases. On the other hand, when FIGS. 17D to 17F are compared, the content ratio of WO ₃ indicated by triangles is 2% or more in Samples A and B, whereas the content ratio in Sample C is 2% or less. is there. This is because the ratio of WO ₂ of sample C is higher than that of samples A and B, as is clear from the graph normalized by WO ₃ .

18 is a waveform analysis diagram of binding energy and spectrum when samples 1 to C are exposed by ion sputtering and 1 (atomic%) of tungsten is detected (Pos. 1). 18A shows the result of Sample A after 4.25 minutes of sputtering treatment, FIG. 18B shows the result of Sample B after 4 minutes of sputtering treatment, and FIG. 18C shows 5 minutes of sputtering treatment. The result of the later sample C is shown. In FIG. 18 (A) ~ (C) , W1 ( tungsten W), W2 (tungsten oxide WO _X, X is nearly 2), W3 (tungsten oxide _{WO X, 2 <X <3} ), W4 ( tungsten oxide WO ₃ Etc.) Area ratio of each of the four states (%)
Corresponds to the composition ratio.

Table 1 shows the area ratios of W1 to W4 of Samples A to C obtained from FIG. Table 1 also shows the ratios of W2 and W3 normalized by W4 (WO ₃ ). In Table 1, the ratio of W1 of samples A and B is about 10%, while that of sample C is 35%, which is high. That is, sample C has a high tungsten ratio and a low tungsten oxide ratio. From the normalized values, it can be seen that Sample C has a higher proportion of W2 (WO ₂ ) in the tungsten oxide than Samples A and B.

Also in Sample C, W2 (WO ₂₎ has become a lot composition ratio of, it is considered that a change in the composition of the oxide layer occurs by performing heat treatment. That is, it is considered that W4 (WO ₃ ) changes in composition to W2 (WO ₂ ) or W3 (WO _x ) and peels between different crystal structures due to the difference in crystal structures.

  Next, the peeled surface on the substrate side after peeling and the peeled surface on the semiconductor film side after peeling were measured by XPS. As a result, the spectrum and the spectrum analysis of the spectrum are shown in FIGS. In addition, for comparison with the oxide layer and the natural oxide film, the result of measuring the sample A by XPS and the waveform analysis are also shown.

  First, FIG. 24 shows the spectrum of the peeled surface measured by XPS. The spectrum of the peeling surface on the semiconductor film side is FIG. 24A, and the spectrum of the peeling surface on the substrate side is FIG. 24B.

  Table 2 shows detection elements and quantitative results obtained from FIG. From Table 2, it can be seen that tungsten remains on the substrate side on the order of about 10 times that of the semiconductor film side.

Subsequently, FIG. 25A is a waveform analysis diagram of the spectrum on the semiconductor film side, and FIG. 25B is a waveform analysis diagram of the spectrum on the substrate side. In FIGS. 25A and 25B, W1 (tungsten W), W2 (tungsten oxide WO _x , X is approximately 2), W3 (tungsten oxide WO _x , 2 <X <3), W4 (tungsten oxide WO _3, etc.) The area ratio (%) of each of the four states corresponds to the composition ratio.

  Further, FIG. 31 shows a spectrum obtained by measuring the sample A on which the natural oxide film is formed by XPS, and the waveform analysis diagram of the spectrum is shown in FIG. 32, and the area ratio of each state of the obtained sample A and each sample Table 3 shows the intensity ratio obtained by standardizing W2 and W3 with W4. Furthermore, Table 3 shows the results of measuring the semiconductor film side surface and the substrate side surface after peeling.

  30A shows a graph showing the intensity ratio of the components W1 to W4 based on Tables 1 and 3, and FIG. 30B standardizes W2 and W3 with W4. The graph showing the intensity ratio is shown.

  On the semiconductor film side after peeling, W1 and W2 are 0%, W3 is 16%, and W4 is 84%, whereas on the substrate side, W1 is 44%, W2 is 5%, W3 is 10%, W4 Is 42%. The spectrum of the natural oxide film shows that W1 is 70, W2 is 6, W3 is 1, and W4 is 23.

  In Sample A, it can be seen that the ratio of W1 (tungsten) is higher than other samples. And it turns out that the ratio of W2-W4 (oxide) is low and the ratio of W3 is quite small.

Moreover, the sum of the WO ₂ and the semiconductor film side and the substrate side after the separation is found to have become less compared to the WO ₂ samples C. This is an energy active (unstable) state with the oxide layer before peeling, and it is going to be stable after peeling, and W4 (WO ₃ ) is the same as a natural oxide film. The main composition is considered.

  FIG. 30 shows that Sample C, which can be peeled off, and Sample A on which a natural oxide film is formed have a large amount of W2 to W4 (oxide).

Accordingly, when peeling is performed at the interface between the oxide layer and the metal film, the interface between the oxide layer and the silicon oxide film, or within the layer of the oxide layer, W1 (metal W) and W2 (WO _X , X 2) all remain on the substrate side, and W4 (WO ₃ etc.) remains 2/3 on the semiconductor film side and 1/3 remains on the substrate side. It can also be seen that the oxide layer and the natural oxide film have different oxide composition ratios. That is, it is considered that the oxide layer easily peels from the boundary between WO ₂ and WO _x or WO ₃ . No WO ₂ on the semiconductor film side at that reason this experiment, WO ₂ was adhered to the substrate side, WO ₂ adheres to the semiconductor film side conversely, it may also be considered when WO ₂ is not on the substrate side.

  In this example, the results of performing secondary ion mass spectrometry (SIMS) on samples A to C will be described with reference to FIGS.

First, paying attention to the profile of hydrogen in the amorphous silicon film, the concentration of hydrogen in samples A and B is about 1.0 × 10 ²² (atoms / cm ³ ), whereas the concentration of hydrogen in sample C is It is about 2.0 × 10 ²¹ (atoms / cm ³ ), which is about double. Further, when the hydrogen profiles in the silicon oxynitride film (SiON) and the silicon oxide film (SiO ₂ ) are observed, the samples A and B show a decreasing tendency when the depth is around 0.2 μm, which is not uniform. Concentration distribution. On the other hand, Sample C has a uniform concentration distribution in the depth direction without a marked decrease. That is, it can be seen that Sample C contains more hydrogen than Samples A and B. From this result, it is considered that the ionization efficiency of hydrogen is different, and it is considered that the composition ratio of the surface is different between the sample C and the samples A and B.

Next, paying attention to the nitrogen concentration at the interface between the silicon oxide film (SiO ₂ ) and the W film, in the samples A and B, the nitrogen concentration is about 1.0 × 10 ²¹ (atoms / cm ³ ). In Sample C, it is about 1.0 × 10 ²² (atoms / cm ³ ), which is different by about one digit. Therefore, it can be seen that Sample C has a different composition of the oxide layer at the interface between the silicon oxide film (SiO ₂ ) and the W film compared to Samples A and B.

  In this example, a light-emitting device including a TFT manufactured over a film substrate by the peeling method of the present invention will be described with reference to FIG.

  FIG. 26A is a top view of the light-emitting device, in which a signal line driver circuit 1201, a scanning line driver circuit 1203, and a pixel portion 1202 are shown over a film substrate 1210.

FIG. 26B is a cross-sectional view taken along line AA ′ of the light-emitting device, and an oxide layer 1250 is provided over the film substrate 1210 with an adhesive 1240 interposed therebetween. The oxide layer is not present as a layer on the back surface of the film substrate, but may be scattered. When the W film is used as the metal film as in the above embodiment, the oxide layer is an oxide containing tungsten as a main component, specifically WO ₃ .

  A signal line driver circuit 1201 including a CMOS circuit having an n-channel TFT 1223 and a p-channel TFT 1224 is shown on a film substrate 1210. Further, the TFT forming the signal line driver circuit or the scanning line driver circuit may be formed of a CMOS circuit, a PMOS circuit, or an NMOS circuit. In this embodiment, the driver integrated type in which the signal line driver circuit and the scanning line driver circuit are formed on the substrate is shown, but it is not always necessary, and it can be formed outside the substrate.

  In addition, the switching TFT 1221 and the current control TFT 1212 are provided. The insulating TFT 1214 that covers the switching TFT and the current control TFT and has an opening at a predetermined position is connected to one wiring of the current control TFT 1212. A light-emitting element 1218 including a first electrode 1213, a layer 1215 containing an organic compound provided over the first electrode, a second electrode 1216 provided opposite to the first electrode 1213, and a light-emitting element using moisture, oxygen, or the like A pixel portion 1220 having a protective layer 1217 provided to prevent deterioration is shown.

  Since the first electrode 1213 is in contact with the drain of the current control TFT 1212, at least the lower surface of the first electrode 1213 is made of a material that can make ohmic contact with the drain region of the semiconductor film, and includes an organic compound. It is desirable to use a material having a high work function on the surface in contact with the layer. For example, when a three-layer structure of a titanium nitride film, a film containing aluminum as a main component, and a titanium nitride film is used, the resistance as a wiring is low and the function can be achieved so that a good ohmic contact can be obtained. Further, the first electrode 1213 may be a single layer of a titanium nitride film, or a stack of three or more layers may be used. Further, when a transparent conductive film is used for the first electrode 1213, a double-sided light-emitting device can be manufactured.

  The insulator 1214 may be formed using an organic resin film or an insulating film containing silicon. Here, the insulator 1214 is formed using a positive photosensitive acrylic resin film.

  In order to improve the coverage of a light-emitting layer that includes an electrode or an organic compound to be formed later, a curved surface having a curvature is preferably formed on the upper end portion or the lower end portion of the insulator 1214. For example, in the case where positive photosensitive acrylic is used as a material for the insulator 1214, it is preferable that only the upper end portion of the insulator 1214 has a curved surface with a radius of curvature (0.2 μm to 3 μm). As the insulator 1214, either a negative type that becomes insoluble in an etchant by photosensitive light or a positive type that becomes soluble in an etchant by light can be used.

  The insulator 1214 may be covered with a protective film. This protective film is mainly composed of silicon nitride or silicon nitride oxide such as an aluminum nitride film, an aluminum nitride oxide film, or a silicon nitride film obtained by a film formation apparatus using a sputtering method (DC method or RF method) or remote plasma. It is an insulating film or a thin film mainly composed of carbon. Moreover, in order to transmit light emission through the protective film, the protective film is preferably as thin as possible.

  Over the first electrode 1213, a layer 1215 containing an organic compound that can emit R, G, and B light is selectively formed by a vapor deposition method using a vapor deposition mask or an inkjet method. Further, a second electrode 1216 is formed over the layer 1215 containing an organic compound.

  When the light emitting element 1218 emits white light, it is necessary to provide a color filter including a colored layer and BM.

  The second electrode 1216 is connected to the connection wiring 1208 through an opening (contact) provided in the insulating film 1214 in the connection region. The connection wiring 1208 is a flexible printed circuit using anisotropic conductive resin (ACF). (FPC) 1209 is connected. Then, a video signal and a clock signal are received from the FPC 1209 serving as an external input terminal. Although only the FPC is shown here, a printed wiring board (PWB) may be attached to the FPC.

  Also, when connecting an FPC using ACF by pressurization or heating, care should be taken not to cause cracks due to the flexibility of the film substrate and softening due to heating. For example, a highly rigid substrate may be disposed as an auxiliary in the adhesion region.

  Further, a sealing material 1205 is provided on the peripheral edge of the substrate, and is bonded to the second film substrate 1204 and sealed. The sealing material 1205 is preferably an epoxy resin.

  In this embodiment, as a material constituting the second film substrate 1204, a plastic substrate made of FRP (Fiberglass-Reinforced Plastics), PVF (polyvinyl fluoride), Mylar, polyester, acrylic, or the like is used in addition to a glass substrate or a quartz substrate. be able to.

  Although not shown, an organic material such as polyvinyl alcohol or an ethylene vinyl alcohol copolymer, or an inorganic material such as polysilazane, aluminum oxide, silicon oxide, silicon nitride, or the like so that water and oxygen do not enter from the film substrate. It is good to cover with the barrier film which consists of lamination.

  Further, a protective layer may be provided on the film substrate in order to protect from chemicals in the manufacturing process. As the protective layer, an ultraviolet curable resin or a thermosetting resin can be used.

  As described above, the light emitting device including the TFT provided on the film substrate is completed. And the light-emitting device provided with the TFT of the present invention is hard to break even when dropped and is lightweight. Further, the film substrate can achieve an increase in size of the light emitting device in mass production.

  In this example, a liquid crystal display device including a TFT manufactured over a film substrate by the peeling method of the present invention will be described with reference to FIG.

  FIG. 27A shows a top view of a liquid crystal display device, in which a signal line driver circuit 1301, a scanning line driver circuit 1303, and a pixel portion 1302 are shown over a first film substrate 1310. FIG.

FIG. 27B is a cross-sectional view taken along the line AA ′ of the liquid crystal display device. An oxide layer 1350 is formed over the film substrate 1310 with an adhesive 1340 interposed therebetween. The oxide layer is not present as a layer on the back surface of the film substrate, but may be scattered. When the W film is used as the metal film as in the above embodiment, the oxide layer is an oxide containing tungsten as a main component, specifically WO ₃ .

  A signal line driver circuit 1301 including a CMOS circuit having an n-channel TFT 1323 and a p-channel TFT 1324 is provided over the film substrate 1310. Note that the TFT forming the signal line driver circuit or the scanning line driver circuit may be formed of a CMOS circuit, a PMOS circuit, or an NMOS circuit. In this embodiment, the driver integrated type in which the signal line driver circuit and the scanning line driver circuit are formed on the substrate is shown, but it is not always necessary, and it can be formed outside the substrate.

  In addition, an interlayer insulating film 1314 that includes a switching TFT 1321 and a storage capacitor 1312, covers the switching TFT and the storage capacitor, and has an opening at a predetermined position, and a pixel portion 1320 are illustrated.

  An alignment film 1317 is provided over the interlayer insulating film 1314 and is rubbed.

  A second film substrate 1304 is prepared as a counter substrate. The second film substrate 1304 is provided with an RGB color filter 1330, a counter electrode 1316, and an alignment film 1317 that has been subjected to a rubbing process in a region partitioned on a matrix by a resin or the like.

  Further, a polarizing plate 1331 is provided on the first and second film substrates and is bonded with a sealant 1305. A liquid crystal material 1318 is injected into the first and second film substrates. Although not shown, a spacer is appropriately provided to hold the first and second film substrates.

  Although not shown, a barrier film made of an organic material such as polyvinyl alcohol or ethylene vinyl alcohol copolymer, an inorganic material such as polysilazane or silicon oxide, or a laminate thereof so that water and oxygen do not enter from the film substrate. Cover it.

  Further, a protective layer may be provided in order to protect from chemicals in the manufacturing process. As the protective layer, an ultraviolet curable resin or a thermosetting resin can be used.

  Then, as in FIG. 26, the wiring and the FPC are connected by anisotropic conductive resin (ACF) to receive a video signal and a clock signal. In addition, it is necessary to be careful not to cause cracks in connection by pressurization or heating using ACF.

  In this way, a liquid crystal display device including a TFT provided on the film substrate is completed. And the liquid crystal display device provided with the TFT of the present invention is hard to break even when dropped and is lightweight. Further, the film substrate can achieve an increase in size of the liquid crystal display device in mass production.

  An embodiment of the present invention will be described with reference to FIG. In this embodiment, a panel in which a pixel portion, a driver circuit for controlling the pixel portion, a memory circuit, and a CPU having a control device and an arithmetic device are mounted on the same insulating surface will be described.

  FIG. 28 shows the appearance of a panel, which has a pixel portion 3000 in which a plurality of pixels are arranged in a matrix on a substrate 3009. Around the pixel portion 3000, a scan line driver circuit 3001 and a signal line driver circuit 3002 for controlling the pixel portion 3000 are provided. The pixel unit 3000 displays an image in accordance with a signal supplied from the drive circuit.

  The counter substrate may be provided only on the pixel portion 3000 and the drive circuits 3001 and 3002, or may be provided on the entire surface. However, it is preferable that the CPU 3008 that may generate heat is disposed so that the heat radiating plate is in contact therewith.

  The panel has VRAM 3003 (video random access memory) for controlling the drive circuits 3001 and 3002, and decoders 3004 and 3005 for controlling the VRAM 3003 around the VRAM 3003. A RAM 3006 and a decoder 3007 for controlling the RAM 3006 and a CPU 3008 are provided around the RAM 3006.

  All elements constituting the circuit over the substrate 3009 are formed of a polycrystalline semiconductor (polysilicon) having a higher field-effect mobility and a higher on-current than an amorphous semiconductor, and hence the same insulating surface. The integrated formation of a plurality of circuits above is realized. In addition, the pixel portion 3001, the driver circuits 3001, 3002, and other circuits are first manufactured over a supporting substrate, and then peeled off and attached by the peeling method of the present invention, so that they are integrally formed on the flexible substrate 3009. is doing. Note that although the configuration of the plurality of pixels arranged in the pixel portion is not limited, the arrangement of the VRAM 3003 and the RAM 3006 may be omitted by arranging an SRAM in each of the plurality of pixels.

  The present invention can be applied to display portions of various electronic devices. Examples of the electronic device include a portable information terminal (a mobile phone, a mobile computer, a portable game machine, an electronic book, etc.), a video camera, a digital camera, a goggle type display, a display display, a navigation system, and the like. Specific examples of these electronic devices are shown in FIGS.

  FIG. 29A shows a display which includes a housing 4001, an audio output portion 4002, a display portion 4003, and the like. The present invention is used for the display portion 4003. The display device includes all information display devices such as a personal computer, a TV broadcast reception, and an advertisement display.

  FIG. 29B illustrates a mobile computer, which includes a main body 4101, a stylus 4102, a display portion 4103, operation buttons 4104, an external interface 4105, and the like. The display device of the present invention is used for the display portion 4103.

  FIG. 29C illustrates a game machine, which includes a main body 4201, a display portion 4202, operation buttons 4203, and the like. The present invention is used for the display portion 4202. FIG. 29D illustrates a mobile phone, which includes a main body 4301, an audio output portion 4302, an audio input portion 4303, a display portion 4304, operation switches 4305, an antenna 4306, and the like. The display device of the present invention is used for the display portion 4304.

  FIG. 29E illustrates an electronic book reader which includes a display portion 4401 and the like. The present invention is used for the display portion 4202.

  As described above, the applicable range of the present invention is so wide that the present invention can be used for electronic devices in various fields. In particular, the present invention that realizes thinness and light weight is very effective for the electronic devices shown in FIGS.

The figure which shows the peeling process of this invention. The figure which shows the experimental sample in this invention. The figure which shows the TEM photograph of the experiment sample A in this invention. The figure which shows the TEM photograph of the experiment sample B in this invention. The figure which shows the TEM photograph of the experiment sample C in this invention. The figure which shows the TEM photograph of the experiment sample D in this invention. The figure which shows the TEM photograph of the experimental sample E in this invention. The figure which shows the EDX spectrum and experimental result of the experimental sample A in this invention. The figure which shows the EDX spectrum and experimental result of the experimental sample B in this invention. The figure which shows the EDX spectrum and experimental result of the experimental sample C in this invention. The figure which shows the experimental sample in this invention. The figure which shows the TEM photograph of the experimental sample (a) in this invention. The figure which shows the TEM photograph of the experiment sample 2 in this invention. The figure which shows the TEM photograph of the experimental sample C in this invention. The figure which shows the TEM photograph of the experiment sample D in this invention. The figure which shows XPS of the experimental samples A to C in this invention. The figure which normalized XPS shown in FIG. The figure which shows EPS of the experimental samples A to C in this invention. The figure which shows the TEM photograph by the side of the board | substrate after peeling of this invention. The figure which shows the TEM photograph by the side of the semiconductor film after peeling of this invention. The figure which shows SIMS of the sample A in this invention. The figure which shows SIMS of the sample B in this invention. The figure which shows SIMS of the sample C in this invention. The figure which shows XPS after peeling of this invention. The figure which analyzed the waveform of XPS shown in FIG. The figure which shows the light-emitting device formed by this invention. The figure which shows the liquid crystal display device formed by this invention. The figure which shows CPU formed by this invention. FIG. 11 is a diagram showing an electronic device formed according to the present invention. The figure which shows the experimental result of this invention. The figure which shows the experimental result of this invention. The figure which shows the experimental result of this invention.

Claims (11)

Forming a tungsten film,
Forming an oxide layer containing tungsten on the tungsten film;
Forming a peelable layer on the oxide layer;
By performing heat treatment to the oxide layer, to crystallize the oxide layer,
The crystallized oxide layer is separated by crystal distortion or lattice defects in the oxide layer or at the boundary between the oxide layer and the film in contact with the oxide layer, A peeling method characterized by peeling a layer to be peeled off. Forming a tungsten film,
By treating the tungsten film with ozone water, an oxide layer having tungsten is formed on the tungsten film,
Forming a peelable layer on the oxide layer;
By performing heat treatment to the oxide layer, to crystallize the oxide layer,
The crystallized oxide layer is separated by crystal distortion or lattice defects in the oxide layer or at the boundary between the oxide layer and the film in contact with the oxide layer, A peeling method characterized by peeling a layer to be peeled off. In claim 1 or claim 2 ,
The peeling method, wherein the lattice defect is a point defect, a line defect, or a surface defect. In claim 3 ,
The surface defect is a surface defect caused by a crystallographic shear surface formed by a collection of oxygen vacancies. Forming a tungsten film,
Forming an oxide layer containing tungsten on the tungsten film;
Forming a peelable layer on the oxide layer;
By performing heat treatment on the oxide layer, the oxide layer is crystallized, and the composition of the crystallized oxide layer is changed.
A separation method characterized by separating the layer to be peeled by separating the separated oxide layer within the layer of the oxide layer or at the boundary between the oxide layer and the film in contact with the oxide layer. . Forming a tungsten film,
By treating the tungsten film with ozone water, an oxide layer having tungsten is formed on the tungsten film,
Forming a peelable layer on the oxide layer;
By performing heat treatment on the oxide layer, the oxide layer is crystallized, and the composition of the crystallized oxide layer is changed,
A separation method characterized by separating the layer to be peeled by separating the separated oxide layer within the layer of the oxide layer or at the boundary between the oxide layer and the film in contact with the oxide layer. . In claim 5 or claim 6 ,
A stripping method, wherein the composition of the oxide layer is changed from WO ₃ to WO ₂ by the heat treatment. In any one of Claims 1 thru | or 7 ,
The peeling method according to claim 1, wherein when the layer to be peeled is peeled off, the oxide layer is non-uniformly in contact with the tungsten film. In any one of Claims 1 to 8 ,
The peeling method, wherein the heat treatment is performed at a temperature of 430 ° C. or higher. In any one of Claims 1 thru | or 9,
The peeling method is characterized in that the layer to be peeled has a semiconductor film. In claim 10,
A stripping method, wherein a TFT, an organic TFT, a thin film diode, a photoelectric conversion element, a silicon resistance element, or a sensor element is formed using the semiconductor film.

Images