JP2007305852A - Semiconductor thin film manufacturing method and semiconductor device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of manufacturing a semiconductor thin film that enables reducing variations in the characteristics of a semiconductor device, and a semiconductor device having a channel region comprising the semiconductor thin film. <P>SOLUTION: The method of manufacturing a semiconductor thin film comprises the steps of: irradiating laser light into a first region of an amorphous semiconductor film through a mask slit, to melt and subsequently crystallize the amorphous semiconductor film; moving a region to be irradiated with the laser light such that laser light is radiated in a second region which superimposes on part of the first region; and radiating laser light into the second region of the amorphous semiconductor film through the mask slit, to melt and subsequently crystallize the amorphous semiconductor film. The mask slit width and the width of the region overlapping the first and second regions are set such that one or more linearly aligned ridge lines are included in the channel region on the surface of the semiconductor thin film. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a method for manufacturing a semiconductor thin film and a semiconductor device, and more particularly to a method for manufacturing a semiconductor thin film capable of reducing variations in characteristics of the semiconductor device and a semiconductor device having a channel region made of a semiconductor thin film obtained by this method. About.

  In general, as a method for manufacturing a semiconductor device, there is a method using single crystal silicon, but there is also a method using a silicon thin film formed on a glass substrate in addition to this manufacturing method. A semiconductor device manufactured using a silicon thin film formed on a glass substrate is used as a part of an image sensor or an active matrix liquid crystal display device.

  Here, in a liquid crystal display device, a semiconductor device is used as a TFT (Thin Film Transistor) arranged as a regular array on a transparent substrate. Each TFT functions as a pixel controller.

  Conventionally, TFTs in liquid crystal display devices have been manufactured by forming an amorphous silicon film on a glass substrate. However, in recent years, a TFT liquid crystal display that uses a polycrystalline silicon film with high electron mobility instead of an amorphous silicon film with low electron mobility to enhance the switching characteristics of the TFT and increase the display speed. Devices are being manufactured. Here, as a method of manufacturing a polycrystalline silicon film, for example, a method of forming a polycrystalline silicon film by irradiating an amorphous silicon film or a microcrystalline silicon film deposited on a glass substrate with excimer laser light. (ELC method).

  As the above-mentioned ELC method, an amorphous or microcrystalline silicon film deposited on a glass substrate is scanned at a constant speed while having a length of about 200 to 400 mm and a width of about 0.2 to 1.0 mm. A method of continuously irradiating linear laser light is common. At this time, the portion irradiated with the laser beam does not melt over the entire thickness direction of the silicon film, but melts while leaving a part of the region. For this reason, crystal nuclei are generated all over the interface between the unmelted region and the molten region of the silicon film, and the crystal grows toward the outermost layer of the silicon film, thereby forming crystal grains with random orientation. The crystal grain size of the polycrystalline silicon film is as very small as 100 to 200 nm.

  Since many unpaired electrons exist in the crystal grain boundary of the polycrystalline silicon film and a potential barrier is formed, electrons are often scattered in the crystal grain boundary of the polycrystalline silicon film. Therefore, in general, a TFT formed of a polycrystalline silicon film with fewer crystal grain boundaries, that is, a larger crystal grain size, has higher electron mobility due to the field effect.

  However, in the above-mentioned ELC method, since it becomes longitudinal crystal growth in which crystallization occurs at random positions at the interface between the unmelted region and the molten region, it is difficult to obtain a polycrystalline silicon film having a large crystal grain size, It has been difficult to obtain a TFT having a high electron mobility due to a field effect.

  In order to obtain a higher performance TFT liquid crystal display device, it is necessary to increase the crystal grain size of the polycrystalline silicon film and to control the crystal orientation. Therefore, many proposals have been made for the purpose of obtaining a polycrystalline silicon film having performance close to that of single crystal silicon. Among them, the laser crystallization technique classified as the lateral growth method is attracting attention because a long crystal having a uniform orientation in the crystal growth direction can be obtained.

  In one of the lateral growth methods described above, the amorphous silicon film is irradiated with a fine-width pulse laser beam through a slit of the mask, and the irradiated portion of the amorphous silicon film is irradiated in the thickness direction of the laser beam irradiation region. There is a method (SLS method) in which crystallization is performed by repeatedly melting and then solidifying the entire region (for example, see Patent Document 1).

  According to the SLS method, since the amorphous silicon film irradiated with the laser beam can be melted over the entire thickness direction, the interface between the unmelted region and the melted region of the silicon film always has a laser beam. Formed at the end of the irradiated region, the crystal grows laterally from the interface. Therefore, in the SLS method, a long crystal having a large crystal grain size in which the orientation is aligned in the crystal growth direction can be obtained.

Specifically, for example, as shown in the schematic top view of FIG. 8A, the boundary between the region F of the amorphous silicon film 300 and the other region is directed to the center of the region F. , The crystal grows. Further, as shown in FIG. 8B, a new region G adjacent to the region F is set so that a portion where the crystal is not formed in the region F is included, and the region G is set in the same manner as the region F. Melt. Then, when the portion melted in the region G is solidified, a crystal grows in the region G as shown in FIG. When such a procedure is repeated and crystallized stepwise along the extending direction of the amorphous silicon film 300, the crystal growth region of the amorphous silicon film 300 is expanded as shown in FIG. can do. Thereby, a crystalline semiconductor film having a large crystal grain size can be formed.
Japanese Patent No. 3204986

  In the SLS method described above, a ridge row in which ridges (projections) in which the surface of the semiconductor thin film is locally high are arranged in a straight line on the surface of the semiconductor thin film corresponding to the vicinity of the central portion of the slit width of the mask. It is formed. This ridge row is formed when crystals grown in the lateral direction collide with each other from the interface between the unmelted region and the molten region corresponding to both ends of the slit width of the mask.

  When a semiconductor thin film is formed by the above-described SLS method and a TFT having a channel region made of the semiconductor thin film is manufactured, the crystal grain size constituting the semiconductor thin film is increased, and as a result, the ridge row is arranged in the channel region. The TFTs that are formed and the TFTs in which the ridge row is not arranged in the channel region are necessarily mixed.

  The characteristics of a TFT including such a ridge row are due to the fact that the gate insulating film formed on the ridge is thinner than the portion other than the ridge row, so that the threshold voltage is particularly higher than that of a TFT not including the ridge row. There is a problem in that the characteristics of TFT vary greatly.

  In particular, when TFTs having greatly different threshold voltages are applied to the same liquid crystal display device, it is difficult to reduce power consumption because the power supply voltage cannot be lowered.

  In view of the above circumstances, an object of the present invention is to provide a method for manufacturing a semiconductor thin film capable of reducing variations in characteristics of semiconductor devices, and a semiconductor device having a channel region made of a semiconductor thin film obtained by this method. There is.

  According to the present invention, the first region of the amorphous semiconductor film is irradiated with laser light through the slit of the mask to melt the amorphous semiconductor film in the first region over its thickness and then crystallize. An amorphous semiconductor through a crystallization step, an irradiation region moving step of moving the laser light irradiation region so that the second region where the first region partially overlaps the first region is irradiated with the laser light, and a mask slit A second crystallization step in which the amorphous semiconductor film in the second region is melted over its thickness by irradiating the second region of the film with laser light, and then crystallized. A method for manufacturing a semiconductor thin film constituting a channel region of a semiconductor device in at least a partial region of a semiconductor film, wherein a ridge row in which ridges formed on the surface of the semiconductor thin film are arranged in a straight line is formed in the channel region 1 or more per Unnecessarily, characterized in that it is a method for manufacturing a semiconductor thin film that sets the width of the region in which the slit width and the first region and the second region of the mask overlaps.

  Here, in the method for manufacturing a semiconductor thin film of the present invention, the maximum value of the difference between the width of the first region and the width of the region where the first region and the second region overlap is the channel in the direction orthogonal to the ridge row. It is preferable to set the slit width of the mask and the width of the region where the first region and the second region overlap so as to be equal to or less than the length of the region.

  In the method of manufacturing a semiconductor thin film according to the present invention, the slit width of the mask and the first region and the first region are set so that the maximum value of the interval between the ridge rows is equal to or less than the length of the channel region in the direction orthogonal to the ridge rows. It is preferable to set the width of the region overlapping the two regions.

  In the method for producing a semiconductor thin film of the present invention, the first crystallization step and the second crystallization step have a first energy amount that melts the amorphous semiconductor film as a laser beam over the thickness of the film. While irradiating a laser beam, you may irradiate the 2nd laser beam which has an energy amount which does not fuse | melt an amorphous semiconductor film over the thickness of the film | membrane.

  Furthermore, the present invention provides a semiconductor device having a channel region made of a semiconductor thin film manufactured by the method for manufacturing a semiconductor thin film according to any one of the above, wherein a ridge row formed on the surface of the semiconductor thin film is formed in the channel region. Are one or more semiconductor devices.

  ADVANTAGE OF THE INVENTION According to this invention, the semiconductor device which has the channel area | region which consists of a semiconductor thin film obtained by this method and the semiconductor thin film which can reduce the dispersion | variation in the characteristic of a semiconductor device, and this method can be provided.

  Embodiments of the present invention will be described below. In the drawings of the present invention, the same reference numerals represent the same or corresponding parts.

(Embodiment 1)
FIG. 1 shows a schematic cross-sectional view of a preferred example of the object to be processed used in the present invention. Here, the target object 101 used in the present invention includes a buffer film 103 formed on the semiconductor substrate 102 and an amorphous semiconductor film 104 formed on the buffer film 103. Here, the semiconductor substrate 102 is preferably insulating, and a glass substrate, a quartz substrate, or the like can be used. However, it is preferable to use a glass substrate because it is inexpensive and a large-area substrate can be easily manufactured. It is.

  The buffer film 103 is formed in order to prevent the semiconductor substrate 102 from being affected by the heat of the melted amorphous semiconductor film 104 mainly when the amorphous semiconductor film 104 is melted and crystallized by laser light. The Further, by forming the buffer film 103, diffusion of impurities from the semiconductor substrate 102 to the amorphous semiconductor film 104 can be prevented. As the buffer film 103, for example, a silicon oxide film, a silicon nitride film, a silicon oxynitride film, or the like formed by vapor deposition, ion plating, sputtering, or the like with a thickness of 100 to 300 nm can be used.

  The amorphous semiconductor film 104 is formed, for example, by plasma enhanced chemical vapor deposition (PECVD), vapor deposition, or sputtering so that the film thickness becomes 10 nm to 100 nm. The amorphous semiconductor film 104 has various characteristics by increasing the crystal growth length of the crystalline semiconductor film (for example, values of electron mobility, threshold voltage, and the like due to a field effect, and variations in these values). It is preferable to use an amorphous silicon film in which the resistance is remarkably improved. The material of the amorphous semiconductor film 104 is not limited to a material made of silicon, and a material mainly containing silicon containing other elements such as germanium can also be used.

  FIG. 2 shows a schematic configuration of a laser beam irradiation apparatus used in the present embodiment. Here, the laser beam irradiation apparatus includes a laser beam oscillator 11, a variable attenuator 12, a field lens 13, a mask 14 with a slit, an imaging lens 15, a sample stage 16, a controller 17, and mirrors 7, 8, and 9. Contains.

  The field lens 13 has a function of causing the laser light that has passed through the mask 14 to enter the imaging lens 15 perpendicularly. Further, the mirrors 7, 8, and 9 are used for folding the laser light, and there are no restrictions on the arrangement location and quantity, and the mirrors 7, 8, and 9 can be appropriately arranged according to the optical design and functional design of the laser light irradiation apparatus. The imaging lens 15 forms an image of the mask 14 on the amorphous semiconductor film on the surface of the workpiece 101 at a magnification determined by the optical design of the laser beam irradiation apparatus. Therefore, the shape of the slit formed in the mask 14 is similar to the image on the amorphous semiconductor film.

  Further, the laser beam oscillator 11 and the sample stage 16 are controlled by the controller 17, and the laser beam irradiation timing and the position of the sample stage 16 can be adjusted. Thereby, the area irradiated with the laser beam can be moved by moving the sample stage 16 in the direction of the arrow 22 in FIG.

  The laser oscillator 11 is not particularly limited as long as the amorphous semiconductor film on the surface of the workpiece 101 can be melted over its thickness. For example, excimer laser light, YAG It is preferable to use one that oscillates laser light having a wavelength in the ultraviolet region, such as a third harmonic wave of laser light oscillated from various solid-state lasers typified by lasers. This is because when the laser beam has a wavelength in the ultraviolet region, the amount of energy of the laser beam increases, and the amorphous semiconductor film tends to be easily melted over its thickness. In particular, it is particularly preferable that the laser light oscillator 11 oscillates excimer laser light having a wavelength of 308 nm capable of pulse emission. When the laser beam oscillator 11 is an excimer laser beam, a laser beam oscillator capable of stably oscillating a laser beam having an energy amount necessary for the present invention can be obtained relatively easily at a low cost. This is because.

  Here, the laser light oscillated from the laser light oscillator 11 has an energy amount for melting the amorphous semiconductor film in a solid state per one irradiation and irradiation area. For example, it is recommended that the laser light oscillated from the laser light oscillator 11 has an energy amount that can heat the amorphous semiconductor film to a temperature equal to or higher than the melting point in the entire film thickness.

  FIG. 3A shows a schematic plan view of a preferred example of the mask used in the present invention. Here, the mask 14 has two adjacent rectangular blocks 23 and 24. As shown in FIG. 3B, the block 23 has a slit 25 that is an opening through which the laser beam passes and a frame 29 that is a region through which the laser beam does not pass, as shown in FIG. As described above, the block 24 includes a slit 26 that is an opening through which laser light passes and a frame portion 29 that is a region through which laser light does not pass. Here, the slits 25 and 26 are formed at predetermined intervals so that the direction orthogonal to the moving direction of the sample stage 16 is the longitudinal direction of the images of the slits 25 and 26.

  An object to be processed 101 shown in FIG. 1 is placed on the sample stage 16 of the laser beam irradiation apparatus. The laser light oscillated from the laser light oscillator 11 passes through the variable attenuator 12 and is reflected by the mirrors 7 and 8. Thereafter, the laser light passes through the field lens 13 and the mask 14. Then, the laser light that has passed through the mask 14 is formed as an image of the mask 14 by the imaging lens 15 and irradiated onto the amorphous semiconductor film 104 on the surface of the object 101 to be processed shown in FIG. This laser light is applied to the first region C shown in FIG. Then, when the amorphous semiconductor film 104 in the first region C irradiated with the laser light is solidified after being melted over the entire thickness direction, an unmelted region and a melted region of the amorphous semiconductor film 104 A crystal grows laterally from the interface and crystallizes (first crystallization step).

  Next, the sample stage 16 is controlled by the controller 17, and the sample stage 16 is moved to the left of the arrow 22, for example. Here, the sample stage 16 is moved so that the second region where the first region C and a part of the region overlap is irradiated with the laser light (irradiation region moving step).

  Then, in the same manner as in the first crystallization step, by irradiating the second region with laser light, the amorphous semiconductor film in the second region irradiated with the laser light extends over the entire thickness direction. When solidifying after melting, the crystal grows laterally from the interface between the unmelted region and the molten region of the amorphous semiconductor film and is crystallized (second crystallization step).

  By performing the first crystallization step, the irradiation region moving step, and the second crystallization step in this order at least once, a desired region of the amorphous semiconductor film of the object 101 can be crystallized. .

  FIG. 4A shows a schematic enlarged plan view of an example of the crystallization state of the semiconductor thin film after the first crystallization step of the present invention. FIG. 4B is a schematic enlarged plan view showing an example of the crystallization state of the semiconductor thin film after the second crystallization step of the present invention.

  Here, as shown in FIG. 4A, after the first crystallization step, the crystals 42 grown to substantially the same length from the end in the width direction of the first region having the width 27 collide with each other. A ridge is formed, and a ridge row 41 in which the ridges are arranged is formed. Then, as shown in FIG. 4B, the ridge row 41 is separated by a predetermined ridge row interval 55 according to the overlap width 28 which is the width of the region where the first region and the second region overlap. Arrange. Here, the ridge row interval 55 is expressed by the difference between the width 27 of the first region and the overlap width 28.

  Next, an unnecessary portion of the crystallization region formed through the first crystallization step, the irradiation region moving step, and the second crystallization step is removed by a photolithography step, and crystallization constituting a channel region of the semiconductor device is performed. By leaving only the region, an island-shaped crystallized film is formed.

  FIG. 5 schematically shows an example of the arrangement relationship of the channel region with respect to the crystallization region. Although the channel region 50 (a region functioning as a channel through which electrons move when a semiconductor device is formed) can be freely arranged with respect to the crystallization region 49, in the present embodiment, electrons are present in the channel region 50. The channel region 50 is arranged so that the moving direction (channel length direction 52) and the direction in which the ridge row 41 extends (ridge row direction 53) are orthogonal to each other.

  In this embodiment, the length 54 of the channel region 50 in the direction orthogonal to the ridge row 41 is S, the ridge row interval 55 is R, the width 27 of the first region is P, and the overlap width is When 28 is Q, the slit width and the overlap width 28 of the mask 14 are set so that the maximum value of R (= P−Q) in the crystallization region 49 is S or less. As a result, the number of ridge rows 41 included in the channel region 50 is always one or more.

  As the number of ridge rows 41 included in the channel region 50 increases, the number of times of laser light irradiation increases and the time required for crystallization increases. Therefore, from the viewpoint of increasing the crystallization throughput, the maximum value of R is preferably as large as possible. Specifically, the maximum value of R is preferably larger than S / 2 (S / 2 <R Maximum value ≦ S). In this case, the number of ridge rows 41 included in the channel region 50 can be one or two.

  Thereafter, although details are omitted, a TFT as a semiconductor device can be formed by forming a source electrode, a drain electrode, a gate electrode, and the like by a conventionally known method.

  Here, two sets of thin film transistors A and B having different ridge row intervals were manufactured by the same method as in this embodiment, and the characteristics were compared. The channel regions of the thin film transistors A and B are both rectangular with a channel length L of 3 μm and a channel width of 20 μm, and crystallization was performed so that the ridge row direction and the channel length direction are orthogonal to each other. Note that the length S of the channel region in the direction orthogonal to the ridge row is equal to the channel length L.

  In the thin film transistor A, the slit width of the mask and the overlap width, which is the width of the region where the first region and the second region overlap, are set, and the width of the first region is 4 μm and the overlap width is 1.5 μm. As a result, the maximum value of the ridge row interval R was 2.5 μm.

  In the thin film transistor B, the slit width of the mask and the overlap width, which is the width of the region where the first region and the second region overlap, are set, and the width of the first region is 4 μm and the overlap width is 0.5 μm. As a result, the maximum value of the ridge row interval R was 3.5 μm.

  As a result, in order to satisfy the relationship of S / 2 <R maximum value ≦ S in the thin film transistor A, the thin film transistor A includes the thin film transistor A-1 having one ridge row included in the channel region, and the channel The thin film transistor A-2 is composed of two ridge rows included in the region.

  Further, in the thin film transistor B, the relationship of S / 2 <maximum value of R ≦ S is not satisfied, and the maximum value of R> S, so that the thin film transistor B has 0 ridge rows included in the channel region. The thin film transistor B-1 and the thin film transistor B-2 having one ridge row included in the channel region are included.

  As a result of setting the drain voltage to 0.1 V and measuring the threshold voltage of the thin film transistor, the threshold voltage of the thin film transistor A-1 is 1.1 V, the threshold voltage of the thin film transistor A-2 is 1.2 V, The threshold voltage of the thin film transistor B-1 was 0.1V, and the threshold voltage of the thin film transistor B-2 was 1.1V.

  From the above results, it was confirmed that the variation in threshold voltage can be reduced for the set of thin film transistors A compared to the set of thin film transistors B. The threshold voltage tends to increase with an increase in the number of ridge rows in the channel region. This tendency is greater when the number of ridge rows in the channel region is changed from 0 to 1. It can be considered that the above result is due to the fact that it becomes larger compared to the case where the number is changed from 2 to 2.

  As described above, in the present embodiment, the slit width of the mask and the width of the region where the first region and the second region overlap are set so as to satisfy the relationship of S / 2 <R maximum value ≦ S. Thus, a semiconductor thin film in which at least one ridge row is included in the channel region can be manufactured. Therefore, by using this semiconductor thin film, it is possible to obtain a semiconductor device with small variations in characteristics.

  In the present embodiment, the case where the ridge row direction and the channel length direction are orthogonal to each other has been described. However, the present invention is not limited to this, and the ridge row direction may have an angle inclined with respect to the channel length direction. . In this case, if the angle between the ridge row direction and the channel length direction is θ, only S = W cos θ + L sin θ, and S / 2 <R maximum value ≦ S as described in the present embodiment. By doing so, the same effects as those described in this embodiment can be obtained.

(Embodiment 2)
In the present embodiment, the first laser beam having an energy amount for melting the amorphous semiconductor film over the thickness of the amorphous semiconductor film on the surface of the object to be processed, and the amorphous semiconductor film Is characterized by irradiating with a second laser beam having an energy amount that does not melt the film over the thickness of the film.

  FIG. 6 shows a schematic configuration of a laser beam irradiation apparatus used in the present embodiment. Here, the laser beam irradiation apparatus is characterized in that it includes the first laser beam oscillator 20, the second laser beam oscillator 21, the optical element group 30, and the mirror 10.

  Here, the first laser oscillator 20 is not particularly limited as long as the amorphous semiconductor film 104 shown in FIG. 1 can be melted over its thickness. For example, an excimer laser beam, It is preferable to use one that oscillates laser light having a wavelength in the ultraviolet region such as a third harmonic wave of laser light oscillated from various solid-state lasers typified by YAG laser. This is because when the laser beam has a wavelength in the ultraviolet region, the amount of energy of the laser beam increases, and thus the amorphous semiconductor film 104 tends to be easily melted over its thickness. Among these, the first laser light oscillator 20 is particularly preferably one that oscillates excimer laser light having a wavelength of 308 nm capable of pulse emission. When the first laser beam oscillator 20 is an excimer laser beam, a laser beam oscillator capable of stably oscillating a laser beam having an energy amount necessary for the present invention is obtained relatively easily and at a low cost. This is because it is possible.

  Here, the first laser beam oscillated from the first laser beam oscillator 20 has an energy amount for melting the amorphous semiconductor film 104 in a solid state per one irradiation and irradiation area. For example, the first laser beam oscillated from the first laser beam oscillator 20 can be used to heat the amorphous semiconductor film 104 to a temperature equal to or higher than the melting point by irradiating the first laser beam alone once. It is recommended to have the amount of energy possible.

  The second laser oscillator 21 is not particularly limited as long as it oscillates the second laser beam that does not melt the amorphous semiconductor film 104 shown in FIG. 1 over its thickness. It is preferable to oscillate the second laser beam having a wavelength in the visible range to the infrared range. As the second laser beam oscillator 21, for example, a laser beam that oscillates a YAG laser beam having a wavelength of 532 nm, a YAG laser beam having a wavelength of 1064 nm, or a carbon dioxide laser beam having a wavelength of 10.6 μm can be used.

  Further, the amount of energy per irradiation and irradiation area of the second laser light oscillated from the second laser light oscillator 21 is less than the energy amount for melting the amorphous semiconductor film 104 in the solid state. In particular, it is recommended that the second laser beam has an energy amount that cannot heat the amorphous semiconductor film 104 to a temperature equal to or higher than the melting point by irradiating the second laser beam alone once. .

  Further, the optical element group 30 can optimize the shape and intensity distribution of the second laser light so that the second laser light emitted from the second laser light oscillator 21 can be irradiated most effectively. .

  The first laser light oscillator 20, the second laser light oscillator 21, and the sample stage 16 are controlled by the controller 17, and adjust the irradiation timing of the first laser light and the second laser light and the position of the sample stage 16. be able to.

  An object to be processed 101 shown in FIG. 1 is placed on the sample stage 16 of the laser beam irradiation apparatus. The second laser light oscillated from the second laser light oscillator 21 passes through the optical element group 30 and is then reflected by the mirror 10 so as to be inclined with respect to the amorphous semiconductor film on the surface of the object 101 to be processed. Irradiated.

  The first laser beam oscillated from the first laser beam oscillator 20 passes through the variable attenuator 12 and is reflected by the mirrors 7 and 8. Thereafter, the first laser light passes through the field lens 13 and the mask 14. Then, the first laser light that has passed through the mask 14 is imaged as an image of the mask 14 by the imaging lens 15, and is irradiated in a direction perpendicular to the amorphous semiconductor film on the surface of the object 101 to be processed. . The first laser light is applied to the first region C shown in FIG.

  FIG. 7 shows a preferred example of the relationship between the irradiation time of the first laser beam and the second laser beam and the output. Here, the pulse waveform V2 of the second laser beam indicates that the second laser beam is irradiated at the time t = 0 and the irradiation stops at the time t = t2. The pulse waveform V1 indicates that the first laser beam is irradiated at time t = t1. Note that t1 is a time between t0 and t2.

  The amorphous semiconductor film 104 in the first region C shown in FIG. 1 irradiated with the first laser light is melted over the entire thickness direction. At this time, the second region is irradiated with the second laser light on the same region as the first region C made of the amorphous semiconductor film 104 in the molten state or the region including the first region C, so that the molten state in the first region C is obtained. Since the time until the amorphous semiconductor film 104 is solidified and crystallized can be extended, a crystal having a larger crystal grain size than the case where the amorphous semiconductor film 104 is crystallized by irradiation with only the first laser beam. Can be obtained.

  In the present embodiment, as described above, the crystal grows in the lateral direction from the interface between the unmelted region and the melted region of the amorphous semiconductor film 104 and is crystallized (first crystallization step). .

  Next, the sample stage 16 is controlled by the controller 17, and the sample stage 16 is moved to the left of the arrow 22, for example. Here, the sample stage 16 is moved so that the first laser beam is irradiated to the second region where the first region C and a part of the region overlap (irradiation region moving step).

  In the same manner as in the first crystallization step, the first laser beam and the second laser beam are irradiated to the second region, so that the amorphous semiconductor film 104 in the second region is entirely formed in the thickness direction. When solidifying after melting, the crystal grows in the lateral direction from the interface between the unmelted region and the molten region of the amorphous semiconductor film 104 and is crystallized (second crystallization step).

  The desired region of the amorphous semiconductor film 104 can be crystallized by performing the first crystallization step, the irradiation region moving step, and the second crystallization step at least once in this order.

  In the present embodiment, in the first crystallization step and the second crystallization step, the first laser beam having an energy amount for melting the amorphous semiconductor film over the thickness of the film is irradiated, and non- Since the second laser beam having an energy amount that does not melt the crystalline semiconductor film over the thickness of the film is irradiated, the crystal growth distance per one irradiation of the first laser beam can be extended. Therefore, in the present embodiment, for example, when the ridge row interval is wide, the crystallization throughput can be increased, and the energy amount of the first laser light necessary to obtain the same crystal growth distance is reduced. Therefore, the maintenance cost of the first laser light oscillator can also be reduced. The description of the second embodiment other than the above is the same as that of the first embodiment.

  In addition, two sets of thin film transistors D and E having different ridge row intervals were manufactured by the same method as in the second embodiment, and the characteristics were compared. The channel regions of the thin film transistors D and E are both rectangular with a channel length L of 3 μm and a channel width of 20 μm, and crystallization was performed so that the ridge row direction and the channel length direction are orthogonal. Note that the length S of the channel region in the direction orthogonal to the ridge row is equal to the channel length L.

  In the thin film transistor D, the slit width of the mask and the overlap width that is the width of the region where the first region and the second region overlap are set, and the width of the first region is 4 μm and the overlap width is 1.5 μm. As a result, the maximum value of the ridge row interval R was 2.5 μm.

  In the thin film transistor E, the slit width of the mask and the overlap width, which is the width of the region where the first region and the second region overlap, are set, and the width of the first region is 4 μm and the overlap width is 0.5 μm. As a result, the maximum value of the ridge row interval R was 3.5 μm.

  As a result, in order to satisfy the relationship of S / 2 <R maximum value ≦ S in the thin film transistor D, the thin film transistor D includes the thin film transistor D-1 in which the number of ridge rows included in the channel region is one, and the channel The thin film transistor D-2 includes two ridge rows included in the region.

  The thin film transistor E does not satisfy the relationship of S / 2 <the maximum value of R ≦ S, and the maximum value of R> S, so that the thin film transistor E has zero ridge rows included in the channel region. A thin film transistor E-1 and a thin film transistor E-2 in which the number of ridge rows included in the channel region is one are included.

  As a result of measuring the threshold voltage of the thin film transistor with the drain voltage set to 0.1 V, the threshold voltage of the thin film transistor D-1 is 1.1 V, and the threshold voltage of the thin film transistor D-2 is 1.2 V. The threshold voltage of the thin film transistor E-1 was 0.1V, and the threshold voltage of the thin film transistor E-2 was 1.1V.

  From the above results, it was confirmed that the variation in threshold voltage can be reduced for the set of thin film transistors D compared to the set of thin film transistors E. The threshold voltage tends to increase with an increase in the number of ridge rows in the channel region. This tendency is greater when the number of ridge rows in the channel region is changed from 0 to 1. It can be considered that the above result is due to the fact that it becomes larger compared to the case where the number is changed from 2 to 2.

  Thus, also in the second embodiment, the slit width of the mask and the width of the region where the first region and the second region overlap are set so as to satisfy the relationship of S / 2 <R maximum value ≦ S. Thus, a semiconductor thin film in which at least one ridge row is included in the channel region can be manufactured. Therefore, by using this semiconductor thin film, it is possible to obtain a semiconductor device with small variations in characteristics.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  ADVANTAGE OF THE INVENTION According to this invention, the semiconductor device which has the channel area | region which consists of a semiconductor thin film obtained by this method and the semiconductor thin film which can reduce the dispersion | variation in the characteristic of a semiconductor device, and this method can be provided.

It is typical sectional drawing of a preferable example of the to-be-processed object used for this invention. It is a typical block diagram of the laser beam irradiation apparatus used by this Embodiment. (A) is a typical top view of a preferable example of the mask used for this invention, (b) is an example of the block in the mask shown to (a), (c) is the mask shown to (a). It is an example of a block inside. (A) is a typical expanded plan view of an example of the crystallization state of the semiconductor thin film after the 1st crystallization process of this invention, (b) is the crystal | crystallization of the semiconductor thin film after the 2nd crystallization process of this invention. It is a typical enlarged plan view of an example of a conversion state. It is the figure which showed typically an example of arrangement | positioning relationship of the channel area | region with respect to the crystallization area | region. It is a typical block diagram of the laser beam irradiation apparatus used by this Embodiment. It is a figure which shows a preferable example of the relationship between the irradiation time of a 1st laser beam and a 2nd laser beam, and an output. (A) is a typical top view of the to-be-processed object used for the conventional SLS method, (b) is a typical upper surface of the to-be-processed object shown to (a) before irradiating the area | region G with a laser beam. It is a figure, (c) is a typical top view of the to-be-processed object shown to (a) after irradiating the laser beam to the area | region G, (d) is (a) after completion | finish of the conventional SLS method. It is a typical top view of the to-be-processed object shown in FIG.

Explanation of symbols

  7, 8, 9, 10 Mirror, 11 Laser light oscillator, 12 Variable attenuator, 13 Field lens, 14 Mask, 15 Imaging lens, 16 Sample stage, 17 Controller, 20 First laser light oscillator, 21 Second laser light Oscillator, 22 arrows, 23, 24 blocks, 25, 26 slits, 27 width, 28 overlap width, 29 frame, 30 optical element group, 41 ridge row, 42 crystal, 49 crystallization region, 50 channel region, 52 channel Long direction, 53 ridge row direction, 55 ridge row interval, 101 object to be processed, 102 semiconductor substrate, 103 buffer film, 104 amorphous semiconductor film, 300 amorphous silicon film.

Claims (5)

A first crystal that crystallizes after irradiating the first region of the amorphous semiconductor film through the slit of the mask with laser light to melt the amorphous semiconductor film in the first region over its thickness. Conversion process,
An irradiation region moving step of moving the laser light irradiation region so that the second region where the first region overlaps with the first region is irradiated with the laser light;
By irradiating laser light into the second region of the amorphous semiconductor film through the slit of the mask, the amorphous semiconductor film in the second region is melted over its thickness and then crystallized. A method for manufacturing a semiconductor thin film comprising a channel region of a semiconductor device in at least a partial region of the amorphous semiconductor film, comprising a second crystallization step,
A region in which the slit width of the mask and the first region and the second region overlap so that one or more ridge rows in which the ridges formed on the surface of the semiconductor thin film are arranged in a straight line are included in the channel region A method for producing a semiconductor thin film, characterized in that the width of the semiconductor thin film is set.   The maximum difference between the width of the first region and the width of the region where the first region and the second region overlap is equal to or less than the length of the channel region in the direction orthogonal to the ridge row. 2. The method of manufacturing a semiconductor thin film according to claim 1, wherein a slit width of the mask and a width of a region where the first region and the second region overlap are set.   The slit width of the mask and the region where the first region and the second region overlap so that the maximum value of the interval between the ridge rows is equal to or less than the length of the channel region in the direction orthogonal to the ridge row. 3. The method for producing a semiconductor thin film according to claim 1, wherein a width is set.   In the first crystallization step and the second crystallization step, the laser beam is irradiated with a first laser beam having an energy amount that melts the amorphous semiconductor film over the thickness of the film, and 4. The method of manufacturing a semiconductor thin film according to claim 1, wherein the second laser beam having an energy amount that does not melt the amorphous semiconductor film over the thickness of the film is irradiated.   5. A semiconductor device having a channel region made of a semiconductor thin film manufactured by the method for manufacturing a semiconductor thin film according to claim 1, wherein the ridge row formed on a surface of the semiconductor thin film is the channel region. One or more semiconductor devices are included in the semiconductor device.

Images