JP2004006840A - Semiconductor thin film - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology of forming a silicon thin film of small trap level density by irradiation with light and to provide another technology of applying the former technology to a substrate of a large area with high reproducibility. <P>SOLUTION: In a semiconductor thin film forming method comprising a light irradiating process, a first process of projecting the image of a pattern formed on an optical mask onto the semiconductor thin film for exposing the thin film so as to modify the prescribed region of the semiconductor thin film and a second process of continuously forming insulating films on the semiconductor thin film are provided. An alignment mark is formed by the use of a color difference between the region modified by irradiation with light and the other region that is not subjected to irradiation with light and has a different color. Positioning operations are carried out in a photolithography process, an etching process, and a film removing process of removing a part of the above laminated insulating films on the semiconductor thin film from the substrate on the basis of the above alignment mark. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a semiconductor thin film used in a semiconductor device and a method for forming the same, and more particularly, to a technique for forming a silicon thin film used for a crystalline silicon thin film transistor by light irradiation, and a field effect by forming a gate insulating film on the silicon film. The present invention also relates to a technique for obtaining a type transistor, and further to a drive circuit for a display, a sensor or the like constituted by the semiconductor thin film or the field effect thin film transistor.
[0002]
[Prior art]
First, the current state of the art in the above technical field will be described in detail, including the indication of its problems. Representative techniques for forming a thin film transistor (TFT) on a glass substrate include a hydrogenated amorphous silicon TFT technique and a polycrystalline silicon TFT technique. The former fabrication process has a maximum temperature of about 300 ° C. and a mobility of 1 cm. ² / Vsec carrier mobility is realized. This technology is used as a switching transistor of each pixel in an active matrix (AM-) liquid crystal display (LCD), and includes a driver integrated circuit (IC, an LSI formed on a single crystal silicon substrate) arranged around a screen. ). Since each pixel has a switching element TFT, it has the feature that crosstalk is reduced and good image quality is obtained compared to a passive matrix LCD that sends an electric signal for driving liquid crystal from a separate peripheral driver circuit. Have.
[0003]
On the other hand, the latter uses, for example, a quartz substrate and a high-temperature process similar to an LSI at about 1000 ° C., so that the carrier mobility is 30 to 100 cm. ² / Vsec. The fact that such a high carrier mobility can be realized is that, for example, when applied to a liquid crystal display, a pixel TFT for driving each pixel and, at the same time, a peripheral drive circuit portion can be simultaneously formed on the same glass substrate. There is an advantage that contributes to reduction of manufacturing process cost and downsizing of equipment. This is because the connection pitch between the AM-LCD substrate and the peripheral driver integrated circuit has to be narrowed with the miniaturization of the device and the increase in resolution, and it cannot be dealt with by the tab connection or the wire bonding method.
[0004]
However, in the polycrystalline silicon TFT technology, in order to use the above-described high-temperature process, an inexpensive low softening point glass that can be used in the former process cannot be used. In order to improve this point, it is necessary to reduce the temperature in the polycrystalline silicon TFT process. In order to cope with this, a low-temperature forming technique of a polycrystalline silicon film using a laser crystallization technique has been researched and developed.
[0005]
As is well known, these laser crystallizations are generally realized using a pulsed laser irradiation device. FIG. 11 is a schematic diagram illustrating an example of the configuration of a pulse laser irradiation device. The laser light supplied from the pulse laser light source (1101) is an optical path defined by optical elements such as a mirror (1102, 1103, 1105) and a beam homogenizer (1104) installed to make the spatial intensity uniform. , And reaches the silicon thin film (1107) on the glass substrate (1109) as the object to be irradiated. In general, since one irradiation range is smaller than that of a glass substrate, laser irradiation is performed on an arbitrary position on the substrate by moving the glass substrate on the xy stage (1109). Instead of the xy stage, a method of moving the above-described optical element group and a method of combining the optical element group and the stage are also performed.
[0006]
See, for example, the literature; Im and R. Sposili, "Crystalline Si films for integrated active-matrix-liquid-crystal displays", Materials Research Society Bulletin, vol. 21, (1996), 39 (publicly known document 1), FIG. 6 shows an example in which a substrate is mounted on an x-direction stage and a homogenizer is mounted on a y-direction stage.
[0007]
The laser irradiation may be performed in a vacuum chamber or in a high-purity gas atmosphere in a vacuum chamber. It also mentions that a cassette (1110) containing a glass substrate with a silicon thin film and a substrate transport mechanism (1111) are provided as needed to mechanically take out and store a substrate between the cassette and the stage.
[0008]
Japanese Patent Publication No. Hei 7-118443 discloses a technique of irradiating a short-wavelength pulsed laser beam to crystallize an amorphous silicon thin film on an amorphous substrate and apply the crystallized thin film to a thin film transistor. According to the disclosed method, since amorphous silicon can be crystallized without raising the temperature of the entire substrate, semiconductor elements and semiconductor integrated circuits can be manufactured on a large-area and inexpensive substrate such as glass such as a liquid crystal display. There is an advantage that you can.
[0009]
However, as described in the publication, 50-500 mJ / cm is required for crystallization of an amorphous silicon thin film by a short wavelength laser. ² It requires a certain irradiation intensity. On the other hand, the light emission output of a pulse laser device which is currently generally available is the maximum and is about IJ / pulse, so that the area that can be irradiated at one time even by simple conversion is 2 to 20 cm. ² Only about. Therefore, for example, in order to carry out laser crystallization on the entire surface of a substrate having a substrate size of 47 × 37 cm, it is necessary to irradiate at least 87 points and 870 points depending on conditions. If the substrate is enlarged, for example, to a substrate size of 1 m square, the number of irradiation locations required correspondingly increases. The laser crystallization in this case is also performed using the pulse laser irradiation apparatus having the configuration shown in FIG. 11 described above.
[0010]
In order to uniformly form a thin film semiconductor element group on a large area substrate by applying the above method, a technique as disclosed in Japanese Patent Application No. 3-315863, that is, the element group is formed by using a laser beam size It is known that a method of dividing a small area into small areas and repeating in order of several pulses irradiation + movement of irradiation area + several pulse irradiation + movement of irradiation area +... By step and repeat is effective. As shown in the laser operation method of FIG. 14A, this is a method of controlling so that laser oscillation and movement of a stage (ie, a substrate) or a beam are performed alternately.
[0011]
However, even if a pulse laser device having an oscillation intensity uniformity of about ± 5 to 10% (at the time of continuous oscillation), which is currently available, is used by this method, for example, irradiation of about 1 to 20 pulses / place is repeated. In the case of the method, the variation in oscillation intensity exceeds ± 5 to 10%, and there is a problem that the resulting polycrystalline silicon thin film and the uniformity of the characteristics of the polycrystalline silicon thin film transistor are not sufficient. Was. In particular, generation of intense light or weak light due to instability of discharge at the beginning of laser oscillation called spiking is one of the causes of non-uniformity.
[0012]
In order to correct this difficulty, there is also known a method of controlling the applied voltage at the next oscillation based on the integrated intensity result. However, this method can suppress the occurrence of spiking but emits weak light instead. There is a problem. That is, as shown in FIG. 13, when the irradiation time and the non-oscillation time are alternately continuous, the first pulse intensity oscillated at each irradiation time is most unstable and is likely to vary. Since the strength histories are different, there arises a problem that sufficient uniformity of the transistor element and the thin film integrated circuit in the substrate surface cannot be obtained.
[0013]
As another method of avoiding such spiking, as shown in FIG. 14B, a method of avoiding laser oscillation by starting before the start of irradiation of the element formation region is known. It cannot be applied to the case where laser oscillation and stage movement are intermittently repeated as shown in FIG.
[0014]
In order to avoid these problems, Japanese Patent Application Laid-Open No. 5-90191 proposes a method in which a pulse laser light source is continuously oscillated and the substrate is not irradiated with a light shielding device during a stage movement period. . That is, as shown in FIG. 14 (c), the laser is continuously oscillated at a certain frequency, and the movement of the stage to a desired irradiation position and the shielding / opening of the optical path are synchronized, so that a laser beam with a stable intensity is generated. It is possible to irradiate a desired irradiation position. However, the disadvantage of this method is that although the laser beam can be stably irradiated on the substrate, the useless laser oscillation that does not contribute to the formation of the polycrystalline silicon thin film increases, and the lifetime of the expensive laser light source and the excitation gas is reduced. The productivity of the polycrystalline silicon thin film and the production efficiency of the polycrystalline silicon thin film with respect to the power required for laser oscillation and the like are reduced, which leads to an increase in production cost.
[0015]
In addition, the substrate to be exposed to the laser may be damaged if the substrate is irradiated with an intense light exceeding a desired value due to a variation in irradiation intensity. In particular, in an imaging device such as an LCD, there is a problem that light transmitted through the substrate causes light scattering or the like in a damaged region on the substrate, thereby deteriorating image quality.
[0016]
Next, a technique for reducing and projecting a pattern on an optical mask onto a silicon thin film and performing laser crystallization is described in, for example, R. H .; Sposili and J.M. Im, "Sequential Lateral Solidification of Films on SiO2", Applied Physics Letters, vol. 69, (1996), 2864 (publicly known document 2) and literature; Im, R .; Sposili and M.S. Crowder, "Single-crystal Si films for thin film transformer devices", Applied Physics Letters, vol. 70, (1997), 3434 (publicly known document 3).
[0017]
According to each of the above documents, a 308 nm excimer laser, a variable-energy actuator, a variable-focus field lens, a patterned-mask, a two-element imaging lens, a projection using a degree of reduction, and a sub-microscope are described. Thus, a beam size on the order of μm and a moving pitch of the substrate stage on the order of μm are realized.
[0018]
However, when this method is used for processing a large-sized substrate as described above, the laser beam irradiated on the optical mask has a spatial intensity profile depending on the light source, and thus, for example, passes through the center and the periphery of the mask. There is a problem that a critical intensity distribution deviation occurs in the exposed exposure pattern, and a crystalline silicon thin film having desired uniformity cannot be obtained. Furthermore, since ultraviolet light having a short wavelength is reduced and projected, the depth of focus of the beam is small, and there is a problem that the irradiation depth is easily shifted due to the warpage or deflection of the substrate. In addition, as the size of the substrate increases, it becomes more difficult to secure the mechanical accuracy of the stage, and there has been a problem that the inclination of the stage and the displacement of the substrate on the stage during movement hinder desired laser irradiation conditions.
[0019]
A method of irradiating a plurality of pulses with a certain delay time when performing laser irradiation as described above is described in Ryoichi Ishihara et al. "Effects of light pulse duration oneximer laser crystallization characteristics of silicon thin films", Japans journal of applied physics. 34, no. 4A, (1995) pp1759 (Publication 4). According to the above-mentioned known literature, the crystallization solidification rate of molten silicon in the laser recrystallization process is 1 m / sec or more, and it is necessary to reduce the solidification rate in order to obtain good crystal growth. By irradiating the second laser pulse immediately after the solidification is completed, it is intended to obtain a recrystallization process with a lower solidification rate by the second irradiation.
[0020]
Now, according to the temperature change (time history curve) of silicon as shown in FIG. 12, the temperature of silicon rises with the irradiation of laser energy (eg, an intensity pulse whose waveform is shown in FIG. 15), and the starting material is a-Si. In this case, when the temperature further rises after passing the melting point of a-Si, and the supply of energy falls below a value required for the temperature rise, cooling starts. At the solidification point of the crystalline Si, after the solidification is completed after a solidification time, it is cooled to the ambient temperature. Here, assuming that the solidification of silicon proceeds in the film thickness direction starting from the silicon-substrate interface, the average value of the solidification rate is expressed by the following equation.
[0021]
Average value of solidification rate = silicon film thickness / solidification time, that is, if the silicon film thickness is constant, it is necessary to increase the solidification time in order to reduce the solidification rate. Therefore, if the process maintains the ideal state of thermal equilibrium, the solidification time can be extended by increasing the ideal input energy, that is, the laser irradiation energy. However, as pointed out in the above-mentioned known documents, there is a problem that an increase in irradiation energy causes the film to become amorphous and microcrystalline. In a realistic melting / recrystallization process, an ideal temperature change as shown in FIG. 12 cannot be obtained. In general, the temperature reaches a stable state through an excessive rise in heating and a supercooling process in cooling. This is because an amorphous solid is formed by rapid cooling and solidification without causing crystallization in the vicinity of the freezing point particularly when the cooling rate at the time of cooling is large and undergoes an excessive supercooling process. As described in the above-mentioned known literature, a thin film may be not amorphous but microcrystalline depending on conditions. The microcrystalline body has an extremely small grain size as compared with a polycrystalline thin film or a single crystal thin film, and therefore has a large number of crystal grain boundaries having a large grain boundary potential. Alternatively, there arises a problem that an off-leak current increases.
[0022]
FIG. 16 shows the maximum cooling rate (Cooling rate, K / sec) obtained from numerical calculation when a silicon thin film having a thickness of 75 nm was irradiated with an excimer laser having a wavelength of 308 nm, and was obtained from SEM observation of the film after laser irradiation. It shows the threshold of irradiation intensity for crystallization-microcrystallization. The emission pulse waveform of the laser used in the experiment is shown in FIG. It exhibits three main peaks and the light emission time is about 120 nsec. Such a pulse waveform has a light emission time that is at least five times that of a rectangular pulse having a pulse width of 21.4 nsec assumed and described in the above-mentioned known document 4, so that even a single pulse irradiation can be performed. The effect of reducing the solidification rate as described in the above-mentioned known document 4 can be expected.
[0023]
FIG. 17 shows a temperature-time curve of silicon obtained from a numerical calculation at the time of laser recrystallization using such a pulse waveform. FIG. 17 shows a silicon film having a thickness of 75 nm and SiO ₂ , XeCl laser (wavelength 308 nm) irradiation intensity 450 mJ / cm ² 3 shows a temperature change of the silicon thin film at the time of (1). About 60 nsec after the second emission peak is almost completed, the temperature reaches the maximum temperature and starts cooling. (Note that in this numerical calculation, the value of amorphous silicon is used as the melting and freezing point, and the behavior near the freezing point is an actual behavior. In particular, when a crystallized film is obtained, crystallization is completed at the freezing point of crystalline silicon). It can be seen that the cooling starts once with a large slope, but the slope at about 100 nsec where the third peak exists is very small. After 120 nsec, when the light emission completely ends, the solidification is again performed through a rapid cooling process. In general, in the case of a solidification process from a liquid that has undergone “quenching” that greatly deviates from the thermal equilibrium process, the solidification time required for forming a crystal structure cannot be obtained, and an amorphous solid is formed. You.
[0024]
FIG. 16 shows the result of estimating the maximum cooling rate after the end of light emission for each irradiation intensity based on the silicon temperature-time curve as shown in FIG. It can be seen that the cooling rate increases as the irradiation intensity increases. On the other hand, when the structure of the silicon thin film after laser irradiation was observed using a scanning electron microscope, as shown in FIG. 16, although the particle size once increased with the increase in irradiation intensity, it was 470 mJ / cm. ² Under the set irradiation intensity conditions of the degree, microcrystallization was observed. Similarly, when the number of irradiation pulses is set to three, 470 mJ / cm ² Even under the set irradiation intensity condition, although the microcrystallized region was partially left, a dramatic increase in the particle size was observed unlike the case of one pulse (FIG. 18; “Laser for each irradiation intensity and irradiation frequency”). Crystal Form of Recrystallized Silicon Thin Film "). Note that the actual irradiation intensity is about 5 to 10% higher than the set value, especially in the first few pulses of the excimer laser, so that the threshold intensity at which microcrystallization occurs is 500 mJ / cm. ² Degree can be estimated. From the above results, it can be seen that 500 mJ / cm in FIG. ² By estimating the cooling rate from the conditions, microcrystallization was about 1.6 × 10 ¹⁰ It has been found that this occurs at a cooling rate condition of at least ° C / sec. When the film to be irradiated is a-Si, about 500 mJ / cm ² Microcrystallization is performed at the above irradiation intensity. Similarly, when the film to be irradiated is poly-Si, when this cooling rate is applied, it is about 30 mJ / cm compared with a-Si. ² High irradiation intensity is suggested. Therefore, a cooling rate of 1.6 × 10 ¹⁰ By controlling the temperature to not more than ° C./sec, microcrystallization and amorphization can be prevented, and a good crystal growth process can be obtained.
[0025]
Next, a case where the second laser light is introduced with a delay to the first laser light will be described. As described above, the laser light in the latter half of the light emission moderates the increase in the cooling rate, and the cooling rate after the light emission ends controls the crystallization. That is, it is considered that the cooling process before that is initialized by the finally input energy. By inputting additional energy, even if amorphization or microcrystallization due to rapid cooling occurs during the previous solidification process, the energy is preserved (for a short time of the order of nanoseconds, It is considered that heat conduction to the atmosphere and radiation to the atmosphere are small. Of course, the time during which sufficient heat can be released is not taken into account), so it is considered that the process is initialized once and the solidification process is repeated again. Therefore, good crystal growth can be expected by paying attention to the cooling rate after the end of the secondary heating by the re-input energy. As shown in FIG. 19, which shows the maximum cooling rate and the cooling rate near the freezing point, the cooling rate is controlled to a desired value by controlling the delay time.
[0026]
On the other hand, a technique for performing a process of forming an a-Si thin film, a laser irradiation process, a plasma hydrogenation process, and a process of forming a gate insulating film sequentially or in a different order without exposure to the air, which is a material to be irradiated with laser, is as follows. Are disclosed in the patent gazettes listed below.
JP-A-5-182923; a step of irradiating a laser after heating an amorphous semiconductor thin film without exposure to air.
Japanese Patent Application Laid-Open No. 7-99321; a substrate having a laser-crystallized polycrystalline silicon thin film is subjected to plasma hydrogenation without exposing to the atmosphere, and is transported to a gate insulating film forming step.
JP-A-9-7911; a substrate having a laser-crystallized polycrystalline silicon thin film is transferred to a gate insulating film forming step without exposing the substrate to the atmosphere.
Japanese Patent Application Laid-Open No. Hei 9-17729; a substrate having a laser-crystallized polycrystalline silicon thin film is conveyed to a gate insulating film forming step without exposing the substrate to atmospheric air to prevent impurities from adhering to the polycrystalline silicon surface.
JP-A-9-148246; formation of an amorphous silicon thin film, laser crystallization, hydrogenation, and formation of a gate insulating film are continuously performed without exposure to the atmosphere.
JP-A-10-116989; formation of an amorphous silicon thin film, laser crystallization, hydrogenation, and formation of a gate insulating film are continuously performed without exposure to the atmosphere.
Japanese Patent Application Laid-Open No. H10-149984; formation of an amorphous silicon thin film, laser crystallization, hydrogenation, and formation of a gate insulating film are continuously performed without exposure to the atmosphere.
JP-A-11-17185; formation of an amorphous silicon thin film, laser crystallization, formation of a gate insulating film, and formation of a gate electrode are continuously performed without exposure to air.
The idea and technology disclosed in each of these publications is that the silicon surface formed by laser crystallization is very active, so that exposure to the atmosphere makes it easier for impurities to adhere to the silicon surface. It is designed to solve the problem of deteriorating characteristics or causing variations in the characteristics.
[0027]
For the above technical evaluation, the applicants performed the excimer laser crystallization technology and the silicon oxide film formation technology in the same device (including transferring the substrate to another device without exposing it to the air), and once exposed to the air The performance of the product was compared with that of the product. As a result, a great effect was confirmed to improve the product yield rate by the effect of preventing dust and particles from adhering. On the other hand, the degree of this effect was equivalent to some degree by increasing the cleanliness of the clean room environment. Was obtained.
[0028]
In order to improve the yield, a device in which a substrate cleaning mechanism is incorporated in the same device as the film forming device has the greatest effect. For example, depending on the formation conditions in the a-Si formation process, some particles adhere to the substrate during the film formation, and once released to the atmosphere, a cleaning process is required. On the other hand, focusing on the performance of the thin film transistor, the difference in the manufacturing process did not bring a significant difference. The reason can be considered as follows.
[0029]
Applicants have described, for example, the literature; Yuda et al. "Improvement of structural and electrical properties in low-temperaturegate-oxides for poly-Si TFTs by controlling 02 / SiH4 ratios", Digest of technical papers 1997 international workshop onactive matrix liquid crystal displays, September 11-12, 1997, Kogakuin Univ. , Tokyo, Japan-87 (Publication 5), fixed oxidation of a silicon oxide film formed using plasma at a temperature of about 300 to 350 ° C. or a silicon oxide film formed through a heat treatment of about 600 ° C. Membrane charge density (10 ¹¹ -10 ¹² cm ^-2 ) And the interface state density with the silicon substrate (シ リ コ ン 6 × 10 ¹⁰ cm ^-2 eV ^-2 ). In this case, the silicon substrate is subjected to acidity such as sulfuric acid / hydrogen peroxide solution, hydrochloric acid / hydrogen peroxide solution / water, ammonia / hydrogen peroxide solution / water, hydrofluoric acid / water, etc. After cleaning and washing with a cleaning liquid, the liquid is introduced into the film forming apparatus. Therefore, the above interface state density value is obtained from a sample which is a single-crystal silicon substrate, but is once exposed to the atmosphere after forming (cleaning) a clean interface and is transferred to a film forming step.
[0030]
Here, attention is paid to the trap level density of one laser crystallized silicon film. Applicants, for example, see the literature; Tanabe et al. , "Excimer laser crystallization of amorphous silicon films", NEC Research and Development, vol. 35, (1994), 254 (Publication 6), a thin film transistor having a laser crystallized silicon film is used to reduce the trap level density (10%) in the crystallized silicon film. ¹² -10 ¹³ cm ^-2 ). Moreover, the field effect mobility of these transistors is 40 to 140 cm ² / Vsec.
[0031]
Incidentally, when comparing the trap state density in the silicon film with the interface state density (or the fixed oxide film charge density), the value of the trap state density is clearly larger. That is, in a sample in which a silicon film / gate insulating film is formed in the same device without exposure to the air, the performance (trap level density) of the silicon film is not sufficient to obtain the effect of cleanliness. It turned out that there is.
[0032]
As a means related to the field according to the present invention, a remote plasma CVD (chemical vapor deposition) method has been proposed as a means for reducing plasma damage and forming a high-quality gate insulating film. For example, JP-A-5-21393 discloses a configuration in which a plasma generation chamber and a substrate processing chamber are separated. By adopting such a configuration, the charge density of the low fixed oxide film (10 ¹¹ -10 ¹² cm ^-2 ) And low interface state density (~ 6 × 10 ¹⁰ cm ^-2 eV ^-2 ) Can be realized, but there is a problem that this effect is limited by the performance of the silicon film formed in advance as described above.
[0033]
[Problems to be solved by the invention]
SUMMARY OF THE INVENTION The present invention has been made to overcome the problems described above in detail, and provides a technique for forming a silicon thin film having a small trap state density by light irradiation, and has a reproducibility on a large-area substrate. The purpose is to provide a technique for applying the technique well. Further, by forming a good-quality gate insulating film on the high-quality silicon film, a good semiconductor-insulating film interface, that is, a field-effect transistor having excellent characteristics is provided.
[0034]
[Means for Solving the Problems]
In the present invention, a semiconductor thin film is formed on a substrate, and a semiconductor thin film in which a region subjected to modification by light irradiation and a region not subjected to light irradiation are mixed, and the composition of the semiconductor thin film is different from that of the semiconductor thin film on the surface of the semiconductor thin film. A structure in which thin films are stacked to form a stacked film is adopted. In particular, the alignment mark is formed by using a color difference between a region which has been modified by the light irradiation and a region which has not undergone the light irradiation having a different optical color.
[0035]
Further, the laminated film has a structure in which a part of the laminated film is removed from the substrate.
[0036]
Furthermore, a configuration is adopted in which the optical color of the region that has undergone the modification by the light irradiation is different from that of the region that has not undergone the light irradiation.
[0037]
Further, a predetermined alignment mark is provided on the substrate, and a region determined based on the alignment mark is modified by light irradiation.
[0038]
In the above-described configuration, by determining the exposure region based on the alignment mark provided in advance, a silicon thin film that has been subjected to exposure modification under a desired exposure condition can be formed at a desired location. Only the exposure can be modified. That is, the source / drain and channel regions can be sequentially patterned in the next step corresponding to the modified regions.
[0039]
Also, in the method of the present invention, in the method of forming a semiconductor thin film having a light irradiation step, light is used to project a pattern formed on the optical mask onto the semiconductor thin film to modify a predetermined region on the semiconductor thin film. And a step of continuously forming an insulating film on the semiconductor thin film.
[0040]
By doing so, it becomes possible to provide a high-performance, multifunctional semiconductor forming apparatus, a low-cost, high-reproducibility thin-film transistor manufacturing process, and a high-performance thin-film transistor. Specifically, 1) provision of a highly stable semiconductor thin film process capable of reducing the number of cleaning steps using a chemical solution, and 2) provision of a multifunctional apparatus capable of processing multiple steps in the same apparatus, thereby setting up a total factory. It is possible to manufacture a low-cost, high-performance thin-film transistor capable of reducing the area (space-saving semiconductor process) and 3) maintaining a clean silicon surface (interface) without using a chemical solution.
[0041]
This makes it possible to designate a device element formation region in a subsequent thin film transistor manufacturing process based on the semiconductor thin film in which a predetermined region on the substrate has been modified by laser irradiation. At this time, since there is no foreign material other than the semiconductor thin film and the insulating film for forming the alignment mark, a clean semiconductor-insulating film interface can be easily formed.
[0042]
Further, the positioning in the photolithography step, the etching step, and the step of removing a part of the laminated film of the semiconductor thin film from the substrate is performed based on the alignment mark.
[0043]
BEST MODE FOR CARRYING OUT THE INVENTION
FIG. 1 shows an example of an embodiment of the present invention, and shows a schematic arrangement of a main part of an exposure apparatus having a reduction projection optical system. The pulsed UV light supplied from the first excimer laser (EL1) and the second excimer laser (EL2) is guided to a homogenizer (opt20 ') via mirrors (opt3, opt3') and lenses (opt4). I will Here, the beam is shaped by an optical mask (opt 21) so as to have a desired uniformity, for example, an in-plane distribution of ± 5%. Since the intensity profile and total energy of the original beam supplied from the excimer laser may change from pulse to pulse, the intensity on the optical mask is made more uniform in terms of spatial distribution and pulse-to-pulse variation. It is desirable to provide a mechanism for performing this. As the homogenizer, one using a fly-eye lens or a cylindrical lens is generally used.
[0044]
The light pattern formed by the optical mask is applied to a substrate (sub0) installed in a vacuum chamber (C0) via a reduction projection exposure apparatus (opt23 ') and a laser introduction window (W0). The substrate is mounted on the substrate stage (S0), and a desired area, for example, a pattern transfer area (ex0) can be exposed to a light pattern by operating the substrate stage.
[0045]
FIG. 1 shows an example of the reduction projection optical system. However, if conditions are met, magnification projection may be performed at the same magnification. An arbitrary region on the substrate (sub0) is irradiated with the laser beam light by moving the substrate stage (the XY direction in the figure). The optical mask (opt 21) is set on a mask stage (not shown). If the optical mask (opt 21) is within an exposure area, the beam irradiated on the substrate can be manipulated by moving the optical mask. Is also possible.
[0046]
Next, an example of a mechanism necessary for irradiating a desired light pattern on a substrate under desired conditions will be described. Since the adjustment of the optical axis requires delicate adjustment, a method of adjusting the position of the substrate while fixing the optical axis once adjusted will be described. For the position of the substrate irradiation surface with respect to the optical axis, it is necessary to correct the position in the focus (Z) direction and the degree of perpendicularity to the optical axis. Therefore, (θxy) tilt correction direction, (θxz) tilt correction direction, (θyz) tilt correction direction, (X) exposure area moving direction, (Y) exposure area moving direction, and (Z) focusing direction shown in the figure. Among them, the degree of perpendicularity to the optical axis is corrected by adjusting the (θxy) tilt correction direction, the (θxz) tilt correction direction, and the (θyz) tilt correction direction. Further, (z) control is performed such that the substrate irradiation surface is arranged at a position corresponding to the depth of focus of the optical system by adjusting the focusing direction.
[0047]
FIG. 2 is a side view illustrating the alignment mechanism for the component to be adjusted and the substrate. The optical mask (opt21), the reduction projection exposure apparatus (opt23 '), and the laser introduction window (W0) are arranged in order with the axis parallel to the exposure axis (L0) as shown in the figure. The substrate (sub0) placed in the vacuum chamber (C0) is placed on a heater (H0) with a substrate suction mechanism and a substrate XYZ-θxy, θxzθpyyz stage (S0 ′). Although a vacuum chamber is used, the actual light irradiation is desirably performed in an atmosphere of an inert gas, hydrogen, oxygen, nitrogen, or the like that has been replaced and filled after evacuation, and the atmospheric pressure is a pressure around atmospheric pressure. There may be.
[0048]
By using a heater with a substrate suction mechanism, it is possible to set a substrate heating condition in a range from room temperature to about 400 ° C. during light irradiation. By setting the atmospheric pressure to about the atmospheric pressure as described above, the substrate can be suctioned by the vacuum chuck function, so that even if the substrate stage is moved in the chamber, the displacement can be prevented, and Even if the substrate has some warpage or deflection, it can be fixed to the substrate stage without fail. Further, it is possible to minimize the deviation of the depth of focus due to the warpage and deflection of the substrate due to heating.
[0049]
In the figure, a laser interferometer (i1, i2) performs alignment of a substrate and precise measurement of the position of the substrate in the Z direction via a length measurement window (Wi) and a length measurement mirror (opt-i). For alignment, an alignment mark provided on the substrate is measured using an off-axis microscope (m0), a light source for a microscope (Lm), and an element for a microscope (opt-m), and substrate position information from a laser interferometer is measured. To measure a desired exposure position.
[0050]
FIG. 2 illustrates the “off-axis method”, but in addition, the “Through The Lens method” and the “Through The Mask (Reticle) method” can be applied. In addition, a method of averaging the measurement errors generated at the time of measurement by determining the linear coordinates from the plurality of measurement points using the least squares method can be adopted.
[0051]
FIGS. 3A to 3C are explanatory diagrams showing the relationship between the mask pattern and the alignment mark. The mask is composed of a first mask (non-exposed portion; mask 1) and a second mask (exposed portion; mask 2). For example, when an excimer laser is used as a light source, a film that absorbs and reflects metals such as aluminum, chromium, and tungsten, and a dielectric multilayer film is formed on a quartz substrate that transmits ultraviolet light, and photolithography and etching techniques are used. Is used to form a predetermined pattern. The silicon film is exposed according to a desired transmission pattern on the mask (indicated by a white portion in FIG. 3A), and the exposed Si is exposed in the unexposed Si (Si1) as shown in FIG. 3B. A portion (Si2) is formed. At this time, if necessary, the alignment is adjusted so that the mark on the mask (mark 1) matches the mark on the substrate (mark 2), and then exposure is performed, thereby exposing a previously designed portion (position) on the silicon thin film. Becomes possible.
[0052]
In the thin film transistor forming step using the silicon thin film, when the exposure process is a first step (first step) requiring positioning (that is, when an alignment mark is not formed in advance), an exposing step on the silicon thin film is performed. At the same time, by simultaneously exposing the exposure forming mark (mark 3), an alignment mark utilizing the optical difference between a-Si and crystalline Si can be formed. The hue difference between a-Si and crystalline Si provides an alignment mark that functions sufficiently for practical use.
[0053]
By performing photolithography and the like in the subsequent process based on the alignment marks obtained in this way, the transistors, desired mechanisms and functions are accurately aligned with the desired regions that have been subjected to exposure modification, and are sequentially formed. Can be included. FIG. 3C shows a state in which a silicon oxide film is formed on the silicon thin film after the exposure step, and a desired region of the silicon layer is removed by etching. The Si-removed portion (Si3) is a region where the laminated silicon film and the Si oxide film are removed by etching, and a Si oxide film (Si4, Si5) is formed on the non-exposed Si portion (Si1) and the exposed Si portion (Si2). The stacked shape is shown. By forming an island-like structure composed of a silicon film covered with an oxide film in this manner, it is possible to form a channel / source / drain region of a thin film transistor separated between elements and a mark necessary for alignment in a later step. .
[0054]
4A and 4B show timing charts of main operations of the apparatus. In the control example 1 of FIG. 4A, the substrate is moved to a desired exposure position by operating the substrate stage. Next, focusing and alignment adjustment operations are performed to precisely adjust the exposure position. At this time, the adjustment is performed so as to reach a desired setting error accuracy, for example, about 0.1 μm to 100 μm. As soon as this position adjustment operation is completed, the substrate is irradiated with light. When these series of operations are completed, the substrate moves to the next exposure area, and all necessary portions on the substrate are irradiated. When the irradiation to all the necessary parts is completed, the substrate is exchanged and a similar series of predetermined processing is performed on the next processing substrate.
[0055]
In the control example 2 shown in FIG. 4B, the substrate is moved to a desired exposure position by the operation of the substrate stage. Next, focusing and alignment operations are performed to precisely adjust the exposure position. At this time, adjustment is made so as to fall within a desired setting error accuracy, for example, about 0.1 μm to 100 μm. When the operation is completed, the movement operation of the mask stage is started (start). 4 is a chart showing control in which light irradiation on the substrate is started after the start of the mask stage operation in order to avoid a variation in the moving step amount at the time of starting. Needless to say, since the stage is moved to expose a point distant from the alignment position, the offset amount must be considered in advance.
[0056]
It is also possible to start the operation of the light source earlier than the light irradiation on the substrate, and to open the shutter or the like to perform light irradiation on the substrate when the stability of the output intensity of the light source increases. In particular, when an excimer laser is used as a light source and a control method is employed in which the oscillation period and the stop period are repeated, it is known that the initial several tens of pulses are particularly unstable. If it is not desired to irradiate the beam, a method of blocking the beam in accordance with the operation of the mask stage can be adopted. When these series of operations are completed, the substrate moves to the next exposure area, and irradiation of all necessary portions on the substrate is completed. Thereafter, the substrate is replaced, and a similar predetermined series of processing is repeated on the second processing substrate.
[0057]
[Experimental Example] A 2 μm × 50 μm beam was scanned at a 0.5 μm pitch in the minor axis direction on an a-Si thin film having a thickness of 75 nm. Using one light source, the laser irradiation intensity is 470 mJ / cm on the irradiation surface. ² As a result, a single-crystal silicon thin film continuous in the scanning direction was obtained. Further, the second light source was set to 150 mJ / cm on the irradiation surface. ² Under the condition of irradiation with a delay of 100 nsec, a single-crystal silicon thin film continuous in the scanning direction was obtained even under the scanning pitch condition of 1.0 μm. The trap level density in the crystallized silicon film is 10 ¹² cm ^-2 It showed a lower value.
[0058]
FIG. 5 is a side view of the semiconductor thin film forming apparatus more specifically showing the embodiment of the present invention. It is composed of a plasma CVD chamber (C2), a laser irradiation chamber (C5), and a substrate transfer chamber (C7). The transfer of the substrate through the gate valves (GV2, GV5) can be performed in a vacuum without contacting the atmosphere outside the apparatus. It is possible in an atmosphere of an active gas, nitrogen, hydrogen, oxygen or the like and in a high vacuum, reduced pressure, or pressurized state. In the laser irradiation chamber (C5), a substrate is set using a chuck mechanism on a substrate stage (S5) that can be heated to about 400 ° C. In the plasma CVD chamber (C2), a substrate is placed on a substrate holder (S2) that can be heated to about 400 ° C. In this example, a silicon thin film (Si1) formed on a glass substrate (Sub0) is introduced into a laser irradiation chamber (C5), and the silicon thin film on the surface is reformed into a crystalline silicon thin film (Si2) by laser irradiation. And a state transferred to the plasma CVD chamber (C2).
[0059]
As for the laser light introduced into the laser irradiation chamber, beams supplied from the excimer laser 1 (EL1) and the excimer laser 2 (EL2) pass through the first beam line (L1) and the second beam line (L2). Laser synthesis optical device (opt1), mirror (opt11), transmission mirror (opt12), laser irradiation optical device (opt2), homogenizer (opt20), optical mask (opt21) fixed to optical mask stage (opt22), projection optics The light reaches the substrate surface via the device (opt23) and the laser introduction window (W1). Although two excimer lasers are shown here, one or more desired light sources can be installed. In addition, the laser beam is not limited to the excimer laser, and may be supplied in pulse form using a pulse laser such as a carbon dioxide gas laser or a YAG laser, or a CW light source such as an argon laser and a high-speed shutter.
[0060]
On the other hand, in the plasma CVD chamber, the plasma formation region (D2) is formed at a position apart from the region where the substrate is arranged by the RF electrode (D1) and the plasma confinement electrode (D3). A silicon oxide film can be formed on a substrate by supplying, for example, oxygen and helium to the plasma formation region and a silane gas using a source gas introduction device (D4).
[0061]
FIG. 6 shows a plan view of a semiconductor thin film forming apparatus showing an embodiment of the present invention. The load / unload chamber (C1), the plasma CVD chamber (C2), the substrate heating chamber (C3), the hydrogen plasma processing chamber (C4), the laser irradiation chamber (C5), and the substrate transfer chamber (C7) are each a gate valve (GV1). To GV6). Laser light supplied from the first beam line (L1) and the second beam line (L2) is applied to the substrate via a laser combining optical device (opt1), a laser irradiation optical device (opt2), and a laser introduction window (W1). Irradiated on the surface. A gas introduction device (gas1 to gas7) and an exhaust device (vent1 to vent7) are connected to each process chamber and transfer chamber, and supply of a desired gas type, setting of process pressure, exhaust, and vacuum are adjusted. You. The processing substrates (sub2, sub6) are arranged on a plane as shown by a dotted line in the figure.
[0062]
FIG. 7 is a schematic side view schematically showing the configuration of the C2 plasma CVD chamber. Power is supplied to a high-frequency electrode (RF2) from a high-frequency power supply (RF1) (a high frequency of 13.56 MHz or more is suitable). Plasma is formed between the electrode with gas supply hole (RF3) and the high-frequency electrode (RF2), and radicals formed by the reaction are guided to the region where the substrate is arranged through the electrode with gas supply hole. Another gas is introduced by the flat gas introduction device (RF4) without exposure to plasma, and a thin film is formed on the substrate (sub2) through a gas phase reaction. The substrate holder (S2) is designed to heat from room temperature to about 500 ° C. by a heater or the like. As shown in the figure, an exhaust device (ven2), a gas introduction device (gas2), an oxygen line (gas21), a helium line (gas22), a hydrogen line (gas23), a silane line (gas24), a helium line (gas25), and an argon line A silicon oxide film can be formed by reacting oxygen radicals with silane gas using (gas 26).
[0063]
When a film was formed under the conditions of a substrate temperature of 300 ° C., a pressure of 0.1 torr, an RF power of 100 W, a silane flow rate of 10 sccm, an oxygen flow rate of 400 sccm, and a helium flow rate of 400 sccm, the fixed oxide film charge density (5 × 10 ¹¹ cm ^-2 ) And the formation of a silicon oxide film having good characteristics was confirmed. In addition, it is possible to form a better oxide film by increasing the flow rate ratio of oxygen to silane. As the form of the plasma CVD chamber, not only the above-described parallel plate type RF plasma CVD apparatus but also a method using no plasma, such as low pressure CVD or normal pressure CVD, or a microwave or ECR (Electron Cyclotron Resonance) effect is used. It is also possible to use a plasma CVD method.
[0064]
For forming the silicon nitride film (Si3N4), N2 (nitrogen) (or ammonia), Ar (argon), SiH4 (silane) as a carrier gas, and argon or the like as a carrier gas can be used. For forming the Si silicon thin film, a source gas such as H2 (hydrogen) and silane, hydrogen (argon as a carrier gas), and SiF4 (silane tetrafluoride, argon as a carrier gas) can be used. Although not a film formation process, hydrogen plasma treatment of a silicon thin film or a silicon oxide film using hydrogen plasma is also possible.
[0065]
(A) to (e), (f1), (f1), (g1), and (g2) of FIG. 8 are process flowcharts when the semiconductor thin film forming apparatus of the present invention is applied to a thin film transistor manufacturing process.
[0066]
(A): A substrate cover film (T1) and a silicon thin film (T2) are sequentially formed on a glass substrate (sub0) from which organic substances, metals, fine particles and the like have been removed by washing. As a substrate cover film, a silicon oxide film of 1 μm is formed at 450 ° C. using silane and oxygen gas as raw materials by a low pressure chemical vapor deposition (LPCVD) method. By using the LPCVD method, it is also possible to cover the entire outer surface of the substrate except for the substrate holding region (not shown). Alternatively, plasma CVD using TEOS (tetraethoxysilane) and oxygen as raw materials, normal pressure CVD using TEOS and ozone as raw materials, plasma CVD as shown in FIG. 8, or the like can be used. Materials that can prevent diffusion of impurities harmful to semiconductor devices, including glass whose concentration is reduced as much as possible and quartz and glass whose surfaces are polished are effective as the substrate cover film. A silicon thin film is formed at a temperature of 500 ° C. and a thickness of 75 nm using disilane gas by LPCVD. In this case, since the concentration of hydrogen atoms contained in the film is 1 atomic% or less, it is possible to prevent the film from being roughened due to hydrogen release in the laser irradiation step. Alternatively, even if a plasma CVD method as shown in FIG. 7 or a widely-used plasma CVD method is used, the hydrogen atom concentration is adjusted by adjusting the substrate temperature, the flow ratio of hydrogen / silane, the flow ratio of hydrogen / 4-fluorinated silane, and the like. Can form a silicon thin film having a low density.
[0067]
(B): The substrate prepared in the above step (a) is introduced into the thin film forming apparatus of the present invention after undergoing a cleaning step for removing organic substances, metals, fine particles, surface oxide films and the like. Irradiation with laser light (L0) modifies the silicon thin film into a crystallized silicon thin film (T2 '). The laser crystallization is performed in an atmosphere of high purity nitrogen of 99.9999% or more and 700 torr or more.
[0068]
(C); The substrate that has gone through the above steps is transferred to the plasma CVD chamber via the substrate transfer chamber after the gas is exhausted. As a first gate insulating film (T3), a silicon oxide film is deposited to a thickness of 10 nm at a substrate temperature of 350 ° C. using silane, helium, and oxygen as source gases. Thereafter, hydrogen plasma treatment or heat annealing is performed as necessary. The processing up to this point is performed in the thin film forming apparatus of the present invention.
[0069]
(D); Next, islands of a silicon thin film and a silicon oxide film stacked film are formed by using photolithography and etching techniques. At this time, it is preferable to select an etching condition in which the etching rate of the silicon oxide film is higher than that of the silicon thin film. As shown in the drawing, when the pattern section is formed in a step shape (or a tapered shape), a gate thin film can be prevented and a highly reliable thin film transistor can be provided.
[0070]
(E): Next, after performing cleaning for removing organic substances, metals, fine particles and the like, a T4 second gate insulating film is formed so as to cover the islands. Here, a 30-nm-thick silicon oxide film was formed at 450 ° C. using silane and oxygen gas as raw materials by an LPCVD method. Alternatively, plasma CVD using TEOS (tetraethoxysilane) and oxygen as raw materials, normal pressure CVD using TEOS and ozone as raw materials, and plasma CVD as shown in FIG. 8 can be used. Next, an n + silicon film of 80 nm and a tungsten silicide film of 110 nm are formed as gate electrodes. The n + silicon film is preferably a crystalline phosphorus-doped silicon film formed by a plasma CVD or LPCVD method. Thereafter, a patterned gate electrode (T5) is formed through photolithography and an etching process.
[0071]
(F1, f2); Next, an impurity implantation region (T6, T6 ') is formed using the gate as a mask. When a CMOS circuit is formed, an n-channel TFT requiring an n + region and a p-channel TFT requiring a p + region are separately formed by using photolithography. Methods such as ion doping without performing mass separation of the impurity ions to be implanted, ion implantation, plasma doping, and laser doping can be employed. At this time, the impurity is introduced while the silicon oxide film on the surface is left or removed as shown in FIGS.
[0072]
(G1) (g2); Deposit interlayer insulating films (T7, T7 '), open contact holes, deposit metal, and form metal wiring (T8) by photolithography and etching. As the interlayer isolation insulating film, a TEOS-based oxide film, a silica-based coating film, or an organic coating film that can planarize the film can be used. The contact hole opening can be formed by photolithography and etching, and the metal wiring can be made of aluminum, copper or an alloy based thereon having a low resistance, or a high melting point metal such as tungsten or molybdenum. By performing the above steps, a thin film transistor with high performance and high reliability can be formed.
[0073]
FIG. 9 shows an embodiment in which an alignment mark is provided in advance and laser irradiation according to the alignment mark is performed, and FIG. 10 shows an embodiment in which an alignment mark is formed simultaneously with laser irradiation based on a TFT manufacturing process flow. explain. Since the description is basically similar to that of FIG. 8, only the differences will be mainly described.
[0074]
FIG. 9A: A substrate cover film (T1) and a tungsten silicide film are sequentially formed on a glass substrate (sub0) from which organic substances, metals, fine particles and the like have been removed by washing. In order to form an alignment mark, patterning is performed by photolithography and etching to form an alignment mark (T9) on the substrate. Next, a mark protection film (T10) is formed to protect the alignment mark, and a silicon thin film is formed.
[0075]
FIG. 9B: At the time of laser beam exposure, a desired region is exposed based on the alignment mark. After that, the alignment in the next step can be performed based on the alignment mater provided in advance or the alignment mark (not shown) formed by patterning the crystallized silicon thin film.
[0076]
FIG. 10B: Simultaneously with the exposure of the silicon thin film, a crystallization alignment mark (T9 ′) is formed on the silicon thin film using the difference in the modification between exposure and non-exposure.
[0077]
FIG. 10D: The alignment at the time of photolithography is performed using the crystallization alignment mark (T9 ′), and an island of a silicon thin film and a silicon oxide film stacked film is formed through an etching process.
[0078]
According to the above-described manufacturing process, by aligning a beam using an alignment function with respect to a mark formed on a substrate on which a silicon film is deposited, a position accuracy of the order of μm or more can be obtained in a desired region. Exposure. As a result, it was possible to minimize the laser irradiation on the region to be finally etched away. In particular, when applied to an imaging device such as an LCD, it has become possible to prevent substrate damage due to intensity variation of a light source and deterioration of image quality due to the substrate damage.
[0079]
【The invention's effect】
Covering the surface of the thin film partially modified by laser irradiation and having a partially active surface with another thin film to prevent contamination of the active surface in the next process and binding of unnecessary substances Is realized, and a thin film transistor with high reliability and high reproducibility can be obtained.
[0080]
Simultaneous modification of the silicon thin film and formation of the alignment mark in the laser irradiation process eliminates the need for an additional alignment formation process, leading to prevention of element contamination and shortening of the manufacturing process, resulting in a significant reduction in manufacturing cost. The industrial effect is great.
[Brief description of the drawings]
FIG. 1 is an overall view showing a schematic arrangement of a main part of an exposure apparatus having a reduction projection optical system according to an embodiment of the present invention.
FIG. 2 is a side view illustrating an alignment mechanism for a component to be adjusted and a substrate in the exposure apparatus.
FIG. 3 is a plan view illustrating a mask and each pattern according to the exposure apparatus.
FIGS. 4A and 4B are timing charts of a control operation in the exposure apparatus.
FIG. 5 is a side sectional view of a semiconductor thin film forming apparatus illustrating one embodiment of the present invention.
FIG. 6 is a plan view of a semiconductor thin film forming apparatus for explaining an embodiment of the present invention.
FIG. 7 is a schematic side view schematically showing a configuration of a plasma CVD chamber.
FIGS. 8 (a) to (e), (f1), (f1), (g1) and (g2) are process flow charts for explaining a process of manufacturing a thin film transistor by the semiconductor thin film forming apparatus according to the present invention.
FIGS. 9 (a) to (e), (f1), (f1), (g1), and (g2) are process flow charts for explaining a manufacturing process of a thin film transistor according to the present invention using an alignment mark.
FIGS. 10 (a) to (e), (f1), (f1), (g1), and (g2) are process flow charts for explaining a manufacturing process of a thin film transistor in which an alignment mark is formed simultaneously.
FIG. 11 is a schematic diagram showing an example of a configuration of a conventional pulse laser irradiation device.
FIG. 12 is a time history curve showing a temperature change of a silicon film.
FIG. 13 is an explanatory diagram showing an example of a distribution of the intensity of a continuous laser pulse between pulses.
FIGS. 14A to 14C are timing charts illustrating a known laser operation method.
FIG. 15 is a waveform diagram showing an intensity waveform of a laser pulse.
FIG. 16 is a diagram showing an example of irradiation intensity, cooling rate, and the like.
FIG. 17 is a diagram showing an example of an experimental value of a silicon thin film temperature change.
FIG. 18 is an electron micrograph illustrating the crystal morphology of a laser recrystallized silicon thin film with respect to each irradiation intensity and irradiation frequency.
FIG. 19 is a diagram showing a maximum cooling rate after a second pulse is injected and a cooling rate near a freezing point.
[Explanation of symbols]
1101: pulse laser light source
1102: Mirror
1103: Mirror
1104: Beam homogenizer
1105: Mirror
1106: Optical path
1107: Silicon thin film
1108: Glass substrate
1109: XY stage
1110: Glass substrate cassette
1111: substrate transfer device
C0: vacuum chamber
C1: Load / unload room
C2: Plasma CVD chamber
C3: Substrate heating chamber
C4: Hydrogen plasma processing chamber
C5: Laser irradiation room
C7: Substrate transfer chamber
D1RF: Electrode
D2: Plasma formation area
D3: Plasma confinement electrode
D4: Source gas introduction device
EL1: First excimer laser
EL2: Second excimer laser
GV1 to GV6: Gate valve
GV2: Gate valve
GV5: Gate valve
H0: heater with substrate suction mechanism
L0: Exposure axis laser beam
L1: First beam line
L2: second beam line
Lm: Light source for microscope
RF1: High frequency power supply
RF2: High frequency electrode
RF3: Electrode with gas supply hole
RF4: Flat type gas introduction device
S0: Substrate stage
S0 ': Substrate XYZθXYθXZθYZ stage
S2: Substrate holder
S5: Substrate stage
Si1: Silicon thin film unexposed Si
Si2: Crystal silicon thin film exposure Si part
Si3: Si removal part
Si4: Si oxide film
Si5: Si oxide film
Sub0: glass substrate
T1: Substrate cover film
T10: Mark protective film
T2: Silicon thin film
T2 ': Crystallized silicon thin film
T3: First gate insulating film
T4: second gate insulating film
T5: patterned gate electrode
T6, T6 ': impurity implantation region
T7, T7 ': interlayer isolation insulating film
T8: Metal wiring
T9: Alignment mark
Wi: window for length measurement
W0: Laser introduction window
W1: Laser introduction window
X: exposure area moving direction
Y: exposure area moving direction
Z: exposure area moving direction
ex0: pattern transfer area
gas1 to gas7: Gas introduction device
gas21: Oxygen line
gas22: Helium line
gas23: hydrogen line
gas24: silane line
gas25: Helium line
gas26: Argon line
i1: Laser interferometer
i2: Laser interferometer
m0: Off-axis microscope
mark1: mark on the mask
mark2: mark on substrate
mark3: Exposure mark
mask1: mask (non-exposed part)
mask2: mask (exposed part)
opt-i: Mirror for length measurement
opt-m: element for microscope
opt1: Laser synthesis optical device
opt11: mirror
opt12: transmission mirror
opt2: Laser irradiation optical device
opt20: Homogenizer
opt20 ': Homogenizer
opt21: Optical mask
opt22: Optical mask stage
opt23: Projection optical device
opt23 ': reduction projection exposure apparatus
opt3: mirrors
opt3 ': mirrors
opt4: Lenses
sub0: glass substrate
sub0: Substrate
sub2, sub6: processing substrate
sub2: substrate holder
vent1-vent7: Exhaust device
θXY: inclination correction direction
θXZ: tilt correction direction
θYZ: tilt correction direction

Claims (1)

A semiconductor thin film formed on a substrate, where a region that has undergone modification by light irradiation and a region that does not undergo light irradiation are mixed, and a thin film having a composition different from that of the semiconductor thin film is stacked on the surface of the semiconductor thin film. A semiconductor thin film comprising: an alignment mark formed by using a color difference between a region that has undergone the modification by the light irradiation and a region that has not undergone the light irradiation that has a different optical color. Semiconductor thin film.

Images