JP5089077B2 - Method for manufacturing semiconductor device - Google Patents

Description

  The present invention relates to a method for manufacturing a semiconductor device by laser irradiation and a laser irradiation apparatus.

  Laser irradiation is often used in the manufacture of semiconductor devices. One reason for this is that the processing time can be significantly reduced as compared with an annealing method using radiant heating or conductive heating. Another reason is that no thermal damage is caused to a substrate that is easily thermally deformed, such as a glass substrate.

  On the other hand, the beam cross section of the laser beam has an energy intensity distribution (hereinafter also referred to as an intensity distribution), and the irradiated object could not be irradiated with a uniform intensity. For example, when the object to be irradiated is crystallized by laser annealing, a semiconductor film having uniform crystallinity cannot be obtained.

  On the other hand, there is a technique in which a laser beam is divided into a plurality of parts and the divided beams are combined to make the intensity distribution uniform.

  However, in this method, the intensity distribution in the beam cross section becomes uniform, but interference between the divided beams (hereinafter also referred to as divided beams) occurs. This is because the beams interfere with each other when the split beams are combined. Therefore, although the initial intensity distribution of the beam cross section disappears, a new intensity distribution is generated due to interference, so that the irradiation surface cannot be irradiated with the laser beam with uniform intensity.

Corresponding to this problem, Patent Document 1 provides an optical path difference equal to or greater than a temporal coherence distance (coherent length) between beams divided using a delay plate to suppress interference caused by beam division.
JP 2003-287703 A

  Since the optical path difference provided in Patent Document 1 is longer than the coherent length, it is necessary to provide a long optical path difference. For example, in the case of a laser beam oscillated from a solid-state laser, the coherent length is several meters to several kilometers, and it is impractical to produce an optical system that provides such an optical path difference.

  In addition, when an optical path difference is provided between the split beams, the time required to reach the irradiation surface differs for each split beam, and the difference in arrival time for each beam increases as the optical path difference increases. Therefore, if the optical path difference provided in the split beam is too long, when the semiconductor film is irradiated, the molten semiconductor film portion is solidified by the non-delayed beam irradiation, and then the delayed beam is irradiated. There was. In that case, since the next split beam is irradiated in the state where the crystal growth by the non-delayed split beam irradiation has been completed, a crystalline semiconductor film with continuous crystal growth cannot be obtained and large crystal grains cannot be formed. It was. Further, when the semiconductor film is crystallized by such an irradiation method, it becomes a thermally discontinuous phenomenon, which is completely different from the step of irradiating a laser beam having a uniform energy distribution.

  In view of the above problems, an object of the present invention is to provide a compact laser beam irradiation apparatus that prevents beam interference while making the intensity of a beam cross section uniform. It is another object of the present invention to provide a method for manufacturing a semiconductor device capable of continuous crystal growth in crystallization of an irradiated object.

  One of the features of the present invention is that the laser beam is divided into a plurality of parts, and in any two divided beams selected from the plurality of divided beams, an optical path difference is provided in the other divided beam with respect to one of the divided beams. This is a method for manufacturing a semiconductor device in which a split beam provided with an optical path difference is irradiated to a common portion of a semiconductor film for different periods to crystallize, and the laser beam has a pulse width of 100 fs to 1 ns and a frequency of 10 MHz or more. The optical path difference is to be greater than or equal to the length corresponding to the pulse width of the laser beam and less than the length corresponding to the coherent length or the pulse oscillation interval.

  One of the characteristics of the present invention is that the laser beam is divided into at least a first beam and a second beam, and the first beam and the second beam are different from each other with respect to the common portion of the semiconductor film. A method for manufacturing a semiconductor device that is crystallized by irradiation, wherein an optical path difference is provided between a first optical path of a first beam and a second optical path of a second beam, and the optical path difference is converted into a pulse of a laser beam. More than the length corresponding to the width and less than the length corresponding to the coherent length or the pulse oscillation interval. The laser beam is a pulse laser having a pulse width of 100 fs or more and 1 ns or less and a frequency of 10 MHz or more.

  One of the features of the present invention is that a laser beam is divided into a plurality of parts, and in any two split beams selected from the plurality of split beams, one split beam is provided with a time difference with respect to the other split beam, A method for manufacturing a semiconductor device in which a split beam provided with a time difference in arrival time to a semiconductor film is irradiated to a common portion of the semiconductor film for different periods to crystallize, and the laser beam has a pulse width of 100 fs to 1 ns, A pulse laser having a frequency of 10 MHz or more is used, and the time difference is set to be equal to or longer than the pulse width of the laser beam and less than the time corresponding to the coherent length or the pulse oscillation interval.

  One of the features of the present invention is that a laser beam having a pulse width of 100 fs or more and 1 ns or less and a frequency of 10 MHz or more is divided into a first laser beam and a second laser beam, and the second laser beam is converted into the first laser beam. The semiconductor film is crystallized by irradiating the first laser beam and the delayed second laser beam to the common part of the semiconductor film in different periods, respectively, and the first laser beam is irradiated. The second laser beam is applied to the semiconductor film while the semiconductor film is melted.

  One of the features of the present invention is that a laser beam oscillated from a laser oscillation source and having a pulse width of 100 fs to 1 ns and a frequency of 10 MHz or more is divided into a plurality of laser beams divided into a plurality of laser beams. In the two split beams, a laser irradiation apparatus having means for providing one of the split beams with an optical path difference with respect to the other split beam and means for irradiating the common part of the irradiated object with the split beam, The means for providing the difference is to provide an optical path difference between the laser beams divided into a plurality of lengths that is equal to or longer than the length corresponding to the pulse width and less than the length corresponding to the coherent length or the pulse oscillation interval.

  Note that in this specification, a megahertz laser beam refers to a laser beam having an extremely short pulse with a frequency of 10 MHz or more and a pulse width of 100 fs or more and 1 ns or less. Note that in the present invention, irradiating a common part with a laser beam does not necessarily mean that the laser beam irradiation areas completely coincide with each other, and the laser beam irradiation areas have a common part. It should be.

  As a laser beam generally used for crystallization of a semiconductor film, an excimer laser having a pulse width of 25 to 200 ns is used. Such an excimer laser has a long pulse width and is suitable for a process that requires time for rearrangement of atoms such as crystallization, and can form crystal grains with a diameter of about 100 to 300 nm. On the other hand, an ultrashort pulse laser with a pulse width of 1 ns or less is not suitable for crystallization because the pulse width is too short, so that crystal grains do not grow until it is sufficiently large.

  However, the present inventor has experimented that even if an ultrashort pulse laser is used, the frequency of the semiconductor film can be increased by increasing the heating time of the semiconductor film, continuously supplying energy to the irradiated object, and the semiconductor film has sufficient crystal grains I found that it was possible to make it larger. This is a megahertz laser beam having a pulse width of 100 fs to 1 ns and a frequency of 10 MHz or more. In the present invention, even if the pulse width is very short, such as 100 fs or more and 1 ns or less, by setting the frequency to 10 MHz or more, the semiconductor film is continuously provided with thermal energy necessary for crystallization, and the semiconductor film is crystallized. Is what you do.

  Furthermore, it has been found that optical interference is suppressed by providing an optical path difference equal to or greater than the pulse width, and optical interference is suppressed by an optical path difference having a very short length when suppressing optical interference in a megahertz laser beam.

  On the other hand, ultrashort pulse lasers are used for shape processing by utilizing the high spire output. However, the laser beam used for shape processing does not melt the irradiated object, but sublimates a part of the irradiated object instantaneously to form a hole or a groove. Therefore, it is different from the laser beam used for crystallization of the semiconductor film, and a laser beam having an energy density much higher than that of the laser beam used for crystallization is used.

  One feature of the present invention is that an optical path difference or a time difference is provided between the split beams. The provision of the optical path difference and the provision of the time difference have the same meaning in terms of shifting the timing at which the divided beams reach the irradiation surface. However, in the present specification, the unit of length is used as the optical path difference, and the unit of time is used as the time difference.

  The present invention divides a megahertz laser beam, adds an optical path difference equal to or longer than the pulse width of the megahertz laser beam and less than a coherent length to the split beam, thereby preventing interference between the split beams. Alternatively, the split beams are provided with a time difference equal to or greater than the pulse width of the megahertz laser beam to prevent interference between the split beams. This enables beam irradiation with a uniform intensity distribution with a slight optical path difference (time difference) without optical interference. Therefore, an optical system for providing an optical path difference can be reduced, and there is an advantage that the laser irradiation apparatus can be downsized.

  By setting the frequency to 10 MHz or higher, heat can be continuously applied to the semiconductor film, so that continuous crystal growth is possible. In addition, by setting the pulse width to 1 ns or less, the optical path difference between the divided beams can be set to several tens cm or less, so that an optical system that can be practically designed can be manufactured. Furthermore, by setting the pulse width to 100 fs or more, the structure of the laser device can be simplified, which is industrially advantageous.

  In a megahertz laser beam having a frequency of 10 MHz or more and a pulse width of 100 fs or more and 1 ns or less, the length corresponding to the pulse width is absolutely shorter when compared with the length corresponding to the pulse width. In addition, interference can be suppressed with a short optical path difference, which is effective.

  In addition, since the optical path difference between the divided lasers is short in the present invention, it is possible to continuously irradiate the semiconductor film with the divided beam while suppressing interference. Diameter crystals can be formed. That is, the megahertz laser beam is divided into the first beam and the second beam, and the second beam can be irradiated before the semiconductor film melted by the first beam is solidified.

  Further, according to the present invention, a semiconductor film can be crystallized by setting the frequency to 10 MHz or more even with an ultrashort pulse laser beam having a pulse width of 100 fs to 1 ns.

  Therefore, the present invention can irradiate the irradiated body with a laser beam with a uniform intensity distribution without optical interference. Further, when the irradiation object is a semiconductor film, the semiconductor film can be continuously grown between the split beams, and a crystal having a uniform and large grain size can be obtained.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the present invention can be implemented in many different modes, and those skilled in the art can easily understand that the modes and details can be variously changed without departing from the spirit and scope of the present invention. Is done. Therefore, the present invention should not be construed as being limited to the description of the embodiment modes.

  Further, Embodiments 1 and 2 shown below can be freely combined within a practicable range.

(Embodiment 1)
1A to 1H show beam cross-sectional shapes and intensity distributions when a megahertz laser beam is divided and irradiated to a common portion of an object to be irradiated in different periods. 1A shows a state before splitting a megahertz laser beam, FIGS. 1B and 1C show a state when splitting a megahertz laser beam, and FIG. 1D shows a case where the megahertz laser beam is different from the common part of the irradiated object. It is each beam cross section when irradiating in a period. 1E is the intensity distribution in the beam cross section before splitting, FIGS. 1F and 1G are the intensity distributions in the beam cross section during splitting, and FIG. 1H is synthesized by irradiation with the split beams 12 and 13. Represents the thermal energy distribution. Here, an example in which the megahertz laser beam is divided into two is shown, but any number of two or more may be divided. The greater the number of divisions, the more uniform the thermal energy intensity of the beam on the irradiated object when the divided beams are irradiated. Note that the beam cross-sectional shapes shown in FIGS. 1A to 1D, the beam intensity distributions shown in FIGS. 1E and 1F, and the thermal energy distribution shown in FIG. It is not limited to.

  The cross section of the beam 11 before the division is circular, and its intensity distribution has a Gaussian distribution (FIGS. 1A and 1E). The beam 11 is split to form split beams 12 and 13, the split beam cross section is semicircular, and the intensity distribution is a distribution obtained by dividing the Gaussian distribution by half (FIGS. 1B and 1C). (F), (G)). Then, the split beams 12 and 13 are irradiated to the common part of the irradiated object in different periods so that the sum of the mutual intensities becomes constant, and a uniform thermal energy distribution is given to the common part of the irradiated object (FIG. 1). (D), (H)). When the beams are divided and combined, the intensity distribution in the beam cross section is relaxed. However, if the split beams 12 and 13 are irradiated to the common part of the irradiated object at the same timing, a new interference occurs. Therefore, as shown in FIG. 2, a time difference is provided in the time when the split beams reach the irradiated object. Will be needed.

  FIG. 2A shows a megahertz laser beam having a pulse oscillation interval of 12.5 ns as an example of the megahertz laser beam. As described with reference to FIGS. 1A to 1H, the spatial profile of the beam 11 is divided into two to form split beams 12 and 13 (FIG. 2B). Then, as shown in FIG. 2C, in order to suppress interference when the irradiated beams 12 and 13 are irradiated on the irradiated object, the split beam 13 is made to have a certain optical path difference (time difference) from the split beam 12. Delay. The time difference at this time is not less than the pulse width of the beam 11 and less than the pulse oscillation interval. If the time difference is less than the pulse width, the pulses of the split beam 12 and the split beam 13 overlap each other, resulting in optical interference. Further, if the time difference is the pulse oscillation interval of the beam 11, the pulse periodically oscillated from the laser source and the pulse of the delayed split beam 13 overlap, causing interference. If the time difference is equal to or longer than the pulse oscillation interval of the beam 11, an optical path difference of 3 m or more must be provided, which is not practical. Therefore, the time difference needs to be less than the pulse oscillation interval. Therefore, by setting the time difference to the split beam, there is no optical interference, and the irradiation surface can be irradiated with a uniform laser beam. In addition, when the beam 11 is divided into three or more, any two divided beams arbitrarily selected from the divided beams may be in the relationship of the divided beams 12 and 13. That is, in two split beams arbitrarily selected from a plurality of split beams, the other split beam may be delayed with respect to one split beam by a time longer than the pulse width and less than the pulse oscillation interval.

Strictly speaking, the upper limit of the time difference to be provided is the time obtained by subtracting the pulse width from the pulse oscillation interval. That is, a <t _d <1 / N−a, where time difference t _d , pulse width a, and pulse oscillation interval 1 / N (N is the frequency of the laser beam).

  In this specification, the pulse oscillation interval is 1 / N, where N is the frequency of the megahertz laser beam.

  When the split beam 12 is irradiated on the irradiated object, the light energy is converted into thermal energy. The thermal energy distribution at this time reflects the intensity distribution of the split beam 12. However, if the optical path difference between the split beams is too long, the diffusion of the thermal energy given to the irradiated object by the split beam 12 starts before the split beam 13 is irradiated, and the thermal energy distribution changes. Therefore, even if the thermal energy distribution by irradiation of the split beam 13 is combined with the thermal energy distribution by irradiation of the split beam 12, a uniform thermal energy distribution cannot be obtained. That is, it becomes impossible to uniformly irradiate the irradiated object with the laser. Therefore, it is preferable to irradiate the split beam 13 until the thermal energy generated by the split beam 12 is diffused.

  In the present invention, since the optical path difference is set to be not less than the length corresponding to the pulse width of the megahertz laser beam and less than the length corresponding to the pulse oscillation interval, the optical path difference can be extremely shortened, and the thermal energy of the split beam 12 is diffused. The split beam 13 can be irradiated before performing. Therefore, uniform energy can be given to the irradiated surface more reliably.

  Further, when optical interference is suppressed by providing an optical path difference longer than the conventional coherent length, since the provided optical path difference is long, the energy of the split beam 13 is reduced before the split beam 13 reaches the irradiation surface. The laser could not be irradiated well. In addition, if the optical path difference is long, after the energy given to the irradiated object by the irradiation of the divided beam 12 is diffused, the divided beam 13 is irradiated, and continuous energy is given to the irradiated object between the divided beams. I couldn't. For example, when a semiconductor is crystallized by laser irradiation, the divided semiconductor 13 is irradiated after the melted semiconductor is solidified by irradiation of the divided beam 12, and the crystal cannot be continuously grown between the divided beams. Crystals could not be obtained.

  On the other hand, in the present invention, even if an optical path difference is provided in the split beam, the optical path difference is very short, more than the length corresponding to the pulse width of the MHz laser. Can be irradiated. Moreover, energy can be continuously given to the irradiated surface by the split beam, and high energy can be efficiently given to the irradiated surface while suppressing optical interference. Therefore, when the present invention is applied to crystallization of a semiconductor, continuous crystal growth can be realized and crystals with a large grain size can be obtained.

  In the present invention, the length corresponding to the pulse width is a value obtained by multiplying the pulse width by the speed of light, and the length corresponding to the pulse oscillation interval is a value obtained by multiplying the pulse oscillation interval by the speed of light.

FIG. 18 shows a photomicrograph of silicon. The silicon when amorphous silicon is crystallized using a YVO ₄ laser with a frequency of 80 MHz and a pulse width of 15 ps is shown. FIGS. 18A and 18B both divide and irradiate the YVO ₄ laser beam. FIG. 18A shows the crystallization under the condition that no time difference is provided between the divided beams. ) Is crystallized with a time difference of about 25 ps between the split beams.

  In the photograph shown in FIG. 18A, periodic interference fringes can be clearly confirmed. On the other hand, FIG. 18B shows a uniform crystal state without interference fringes.

  As described above, the present invention can eliminate the optical interference between the split beams by setting a very short optical path difference, and enables the crystallization of the semiconductor film by the megahertz laser beam that gives a uniform thermal energy distribution. is there.

(Embodiment 2)
In this embodiment mode, a mode in which a laser beam is irradiated onto an irradiation object on a glass substrate or a quartz substrate will be described. When an object to be irradiated is present on a glass substrate or a quartz substrate, a laser beam that could not be absorbed by the object to be irradiated may be reflected on the back surface of the substrate and irradiated again. In that case, it is necessary to consider the behavior of the reflected light from the back surface of the substrate when determining the optical path difference for preventing light interference. In this embodiment, setting of the optical path difference in consideration of the reflected light on the back surface of the substrate will be described with reference to FIGS. The time difference t _d is the oscillation interval less than 1 / N of the laser beam. The reason is that if the time difference is longer than the pulse oscillation interval of the laser beam, the current technology must provide an optical path difference of 3 m or more, which is not practical. Another reason is that if the time difference is the pulse oscillation interval of the laser beam, the pulse periodically oscillated from the laser source overlaps with the delayed split beam pulse, causing interference.

  3 and 4, one of the divided beams formed by dividing the megahertz laser beam into two is a first laser beam 401, and the other divided beam is a second laser beam 402. Alternatively, when the megahertz laser beam is divided into a plurality of beams, attention is paid to arbitrary two laser beams among the plurality of divided beams. Of the two laser beams, the first irradiated beam is the first laser beam 401, and the second irradiated beam is the second laser beam 402.

When the first laser beam 401 and the second laser beam 402 are irradiated, the condition for preventing interference between the first and second laser beams in the irradiated object 403 on the substrate 404 is the first laser beam. It can be classified into the following three types according to times t ₁ to t ₃ during which 401 propagates through the substrate 404. Note that in this embodiment mode, an amorphous silicon film is used as the irradiation object 403, and a glass substrate is used as the substrate 404.

  The first laser beam 401 passes through the irradiated body 403, is reflected by the back surface of the substrate 404, passes through the irradiated body 403 again, and then the delayed second laser beam 402 is irradiated onto the irradiated body 403. The case will be described with reference to FIG. FIG. 3A shows the moment when the irradiated body 403 is irradiated with the first laser beam 401, and the pulse tip 405 of the first laser beam 401 reaches the irradiated body 403. At this time, time t = 0.

FIG. 3B shows a state when t ₁ second has elapsed from the state of FIG. FIG. 3B is a diagram of the moment when the pulse tip 405 of the first laser beam 401 passes through the object to be irradiated 403 and is reflected by the back surface of the substrate and then irradiated again to the object to be irradiated 403. Here, the time t ₁ during which the pulse tip 405 of the first laser beam propagates through the substrate 404 is t ₁ = 2nD / c, where D is the thickness of the substrate, n is the refractive index of the substrate, and c is the speed of light. expressed.

Next, FIG. 3C shows a state when t ₂ seconds have elapsed from the state of FIG. FIG. 3C shows a state in which the pulse end 406 of the first laser beam 401 has passed through the irradiation object 403, and t ₂ corresponds to the pulse width of the first laser beam 401. In FIG. 3C, the irradiated object 403 holds the phase information of the first laser beam 401.

Next, FIG. 3 (D) from the state of FIG. 3 (C), the shows the state in which the phase relaxation time t ₃ of the object to be irradiated 403 has elapsed. In FIG. 3D, since the irradiation object 403 does not hold the phase information of the first laser beam 401, even if the second laser beam 402 is irradiated here, the first laser beam 401 and the first laser beam 401 There will be no interference between the two laser beams 402. At this time, after the first laser beam 401 is irradiated to the irradiation object 403 (the state in FIG. 3A), the second laser beam 402 is irradiated to the irradiation object 403 (FIG. 3D). )))) Is t ₁ + t ₂ + t ₃ . Therefore, in order to suppress optical interference between the first laser beam and the second laser beam, the time difference t _d between the first laser beam 401 and the second laser beam 402 is set to the time t ₁ described above. It may be larger than + t ₂ + t ₃ . That is, the following formula 1 is established.
t ₁ + t ₂ + t ₃ <t _d (Formula 1)

However, since the laser pulse is emitted from the laser oscillator at an oscillation interval of 1 / N seconds (N is the frequency of the laser beam), the second laser beam 402 is not only different from the first laser beam 401. The difference from the third laser beam emitted after 1 / N seconds from the first laser beam 401 must also be taken into consideration. For that purpose, after the phase information by the second laser beam 402 disappears in the irradiated body 403, the third laser beam emitted 1 / N second after the first laser beam 401 is applied to the irradiated body 403. It is better to reach. That is, the second laser beam 402 is reflected on the back surface of the substrate, is irradiated again on the irradiated object 403, passes through the irradiated object 403, and then irradiated with the third laser beam. In FIG. 3, the time from when the irradiated body 403 is irradiated with the first laser beam 401 to when the second laser beam 402 passes through the irradiated body 403 again after being reflected by the back surface of the substrate is t _d + t ₁ + t. represented by ₂ + _{t 3.} Here, if the oscillation interval 1 / N of the laser pulse is larger than the above-described time t _d + t ₁ + t ₂ + t ₃ , the second laser beam 402 and the third laser beam are temporally transmitted in the irradiated body 403. There is no overlap, and interference between laser beams can be prevented. Under the above conditions, the time difference t _d can be expressed by the following Equation 2.
t _d <1 / N−t ₁ −t ₂ −t ₃ (Formula 2)

Summarizing the above formulas 1 and 2, the condition of the time difference t _d that can prevent the interference between the first laser beam 401 and the second laser beam 402 is expressed by the following formula 3. Can do.
t ₁ + t ₂ + t ₃ <t _d <1 / N−t ₁ −t ₂ −t ₃ (Formula 3)

Note that the phase relaxation time refers to the time from when the laser beam is absorbed by the irradiated object until the phase information amount of the laser beam becomes 1 / e (e is a natural logarithm). Therefore, the phase relaxation time t ₃ is the time to phase information from the first laser beam 401 is absorbed by the object to be irradiated 403 is 1 / e. However, this time is very short compared to the pulse width and may be regarded as t ₃ ≈0. Therefore, Formula 3 can be expressed by Formula 4.
t ₁ + t ₂ <t _d <1 / N−t ₁ −t ₂ (Formula 4)

In other words, if a time difference t _d that satisfies the above formula 3 or 4 is provided between the first laser beam 401 and the second laser beam of the split beam, the first laser beam 401 reflected from the back surface of the substrate is also used. The object to be irradiated can be irradiated with the second laser beam 402 without causing interference. t ₁ is 2nD / c, _{t 2} because the pulse width, the 1 / N is a pulse oscillation interval, set to calculate the from _{t d} These values time difference (optical path difference). When the megahertz laser beam is divided into three or more, interference can be prevented by providing a time difference t _d satisfying Equation 3 or Equation 4 for any two of the divided beams.

As described above, when the reflected light on the back surface of the substrate is taken into consideration, a value obtained by adding the time t ₁ during which the first laser beam propagates through the substrate to the pulse width t ₂ may be used as the lower limit of the provided time difference (optical path difference). . In addition, as an upper limit of the provided time difference (optical path difference), a value obtained by subtracting not only the pulse width t ₂ but also the time t ₁ during which the first laser beam propagates through the substrate from the oscillation interval may be used.

The Formula 3 or Formula 4 obtained above, can determine the time difference t _d between the first laser beam and the second laser beam, an optical path difference required for an optical system can be calculated. For example, the pulse width t ₂ of the laser beam emitted from the laser oscillator 101 is 10 ps, the phase relaxation time t ₃ of the irradiated body 403 that is the irradiation surface is 0.1 ps, the substrate thickness is 1 mm, and the refractive index of the substrate is 1. In the case of 5, t ₁ + t ₂ + t ₃ = 20.1 ps. Thus if the time difference t _d between the laser beam from the formula 3 or more 20.1Ps, it is possible to prevent light interference on the irradiated surface. Also, is calculated by _{1 / N-t 1 -t 2} -t 3 is the upper limit of the time difference _{t d,} when the pulse oscillation interval 1 / N is 10 ns, the upper limit value of the delay time is 9.98ns, and the time difference _{t d} Is set to 20.1 ps to 9.98 ns, optical interference can be prevented.

Note that when the object to be irradiated is crystallized with a megahertz laser beam, the time difference t _d may be set within the solidification time of the object to be irradiated. If the delay time is set within the solidification time, the irradiated object can be continuously irradiated with the first laser beam and the second laser beam, and continuous crystal growth between the divided beams can be realized. Therefore, the time difference t _d is good above formula 3 or practitioner depending on the melting time of the object to be irradiated within the scope of Formula 4 is timely adjusted. For example, since the setting time of a silicon is 100 ns, is set as the time difference t _d is within 100 ns, can be continuously grown silicon.

  Next, another example of preventing interference between the first laser beam and the second laser beam will be described with reference to FIG. The same reference numerals and symbols as those in FIG. 3 are used for the same components as those in FIG.

  FIG. 4 shows the second laser beam after the first laser beam 401 passes through the irradiated body 403 and before the reflected light reflected by the back surface of the glass substrate is irradiated onto the irradiated body 403 again. It is a figure which irradiates 402 to be irradiated 403. FIG. FIG. 4A shows a state at the moment when the first laser beam 401 is irradiated on the irradiation object 403, and time t = 0. At this time, the pulse tip 405 of the first laser beam 401 has reached the irradiated object 403.

_Let t _{d be} the time difference between the first laser beam 401 and the second laser beam 402. After t _d seconds, the pulse tip 407 of the second laser beam 402 reaches the irradiated body 403. The state at this time is shown in FIG. At this time, in order to prevent the interference between the first laser beam 401 and the second laser beam 402, the pulse end 406 of the first laser beam 401 passes through the irradiated body 403, and further the first laser beam. After the phase information 401 disappears, it is necessary to irradiate the irradiated body 403 with the second laser beam 402. Therefore, the time difference t _d between the first laser beam 401 and the second laser beam 402 can be expressed by the following formula 5 using the pulse width t ₂ and the phase relaxation time t ₃ .
t ₂ + t ₃ <t _d (Formula 5)

FIG. 4C is a diagram after the pulse width t ₂ seconds of the second laser beam 402 has elapsed from FIG. 4B, and the pulse end 408 of the second laser beam 402 passes through the irradiation object 403. It shows how it was done.

Further in FIG. 4 (C) phase relaxation time t ₃ from elapses, so that the phase information according to the second laser beam 402 at the irradiation object 403 disappears. The state at this time is shown in FIG. After the phase information of the second laser beam disappears, if the irradiated body 403 is irradiated again with the first laser beam 401 reflected from the back surface of the substrate, the first laser beam 401 and the second laser beam 402 There is no interference between the two. Note that the time from the irradiation of the pulse tip 405 of the first laser beam 401 to the irradiated object 403 (FIG. 4A) to the irradiation of the irradiated object 403 again (FIG. 4D) is as follows. , Represented by t ₁ as shown in FIG. Therefore, in order to satisfy the condition in which no interference occurs in the state of FIG.
t _d + t ₂ + t ₃ <t ₁ (Formula 6)

Summarizing Formulas 5 and 6, a condition that can prevent interference between the first laser beam 401 and the second laser beam 402 can be expressed by Formula 7.
t ₂ + t ₃ <t _d <t ₁ −t ₂ −t ₃ (Formula 7)

Finally, the second laser beam 402 passes through the irradiated body 403 and is reflected on the back surface of the substrate, and the reflected light is irradiated again on the irradiated body 403. A case will be described in which the irradiated body 403 is irradiated with a third laser beam emitted after N seconds. After the first laser beam 401 is irradiated to the irradiation object 403 (FIG. 4A), the second laser beam 402 is irradiated to the irradiation object 403, and the second laser beam 402 is irradiated to the irradiation object 403. The time until the phase information due to disappears (FIG. 4D) is represented by t _d + t ₂ + t ₃ . Here, when the laser pulse oscillation interval 1 / N is larger than the above-described time t _d + t ₂ + t ₃ , the second laser beam 402 and the third laser beam are temporally transmitted to the irradiated object 403. There is no overlap, and interference between laser beams can be prevented. From the above relationship, the following Equation 8 regarding the time difference t _d is derived.
t _d <1 / N−t ₂ −t ₃ (Formula 8)

Further, after the first laser beam 401 is irradiated on the irradiated body 403, the third laser beam is irradiated on the irradiated body 403, and the phase information by the third laser beam disappears in the irradiated body 403. Is expressed by 1 / N + t ₂ + t ₃ . When the second laser beam 402 reflected from the back surface of the substrate reaches the irradiated body 403 after the time 1 / N + t ₂ + t ₃ has elapsed, the third laser beam and the second laser beam are irradiated with the irradiated body. In 403, there is no time overlap, and interference between laser beams can be prevented. Note that the time from when the irradiated body 403 is irradiated with the first laser beam 401 to when the irradiated body is irradiated with the second laser beam, reflected by the back surface of the substrate, and reaches the irradiated body 403 again. It is represented by t _d + t ₁ . Therefore, when the following Expression 9 is satisfied, the phase information of the third laser beam disappears from the irradiated body 403, and then the second laser beam reflected by the back surface of the substrate is irradiated again. Thus, interference can be prevented.
1 / N + t ₂ + t ₃ <t _d + t ₁ (Formula 9)

Summarizing Expressions 8 and 9, the condition that can prevent interference between the second laser beam 402 and the third laser beam irradiated 1 / N seconds after the first laser beam can be expressed by Expression 10. it can.
1 / N−t ₁ + t ₂ + t ₃ <t _d <1 / N−t ₂ −t ₃ (Formula 10)

As in FIG. 3, the phase relaxation time t ₃ is the time from when the first or second laser beam is absorbed by the irradiated body 403 until the phase information becomes 1 / e. Therefore, it can be approximated as t ₃ ≈0. That is, it can be expressed by the following formula 11.
1 / Nt ₁ + t ₂ <t _d <1 / Nt ₂ (Formula 11)

When the megahertz laser beam is divided into three or more, interference can be prevented by providing a time difference t _d satisfying Equation 10 or Equation 11 for any two divided beams of the divided beams.

Note that when the object to be irradiated is crystallized with a megahertz laser beam, the time difference t _d may be set within the solidification time of the object to be irradiated. If the delay time is set within the solidification time, the irradiated object can be continuously irradiated with the first laser beam and the second laser beam, and continuous crystal growth between the divided beams can be realized. Therefore, the time difference t _d is good practitioner depending on the melting time of the object to be irradiated in the range of the formula 10 or formula 11 is timely adjusted.

Thus, to calculate the time difference t _d from equation may be designed to reflect the optical system providing an optical path difference. By doing so, it is possible to design an optical path difference in consideration of the behavior of the reflected light reflected from the back surface of the substrate. By providing the optical path difference of the present invention, the amorphous silicon film can be irradiated without optical interference, a uniform thermal energy distribution can be given to the amorphous silicon film, and continuous crystal growth can be performed. That is, in order to uniformly irradiate the semiconductor film without optical interference, it is sufficient that two laser beams do not exist in the semiconductor film, and as a result, the phase information of the two laser beams does not exist in the semiconductor film. .

Although this embodiment describes the time difference t _d in units of time, to convert the optical path difference may be multiplied by the speed of light delay time.

  The following Examples 1 to 8 can be freely combined within a practicable range. In addition, the following Examples 1 to 8 can be freely combined with Embodiment 1 or 2 as long as practicable.

  Example 1 shows a laser irradiation apparatus of the present invention. 5A is a laser irradiation apparatus, FIG. 5B is a beam cross-sectional shape before splitting, FIGS. 5C and 5D are beam cross-sectional shapes at splitting, and FIG. 5E is an irradiation surface. This is a beam cross-sectional shape when the divided beams are irradiated to the common parts 110 in different periods. FIG. 5 (F) shows the intensity distribution of the beam cross section before splitting, FIGS. 5 (G) and (H) show the intensity distribution of the beam cross section at the time of splitting, and FIG. 5 (I) shows the split beam irradiated to the common part of the irradiation surface. Shows the thermal energy distribution synthesized.

  In FIG. 5A, a laser beam 107 oscillated from a laser oscillator 101 is divided into a first laser beam 108 and a second laser beam 109 by a dividing means 102. The laser beams 107, 108, and 109 are megahertz laser beams. A dividing mirror is used as the dividing unit 102. The laser beam 107 has a circular beam cross-sectional shape and has a Gaussian intensity (FIGS. 5B and 5F). The beam cross-sectional shapes of the first laser beam 108 and the second laser beam 109 are divided by a dividing mirror into semicircles (FIGS. 5C and 5D). The intensity is a distribution obtained by dividing the Gaussian distribution at the center, and is a line contrasting distribution (FIGS. 5G and 5H).

  The first laser beam 108 is reflected by the mirror 104, the second laser beam 109 is reflected by the mirror 103 and deflected, and the first laser beam 108 and the second laser beam 109 are respectively applied to the irradiation surface 110. Irradiated for different periods. When each of the first laser beam and the second laser beam is irradiated, the intensity distributions of the first laser beam 108 and the second laser beam 109 are relaxed as shown in FIG. Irradiate in this way. By the irradiation with the first laser beam 108 and the second laser beam 109, the thermal energy synthesized on the irradiation surface becomes uniform as shown in FIG. The cylindrical lenses 105 and 106 form a split beam.

  Comparing the optical path length d1 from the dividing means 102 to the mirror 104 and the optical path length d2 from the dividing means 102 to the mirror 103, d1 <d2, and this optical path length difference d (d = d2-d1) is the first. This is the optical path difference between the laser beam 108 and the second laser beam 109. The optical path difference d is set to be equal to or longer than the pulse width of the laser beam 107 and less than the length corresponding to the pulse oscillation interval. In this embodiment, since a megahertz laser beam is used, the optical path difference d may be several mm to several tens mm, and it is not necessary to provide a long optical path difference. For example, if the pulse width is 10 ps, the optical path difference d is 3 mm, which is a very short optical path difference.

Also, when considering the reflection light from the rear surface of the substrate as in the embodiment 2, it may be set to the optical path difference d multiplied by the speed of light t _d calculated from the formula.

  Next, an example of a laser irradiation apparatus combining a 1 / 2λ wavelength plate and a beam splitter will be described with reference to FIG. The same components as those in FIG.

  The laser beam 107 output from the laser oscillator 101 that oscillates the megahertz laser beam is magnified by about 5 times in both the long side direction and the short side direction by the beam expander 201 including the spherical lenses 201a and 201b. The spherical lens 201a has a radius of 50 mm, a thickness of 7 mm, and a curvature radius of the first surface of −220 mm. The spherical lens 201b has a radius of 50 mm, a thickness of 7 mm, and a curvature radius of the second surface of −400 mm. The beam expander 201 is effective when the shape of the laser beam emitted from the laser oscillator 101 is small, but may not be used when the shape of the laser beam is large, and may be appropriately selected by the practitioner. In this specification, with respect to the radius of curvature of the lens, the case where the center of curvature is on the light exit side with respect to the lens surface is shown as positive, and the case where the center of curvature is on the incident side with respect to the lens surface is shown as negative. In the lens, a surface on which light is incident is a first surface, and a surface on which light is emitted is a second surface.

  The laser beam 107 expanded by the beam expander 201 is reflected by the dividing unit 102. As shown in FIGS. 5C, 5D, and 5H, the dividing unit 102 has a first laser beam 107 having a line-symmetric intensity distribution and a semicircular cross-sectional shape. The laser beam 108 and the second laser beam 109 are divided into two.

  The first laser beam 108 is reflected by the mirror 104 and passes through the ½λ wavelength plate 203. By passing through the ½λ wavelength plate 203, the polarization direction of the first laser beam 108 becomes s-polarized light. Furthermore, the irradiation position on the irradiation surface 110 is adjusted by the first laser beam 108 entering the polarization beam splitter 205.

  On the other hand, the second laser beam 109 divided by the dividing means 102 is reflected by the mirror 103 and passes through the ½λ wavelength plate 202. By passing through the 1 / 2λ wavelength plate 202, the polarization direction of the second laser beam becomes p-polarized light. Further, the irradiation position on the irradiation surface 110 of the second laser beam 109 reflected by the mirror 204 is adjusted by the polarization beam splitter 205.

  The optical path of the second laser beam 109 is longer than the optical path of the first laser beam 108 by the distance d between the mirror 204 and the polarization beam splitter 205, and the optical path difference d is provided in the second laser beam 109. ing. Therefore, the second laser beam 109 is irradiated onto the irradiation surface 110 with a delay of the optical path difference d from the first laser beam 108. The optical path difference d is set to be not less than the length corresponding to the pulse width of the laser beam 107 and less than the length corresponding to the pulse oscillation interval. In addition, when the reflected light on the back surface of the substrate is taken into consideration, the optical path difference calculated from the formula obtained in the second embodiment is set.

  By adjusting the irradiation region of the first laser beam and the second laser beam by the polarization beam splitter 205 so that the irradiation surface 110 overlaps, a uniform thermal energy distribution can be synthesized on the irradiation surface. The cylindrical lens 105 has a focal length of 150 mm and a thickness of 5 mm. As a result, the laser beam can be shaped so that the length in the major axis direction of the beam cross section on the irradiation surface 110 is 500 μm. The focal length of the cylindrical lens 106 is 20 mm. As a result, the laser beam is shaped such that the length in the minor axis direction of the beam cross section at the irradiation surface 110 is 10 μm. As described above, a linear beam can be formed on the irradiated surface.

  In this embodiment, it is possible to irradiate the irradiated surface of the split beam while suppressing the interference of the laser beam, so that the thermal energy distribution on the irradiated surface can be made uniform. Therefore, uniform crystal growth can be realized. Further, continuous crystal growth can be performed by irradiation with a split beam, and a crystal having a large grain size can be obtained.

  For example, it is assumed that a YAG laser having a pulse width of 10 ps and a frequency of 10 MHz is divided, and the divided beams are irradiated to the common portions of the irradiated surface in different periods to crystallize silicon. The solidification time of silicon is generally 100 ns. When an optical path difference corresponding to the coherent length is provided in this YAG laser to suppress optical interference, it is usually necessary to provide a time difference of 300 ns or more between the divided beams. Therefore, since the time difference between the split beams is longer than the solidification time of silicon, silicon cannot be continuously crystallized between the split beams. On the other hand, when an optical path difference equal to or greater than the pulse width is provided in this YAG laser, a time difference of at least 10 ps or more occurs between the split beams. Since 10 ps is much shorter than the solidification time of silicon, silicon can be continuously crystallized between split beams.

  In Example 2, a laser irradiation apparatus having a configuration different from that of Example 1 will be described with reference to FIG. FIG. 7A is a plan view of the laser irradiation apparatus, and FIG. 7B is a side view. The same components as those in FIG.

  The laser beam 107 output from the laser oscillator 101 that oscillates the megahertz laser beam is expanded in the long side direction and the short side direction of the beam cross section by the beam expander 201 configured by the spherical lenses 201a and 201b.

  The laser beam 107 that has passed through the beam expander 201 is divided by a cylindrical lens array 701. Here, an example of dividing into two parts is shown, but a cylindrical lens array divided into two or more parts may be used. The cylindrical lenses constituting the cylindrical lens array 701 have a thickness of 5 mm and a curvature radius of the first surface of 40 mm.

Here, the divided laser beams are referred to as a first laser beam 108 and a second laser beam 109, respectively. Then, only the second laser beam 109 is allowed to pass through the delay means 702 disposed behind the cylindrical lens array 701. The delay means 702 only needs to have a function of delaying the second laser beam 109 with respect to the first laser beam 108. Here, a quartz plate is used as the delay means 702. When the thickness of the quartz plate to be used is d ₁ , the speed of light is c, and the refractive index of the quartz plate is n, the optical path difference between the divided laser beams is represented by d ₁ (n−1) / c. The optical path difference can be selected by selecting the thickness.

  In addition to the quartz plate, the delay means 702 may be a medium that does not absorb megahertz laser beams. For example, there are crystals made of glass (BK7), water, or fluoride. Quartz also has no absorbability for megahertz laser beams.

  The first laser beam and the second laser beam are condensed by passing through the cylindrical lens 703. The cylindrical lens 703 has a lens first surface with a radius of curvature of 70 mm and a thickness of 5 mm. As a result, the irradiation regions of the first laser beam and the second laser beam overlap each other behind the cylindrical lens 703, and a uniform thermal energy distribution can be synthesized on the irradiation surface.

  Then, the first laser beam 108 and the second laser beam 109 are projected onto the irradiation surface 110 by the cylindrical lenses 105 and 106. With the cylindrical lens 105, the length of the beam cross section in the long axis direction is formed to 500 μm. Further, the length in the minor axis direction of the beam cross section on the irradiation surface is formed by the cylindrical lens 106. The cylindrical lens 106 has a focal length of 20 mm, and is shaped so that the length in the minor axis direction of the beam cross section on the irradiation surface is 10 μm. A linear beam spot can be formed on the irradiated surface by the optical system.

  In this specification, “linear” does not mean “line” in a strict sense, but means a rectangle (or oval) having a large aspect ratio. For example, those having an aspect ratio of 2 or more (preferably 10 to 10000) are called linear.

  In the present embodiment, the method of providing the optical path difference using the refractive index by the delay means has been described. In this embodiment, the optical path difference may be set in combination with the method of providing the optical path difference without using the refractive index as in the first embodiment.

  In this embodiment, by using an ultrashort pulse megahertz laser beam, it is possible to divide the megahertz laser beam while irradiating the irradiated object while reducing the influence of the interference of the laser beam. The strength can be made uniform.

  In this embodiment, an example of a laser irradiation apparatus using a beam splitter for splitting an ultrashort pulse megahertz laser beam will be described with reference to FIG. The same components as those in FIG. 5A are denoted by the same reference numerals and description thereof is omitted. 8A is a laser irradiation apparatus, FIG. 8B is a beam cross-sectional shape before division of the megahertz laser beam, FIGS. 8C and 8D are beam cross-sectional shapes at the time of division, and FIG. It is the cross-sectional shape of the beam irradiated to the common part of the to-be-irradiated surface at different periods. 8F shows the intensity distribution of the beam cross section before splitting, FIGS. 8G and 8H show the intensity distribution of the beam cross section at the time of splitting, and FIG. 8I shows the synthesis on the irradiated surface by irradiation of the split beam. The thermal energy distribution is shown.

  The laser beam 107 output from the laser oscillator 101 is converted into a second harmonic by the nonlinear optical element 801. The laser beam that has passed through the nonlinear optical element 801 is expanded in both the long-side direction and the short-side direction by a beam expander 201 composed of spherical lenses 201a and 201b. Note that the beam expander may not be used depending on the size of the laser beam.

  The laser beam 107 expanded by the beam expander passes through either the first 1 / 2λ wavelength plate 806a or the second 1 / 2λ wavelength plate 806b. Half of the laser beam 107 is polarized in the first polarization direction by passing through the first 1 / 2λ wavelength plate 806a, and half of the laser beam 107 is passed through the second 1 / 2λ wavelength plate 806b. The light is polarized in a second polarization direction that is 90 ° different from the first polarization direction. The first 1 / 2λ wavelength plate 806a and the second 1 / 2λ wavelength plate 806b are obtained by changing the cross-sectional shapes of the first laser beam 108 and the second laser beam 109 with reference to FIG. A semicircle is provided as shown in (D), and the intensity distribution is provided in order to obtain a distribution in which the Gaussian distribution is line-symmetrically divided as shown in FIGS.

  Next, the laser beam 107 is separated into the first laser beam 108 and the second laser beam 109 by the polarization beam splitter 802. The first laser beam 108 passes through the polarization beam splitter 802 and is reflected by the mirror 104. Further, the light enters the polarizing beam splitter 804 and the irradiation position on the irradiation surface 110 is adjusted. The irradiation region of the first laser beam is adjusted so that the irradiation region overlaps with the irradiation region of the second laser beam, and uniform thermal energy is synthesized between the irradiation region and the second laser beam.

  On the other hand, the second laser beam is reflected by the polarization beam splitter 802 and reflected by the mirror 103. Further, the incident position on the irradiation surface 110 is adjusted by entering the polarization beam splitter 804. The irradiation region of the second laser beam is adjusted so that the irradiation region overlaps with the irradiation region of the first laser beam, and uniform thermal energy is synthesized between the irradiation region and the first laser beam.

  Here, a delay means 803 is disposed between the mirror 103 and the polarization beam splitter 804. Here, a quartz plate is used as the delay means 803. As a result, an optical path difference is generated between the first laser beam 108 and the second laser beam 109, and interference between the first laser beam and the second laser beam is prevented on the irradiation surface. The The optical path difference can be set in any way by selecting the thickness of the quartz plate. Further, the thermal energy distribution on the irradiated surface can be made uniform by irradiating the common portion of the irradiated surface with the first laser beam and the second laser beam in different periods.

  The first laser beam 108 and the second laser beam 109 are incident on the cylindrical lens 105 so that the length in the major axis direction of the beam cross section is formed. By entering the cylindrical lens 106, the minor axis direction of the beam cross section is obtained. The length of is molded. Note that the first laser beam 108 and the second laser beam 109 immediately after being split by the polarization beam splitter 802 have an intensity distribution of the laser beam 107 as shown in FIGS. 8G and 8H. The intensity distribution is divided into line symmetry, and as the beam propagates, the cut-out occurs and the intensity distribution of the laser beam 107 is destroyed. However, this problem can be solved by setting the position of the polarization beam splitter 802 and the position of the irradiation surface 110 to be conjugate to the cylindrical lens 105 acting in the long axis direction of the beam cross section.

  In the present embodiment, the method of providing the optical path difference using the refractive index by the delay means has been described. In addition, the optical path difference may be set in combination with the method of providing the optical path difference without using the refractive index as in the first embodiment.

  In this embodiment, the megahertz laser beam is divided while reducing the influence of the interference of the megahertz laser beam by using the ultra-short pulse megahertz laser beam, and the divided beams have different periods with respect to the common part of the irradiated surface. Therefore, the thermal energy distribution of the megahertz laser beam on the irradiated surface can be made uniform.

  In this embodiment, a laser beam irradiation method different from that in Embodiment 3 will be described with reference to FIG. 9A shows a laser irradiation apparatus, FIGS. 9B and 9C show first and second laser beam cross-sectional shapes, and FIG. 9D shows different periods with respect to a common portion of the irradiated surface. 9 (E) to 9 (F) show intensity distributions corresponding to FIGS. 9 (B) to 9 (C), and FIG. 9 (G) shows the irradiation by split beam irradiation. The thermal energy distribution synthesized on the surface is shown.

  The laser beam 107 output from the laser oscillator 101 is converted into a second harmonic by the nonlinear optical element 801. The laser beam that has passed through the nonlinear optical element 801 is expanded in both the long-side direction and the short-side direction by a beam expander 201 composed of spherical lenses 201a and 201b.

  The laser beam 107 expanded by the beam expander is separated into a first laser beam 108 and a second laser beam 109 by a polarization beam splitter 802. The divided first laser beam 108 and second laser beam 109 have a circular cross-sectional shape and an intensity distribution having a Gaussian distribution (FIGS. 9B, 9C, 9E, 9F). )).

  The first laser beam 108 passes through the polarization beam splitter 802 and is reflected by the mirror 104. Further, the light enters the polarizing beam splitter 804.

  On the other hand, the second laser beam is reflected by the polarization beam splitter 802 and reflected by the mirror 103. Further, the light enters the polarizing beam splitter 804.

  Specifically, the optical path length of the second laser beam 109 from the polarizing beam splitter 802 to the polarizing beam splitter 804 is larger than the optical path length of the first laser beam 108 from the polarizing beam splitter 802 to the polarizing beam splitter 804. The length is increased by a length not more than 0.5 times the beam diameter. In this way, when the first laser beam 108 and the second laser beam 109 are irradiated to the common portion of the irradiated surface in different periods, a uniform thermal energy distribution can be synthesized. When split by the polarization beam splitter 802, the intensity distributions of the first laser beam 108 and the second laser beam 109 are Gaussian as shown in FIGS. Therefore, when the irradiation positions of the first and second laser beams on the irradiated surface are adjusted by the polarizing beam splitter 804, the first and second laser beams maintain the intensity of the Gaussian distribution when they are completely overlapped. A uniform energy distribution cannot be obtained. However, by irradiating the irradiation surfaces of the first laser beam 108 and the second laser beam 109 with a slight shift as in this embodiment, the thermal energy distribution on the irradiation surface can be made uniform.

  In addition, the optical path length of the second laser beam 109 from the polarizing beam splitter 802 to the polarizing beam splitter 804 is made larger than the optical path length of the first laser beam 108 from the polarizing beam splitter 802 to the polarizing beam splitter 804. The length may be increased by a length of 0.5 to 0.7 times. When an optical path difference of 0.5 times or more of the beam diameter is provided, the thermal energy distribution synthesized on the irradiated surface is not uniform as shown in FIG. However, the beam cross-sectional area can be increased, and the irradiated surface can be irradiated more efficiently. However, if an optical path difference of 0.7 times or more of the beam diameter is provided, an energy density is weak in the beam cross section and a region that is not partially crystallized appears. It should be set to 7 times longer.

  In this embodiment, a method for manufacturing a semiconductor device using the present invention will be described.

  A method for manufacturing a semiconductor device will be described with reference to FIGS. First, as shown in FIG. 10A, a base film 71 and an amorphous semiconductor film 72 are sequentially stacked over an insulating substrate 70. Then, the amorphous semiconductor film 72 is irradiated with a megahertz laser beam according to the present invention (FIG. 10B). By uniformly irradiating the amorphous semiconductor film with a megahertz laser beam, the amorphous semiconductor film can be uniformly crystallized, and a crystalline semiconductor film 73 having a large grain size can be obtained. Therefore, using the crystalline semiconductor film 73, a thin film transistor (TFT) having almost no crystal grain boundary in the channel carrier moving direction can be formed.

  After that, as illustrated in FIG. 10C, the crystalline semiconductor film 73 is etched to form island-shaped semiconductor films 74 to 77. Next, a gate insulating film 78 is formed so as to cover the island-shaped semiconductor films 74 to 77. For the gate insulating film 78, for example, silicon oxide, silicon nitride, silicon nitride oxide, or the like can be used. As a film formation method at that time, a plasma CVD method, a sputtering method, or the like can be used. For example, an insulating film containing silicon with a thickness of 30 nm to 200 nm may be formed by a sputtering method.

  Next, a conductive film is formed over the gate insulating film 78 and etched to form a gate electrode. After that, using the gate electrode as a mask, an impurity imparting n-type or p-type conductivity is selectively added to the island-shaped semiconductor films 74 to 77 to form a source region, a drain region, an LDD region, and the like. . Through the above steps, the N-type transistors 710 and 712 and the P-type transistors 711 and 713 can be formed over the same substrate (FIG. 10D). Subsequently, an insulating film 714 is formed as a protective film thereof. As this insulating film 714, a plasma CVD method or a sputtering method may be used to form an insulating film containing silicon with a thickness of 100 nm to 200 nm as a single layer or a stacked structure. For example, a silicon oxynitride film with a thickness of 100 nm may be formed by a plasma CVD method.

  Next, an organic insulating film 715 is formed over the insulating film 714. As the organic insulating film 715, an organic insulating film such as polyimide, polyamide, BCB, or acrylic applied by an SOG (Spin On Glass) method is used. In the case where it is desired to reduce unevenness due to the thin film transistor formed over the insulating substrate 70 and planarize, it is preferable to use a material having excellent flatness as the organic insulating film 715. Further, the insulating film 714 and the organic insulating film 715 are patterned by photolithography to form contact holes that reach the impurity regions.

  Next, a conductive film is formed, and the conductive film is patterned to form wirings 716 to 723. After that, when an insulating film 724 is formed as a protective film, a semiconductor device as illustrated in FIG. 10D is completed. Note that a method for manufacturing a semiconductor device using the laser annealing method of the present invention is not limited to the structure of the thin film transistor described above. This embodiment is characterized in that a crystalline semiconductor film obtained by splitting a megahertz laser beam and irradiating it with different periods is used as an active layer of a TFT. As a result, a crystal having a large grain size can be obtained, and the electron mobility of the semiconductor device can be increased.

  In addition, a step of adding a metal element to the amorphous semiconductor film and crystallizing may be provided before crystallization with laser light. As metal elements, nickel (Ni), germanium (Ge), iron (Fe), palladium (Pd), tin (Sn), lead (Pb), cobalt (Co), platinum (Pt), copper (Cu), An element such as gold (Au) can be used. When laser light is further irradiated after crystallization with the addition of a metal element, crystals formed during crystallization with the metal element are recrystallized as crystal nuclei. Therefore, the crystallinity of the semiconductor film can be further improved as compared with the crystallization process only by laser light irradiation, and the surface roughness of the semiconductor film after crystallization by laser light irradiation can be suppressed. Accordingly, variation in characteristics of semiconductor elements formed later, typically TFTs, can be further suppressed, and off-current can be suppressed.

  Note that after adding a metal element, crystallization may be performed by heat treatment, and crystallinity may be further improved by irradiation with laser light, or crystallization may be performed by laser irradiation without adding heat treatment after the addition of the metal element. .

  In this embodiment, an example in which the laser irradiation method of the present invention is used for crystallization of a semiconductor film is shown; however, it may be used for activating an impurity element doped in a semiconductor film. In the case of manufacturing a semiconductor device using a functional circuit such as a driver or a CPU in this embodiment, it is preferable to form a thin film transistor having an LDD region or a thin film transistor having a structure in which the LDD region overlaps with a gate electrode. is there. In order to increase the speed of the semiconductor device, it is preferable to miniaturize the thin film transistor. Since the N-type transistors 710 and 712 and the P-type transistors 711 and 713 of the semiconductor device completed according to this embodiment have an LDD structure having an LDD region, they are preferably used for a driver or a CPU.

  In this embodiment, a method for manufacturing an EL display device using the semiconductor device manufactured in Embodiment 5 will be described with reference to FIGS.

  A pixel electrode 504 is formed so as to be in contact with the wiring 722. The pixel electrode 504 is formed by etching a transparent conductive film. As the transparent conductive film, a compound of indium oxide and tin oxide, a compound of indium oxide and zinc oxide, zinc oxide, tin oxide, or indium oxide can be used.

  After the pixel electrode is formed, a partition wall 505 made of a resin material is formed. The partition wall 505 is formed so as to expose a part of the pixel electrode 504 by etching an acrylic film or a polyimide film having a thickness of 1 to 2 μm. Note that a black film serving as a shielding film (not shown) may be appropriately formed below the partition 505.

  Next, an EL layer 506 is formed. When the light-emitting material of the EL layer 506 is an organic compound, an organic EL element is formed. When the light-emitting material is an inorganic compound, an inorganic EL element is formed.

  Inorganic EL elements are classified into dispersion-type inorganic EL elements and thin-film inorganic EL elements depending on the element structure. A dispersion-type inorganic EL element has a light emitting layer in which particles of a light emitting material are dispersed in a binder. The thin film type inorganic EL element has a light emitting layer made of a thin film of a fluorescent material. In both of the light emission mechanisms, light emission can be obtained by collision excitation of a base material or a light emission center by electrons accelerated by a high electric field. In the case of forming an inorganic EL element, an insulating layer in which a light-emitting material is dispersed is provided as an EL layer between the pixel electrode 504 and the electrode 507, or a light-emitting layer sandwiched between insulating layers is preferably provided. For example, zinc sulfide (ZnS) or strontium sulfide (SrS) can be used as the light emitting material. The EL layer of the inorganic EL element can be formed by screen printing or vapor deposition.

  Below, the example in the case of using an organic EL element is demonstrated.

  The EL layer 506 and the electrode (MgAg electrode) 507 are continuously formed using the vacuum deposition method without being released to the atmosphere. Note that the EL layer 506 may have a thickness of 100 nm to 1 μm, and the electrode 507 may have a thickness of 180 to 300 nm (typically 200 to 250 nm). In addition, the EL layer may be formed by inkjet, screen printing, or the like.

  In this step, an EL layer and a cathode are sequentially formed for a pixel corresponding to red, a pixel corresponding to green, and a pixel corresponding to blue. However, since the EL layer has poor resistance to the solution, it has to be formed individually for each color without using a photolithography technique. Therefore, it is preferable to hide other than the desired pixels using a metal mask, and selectively form the EL layer and the cathode only at necessary portions. At least one color development of each color is performed with a triplet compound. Since the triplet compound is brighter than the singlet compound, it is preferable to form pixels corresponding to red that appear dark with the triplet compound and other pixels with the singlet compound.

  That is, first, a mask that hides all pixels other than those corresponding to red is set, and the EL layer and electrodes that emit red light are selectively formed using the mask. Next, a mask that hides all but the pixels corresponding to green is set, and a green light emitting EL layer and electrodes are selectively formed using the mask. Next, similarly, a mask for hiding all but the pixels corresponding to blue is set, and the EL layer and the electrode emitting blue light are selectively formed using the mask. Note that although all the different masks are described here, the same mask may be used. Further, it is preferable to perform processing without breaking the vacuum until the EL layer and the electrode are formed on all the pixels.

  Note that a known material can be used for the EL layer 506. As the known material, it is preferable to use an organic material in consideration of the driving voltage. For example, a four-layer structure including a hole injection layer, a hole transport layer, a light emitting layer, and an electron injection layer may be used as the EL layer. A film in which molybdenum oxide and α-NPD are mixed may be used as the EL layer. A hybrid layer in which an organic material and an inorganic material are combined may be used as the EL layer. When an organic material is used for the EL layer, each of a low molecular material, a medium molecular material, and a high molecular material can be used. In this embodiment, an example in which an MgAg electrode is used as a cathode of an EL element is shown, but other known materials may be used.

  When the electrodes 507 are formed, the light emitting element 508 is completed. Thereafter, a protective film 509 is provided so as to completely cover the light emitting element 508. As the protective film 509, an insulating film including a carbon film, a silicon nitride film, or a silicon nitride oxide film can be used, and these insulating films can be used as a single layer or a stacked layer.

  Further, a sealing material 510 is provided so as to cover the protective film 509, and a cover material 511 is attached thereto. The sealing material 510 is an ultraviolet curable resin, and it is preferable to use a substance having a hygroscopic effect or a substance having an antioxidant effect inside. In this embodiment, a glass substrate, a quartz substrate, or a plastic substrate can be used as the cover material 511. Although not illustrated, a polarizing plate may be provided between the sealing material 510 and the cover material 511. By providing a polarizing plate, a display with high contrast can be provided.

  In this way, an active matrix having a structure having a P-type transistor 711 and an N-type transistor 710 constituting a drive circuit portion as shown in FIG. 17, an N-type transistor 712 for switching and a P-type transistor 713 for current control constituting a pixel portion. A type EL display device is completed. In this embodiment mode, a TFT structure having an LDD region that does not overlap with the gate electrode has been described; however, the TFT is not limited to this structure. The LDD region may not be provided, or a part or all of the LDD region may be formed so as to overlap with the gate electrode.

  Through the above process, an EL display device can be manufactured.

  In this embodiment, a thin film integrated circuit or a non-contact type thin film integrated circuit device (wireless chip, wireless IC tag, RFID (also referred to as radio frequency identification)) is replaced with the laser irradiation device of any of FIGS. A method of using these is shown in FIG.

  First, a separation layer 1701 is formed over a glass substrate (first substrate) 1700 by a sputtering method. The separation layer 1701 can be formed by a sputtering method, a low pressure CVD method, a plasma CVD method, or the like. In this embodiment, amorphous silicon having a thickness of about 50 nm is formed by a sputtering method and used as the peeling layer 1701. Note that the peeling layer 1701 is not limited to silicon, and may be formed using a material that can be selectively removed by etching (for example, W, Mo, or the like). The thickness of the release layer 1701 is desirably 50 to 60 nm.

  Next, a base insulating film 1702 is formed over the peeling layer 1701. The base insulating film 1702 is provided to prevent an alkali metal such as Na or an alkaline earth metal contained in the first substrate from diffusing into the semiconductor film and adversely affecting the characteristics of a semiconductor element such as a TFT. The base insulating film 1702 also has a role of protecting the semiconductor element in a process of peeling the semiconductor element later. The base insulating film 1702 may be a single layer or a stack of a plurality of insulating films. Therefore, an insulating film such as silicon oxide, silicon nitride, silicon oxide containing nitrogen (SiON), or silicon nitride containing oxygen (SiNO) that can suppress diffusion of alkali metal or alkaline earth metal into the semiconductor film is used. Form.

  Next, an amorphous semiconductor film 1703 is formed over the base insulating film 1702. The amorphous semiconductor film 1703 is preferably formed without being exposed to the air after the base insulating film 1702 is formed. The thickness of the amorphous semiconductor film 1703 is 20 to 200 nm (desirably 40 to 170 nm, preferably 50 to 150 nm).

  Then, the substrate manufactured in the above process is irradiated with a megahertz laser beam using the laser irradiation apparatus illustrated in FIGS. 5 to 9 to uniformly crystallize the amorphous semiconductor film 1703. Then, a crystalline semiconductor film 1704 is formed (see FIG. 11A).

  Next, as illustrated in FIG. 11B, the crystalline semiconductor film 1704 is etched to form island-shaped semiconductor layers 1705 to 1707, and then a gate insulating film 1708 is formed. The gate insulating film 1708 can be formed using a single layer or a stacked layer of silicon nitride, silicon oxide, silicon oxide containing nitrogen, or silicon nitride containing oxygen by a plasma CVD method, a sputtering method, or the like.

  Note that after the gate insulating film 1708 is formed, heat treatment is performed at 300 to 450 ° C. for 1 to 12 hours in an atmosphere containing 3 to 100% hydrogen to hydrogenate the island-shaped semiconductor layers 1705 to 1707. May be performed. Further, plasma hydrogenation (using hydrogen excited by plasma) may be performed as another means of hydrogenation.

  Next, as shown in FIG. 11C, gate electrodes 1709 to 1711 are formed. Here, after forming Si and W so as to be stacked by a sputtering method, gate electrodes 1709 to 1711 are formed by performing etching using the resist 1712 as a mask. Needless to say, the conductive material, structure, and manufacturing method of the gate electrodes 1709 to 1711 are not limited to these, and can be selected as appropriate. For example, a stacked structure of Si and NiSi (nickel silicide) doped with an n-type impurity (phosphorus, arsenic, or the like) or a stacked structure of TaN (tantalum nitride) and W (tungsten) may be used. Alternatively, a single layer may be formed using various conductive materials. In the case where the gate electrode and the antenna are formed at the same time, materials may be selected in consideration of their functions.

  A mask made of silicon oxide or the like may be used instead of the resist mask. In this case, a step of etching to form a mask (referred to as a hard mask) of silicon oxide, silicon oxide containing nitrogen, or the like is added, but since the film thickness of the mask during etching is less than that of the resist, a desired width is obtained. Gate electrodes 1709 to 1711 can be formed. Alternatively, the gate electrodes 1709 to 1711 may be selectively formed by a droplet discharge method without using the resist 1712.

  Next, as illustrated in FIG. 11D, an island-shaped semiconductor layer 1706 to be a p-channel TFT is covered with a resist 1713, and the gate-shaped electrodes 1709 and 1711 are used as masks to form island-shaped semiconductor layers 1705 and 1707. An impurity element imparting a mold (typically P (phosphorus) or As (arsenic)) is doped. By this doping step, doping is performed through the gate insulating film 1708, and a pair of low-concentration impurity regions 1716 and a channel formation region 1719 a are formed in the island-shaped semiconductor layer 1705, and a pair of low-concentration semiconductor layers 1707 is formed in the island-shaped semiconductor layer 1707. A concentration impurity region 1717 and a channel formation region 1719b are formed. Note that this doping step may be performed without covering the island-shaped semiconductor layer 1706 to be a p-channel TFT with the resist 1713.

  Next, as shown in FIG. 11E, after the resist 1713 is removed by ashing or the like, a resist 1718 is newly formed so as to cover the island-shaped semiconductor layers 1705 and 1707 to be n-channel TFTs. Using the electrode 1710 as a mask, the island-shaped semiconductor layer 1706 is doped with an impurity element imparting p-type conductivity (typically, B (boron)). By this doping step, doping is performed through the gate insulating film 1708, so that a pair of p-type high-concentration impurity regions 1720 and a channel formation region 1719c are formed in the island-shaped semiconductor layer 1706.

  Next, after the resist 1718 is removed by ashing or the like, an insulating film 1721 that covers the gate insulating film 1708 and the gate electrodes 1709 to 1711 is formed as illustrated in FIG.

After that, the insulating film 1721 is etched by an etch back method, and sidewalls 1722 to 1724 that are in contact with both side walls of the gate electrodes 1709 to 1712 are formed as shown in FIG. As an etching gas, a mixed gas of CHF ₃ and He was used.

  Next, as shown in FIG. 12C, a resist 1726 is newly formed so as to cover the island-shaped semiconductor layer 1706 to be a p-channel TFT, and the gate electrodes 1709 and 1711 and the sidewalls 1722 and 1724 are masked. As described above, an impurity element imparting n-type (typically P or As) is doped. In this doping step, doping is performed through the gate insulating film 1708, and a pair of n-type high concentration impurity regions 1727 and 1728 are formed in the island-shaped semiconductor layers 1705 and 1707.

  Next, after removing the resist 1726 by ashing or the like, the impurity region may be thermally activated. At this time, for example, after a silicon oxide film containing nitrogen of 50 nm is formed, heat treatment may be performed in a nitrogen atmosphere at 550 ° C. for 4 hours. Alternatively, after a silicon nitride film containing hydrogen is formed to a thickness of 100 nm, heat treatment may be performed at 410 ° C. for 1 hour in a nitrogen atmosphere. By these heat treatments, defects in the polycrystalline semiconductor film can be improved. This terminates dangling bonds existing in the polycrystalline semiconductor film, which is called a hydrogenation process.

  Through the series of steps described above, an n-channel TFT 1730, a p-channel TFT 1731, and an n-channel TFT 1732 are formed. In the manufacturing process, a miniaturized TFT having a channel length of 0.2 μm to 2 μm can be formed by appropriately changing the conditions of the etch back method and adjusting the size of the sidewall. Further, after that, a passivation film for protecting the TFTs 1730 to 1732 may be formed.

  Next, as illustrated in FIG. 12D, a first interlayer insulating film 1733 is formed so as to cover the TFTs 1730 to 1732. Further, a second interlayer insulating film 1734 is formed over the first interlayer insulating film 1733. Note that a filler may be mixed in the first interlayer insulating film 1733 or the second interlayer insulating film 1734. This is because the first interlayer insulating film 1733 or the second interlayer insulating film 1734 and the first interlayer insulating film 1733 or the like due to a stress generated from a difference in thermal expansion coefficient between a conductive material or the like constituting a wiring to be formed later. This is to prevent the second interlayer insulating film 1734 from peeling off or cracking.

  Next, as illustrated in FIG. 12D, contact holes are formed in the first interlayer insulating film 1733, the second interlayer insulating film 1734, and the gate insulating film 1708, and wirings 1735 to 1739 connected to the TFTs 1730 to 1732 are formed. Form. Note that the wirings 1735 and 1736 are in the high-concentration impurity region 1727 of the n-channel TFT 1730, the wirings 1736 and 1737 are in the high-concentration impurity region 1720 of the p-channel TFT 1731, and the wirings 1738 and 1739 are high-concentration impurity regions of the n-channel TFT 1732. 1728, respectively. Further, the wiring 1739 is also connected to the gate electrode 1711 of the n-channel TFT 1732. The n-channel TFT 1732 can be used as a memory element of a random number ROM.

  Next, as illustrated in FIG. 13A, a third interlayer insulating film 1741 is formed over the second interlayer insulating film 1734 so as to cover the wirings 1735 to 1739. The third interlayer insulating film 1741 is formed so that the wiring 1735 is partially exposed. Note that the third interlayer insulating film 1741 can be formed using a material similar to that of the first interlayer insulating film 1733.

  Next, an antenna 1742 is formed over the third interlayer insulating film 1741. The antenna 1742 can be formed using a conductive material containing a metal or a metal compound such as Ag, Au, Cu, Pd, Cr, Mo, Ti, Ta, W, Al, Fe, Co, Zn, Sn, or Ni. The antenna 1742 is connected to the wiring 1735. Note that in FIG. 13B, the antenna 1742 is directly connected to the wiring 1735; however, the wireless IC tag of the present invention is not limited to this structure. For example, the antenna 1742 and the wiring 1735 may be electrically connected using a separately formed wiring.

  The antenna 1742 can be formed by a printing method, a photolithography method, an evaporation method, a droplet discharge method, or the like. In FIG. 13A, the antenna 1742 is formed using a single-layer conductive film; however, the antenna 1742 can be formed by stacking a plurality of conductive films. For example, the antenna 1742 may be formed by coating a wiring formed of Ni or the like with electroless plating. The droplet discharge method means a method of forming a predetermined pattern by discharging droplets containing a predetermined composition from the pores, and includes an ink jet method and the like in its category. The printing method includes a screen printing method and an offset printing method. By using a printing method or a droplet discharge method, the antenna 1742 can be formed without using an exposure mask. In addition, unlike the photolithography method, there is no waste of material that is removed by etching in the droplet discharge method and the printing method. Further, since it is not necessary to use an expensive mask for exposure, the cost for manufacturing a wireless IC tag can be suppressed.

  In the case of using a droplet discharge method or various printing methods, for example, conductive particles in which Cu is coated with Ag can be used. Note that in the case where the antenna 1742 is formed by a droplet discharge method, it is preferable to perform treatment on the surface of the third interlayer insulating film 1741 so as to increase the adhesion of the antenna 1742. As a method for improving the adhesion, specifically, for example, a method of attaching a metal or a metal compound that improves the adhesion of the conductive film or the insulating film to the surface of the third interlayer insulating film 1741 by catalytic action, a conductive film formed Alternatively, an organic insulating film with high adhesion to the insulating film, a method of attaching a metal or a metal compound to the surface of the third interlayer insulating film 1741, or plasma on the surface of the third interlayer insulating film 1741 under atmospheric pressure or reduced pressure. Examples of the method include performing treatment to modify the surface.

  When the metal or metal compound attached to the third interlayer insulating film 1741 has conductivity, the sheet resistance is controlled so that the normal operation of the antenna is not hindered. Specifically, the average thickness of the conductive metal or metal compound is controlled to be, for example, 1 to 10 nm, or these metals or metal compounds are partially or entirely insulated by oxidation. Or just make it. Alternatively, the deposited metal or metal compound may be selectively removed by etching except for the region where the adhesion is desired to be improved. Alternatively, the metal or the metal compound may be selectively attached only to a specific region by using a droplet discharge method, a printing method, a sol-gel method, or the like, instead of attaching the metal or the metal compound to the entire surface of the substrate in advance. Note that the metal or metal compound does not need to be a completely continuous film on the surface of the third interlayer insulating film 1741, and may be dispersed to some extent.

  Then, as shown in FIG. 13B, after the antenna 1742 is formed, a protective layer 1745 is formed over the third interlayer insulating film 1741 so as to cover the antenna 1742. The protective layer 1745 is formed using a material that can protect the antenna 1742 when the peeling layer 1701 is removed later by etching. For example, the protective layer 1745 can be formed by applying an epoxy resin, an acrylate resin, or a silicon resin soluble in water or alcohols to the entire surface.

  Next, as shown in FIG. 13C, a groove 1746 is formed in order to separate the wireless IC tags individually. The groove 1746 may be formed so long as the peeling layer 1701 is exposed. The groove 1746 can be formed by dicing, scribing, or the like. Note that in the case where it is not necessary to separate the wireless IC tag formed over the first substrate 1700, the groove 1746 is not necessarily formed.

Next, as illustrated in FIG. 13D, the peeling layer 1701 is removed by etching. Here, halogen fluoride is used as an etching gas, and this gas is introduced from the groove 1746. For example, ClF ₃ (chlorine trifluoride) is used, the temperature is 350 ° C., the flow rate is 300 sccm, the atmospheric pressure is 798 Pascal (798 Pa), and the treatment time is 3 hours. Further, a gas in which nitrogen is mixed with ClF ₃ gas may be used. By using halogen fluoride such as ClF ₃ , the separation layer 1701 is selectively etched, and the first substrate 1700 can be separated from the TFTs 1730 to 1732. The halogen fluoride may be either a gas or a liquid.

Alternatively, the first substrate 1700 may be peeled off by heat treatment without using an etchant. For example, a tungsten (W) film is used as the peeling layer 1701 and heat treatment is performed, so that a tungsten oxide (WO _X ) film is formed on the upper surface of the tungsten film. Then, since the tungsten oxide film is formed, the space between the peeling layer 1701 and the base insulating film 1702 becomes brittle, so that the glass substrate can be easily peeled off.

  Next, as illustrated in FIG. 14A, the peeled TFTs 1730 to 1732 and the antenna 1742 are attached to the second substrate 1751 with an adhesive 1750. As the adhesive 1750, a material capable of bonding the second substrate 1751 and the base insulating film 1702 is used. As the adhesive 1750, for example, various curable adhesives such as a reactive curable adhesive, a thermosetting adhesive, a photocurable adhesive such as an ultraviolet curable adhesive, and an anaerobic adhesive can be used.

  Note that the second substrate 1751 can be formed using an organic material such as flexible paper or plastic.

  Next, as shown in FIG. 14B, after the protective layer 1745 is removed, an adhesive 1752 is applied over the third interlayer insulating film 1741 so as to cover the antenna 1742, and a cover material 1753 is attached thereto. The cover material 1753 can be formed using a flexible organic material such as paper or plastic, like the second substrate 1751. The thickness of the adhesive 1752 may be, for example, 10 to 200 μm.

  The adhesive 1752 is formed using a material that can bond the cover material 1753 to the third interlayer insulating film 1741 and the antenna 1742. As the adhesive 1752, for example, various curable adhesives such as a reactive curable adhesive, a thermosetting adhesive, a photocurable adhesive such as an ultraviolet curable adhesive, and an anaerobic adhesive can be used.

  The wireless IC tag is completed through the above-described steps. By the above manufacturing method, an extremely thin integrated circuit with a total film thickness of 0.3 μm to 3 μm, typically about 2 μm, can be formed between the second substrate 1751 and the cover material 1753.

Note that the thickness of the integrated circuit includes not only the thickness of the semiconductor element itself but also the thicknesses of various insulating films and interlayer insulating films formed between the adhesive 1750 and the adhesive 1752. Further, the planar area occupied by the integrated circuit in the wireless IC tag, 5 mm square (25 mm ²⁾ or less, and more preferably may be 0.3mm square (0.09 mm ²⁾ to 4 mm square (16 mm ²⁾ degree.

  Note that although this embodiment mode describes a method in which a separation layer is provided between the first substrate 1700 with high heat resistance and the integrated circuit, and the separation layer is removed by etching, whereby the substrate and the integrated circuit are separated. The manufacturing method of the wireless IC tag of the present invention is not limited to this configuration. For example, a metal oxide film may be provided between a substrate having high heat resistance and the integrated circuit, and the integrated circuit may be peeled by weakening the metal oxide film by crystallization. Alternatively, a release layer using an amorphous semiconductor film containing hydrogen is provided between the substrate with high heat resistance and the integrated circuit, and the substrate and the integrated circuit are separated by removing the release layer by laser beam irradiation. You may do it. Alternatively, the integrated circuit may be separated from the substrate by mechanically removing the highly heat-resistant substrate on which the integrated circuit is formed or removing the substrate by etching with a solution or gas.

  Note that although an example in which the antenna is formed over the same substrate as the integrated circuit has been described in this embodiment, the present invention is not limited to this structure. An antenna formed over another substrate and the integrated circuit may be bonded later to be electrically connected.

  Note that the frequency of radio waves generally used in RFID (radio frequency identification) is 13.56 MHz, 433 MHz, 860 to 960 MHz, and 2.45 GHz, so that radio waves of these frequencies can be detected. Forming a wireless IC tag is very important for enhancing versatility.

  The wireless IC tag of this embodiment has an advantage that radio waves are less likely to be shielded than an RFID tag formed using a semiconductor substrate, and the signal can be prevented from being attenuated by shielding the radio waves. Accordingly, since a semiconductor substrate is not necessary, the cost of the wireless IC tag can be significantly reduced.

  Note that although an example in which the integrated circuit is separated and attached to a flexible substrate is described in this embodiment, the present invention is not limited to this structure. For example, in the case of using a substrate having a heat resistant temperature that can withstand heat treatment in a manufacturing process of an integrated circuit such as a glass substrate, the integrated circuit is not necessarily peeled off.

  In this embodiment, a specific example using a semiconductor device manufactured according to the present invention will be described.

  FIG. 15A illustrates a display device, which includes a housing 2201, a support base 2202, a display portion 2203, a speaker portion 2204, a video input terminal 2205, and the like. The display portion 2203 includes a thin film transistor to form a pixel, and the EL display device described in Embodiment 6 may be used. By manufacturing the thin film transistor included in the display portion 2203 using the present invention, a bright display with few defects can be realized. The display portion 2203 may include a memory, a driver circuit portion, and the like, and the semiconductor device of the present invention may be applied to the memory, the driver circuit portion, and the like. In addition, as a display unit, a liquid crystal display device using an electro-optic effect of liquid crystal, a display device using a light emitting material such as electroluminescence, a display device using an electron source element, a contrast medium whose reflectivity changes by applying an electric field Various display devices using a combination of a thin film transistor and various display media such as a display device using electronic ink (also referred to as electronic ink) are included. The usage forms include all information display devices such as computers, televisions, information display devices such as electronic books, advertisement displays, and guidance displays.

  FIG. 15B illustrates a computer, which includes a housing 2211, a display portion 2212, a keyboard 2213, an external connection port 2214, a pointing mouse 2215, and the like. A thin film transistor is included in the display portion 2212, a CPU, a memory, a driver circuit portion, and the like attached to the computer. By using a semiconductor device manufactured using the present invention for the display portion 2212, a CPU attached to a computer, a memory, a driver circuit portion, and the like, performance can be improved and driving ability can be increased.

  FIG. 15C illustrates a mobile phone, which is a typical example of a mobile terminal. This mobile phone includes a housing 2221, a display portion 2222, operation keys 2223, and the like. Thin film transistors are included in a functional circuit portion such as a CPU and a memory attached to the display portion 2222 and the mobile phone. When a semiconductor device manufactured using the present invention is used for a functional circuit portion such as a CPU or a memory attached to the display portion 2222 or a mobile phone, performance can be improved and driving ability can be increased. A semiconductor device manufactured using the present invention can be used for electronic devices such as PDAs (Personal Digital Assistants), digital cameras, and small game machines, in addition to the above mobile phones.

  FIGS. 15D and 15E show a digital camera. Note that FIG. 15E is a diagram illustrating the back side of FIG. This digital camera includes a housing 2231, a display portion 2232, a lens 2233, operation keys 2234, a shutter 2235, and the like. Thin film transistors are provided in the display portion 2232, a driver circuit portion for controlling the display portion 2232, and the like. When the thin film transistor manufactured according to the present invention is used for the display portion 2232, a driver circuit portion for controlling the display portion 2232, another circuit, or the like, performance can be improved and driving ability can be increased.

  FIG. 15F illustrates a digital video camera. This digital video camera includes a main body 2241, a display portion 2242, a housing 2243, an external connection port 2244, a remote control receiving portion 2245, an image receiving portion 2246, a battery 2247, an audio input portion 2248, operation keys 2249, an eyepiece portion 2250, and the like. . Thin film transistors are provided in the display portion 2242 and a driver circuit portion that controls the display portion 2242. By using the thin film transistor manufactured using the present invention for the display portion 2242, a driver circuit portion for controlling the display portion 2242, other circuits, and the like, performance can be improved and driving ability can be increased.

  In addition, a thin film transistor manufactured using the present invention can be used as a thin film integrated circuit or a non-contact thin film integrated circuit device (also referred to as a wireless IC tag or RFID (radio frequency identification)). A thin film integrated circuit and a non-contact thin film integrated circuit manufactured by using the manufacturing method described in another embodiment can be used for a tag or a memory.

  FIG. 16A illustrates a state where the wireless IC tag 2302 is attached to the passport 2301. Further, the wireless IC tag 2302 may be embedded in the passport 2301. Similarly, affixing or embedding a wireless IC tag on a driver's license, credit card, banknote, coin, securities, gift certificate, ticket, traveler's check (T / C), health insurance card, resident's card, family register copy, etc. Can do. In this case, only information indicating authenticity is input to the wireless IC tag, and an access right is set so that information cannot be read or written illegally. By using it as a tag in this way, it becomes possible to distinguish it from a forged one.

  In addition, a wireless IC tag can be used as a memory. FIG. 16B illustrates an example in which the wireless IC tag 2311 is embedded in a label attached to a vegetable package. Further, a wireless IC tag may be attached or embedded in the package itself. The wireless IC tag 2311 includes a production stage process such as production place, producer, date of manufacture, processing method, product distribution process, price, quantity, usage, shape, weight, expiration date, various authentication information, etc. Can be recorded. Information from the wireless IC tag 2311 is received and read by the antenna unit 2313 of the wireless reader 2312 and displayed on the display unit 2314 of the reader 2312 so that the wholesaler, retailer, and consumer can easily grasp the information. become. In addition, by setting access rights for each of producers, traders, and consumers, a system is incapable of reading, writing, rewriting, and erasing without access rights.

  The wireless IC tag can be used as follows. At the time of accounting, the fact that accounting has been completed is entered in the wireless IC tag, and a check means is provided at the exit to check whether accounting has been written on the wireless IC tag. If you try to leave the store without checking out, an alarm will sound. This method can prevent forgetting to pay and shoplifting.

  Furthermore, in consideration of customer privacy protection, the following method can be used. At the stage of accounting at the cash register, (1) lock the data input to the wireless IC tag with a password, (2) encrypt the data itself input to the wireless IC tag, (3) wireless Either the data input to the IC tag is deleted, or (4) the data input to the wireless IC tag is destroyed. Then, by providing a check means at the exit, it is checked whether any of the processes (1) to (4) has been performed, or whether the wireless IC tag data has not been processed. Check for accounting. In this way, it is possible to check whether or not there is a transaction in the store, and it is possible to prevent information on the wireless IC tag from being read outside the store against the will of the owner.

  Note that there are several methods for destroying the data input to the wireless IC tag (4). For example, (a) a method of writing only “0 (off)” or “1 (on)” or both “0” and “1” into at least a part of electronic data of the wireless IC tag and destroying only the data Alternatively, (b) a method in which a current is excessively supplied to the wireless IC tag and a part of wiring of a semiconductor element included in the wireless IC tag is physically destroyed can be used.

  Since the wireless tag mentioned above is higher in manufacturing cost than a conventionally used barcode, it is necessary to reduce the cost. By using the present invention, uniform laser annealing of a semiconductor film is possible, so that a semiconductor device with good quality and no variation can be efficiently manufactured, which is effective for cost reduction. Further, any wireless tag can be manufactured with high quality and high reliability without any performance variation.

  As described above, the applicable range of the semiconductor device manufactured according to the present invention is so wide that the semiconductor device manufactured according to the present invention can be used for electronic devices in various fields.

FIG. 6 illustrates Embodiment 1. FIG. 6 illustrates Embodiment 1. FIG. 9 illustrates Embodiment 2. FIG. 9 illustrates Embodiment 2. FIG. 6 is a diagram illustrating Example 1; FIG. 6 is a diagram illustrating Example 1; FIG. 6 is a diagram illustrating a second embodiment. FIG. 6 is a diagram illustrating Example 3; FIG. 6 is a diagram illustrating Example 4; FIG. 6 is a diagram illustrating Example 5; FIG. 10 is a diagram illustrating Example 7. FIG. 10 is a diagram illustrating Example 7. FIG. 10 is a diagram illustrating Example 7. FIG. 10 is a diagram illustrating Example 7. FIG. 10 is a diagram illustrating Example 8. FIG. 10 is a diagram illustrating Example 8. FIG. 6 is a diagram illustrating Example 6; Photomicrograph of megahertz laser beam.

Explanation of symbols

11 beam 12 split beam 13 split beam 401 first laser beam 402 second laser beam 403 irradiated object 404 substrate 405 pulse tip 406 pulse end 407 pulse tip 408 pulse end

Claims (10)

A laser beam having a pulse width of 100 fs to 1 ns and a frequency of 10 MHz
In any two split beams selected from a plurality of split beams, with respect to one split beam, the other split beam is provided with an optical path difference that is greater than or equal to the length corresponding to the pulse width and less than the coherent length.
A method for manufacturing a semiconductor device, wherein the semiconductor film is crystallized by irradiating the semiconductor film with the plurality of divided beams provided with the optical path difference. A laser beam having a pulse width of 100 fs or more and 1 ns or less and a frequency of 10 MHz or more is divided into a plurality of parts,
In any two split beams selected from a plurality of split beams, for one split beam, the other split beam is longer than the length corresponding to the pulse width and less than the length corresponding to the pulse oscillation interval. An optical path difference equal to or longer than the pulse width and less than the coherent length is provided,
A method for manufacturing a semiconductor device, wherein the semiconductor film is crystallized by irradiating the semiconductor film with the plurality of divided beams provided with the optical path difference. A laser beam having a pulse width of 100 fs to 1 ns and a frequency of 10 MHz or more is divided into a first laser beam and a second laser beam;
Delaying the second laser beam relative to the first laser beam;
By irradiating the semiconductor film with the first laser beam and the delayed second laser beam, the semiconductor film is crystallized,
The method for manufacturing a semiconductor device, wherein the second laser beam is delayed with respect to the first laser beam within a range equal to or greater than the pulse width and less than a time corresponding to a coherent length. A laser beam having a pulse width of 100 fs to 1 ns and a frequency of 10 MHz or more is divided into a first laser beam and a second laser beam;
Delaying the second laser beam relative to the first laser beam;
By irradiating the semiconductor film with the first laser beam and the delayed second laser beam, the semiconductor film is crystallized,
The second laser beam is delayed with respect to the first laser beam within a range not less than the pulse width and less than the pulse oscillation interval, and not less than the pulse width and less than the time corresponding to the coherent length. A method for manufacturing a semiconductor device. In claim 1 or claim 2,
A method for manufacturing a semiconductor device, wherein each region of the semiconductor film irradiated with the plurality of split beams has a common portion. In claim 3 or claim 4 ,
A method for manufacturing a semiconductor device, wherein a region of the semiconductor film irradiated with the first laser beam and a region of the semiconductor film irradiated with the second laser beam have a common portion. In claim 1 or claim 2,
The method of manufacturing a semiconductor device, wherein the plurality of divided beams form a linear beam on an irradiation surface. In claim 3 or claim 4 ,
A manufacturing method of a semiconductor device, wherein the first laser beam and the second laser beam form a linear beam on an irradiation surface. In claim 1, claim 2, claim 5 or claim 7,
The method for manufacturing a semiconductor device, wherein the plurality of divided beams cause interference when the same region is irradiated at the same timing. Claim 3, claim 4, claim 6 or claim 8,
The method for manufacturing a semiconductor device, wherein the first laser beam and the second laser beam cause interference when irradiated to the same region at the same timing.