JP4588153B2 - Laser irradiation device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The invention disclosed in this specification relates to a technique for uniformly irradiating a large area with laser light. It also relates to its application.
[0002]
[Prior art]
In recent years, an amorphous semiconductor film or a crystalline semiconductor film (a semiconductor film having crystallinity such as polycrystalline or microcrystalline) that is formed over an insulating substrate such as glass, that is, a non-single-crystal semiconductor film is used. On the other hand, a technique for performing laser annealing to crystallize or improve crystallinity has been widely studied. A silicon film is often used as the semiconductor film.
[0003]
The glass substrate is inexpensive and rich in workability as compared with a quartz substrate that has been often used conventionally, and has an advantage that a large-area substrate can be easily manufactured. This is the reason for the above research. In addition, the reason why laser is used for crystallization is that the melting point of the glass substrate is low. The laser can give high energy only to the non-single crystal film without changing the temperature of the substrate so much.
[0004]
Since a crystalline silicon film formed by laser annealing has high mobility, a thin film transistor (TFT) is manufactured using this crystalline silicon film.
For example, it is actively used for a monolithic liquid crystal electro-optical device or the like in which TFTs for driving pixels and driving circuits are formed on a single glass substrate. Since the crystalline silicon film is made of many crystal grains, it is called a polycrystalline silicon film or a polycrystalline semiconductor film.
[0005]
In addition, a high-power excimer laser or other pulsed laser beam is processed by an optical system so that the irradiated surface has a rectangular shape of several centimeters square or a linear shape with a width of several mm and a length of 10 cm or more. The method of performing laser annealing by scanning the beam (moving the laser beam irradiation position relative to the irradiated surface) is preferred because it is mass-productive and industrially superior. .
[0006]
In particular, when a linear laser beam is used, unlike the case of using a spot laser beam that requires front / rear / left / right scanning, the entire irradiated surface is scanned by scanning only in a direction perpendicular to the linear direction of the linear laser. Since irradiation can be performed, high mass productivity is obtained. The reason for scanning in the direction perpendicular to the line direction is that it is the most efficient scanning direction. Due to this high mass productivity, at present, it is becoming mainstream to use a linear laser beam obtained by processing an excimer laser beam with an appropriate optical system for laser annealing.
[0007]
Recently, a continuous wave laser such as an Ar laser having a higher output has been developed. There is also a report that an Ar laser was used for annealing the semiconductor film and a good result was obtained. In this case, the irradiation surface is spot-like because the output of the Ar laser is not sufficient.
[0008]
[Problems to be solved by the invention]
As a laser widely used for crystallization, an excimer laser is known as a gas laser, and as a solid laser, an Nd: YAG laser or an Nd: YVO is used. _Four Lasers and argon lasers are known.
[0009]
Since the continuous wave argon laser has a wavelength around 500 nm, the absorption coefficient of the argon laser with respect to the silicon film is 10 ^Five About / cm. On the other hand, since the excimer laser is ultraviolet light of 400 nm or less, the absorption coefficient is 10 ⁶ / cm, about one digit higher than argon laser. For this reason, the argon laser attenuates the intensity to 1 / e (e is a natural logarithm) when it passes through the silicon film by 100 nm, and the excimer laser attenuates the intensity to 1 / e when it passes through the silicon film by 10 nm. .
[0010]
Generally, in the TFT field, the thickness of the polycrystalline silicon film is appropriately around 50 nm. If the silicon film is thicker than 50 nm, the off characteristics tend to deteriorate, and if it is thin, the reliability is affected. When a 50 nm thick silicon film is irradiated with an argon laser, more than half of the laser light passes through the silicon film and is absorbed by the glass substrate, heating the glass substrate more than necessary. Actually, a 200 nm silicon oxide film and a 50 nm silicon film were sequentially formed on a Corning 1737 substrate and crystallization was attempted with an argon laser. However, the glass was deformed before the silicon film was sufficiently crystallized.
[0011]
On the other hand, when the excimer laser is irradiated, most of the energy is absorbed by the 50 nm silicon film, and most of the laser light can be used for crystallization of the silicon film. Thus, the advantage of using the excimer laser for crystallization of the silicon film is the high absorption coefficient of the silicon film with respect to the excimer laser.
[0012]
The diagram shown in FIG. 24A is a top view of a silicon film irradiated while scanning with a conventional pulsed excimer laser. FIG. 24B is a cross-sectional view of the silicon film taken along a cross section parallel to the scanning direction of the pulsed excimer laser (a plane perpendicular to the silicon film including the line segment EF). FIG. 24C is a cross-sectional view of the silicon film cut along a plane perpendicular to the plane (plane perpendicular to the silicon film including the line segment GH).
[0013]
As can be seen from FIG. 24B, the pulse laser irradiation traces generate undulations of the same order as the silicon film thickness. On the other hand, the undulations shown in FIG. 24C are much smaller than those shown in FIG. 24B, but periodic undulations appear. As will be described later, this is due to interference of the linear beam shaped by the beam homogenizer.
[0014]
An optical system that plays a role in homogenizing the energy distribution (light intensity) in the linear beam is called a beam homogenizer. FIG. 25 shows an example of a beam homogenizer.
[0015]
On the optical path, cylindrical lens groups (also referred to as multi-cylindrical lenses or cylindrical lens arrays) 12 and 13, cylindrical lenses 14, slits 15, cylindrical lenses 16, mirrors are arranged on the optical path from the emission side of the laser device 11 which is a light source of excimer laser. 17 are arranged sequentially, and a cylindrical lens 18 is arranged on the optical path in the reflection direction of the mirror 17.
[0016]
The cylindrical lens 12 divides the beam into a plurality of predetermined directions (a direction perpendicular to the paper surface in the side view), and the cylindrical lens 16 combines the beams divided in this direction. On the other hand, the cylindrical lens group 13 also divides the beam into a plurality of predetermined directions (in the side view, the direction parallel to the paper surface), and the cylindrical lens 14 synthesizes the beams divided in the dividing direction of the cylindrical lens group 13.
[0017]
Therefore, the laser beam emitted from the oscillator is two-dimensionally divided by the cylindrical lens groups 12 and 13, enters the cylindrical lens 14, and is combined into a plurality of beams in a predetermined direction (a direction perpendicular to the paper in the side view). A plurality of beams divided in one direction (direction parallel to the paper surface) are passed through the slit 15 and condensed by the cylindrical lens 16 so as to become one beam again. The condensed beam is reflected by the mirror 17, collected by the cylindrical lens 18, and irradiated as a linear beam 19 (having a longitudinal direction in a direction perpendicular to the paper surface in a side view).
[0018]
In the homogenizer of FIG. 25, in the cylindrical lens groups 12 and 13, the beam splitting directions are orthogonal, and the cylindrical lenses 14 and 16 condense the beams are orthogonal to each other. According to the cylindrical lens group 12, the intensity distribution in the longitudinal direction of the linear laser beam 19 is made uniform by the combination of the cylindrical lens 16, and the width direction of the linear laser beam 19 is combined by the combination of the cylindrical lens group 13 and the cylindrical lens 14. In other words, the intensity distribution of the beam is made uniform, that is, the beams are two-dimensionally divided and overlapped again to make the energy of the linear beam uniform.
[0019]
Therefore, it seems that the energy distribution becomes more uniform as the number of beams divided by the cylindrical lens groups 12 and 13 increases. However, regardless of the fineness of the division, a stripe pattern, which is an irradiation trace of a linear laser beam, was formed on the silicon film. As shown in FIG. 24A, stripe patterns appear innumerably so as to be orthogonal to the longitudinal direction of the linear laser beam (or the scanning direction of the linear beam, the GH direction), and are shown in FIG. Thus, peaks appear periodically in the silicon film. The cause of the stripe pattern is expected to be either the beam before entering the beam homogenizer or the optical system of the beam homogenizer.
[0020]
Therefore, the present inventor conducted a simple experiment to find out the cause of the striped pattern. Before the rectangular laser beam was incident on the beam homogenizer, it was examined how the vertical stripes changed by rotating the laser beam. The result was that the striped pattern did not change at all. It was found that the stripe pattern was caused by the beam homogenizer, not the beam before entering the beam homogenizer. A beam homogenizer splits and re-superimposes the energy of a single-wavelength phase-matched beam (the laser is the light that has the same phase because the laser gains intensity by matching the phase). Therefore, it can be explained that the stripe pattern has interference of light when the light is superimposed.
[0021]
An object of the present invention is to eliminate the problem of interference of beams having the same phase as the laser light as described above, and to make the energy distribution in the longitudinal direction of the linear laser light uniform.
[0022]
[Means for Solving the Problems]
[Process to Lead to Invention] FIG. 26 schematically shows a state of optical interference fringes of the linear laser beam 19 formed by the beam homogenizer of FIG. In FIG. 26, the vertical axis represents the laser intensity I. As shown in the figure, peaks 302 appear periodically in the laser intensity in the linear laser beam 301. This peak 302 is present until interference occurs.
[0023]
The occurrence of the peak 302 is caused when the beams divided by the cylindrical lens groups 12 and 13 of the homogenizer are combined by the cylindrical lenses 14 and 16, and the beams interfere to form a standing wave in the beam. That is, since the divided laser beams are overlapped with the same region of the irradiated surface, it is considered that a large peak 302 is periodically generated.
[0024]
As shown in FIG. 26, in the homogenizer of FIG. 25, three waves are formed for each period in the longitudinal direction of the linear beam 19. The number of waves n (which may be referred to as the number of bright lines for one period of interference) and the number of lenses s of the cylindrical lens group 12 satisfy the following relational expression.
[0025]
n = (s-1) / 2 (s is an odd number)
[0026]
n = s ÷ 2 (s is an even number)
[0027]
In the homogenizer of FIG. 25, the number of lenses of the cylindrical lens group 12 is s = 7 (odd number), and n = 3.
[0028]
The inventor calculated the relationship between the shape of the longitudinal wave of the linear beam at a certain time of the linear laser beam and the number of lenses s by a computer. FIG. 27 shows the calculation result. FIG. 27A shows the case where s = 7 and n = 3, and FIG. 27B shows the case where s = 8 and n = 4.
[0029]
In FIG. 27, the horizontal axis represents the phase (position) in the longitudinal direction of the linear laser beam, and the vertical axis represents the amplitude of the wave. The square of the amplitude (value on the vertical axis) is the light intensity (the degree to which in-phase light is intensified). d is the length of one period of the wave, and is the distance between the brightest bright lines of the interference fringe. FIG. 27A shows an interference wave corresponding to FIG. 26, and d is the interval between the peaks 302 having the highest intensity.
[0030]
FIG. 27 shows the result of computer simulation. In an actual linear laser beam, the intensity of light intensity is not as clear as in simulation. This is presumed to be caused by subtle misalignment of the optical system, material of the optical member, processing error, and energy dispersion due to heat conduction in the semiconductor film. The actual intensity of the laser beam is as shown in FIG.
[0031]
In FIG. 25, when the cylindrical lens 16 is divided into two by the broken line 20 and the optical axis (principal point) is shifted in the direction perpendicular to the paper surface in the side view, the cylindrical lens 16 passes through the upper half cylindrical lens 16a. Since the beam passing through the lower half cylindrical lens 16b overlaps with an appropriate deviation on the irradiation surface, the interference fringe pattern changes. That is, the intensity (energy) distribution in the longitudinal direction of the linear beam changes. By making good use of this phenomenon, the light intensity can be averaged by optimizing the distance that the divided cylindrical lenses 16a and 16b are displaced from the principle of wave superposition.
[0032]
The present invention utilizes this phenomenon to design a beam homogenizer. Therefore, the beam homogenizer parameters were changed, and the change in the waveform of the linear beam was simulated by a computer. 28 and 29 show the simulation results. The graphs of FIGS. 28 and 29 show the relationship between the phase in the longitudinal direction of a linear laser beam at a certain time and the light intensity, as in FIG. FIG. 28 shows a case where the number of lenses s of the cylindrical lens group 12 of the homogenizer of FIG. 25 is 7, and FIG. 29 shows a case where the number of lenses s is 9.
[0033]
Hereinafter, the case of s = 7 will be considered using FIG. In the case of s = 7, it has been found that in order to average the intensity distribution in the longitudinal direction, it is preferable to superimpose waves whose phases are shifted by a half cycle (d / 2).
[0034]
The wave shown in FIG. 28B is a wave whose phase is shifted by a half period from FIG. When the wave of FIG. 28 (A) and the wave of FIG. 28 (B) are superimposed, the waveform shown in FIG. 28 (C) is obtained. A pattern is obtained.
[0035]
In the wave of FIG. 28C, amplitude fluctuations are averaged, and the wave period d ′ is smaller than the wave period d of FIG. If the values obtained by squaring the amplitudes of the waves in FIGS. 28A and 28C are compared, in FIG. 28C, the light intensity fluctuation at the position in the longitudinal direction is reduced and averaged. Understandable.
[0036]
Next, the case of s = 9 will be considered using FIG. In the case of s = 9, it has been found that the intensity in the longitudinal direction of the linear beam is most averaged, and it is preferable to superimpose three waves whose phases are shifted by 1/3 period (d / 3).
[0037]
In FIG. 29B and FIG. 29C, the phase of the wave in FIG. 29A is shifted by 1/3 period. FIG. 29D shows a combined wave obtained by superimposing the waves shown in FIGS. 29A to 29C.
[0038]
Similarly to the composite wave of FIG. 28C, the amplitude fluctuation of the wave of FIG. 29D is averaged, and the wave period d ′ is smaller than the wave period d of FIG. 29A. Yes. Comparing the values obtained by squaring the amplitudes of the waves in FIGS. 29A and 29D, the waves in FIG. 29D are averaged because the light intensity fluctuations at the longitudinal positions are reduced. I understand that. Furthermore, the energy distribution in the longitudinal direction of the wave in FIG. 29D is averaged more than in FIG.
[0039]
In order to form the waveforms as shown in FIG. 28C and FIG. 29D, the phase (position) of the wave is set at the same time as in the relationship between FIG. 28A and FIG. It can be seen that M beams shifted by d / M (d is a period and M is a natural number) are superposed.
[0040]
In the conventional beam homogenizer shown in FIG. 25, the laser beam that is generated through the combination of the cylindrical lens group 12 and the cylindrical lens 16 is divided by the cylindrical lens group 13.
[0041]
Therefore, if the phases of the respective laser beams divided by the cylindrical lens group 13 are shifted and superimposed at the same irradiation position, the energy in the longitudinal direction as shown in FIGS. 28 (C) and 29 (D). A linear laser beam having a uniform distribution can be formed.
[0042]
In FIG. 25, the beam transmitted through the cylindrical lens 16a obtained by dividing the cylindrical lens 16 into two parts and the beam transmitted through the cylindrical lens 16b can be obtained by shifting the positions of the cylindrical lenses 16a and 16b from each other. Similarly to the relationship between the wave and the wave in FIG. 28B, the phase can be shifted in the longitudinal direction of the linear beam 19 (the direction perpendicular to the paper surface in the side view). Therefore, by optimizing the phase shift, it is possible to average the intensity distribution in the longitudinal direction of the beam obtained by superimposing these beams, similarly to the wave in FIG.
[0043]
The invention disclosed herein provides an optimal combination of parameters for a beam homogenizer in order to optimize the phase shift.
[0044]
The beam homogenizer according to the present invention includes a first splitting optical lens that splits one beam into (2n + 1) beams in a first direction, and one beam perpendicular to the first direction. A second splitting optical lens that splits N (n-1) beams in a second direction; and a plurality of beams that are focused in the second direction and split in the second direction. A first synthesizing lens to synthesize, and a second synthesizing lens that synthesizes the plurality of beams that are condensed in the first direction and divided in the first direction,
The second combining lens is composed of (n′−1) cylindrical lenses,
Images obtained by orthogonal projection of the principal points of the (n′−1) cylindrical lenses on a plane orthogonal to the second direction are arranged on the same straight line at an interval of d / (n′−1). (N'-1) points,
The d is an interval between peaks of interference fringes formed by a beam transmitted through one cylindrical lens of the second synthesizing lens on an irradiation surface,
The N is a natural number, the n is an integer of 3 or more, and the n ′ is an integer satisfying 3 ≦ n ′ ≦ n.
[0045]
As described with reference to FIGS. 28 and 29, the intensity of interference fringes on the irradiated surface of the linear beam is synthesized by shifting the longitudinal phases of the linear beams by a predetermined amount and combining them. Can be made uniform.
[0046]
When the mirror is inserted in the beam homogenizer of the present invention, the direction of the optical path changes, so the direction in which the cylindrical lens divides the light beam and the direction of light collection are also changed.However, in the present invention, the change in direction by the mirror is ignored. Suppose that there is no mirror.
[0047]
In the present invention, in the second combining lens, the phases of the plurality of beams are shifted, and the plurality of beams are condensed so that the same region is irradiated. Therefore, in the second synthesizing lens, the principal point of each cylindrical lens is shifted by d / (n′−1) in a direction perpendicular to the optical axis.
[0048]
The first splitting lens according to the present invention includes a cylindrical lens group in which (2n + 1) cylindrical lenses having parallel optical axes are connected in a row (array). Furthermore, although the beam is divided into odd (2n + 1) beams in the first split cylindrical lens group here, it may be divided into even (2n) beams. In this case, 2n cylindrical lenses whose optical axes are parallel to each other may be connected in a row.
[0049]
The second splitting lens of the present invention may be constituted by a cylindrical lens group in which N (n−1) cylindrical lenses whose optical axes are parallel to each other are connected in a row. A cylindrical lens can be used as the first lens for synthesis.
[0050]
The homogenizer of the present invention exhibits a remarkable effect when a coherent beam is shaped into a linear shape, and can average the light intensity in the longitudinal direction of the linear beam. A laser device such as a gas laser or a solid-state laser is used as a light source for the coherent light. A continuous light emission type argon laser apparatus or a pulse oscillation type excimer laser apparatus can be used.
[0051]
An excimer laser is an example of the gas laser. Excimer lasers are widely recognized as pulsed lasers, but recently, continuous emission excimer laser oscillators have been developed. In order to emit light continuously, excitation of the oscillation gas is promoted using a microwave.
[0052]
By irradiating the oscillation gas with gigahertz-order microwaves, the reaction that is the rate-determining factor of oscillation is promoted, so that the excimer laser can emit light continuously. An excimer laser having a high absorption coefficient in a silicon film becomes increasingly important for crystallization of a semiconductor film when a continuous emission laser is put into practical use. If a continuous emission excimer laser is used, the irradiation trace of the pulse laser can be eliminated, so that the effect of the laser irradiation treatment can be made very uniform.
[0053]
As the excimer laser, for example, KrF laser (wavelength 248 nm), XeCl excimer laser (wavelength 308 nm), XeF laser (wavelength 351 nm, 353 nm), ArF excimer laser (wavelength 193 nm), XeF laser (wavelength 483 nm), or the like may be used.
[0054]
Pulse oscillation type Nd: YAG laser or Nd: YVO as solid laser _Four A laser can be used. In particular, when a pulsed laser device of a laser diode excitation type is used, a high output and a high pulse oscillation frequency can be obtained. Nd: YAG laser or Nd: YVO _Four The fundamental frequency of the laser is 1064 nm, but not only the fundamental frequency but also any of the second harmonic (532 nm), the third harmonic (354.7 nm), and the fourth harmonic (266 nm) can be used. it can.
[0055]
Embodiment
Embodiments of the present invention will be described below with reference to the drawings.
[0056]
1 and 2 show an optical system of the beam homogenizer of this embodiment. In this embodiment, a cylindrical lens is used for each lens of the homogenizer. 1A is a top view, FIG. 1B is a side view, and FIG. 2 is a perspective view.
[0057]
From the emission side of the laser generating means 201, cylindrical lens groups 202 and 203 for dividing the beam, a cylindrical lens 204 for overlapping the divided beams, and a cylindrical lens group 206 are sequentially arranged. Further, a slit 205 is disposed on the optical path between the cylindrical lens 204 and the cylindrical lens group 206, and a mirror 207 and a cylindrical lens 208 are sequentially disposed on the optical path on the transmission side of the synthesizing cylindrical lens group 206. Has been. For convenience of explanation, the mirror 207 is omitted in FIG.
[0058]
The cylindrical lens group 202 is composed of (2n + 1) cylindrical lenses whose optical axes are parallel to each other, and the direction in which one beam is divided into (2n + 1) beams is a direction perpendicular to the paper surface in the side view (first view). Direction).
[0059]
The cylindrical lens group 203 is composed of N (n−1) cylindrical lenses, and the direction in which one incident beam is divided into N (n−1) beams is a direction perpendicular to the paper surface in the top view (first order). 2 direction).
[0060]
The synthesizing cylindrical lens 204 is paired with the dividing cylindrical lens group 203 and is a lens for condensing the beam divided in the direction perpendicular to the paper surface in the top view. The cylindrical lens constituting the cylindrical lens group 203 and the generatrix of the cylindrical lens 204 are parallel.
[0061]
The synthesizing cylindrical lens group 206 is composed of (n′−1) cylindrical lenses whose optical axes are parallel to each other, and is paired with the cylindrical lens group 202, and is divided in a direction perpendicular to the paper surface in a side view. It is a lens for condensing the focused beam. The cylindrical lenses constituting the cylindrical lens groups 202 and 206 have generatrices parallel to each other.
[0062]
1 and 2 show an optical system in the case where N = 2, n = 3, and n ′ = 3. The number of lenses (2n + 1) of the cylindrical lens group 202 is 7, the number of lenses N (n−1) of the cylindrical lens group 203 is 4, and the number of lenses (n′−1) of the cylindrical lens group 206 is 2.
[0063]
The configuration of the cylindrical lens 206 group is shown in FIG. 3A is a perspective view, and FIG. 3B is a top view. As shown in FIG. 3, the cylindrical lens group 206 includes cylindrical lenses 206a and 206b. The principal points of the lenses 206a and 206b are shifted from each other by a predetermined length ΔD = d / 2.
[0064]
More specifically, as shown in FIG. 3B, the principal planes 100a and 100b of the cylindrical lenses 206a and 206b are the same plane, and the principal points 101a and 101b of the lenses are planes formed by these principal planes (or The images orthogonally projected on a plane orthogonal to the generatrix of each lens are on the same straight line and are arranged at intervals ΔD. The orthogonally projected images correspond to principal points 101a and 101b on the paper surface. Alternatively, the cylindrical lens group 206 is configured such that images obtained by orthogonally projecting the rear focal points 103a and 103b of the respective lenses onto a plane perpendicular to the plane (or the generatrix of each lens) formed by the main plane are on the same straight line. ΔD is lined up. Alternatively, an image obtained by orthogonally projecting the optical axes 102a and 102b of the respective lenses onto a plane orthogonal to the plane formed by the main plane is a parallel line having a distance ΔD.
[0065]
In FIG. 3, although the cylindrical lenses 206a and 206b have only one principal plane (principal point) in the lens, strictly speaking, there are two principal planes. One of the main plane and the object side main plane is considered.
[0066]
As described with reference to FIGS. 27 to 29, d is the period of the wave in the longitudinal direction, and interference formed by the beam transmitted through the cylindrical lens (206a or 206b) constituting the cylindrical lens group 206 on the irradiated surface. It is a cycle of a stripe.
[0067]
In order to measure the period d of interference fringes, in the cylindrical lens group 206, only one cylindrical lens 206a is left, and the other cylindrical lens 206b is directly observed with the linear laser beam in a state where light does not pass through 206b. The period may be measured. Moreover, it can measure indirectly by the annealing effect by the linear laser beam. For example, as described with reference to FIG. 24, vertical stripes appear in the silicon film irradiated with the linear laser, and the interval between the vertical stripes may be measured. Moreover, it can also obtain | require from a simple calculation formula so that it may show later.
[0068]
Hereinafter, a method for obtaining the optimum value of the cylindrical lens shift ΔD in the cylindrical lens group 206 will be described.
[0069]
In the present embodiment, the number of cylindrical lenses 202 is 7 (2n + 1), and n = 3. Therefore, the number of cylindrical lenses in the cylindrical lens group 203 is an integral multiple of (n-1) (N (n-1)). That is, an even number is sufficient. At this time, ΔD may be d / 2.
[0070]
The effect of this embodiment is demonstrated using FIG. The laser beam from the laser generator 201 is divided into seven beams in the cylindrical lens group 202. Here, one beam is traced for simplicity. The beam transmitted through the cylindrical lens group 202 is divided into four by the cylindrical lens group 203. The four beams are collected by the cylindrical lens 204 so as to be combined into one beam, shaped into a linear shape by the slit 205, and incident on the cylindrical lens group 206.
[0071]
In the cylindrical lens group 206, the slit 205 emits two beams whose phase in the longitudinal direction is shifted by ΔD (d / 2), is condensed by the cylindrical lens 208, and is irradiated as a linear beam 210.
[0072]
Here, among the beams irradiated on the irradiation surface, a linear beam transmitted through the cylindrical lens 206a is denoted as 211a, and a linear beam transmitted through the cylindrical lens 206b is denoted as 211b. The interference fringes of the linear beams 211a and 211b are schematically shown in black and white shading. Black is the bright line. When the linear beam 211a or 211b having such a strong interference fringe is irradiated while scanning in the width direction, as shown in FIG. 24, a stripe pattern due to the difference in the light intensity of the linear beam appears.
[0073]
Therefore, in this embodiment, by shifting the linear beams 211a and 211b by d / 2 in the longitudinal direction, the high intensity portion (black portion) and the low portion (white portion) overlap each other, As with the linear beam 210, interference fringes are eliminated or obscured.
[0074]
Here, when the number of lenses of the cylindrical lens group 202 is 9, since n = 4, if the number of cylindrical lens groups 203 is an integral multiple of 3 (corresponding to n−1), for example, 6, the linear beam 210 The energy distribution in the longitudinal direction can be made uniform. In this case, the number of lenses of the cylindrical lens group 206 can be made more uniform when the number of lenses is 3, rather than 2, so that instead of the cylindrical lens group 206, three cylindrical lenses as shown in FIG. A cylindrical lens group 220 may be used.
[0075]
As shown in FIG. 4, in the cylindrical lens group 220, the interval between the orthogonal projections of the principal points of the cylindrical lenses 220a, 220b, and 220c may be set so that ΔD is d / 3.
[0076]
However, the combination of three cylindrical lenses becomes expensive because the structure is complicated, and the alignment of the optical system becomes more difficult, so it may be configured with two cylindrical lenses.
[0077]
From the above consideration and calculation, when the number of cylindrical lenses in the cylindrical lens group 202 is an odd number, the number of cylindrical lenses in the cylindrical lens group 206 is (n′−1), and the principal points of the (n′−1) cylindrical lenses. It is understood that it is only necessary to shift d / (n′−1) sequentially. Here, n ′ is an integer that satisfies 3 ≦ n ′ ≦ n, and the number of lenses of the cylindrical lens group 203 is N (n′−1) (N is a natural number), that is, an integer of the number of lenses of the cylindrical lens group 206. It was good to be double.
[0078]
By doing so, the light intensity in the longitudinal direction within the linear beam can be made uniform by shifting the phases of the laser beams divided by the cylindrical lens group 203 and superimposing them.
[0079]
The configuration of the beam homogenizer shown in FIGS. 1 and 2 is basic, and other optical systems may be arranged. Further, a part of the optical member may be replaced with another optical member having the same action. The illustrated optical system may be used as a part of the whole.
[0080]
For example, although the cylindrical lens group 202 and the cylindrical lens group 203 shown in FIG. 1 are lens arrays made of convex lenses, a concave lens group or a concave / convex mixed lens array may be used. For example, a cylindrical lens array in which concave convex shapes are combined as shown in FIG. 7 can be used.
[0081]
Further, the cylindrical lens group 203 and the cylindrical lens 204 may be replaced with a multi-phase lens 231 shown in FIG. FIG. 9 shows an optical system when replaced. 9, the same reference numerals as those in FIG. 1 denote the same members.
[0082]
However, when using lens groups that are not congruent as represented by the concave-convex mixed lens group, the parallel light beams processed by these lenses are composed of lens groups having the same spread angle after processing. Better.
[0083]
Otherwise, when the split beams are recombined, the individual beams overlap in different sizes and shapes and the beam contours are unclear.
[0084]
The present embodiment is particularly effective when processing a rectangular laser beam having an aspect ratio of 100 or more into a linear laser beam having an aspect ratio that is not so large.
[0085]
Moreover, the number of lenses of the cylindrical lens group 202 can be an even number (2n). Other conditions may be the same as in the case of an odd number (2n + 1). However, a remarkable effect is obtained when the number of lenses of the cylindrical lens group 202 is an odd number. Even if the number of lenses of the cylindrical lens group 202 is an even number, if an odd number of beams are incident on the cylindrical lens group 203, the cylindrical lens group 202 is considered to be configured by an odd number of lenses. Needless to say, it is good.
[0086]
When the number of lenses of the cylindrical lens group 202 is an odd number, the longitudinal waveform of the linear beam on the irradiation surface can be a sine wave as shown in FIG. 28 (A) or FIG. 29 (D). The energy in the longitudinal direction can be made more uniform.
[0087]
When the number of lenses of the cylindrical lens group 202 is two or three, even a conventional homogenizer as shown in FIG. 25 can be a sine wave. However, since the number of beam divisions is small, the energy on the irradiation surface is small. However, it is difficult to obtain a uniform beam. Therefore, in the present invention, it is preferable that the number of lenses of the cylindrical lens group is at least six.
[0088]
When the number of lenses of the cylindrical lens group 202 is an even number, a beam in which the light intensity is dispersed cannot be obtained as in the case of an odd number. However, compared with a conventional optical system using a non-divided cylindrical lens, energy averaging has a significant effect.
[0089]
By the way, as shown in FIG. 27, since d is defined by the period of interference fringe, as described above, a value can be obtained by actually measuring the period of interference fringe. You can also.
[0090]
Using the wavelength λ of the laser light generated by the laser device 201, the focal length f of one cylindrical lens of the cylindrical lens group 206, and the width L of the cylindrical lens constituting the cylindrical lens group 202, the formula d = λf / L It can be calculated by
[0091]
Further, as apparent from the above description, it is preferable that the interference fringe period d is constant in the linear laser beam. That is, it is preferable that interference fringes appear along the longitudinal direction of the linear beam at a constant period as shown in FIG.
[0092]
However, in the beam homogenizer of FIG. 1, the interference fringe period of the linear beam 210 is not uniform except in a special case. This is because the linear beam 210 combines spherical waves into a linear shape. (See FIG. 10A. When spherical waves are cut in a straight line, the intervals between the same phases are not constant.) If it is desired to make the interference peak interval substantially constant, the plane wave may be synthesized into a linear shape. (FIG. 10B shows an optical system that forms such a light wave in which the interval between the same phases is constant when the plane wave is cut obliquely by a straight line.
[0093]
10B is different from FIG. 10A in whether the laser beam divided by the cylindrical lens group 202 on the beam incident side is all processed into parallel rays by the subsequent cylindrical lens group 206 or not. .
[0094]
The optical system as shown in FIG. 10B can be easily obtained by appropriately selecting the distance between the front cylindrical lens group 202 for division and the rear cylindrical lens group 206 for synthesis. 10B, any beam divided by the cylindrical lens group is processed into a plane wave by the cylindrical lens group 206. FIG. When a beam processed by the present optical system is used, the distance between the vertical stripes becomes substantially constant. An optical system having such an arrangement is most suitable for the present invention.
[0095]
However, even a linear beam obtained by synthesizing a spherical wave has a sufficiently large radius of curvature of the spherical wave, and can be regarded as a parallel light beam, so that the present invention can be applied. However, when a spherical wave is synthesized, the interference fringe period d is defined by the average value of the entire linear beam.
[0096]
As described above, by utilizing the invention disclosed in this specification, the energy distribution in the longitudinal direction of the linear laser beam is dramatically homogenized. In particular, when the number of lenses constituting the cylindrical lens group 202 is an odd number, the waveform in the longitudinal direction of the linear laser beam can be shaped into a sine wave (see FIGS. 28C and 29D). Since this is possible, the present invention works most effectively.
[0097]
However, it is very difficult to make the longitudinal energy distribution completely constant. Depending on the irradiation condition of the laser beam, the fluctuation of the energy distribution may be more emphasized.
[0098]
In such a case, it can be improved by finely adjusting the scanning direction of the laser beam. In the fine adjustment, laser processing is performed while the linear laser beam is scanned in a direction perpendicular to the linear direction of the beam and shifted from the direction including the surface formed by the linear laser beam by an angle y in the plane. To do. This angle y can be found in the range of | tan y | ≦ 0.1. (However, | tan y | ≠ 0)
[0099]
Laser annealing of the semiconductor film through the optical system described in the present invention to form a polycrystalline semiconductor film, for example, a device such as a TFT liquid crystal display, can suppress variations in characteristics of individual TFTs, resulting in high image quality. You can get something. The laser annealing of the present invention can be used not only for crystallizing a semiconductor but also for activating impurities such as phosphorus and boron doped to control the conductivity type of the semiconductor film.
[0100]
When the invention disclosed in this specification is used for laser annealing in manufacturing a semiconductor integrated circuit, characteristics of elements formed on the same substrate can be aligned, and a circuit having high performance can be obtained.
[0101]
【Example】
Embodiments of the present invention will be described with reference to the drawings.
[0102]
[Example 1]
In the manufacturing process of the example, first, a manufacturing method of a film irradiated with a laser will be described. The films irradiated with the laser are three kinds of amorphous silicon films in this specification. The present invention is effective for any film.
[0103]
First, all three types of amorphous silicon films are films formed by plasma CVD on a 127 nm square Corning 1737 glass substrate and a 200 nm thick silicon oxide film as a base film. The thickness of these amorphous silicon films is 50 nm. Hereinafter, this amorphous silicon film is referred to as a starting film.
[0104]
(Procedure for Membrane A)
The starting film is subjected to a thermal anneal at 450 ° C. for 1 hour. This step is a step for reducing the hydrogen concentration in the amorphous silicon film. This process is necessary because the film cannot withstand the laser energy if there is too much hydrogen in the film. The density of hydrogen in the film is 10 ²⁰ atoms / cm ^Three The order is appropriate. This film is called a non-single crystal silicon film A.
[0105]
(Producing procedure of membrane B)
A 10 ppm nickel acetate aqueous solution is applied onto the starting film by a spin coating method to form a nickel acetate layer. It is more preferable to add a surfactant to the aqueous nickel acetate solution. Since the nickel acetate layer is extremely thin, it is not always a film, but there is no problem in the subsequent steps.
[0106]
Next, thermal annealing is performed at 600 ° C. for 4 hours. Then, the amorphous silicon film is crystallized to form a crystalline silicon film B which is a non-single-crystal silicon film. At this time, nickel as a catalytic element plays a role of crystal growth nucleus, and crystallization is promoted. The Nickel can be crystallized at 600 ° C for 4 hours at a low temperature. Action by. Details are described in JP-A-6-244104.
[0107]
The concentration of the catalytic element is 1 × 10 ¹⁵ -10 ¹⁹ Atom / cm ^Three Is preferable. 1 × 10 ¹⁹ Atom / cm ^Three At the above high concentration, metallic properties appear in the crystalline silicon film and the characteristics as a semiconductor are lost. In this embodiment, the concentration of the catalytic element in the crystalline silicon film is the minimum value in the film and is 1 × 10. ¹⁷ ~ 5x10 ¹⁸ Atom / cm ^Three It is. These values are analyzed and measured by secondary ion mass spectrometry (SIMS).
[0108]
(Producing procedure of membrane C)
A silicon oxide film is formed to a thickness of 70 nm on the starting film. A plasma CVD method is used as a film forming method.
[0109]
Next, a part of the silicon oxide film is completely opened by a photolithography patterning process.
[0110]
Further, UV light is irradiated for 5 minutes in an oxygen atmosphere in order to form a thin oxide film in the opening. This thin oxide film is formed in order to improve the wettability of the opening portion with respect to the nickel aqueous solution to be introduced later.
[0111]
Next, a 100 ppm nickel acetate aqueous solution is applied onto the film by a spin coating method, and nickel acetate enters the hole portion. It is more preferable to add a surfactant to the aqueous nickel acetate solution.
[0112]
Next, thermal annealing is performed at 600 ° C. for 8 hours, and crystals grow laterally from the nickel-introduced portion. At this time, the role played by nickel is the same as that of the film B. Under these conditions, the crystal growth distance is about 40 μm.
[0113]
In this way, the amorphous silicon film is crystallized to form a crystalline silicon film C which is a non-single crystal silicon film. Thereafter, the silicon oxide film on the crystalline silicon film is peeled and removed using buffer hydrofluoric acid.
[0114]
The non-single crystal silicon films A, B, and C thus obtained are crystallized.
[0115]
Next, in order to further enhance the crystallinity, laser annealing is performed using an excimer laser.
[0116]
FIG. 5 shows a laser irradiation system in the embodiment. FIG. 5 is an overview of the laser irradiation system.
[0117]
In FIG. 5, the laser irradiation system is irradiated from a laser device 201, and after adjusting the laser traveling direction by two pairs of reflecting mirrors 901, the cross-sectional shape is processed into a linear shape by a beam homogenizer 902 disclosed in the present invention. The pulsed laser beam has a function of irradiating the substrate 904 to be processed while being reflected by the mirror 207 and condensed by the cylindrical lens 208. A beam expander that can suppress the spread angle of the laser beam and adjust the beam size may be inserted between the two pairs of reflection mirrors 901.
[0118]
The optical system 902 is an optical system on the optical path from the cylindrical lens group 202 to the cylindrical lens group 206 shown in FIG. 1, and the mirror 207 and the cylindrical lens 208 also conform to the structure shown in FIG. All the linear laser beams used in the present invention conform to the optical system shown in FIG. The role of the lens group of the type shown in FIG. 1 will be described below.
[0119]
In the beam homogenizer used in this embodiment, the number of cylindrical lens groups 202 is 7 (2n + 1) in this embodiment, so the number of lenses in the cylindrical lens group 206 in the structure shown in FIG. n-1).
[0120]
FIG. 3 shows a configuration diagram of the cylindrical lens group 206. Hereinafter, a method for determining the distance ΔD between the principal points of the cylindrical lenses 206a and 206b constituting the cylindrical lens group 206 will be described.
[0121]
In the case of the present embodiment, optical interference distributed in a linear laser beam formed through one lens 206a arbitrarily selected from the cylindrical lens group 206 and a lens other than the cylindrical lens group 206 in FIG. The period was 0.1 mm. This value corresponds to the parameter d used in the invention.
[0122]
As described above, the distance calculated from the formula d / (n-1) is set to a distance that shifts the principal points of the cylindrical lenses 206a and 206b, thereby making the energy distribution in the longitudinal direction of the linear beam most uniform. Can do.
[0123]
Here, the values of d and n are substituted into the equation. In this embodiment, since n = 3, the distance to be obtained is 0.05 mm. Needless to say, the effect is the same even if the distance is changed to 0.15 mm, 0.25 mm, 0.3 mm,... The wider the length, the shorter the effective length of the linear beam in the longitudinal direction.
[0124]
That is, if the principal points of the cylindrical lenses 206a and 206b are shifted from each other in the direction orthogonal to the optical axis, both ends in the longitudinal direction of the linear laser beam 210 are blurred by the shifted distance. (Refer to FIG. 2. The white portions at both ends indicate a blurred portion.) Since it is easy to set the portion irradiated with both ends in the longitudinal direction of the linear laser beam as an element region, some blurring is not treated at all. Does not affect. On the other hand, since both ends in the width direction are not blurred at all, there is no adverse effect even if they are struck by the element region.
[0125]
In this embodiment, since n = 3, the number of dividing the laser beam in the vertical direction (width direction of the linear beam) is determined by a multiple of (3-1). In the case of this example, N = 4 and eight divisions. Further, the number of dividing the laser beam in the horizontal direction (longitudinal direction of the linear beam) is (2 × 3 + 1) = 7. The beam emitted from the laser device 201 is divided into eight parts in the vertical direction (width direction of the linear laser beam) and seven parts in the horizontal direction (longitudinal direction of the linear laser beam).
[0126]
The linear laser beam 210 is a combination of 56 (7 × 8) divided beams. By doing so, the energy distribution of the beam is averaged.
[0127]
The ratio of the vertical and horizontal lengths of the beam is variable due to the structure of the lens group, but the easy-to-create beam shape is limited by the combination of the lens size and focal length. In this optical system, the length of the long side of the beam cannot be changed.
[0128]
Here, the laser oscillation device 201 uses a device that oscillates a XeCl excimer laser (wavelength 308 nm). In addition, a KrF excimer laser (wavelength 248 nm) or the like may be used.
[0129]
The substrate to be processed 904 is disposed on the stage 905. The stage 905 can be moved in one axis direction by a moving mechanism 903. In actual processing, the stage 905 is moved in parallel to the direction perpendicular to the line width direction of the linear laser beam (including a plane including the linear laser beam).
[0130]
FIG. 6 is a configuration diagram of a laser annealing apparatus provided with the laser irradiation apparatus of FIG. In the load / unload chamber 915, a cassette 913 containing a large number of substrates to be processed 904, for example, 20 sheets, is disposed. One substrate is moved from the cassette 913 to the alignment chamber 912 by the robot arm 914.
[0131]
The alignment chamber 912 is provided with an alignment mechanism for correcting the positional relationship between the substrate to be processed 904 and the robot arm 914. The alignment chamber 912 is connected to the load / unload chamber 915.
[0132]
The substrate is transferred to the substrate transfer chamber 911 by the robot arm 914 and further transferred to the laser irradiation chamber 916 by the robot arm 914.
[0133]
In FIG. 5, the linear laser beam irradiated onto the substrate to be processed 904 is 0.4 mm wide × 135 mm long.
[0134]
The energy density of the laser beam on the irradiated surface is 100 mJ / cm ² ~ 500mJ / cm ² For example, 300 mJ / cm ² And A linear laser beam is scanned by moving the stage 905 in one direction at 1.2 mm / s. When the laser oscillation frequency is 30 Hz and attention is paid to one point of the irradiated object, a 10-shot laser beam is irradiated. The number of shots is appropriately selected in the range of 5 shots to 50 shots.
[0135]
After the laser irradiation is completed, the substrate 904 to be processed is pulled back to the substrate transfer chamber 912 by the robot arm 914.
[0136]
The substrate 904 to be processed is transferred to the load / unload chamber 915 by the robot arm 914 and stored in the cassette 913.
[0137]
Thus, the laser annealing process is completed. In this way, by repeating the above steps, a large number of substrates can be successively processed one by one.
[0138]
In this embodiment, a linear laser is used. However, any beam shape ranging from a linear shape to a square shape can be used in the present invention.
[0139]
When a TFT using the semiconductor film subjected to laser annealing as an active layer is manufactured, both an N channel type and a P channel type can be manufactured.
[0140]
It is also possible to obtain a structure in which an N channel type and a P channel type are combined. In addition, an electronic circuit can be configured by integrating a large number of TFTs.
[0141]
The above is also true for a semiconductor film that has been laser-annealed through the optical system shown in the other embodiments. When a liquid crystal display composed of TFTs is manufactured using a semiconductor film that has been laser-annealed through the optical system of the present invention, a high-quality display with little variation in individual TFT characteristics can be obtained.
[0142]
[Example 2] In Example 1, when the stripe pattern does not disappear well, it is because the arrangement of the optical system is not appropriate, or the superposition of the linear laser beams is inappropriate. In this case, the scanning direction changing device 906 finely adjusts the scanning direction of the substrate, and a scanning direction in which interference fringes are less noticeable may be selected.
[0143]
That is, it is preferable that the laser beam is scanned and irradiated with a slight angle with respect to the width direction of the linear laser beam.
[0144]
Example 3 In Example 1, the pitch d of the interference fringe when the arrangement of the optical system shown in FIG. 1 is adopted can be easily derived by calculation. In this embodiment, the calculation method will be described with reference to FIGS.
[0145]
First, it is assumed that the divided lenses of the cylindrical lens group 206 are not shifted from each other. For convenience, the cylindrical lens group 206 in this state is referred to as a cylindrical lens 1206.
[0146]
The optical system shown in FIG. 11 may be considered to show a cross section of the cylindrical lens group 202 and the cylindrical lens 1206 shown in FIG.
[0147]
When the arrangement of the optical system in FIG. 11 is employed, the beams synthesized by the cylindrical lens 1206 may be referred to as plane waves.
[0148]
In this case, among the lenses constituting the cylindrical lens group 202, the laser light beam incident on the cylindrical lens 1206 via the two lenses 1201 adjacent to the central lens intersects the irradiation surface 1204 at an angle α.
[0149]
Here, since the wavefront 1205 of the laser is a straight line, a straight line drawn at intervals of the wavelength λ of the wavefront cuts the irradiation surface 1204 at an interval β. (See FIG. 12.)
[0150]
The relational expression between the angle α and the interval β can be expressed using the wavelength λ. That is, it can be expressed as β = λ / sin α.
[0151]
The two lenses 1201 form a standing wave with an interval β on the irradiation surface 1204. Further, the two lenses 1202 form a standing wave with an interval β / 2 on the irradiation surface 1204. Further, the two lenses 1203 form a standing wave with an interval β / 3 on the irradiation surface 1204. These standing waves are synthesized at the irradiation surface 1204 to form standing waves as shown in FIGS. 28C and 29D. Therefore, β coincides with the wave period d in the longitudinal direction, as shown in FIGS. This can be understood by a simple calculation.
[0152]
Further, even if the position of the cylindrical lens 1206 is moved with respect to the cylindrical lens group 202 along a direction parallel to the main plane of the lens 1206 indicated by an arrow 1207 (direction perpendicular to the optical axis), the period d does not change. This can be understood by simple calculation. This suggests that when the cylindrical lens 1206 is returned to the state of the cylindrical lens group 206, even if each of the divided cylindrical lenses 1206 is moved in the direction indicated by the arrow 1207, the essence of the present invention is not affected at all. To do.
[0153]
In this case, if the focal length f of the cylindrical lens 1206 and the width L per lens of the cylindrical lens group 202 are set, tan α = L / f is established.
[0154]
Since the angle α is sufficiently small, tan α≈sin α is established. Therefore, β≈λf / L is established. As described above, since β = d in general, d≈λf / L.
[0155]
Therefore, if the focal length f of the cylindrical lens 1206, the width L per lens of the cylindrical lens group 202, and the wavelength λ of the laser light are known, the calculation can be performed without actually measuring the period d of the interference fringe of the linear laser. Can be obtained.
[0156]
When the optical system having the arrangement shown in FIG. 10A is adopted, the beam that has passed through the cylindrical lens group 206 becomes a spherical wave, and the above-described mathematical formula is not completely established. In this case, d is calculated by numerical calculation using a computer.
[0157]
Even in this case, if the sum of the focal length f of the cylindrical lens group 206 and the focal length of the cylindrical lens group 202 is close to the interval between the cylindrical lens group 206 and the cylindrical lens group 202, d obtained by the above formula is used. be able to.
[0158]
[Embodiment 4] In the above-described embodiment, a pulsed excimer laser is used. However, in this embodiment, a continuous emission excimer laser is used. In the case of the continuous emission type, since the scanning speed of the linear laser is slower than that of the pulse oscillation type, the heat of the laser light is easily transmitted to the substrate. Therefore, it is desired to use a quartz substrate having a high strain point temperature for the substrate. Even if the quartz substrate is heated to the melting temperature of the silicon film, it is not deformed or altered at all. Therefore, the beam size can be expanded.
[0159]
In this embodiment, an example in which a 1000 W continuous light emitting excimer laser is processed into a linear beam (size: 125 mm × 0.4 mm) is shown. FIG. 13 shows the configuration of the beam homogenizer of this embodiment. The beam homogenizer in FIG. 13 corresponds to the one in which the slit 205 is omitted in FIG. The cylindrical lens group 407 corresponds to the cylindrical lens group 202, the cylindrical lens group 408 corresponds to the cylindrical lens group 203, the cylindrical lens 409 corresponds to the cylindrical lens 204, and the cylindrical lens group 410 corresponds to the cylindrical lens group 206.
[0160]
All the above optical members are made of quartz. Quartz was used because the excimer laser has a sufficiently high transmittance in the wavelength region. Further, an appropriate coating was applied to the surface of the optical system in accordance with the wavelength of the excimer laser to be used (in this specification, 248 nm). As a result, a transmittance of 99% or more was obtained with the lens alone. The durability of the lens has also increased.
[0161]
All the lenses had a curvature in the width direction, and all were spherical lenses. The lens material was synthetic quartz, and AR coating was applied so that a transmittance of 99% or more was obtained at a transmitted light wavelength of 248 nm.
[0162]
Similar to the first embodiment, the combination of the cylindrical lens group 407 and the cylindrical lens group 410 has a function of making the intensity distribution in the longitudinal direction of the linear laser beam uniform, and the combination of the cylindrical lens group 408 and the cylindrical lens 409 is as follows. And has a function of making the intensity distribution in the width direction of the linear laser beam uniform.
[0163]
A combination of the cylindrical lens group 408 and the cylindrical lens 409 forms a beam having a beam width w once. By arranging a doublet cylindrical lens 412 further via the mirror 411, a thinner linear laser beam (thinner than the beam width w) can be obtained.
[0164]
The apparatus shown in FIG. 13 has a function of irradiating a laser beam emitted from the laser apparatus 406 as a linear beam 405 through an optical system indicated by cylindrical lenses 407, 408, 409, 410, and 412. The stage 413 is a single axis stage that operates in one direction. By scanning this, the substrate placed on the stage 413 is irradiated with laser.
[0165]
The size of the laser beam emitted from the laser device 406 is originally a 0.3 mm diameter circular beam, but this is expanded into an ellipse of approximately 10 × 35 mm using two sets of beam expanders (not shown). Reference numeral 411 denotes a mirror.
[0166]
The energy distribution of the linear laser beam formed by the optical system in FIG. 13 showed a rectangular distribution when viewed in the cross section in the width direction. That is, it was possible to obtain a linear laser beam with very high energy density.
[0167]
At this time, the cylindrical lens group 407 used seven cylindrical lenses having a focal length of 41 mm, a width of 5 mm, a length of 30 mm, and a center thickness of 5 mm. The cylindrical lens group 408 used four cylindrical lenses having a focal length of 250 mm, a width of 2 mm, a length of 60 mm, and a center thickness of 5 mm. As the cylindrical lens 409, a cylindrical lens having a focal length of 200 mm, a width of 30 mm, a length of 120 mm, and a center thickness of 10 mm was used. The cylindrical lens group 410 uses two cylindrical lenses having a focal length of 1022 mm, a width of 180 mm, a length of 20 mm, and a center thickness of 35 mm.
[0168]
The doublet cylindrical lens 412 used was a set of two cylindrical lenses having a width of 90 mm, a length of 160 mm, and a center thickness of 16 mm, and a combined focal length of 220 mm.
[0169]
In addition, the cylindrical lens group 407 was disposed near the 2100 mm laser device from the irradiation surface along the optical path. The cylindrical lens group 408 was disposed near the 1980 mm laser device 406 from the irradiation surface along the optical path. The cylindrical lens 409 was disposed near the 1580 mm laser device 406 from the irradiation surface along the optical path.
[0170]
The cylindrical lens 410 was disposed near the 1020 mm laser device 406 from the irradiation surface along the laser optical path. The doublet cylindrical lens 412 was disposed near the 275 mm laser from the irradiation surface along the optical path. The above numbers are approximate, and depend on the accuracy of lens creation.
[0171]
The entire surface of the silicon film is crystallized by scanning a linear continuous light emitting excimer laser beam processed to the above size by a method as shown in FIG. Since the length of the long side of the linear laser beam is equal to or longer than the length of the short side of the silicon film, the entire surface of the substrate can be crystallized by one scan. In FIG. 14, 401 is a substrate, 402 is a source driver region, 403 is a gate driver region, and 404 is a pixel. As can be seen from FIG. 14, the entire silicon film is crystallized only by scanning the linear laser beam 405 once in one direction.
[0172]
The practitioner may determine the scanning speed as appropriate, but the standard is selected in the range of 0.5 to 100 mm / s. At this time, it is necessary to run the uniaxial stage 413 before irradiation until the scanning speed reaches a desired speed.
[0173]
[Example 5] When mass-producing liquid crystal panels, a method of forming a plurality of panels on one substrate and cutting the substrate after the process is generally performed.
In this embodiment, an example in which a linear laser beam is irradiated on such a multi-planar substrate using a continuous light emitting excimer laser oscillation device as a light source is shown. In this embodiment, the size of the multi-sided substrate is 600 mm × 720 mm.
[0174]
Various methods of irradiating a multi-sided substrate with a linear laser are conceivable. In this embodiment, a representative one will be described.
[0175]
The method used in this example is shown in FIG. The laser light emitted from the continuous emission excimer laser oscillation device 1301 is converted into a linear laser beam 1304 on the irradiation surface (substrate 1306) through the optical system 1302 and Mira-1303. As the optical system 1302, the one shown in the previous embodiment, for example, the one shown in FIG. 13 is used.
[0176]
In this embodiment, 5 × 6, that is, 30 3.5 inch liquid crystal panels are formed on the substrate 1306. Since the size of the multi-sided substrate is 600 mm × 720 mm, the area that one panel can hold is a square of 120 mm × 120 mm. FIG. 15 shows only four liquid crystal panels for simplicity. Of these, a region 1307 to be a source driver, a region 1308 to be a gate driver, and a region 1309 to be a pixel are illustrated.
[0177]
Since the length of the linear laser beam formed by the optical system shown in FIG. 13 is 125 mm, it is longer than the length of one side of the area occupied by one panel (120 mm square). Therefore, it is possible to process an area for one column of the panel only by scanning the linear laser beam once in one direction. Since the panels are arranged in 6 rows and 5 columns on the multi-sided substrate 1306, the entire surface of the substrate can be irradiated with laser by five scans. Substrate scanning is performed by moving an XY stage 1305 that is movable in two orthogonal directions. The scanning direction of the substrate is, for example, a direction indicated by a dotted arrow in FIG.
[0178]
[Embodiment 6] In this embodiment, another example of irradiating a multi-chamfer substrate with a linear laser beam using a continuous emission excimer laser oscillation device as a light source will be described. In this embodiment, the size of the multi-sided substrate is 600 mm × 720 mm.
[0179]
The method used in this example is shown in FIG. The laser light emitted from the continuous emission excimer laser oscillation device 1401 becomes a linear laser beam 1404 on the irradiation surface (substrate 1406) through the optical system 1402 and Mira-1403. As the optical system 1402, the one shown in the previous embodiment, for example, the one shown in FIG. 13 is used.
[0180]
In this embodiment, 10 × 12, that is, 120 2.6-inch liquid crystal panels are formed on the substrate 1406. Since the size of the multi-sided substrate is 600 mm × 720 mm, the area that one panel can hold is a square of 60 mm × 60 mm. FIG. 16 shows only four liquid crystal panels for simplicity. Of these, an area 1407 to be a source driver, an area 1408 to be a gate driver, and an area 1409 to be a pixel are illustrated.
[0181]
Since the length of the linear laser beam formed by the optical system shown in FIG. 13 is 125 mm, from the length of one side when the above four panels are arranged in 2 rows and 2 columns (120 mm square). Also long. Therefore, it is possible to process a region corresponding to two rows of panels only by scanning the linear laser beam once in one direction.
[0182]
Since the panel is arranged in 12 rows and 10 columns on the multi-sided substrate 1406, the entire surface of the substrate can be irradiated with laser by five scans. The substrate is scanned by moving an XY stage 1405 that is movable in two orthogonal directions. The scanning direction of the substrate is, for example, a direction indicated by a dotted arrow in FIG.
[0183]
As the length of the linear laser beam becomes longer and the panel becomes smaller, the number of columns of the panel that can be irradiated with laser by one scanning of the linear laser beam increases. Depending on the length of the linear laser beam and the panel size, it is possible to irradiate the laser for three rows of panels or more by one scanning of the linear laser beam.
[0184]
In the above embodiment, an example using an excimer laser is described, but a pulse oscillation type YAG laser or YVO is used. _Four A laser can be used, and in particular, when a laser device of a laser diode excitation type is used, a high output and a high pulse oscillation frequency can be obtained.
[0185]
[Example 7] In this example, an example in which a TFT (thin film transistor) is manufactured using the crystalline silicon film obtained in Example 1 is shown. The steps of this example are shown in FIGS.
[0186]
First, a quartz substrate 701 is prepared as a substrate. The reason why the quartz substrate is used is that a continuous emission excimer laser is used as a crystallization means. A 200-nm-thick silicon oxide film (also referred to as a base film) 702 and a 55-nm-thick amorphous silicon film 703a were continuously formed thereon without being released to the atmosphere. (FIG. 17A) This can prevent impurities such as boron contained in the atmosphere from adsorbing on the lower surface of the amorphous silicon film 703a.
[0187]
In this embodiment, an amorphous silicon film is used as the amorphous semiconductor film, but other semiconductor films may be used. An amorphous silicon germanium film may be used. Further, as a method for forming the base film and the semiconductor film, a PCVD method, an LPCVD method, a sputtering method, or the like can be used. After that, when the hydrogen concentration is high, heat treatment for reducing the hydrogen concentration may be performed.
[0188]
Next, the amorphous silicon film 703a is crystallized. Laser crystallization was performed using the laser irradiation method shown in Example 4. In this way, laser irradiation was performed for crystallization to form a region 704a made of a crystalline silicon (polysilicon) film. (Fig. 17 (B))
[0189]
Other methods include pulse oscillation type YAG laser and YVO. _Four There is a method using a laser. In particular, when a laser diode excitation type laser device is used, a high output and a high pulse oscillation frequency can be obtained. Laser annealing for crystallization uses one of the second harmonic (532 nm), third harmonic (354.7 nm), and fourth harmonic (266 nm) of these solid-state lasers. For example, laser pulse oscillation Frequency 1 to 20000 Hz (preferably 10 to 10000 Hz), laser energy density 200 to 600 mJ / cm ² (Typically 300-500mJ / cm ² ). Then, the linear beam is irradiated over the entire surface of the substrate, and the linear beam superposition ratio (overlap ratio) at this time is set to 80 to 90%. When the second harmonic is used, heat is uniformly transferred to the inside of the semiconductor layer, and crystallization is possible even if the irradiation energy range varies somewhat. As a result, a processing margin can be obtained, and variations in crystallization are reduced. Further, since the pulse frequency is high, the throughput is improved.
[0190]
The formed crystalline silicon (polysilicon) film was patterned to form a TFT semiconductor layer 704b. (Fig. 17 (C))
[0191]
Note that an impurity element (phosphorus or boron) for controlling the threshold voltage of the TFT may be added to the crystalline silicon film before and after the formation of the semiconductor layer 704b. This process may be performed only for NTFT or PTFT, or for both.
[0192]
Next, an insulating film 705 is formed by a sputtering method or a plasma CVD method, and a first conductive film 706a and a second conductive film 706b are stacked by a sputtering method. (Fig. 17 (D))
[0193]
This insulating film 705 is an insulating film that functions as a gate insulating film of the TFT, and has a thickness of 50 to 200 nm. In this example, a silicon oxide film having a thickness of 100 nm was formed by sputtering using silicon oxide as a target. Further, not only the silicon oxide film but also a stacked structure in which a silicon nitride film is provided over the silicon oxide film, or a silicon oxynitride film in which nitrogen is added to the silicon oxide film may be formed.
[0194]
In this embodiment, the amorphous silicon film is laser-crystallized and then patterned to form a gate insulating film. However, the process order is not particularly limited, and the amorphous silicon film and the gate are formed. An insulating film may be continuously formed by sputtering, followed by laser crystallization and patterning. When the film is continuously formed by sputtering, good interface characteristics can be obtained.
[0195]
The first conductive film 706a is formed using a conductive material containing an element selected from Ta, Ti, Mo, and W as a main component. The thickness of the first conductive film 706a may be 5 to 50 nm, preferably 10 to 25 nm. On the other hand, the second conductive film 707a is formed using a conductive material containing Al, Cu, or Si as a main component. The second conductive film 707a may be formed with a thickness of 100 to 1000 nm, preferably 200 to 400 nm. The second conductive film 707a is provided to reduce the wiring resistance of the gate wiring or the gate bus line.
[0196]
Next, unnecessary portions of the second conductive film 707a are removed by patterning to form an electrode 707b to be a part of the gate bus line in the wiring portion, and then resist masks 708a to 708d are formed. The resist mask 708a covers the PTFT, and the resist mask 708b is formed so as to cover the channel formation region of the NTFT of the driver circuit. The resist mask 708c covers the electrode 707b, and the resist mask 708d is formed to cover the channel formation region of the pixel matrix circuit. Thereafter, an impurity element imparting n-type conductivity is added using the resist masks 708a to 708d as masks, so that impurity regions 710 and 711 are formed. (FIG. 18 (A))
[0197]
In this embodiment, phosphorus is used as an impurity element imparting n-type, and phosphine (PH _Three ) Using an ion doping method. In this step, in order to add phosphorus to the semiconductor layer 704b thereunder through the gate insulating film 705 and the first conductive film 706a, the acceleration voltage was set to 80 keV and high. The concentration of phosphorus added to the semiconductor layer 704b is 1 × 10 ¹⁶ ~ 1x10 ¹⁹ atoms / cm ^Three In the range of 1 × 10 ¹⁸ atoms / cm ^Three It was. Then, regions 710 and 711 in which phosphorus was added to the semiconductor layer were formed. Part of the region added with phosphorus formed here functions as an LDD region. In addition, a part of the region covered with the mask and not doped with phosphorus (regions 709 and 712 made of a crystalline silicon film) functions as a channel formation region.
[0198]
Note that in the phosphorus addition step, an ion implantation method in which mass separation is performed may be used, or a plasma doping method in which mass separation is not performed may be used. The practitioner may set optimum values for the acceleration voltage, the dose amount, and the like.
[0199]
Next, after removing the resist masks 708a to 708d, an activation process is performed if necessary. Then, a third conductive film 713a was formed by a sputtering method. (FIG. 18B) The third conductive film 713a is formed using a conductive material containing an element selected from Ta, Ti, Mo, and W as a main component. The thickness of the third conductive film 713a is 100 to 1000 nm, preferably 200 to 500 nm.
[0200]
Next, resist masks 714a to 714d are newly formed and patterned to form the gate electrodes 706b and 713b of the PTFT and the wirings 706c and 713c. Then, p-type is imparted using the masks 714a to 714d as they are. An impurity element to be added is added to form a source region and a drain region of the PTFT. (FIG. 18C) Here, boron is used as the impurity element, and diborane (B ₂ H ₆ ) Using an ion doping method. Again, the acceleration voltage is 80 keV and 2 × 10 ²⁰ atoms / cm ^Three Boron was added to a concentration of.
[0201]
Next, the resist masks 714a to 714d are removed, and new resist masks 718a to 718e are formed. Then, etching is performed using the resist masks 718a to 718e as a mask, and the gate wirings 706d and 713d of the NTFT and the gates of the TFTs of the pixel matrix circuit are performed. Wirings 706e and 713e and storage capacitor upper wirings 706f and 713f are formed. (Figure 18 (D))
[0202]
Next, after removing the resist masks 718a to 718e and newly forming a resist mask 719, an impurity element imparting n-type conductivity is added to the source region and drain region of the NTFT to form impurity regions 720 to 725. (FIG. 19A) Here, the phosphine (PH _Three ) Using an ion doping method. The concentration of phosphorus added to the impurity regions 720 to 725 is higher than that of the step of adding the impurity element imparting n-type, and is 1 × 10 6. ¹⁹ ~ 1x10 ^{twenty one} atoms / cm ^Three Is preferred, here 1 × 10 ²⁰ atoms / cm ^Three It was.
[0203]
Thereafter, after removing the resist mask 719, a protective film 727 made of a silicon nitride film with a thickness of 50 nm is formed, and the state of FIG. 19B is obtained.
[0204]
Next, activation processing for activating the added impurity element imparting n-type or p-type is performed. This process includes thermal annealing using an electric heating furnace, the excimer laser, YAG laser, and YVO described above. _Four A laser annealing method using a laser or a rapid thermal annealing method (RTA method) using a halogen lamp may be used. In the case of heat treatment, heat treatment was performed at 300 to 700 ° C., preferably 350 to 550 ° C., and in this embodiment, 450 ° C. in a nitrogen atmosphere for 2 hours.
[0205]
YAG laser and YVO _Four When using a laser, one of its fundamental wave (1064 nm), second harmonic (532 nm), third harmonic (354.7 nm), or fourth harmonic (266 nm) is used, for example, laser pulse oscillation Frequency 1 to 20000 Hz (preferably 10 to 10000 Hz), laser energy density 200 to 600 mJ / cm ² (Typically 300-500mJ / cm ² ). Then, the linear beam is irradiated over the entire surface of the substrate, and the linear beam superposition ratio (overlap ratio) at this time is set to 80 to 90%. Further, since the pulse frequency is high, the throughput is improved.
[0206]
Next, after forming the first interlayer insulating film 730, contact holes are formed, and source and drain electrodes 731 to 735 and the like are formed by a known technique.
[0207]
Thereafter, a passivation film 736 is formed. As the passivation film 736, a silicon nitride film, a silicon oxynitride film, a silicon nitride oxide film, or a stacked film of these insulating films and a silicon oxide film can be used. In this embodiment, a silicon nitride film having a thickness of 300 nm is used as a passivation film.
[0208]
In this embodiment, plasma treatment using ammonia gas is performed as a pretreatment for forming the silicon nitride film, and the passivation film 736 is formed as it is. Since hydrogen activated (excited) by plasma by this pretreatment is confined by the passivation film 736, hydrogen termination of the active layer (semiconductor layer) of the TFT can be promoted.
[0209]
Further, when a nitrous oxide gas is added in addition to a gas containing hydrogen, the surface of the object to be processed is cleaned by the generated moisture, and contamination by boron or the like contained in the atmosphere can be effectively prevented.
[0210]
After the passivation film 736 was formed, an acrylic film having a thickness of 1 μm was formed as the second interlayer insulating film 737, and then patterned to form a contact hole, thereby forming a pixel electrode 738 made of an ITO film. Thus, an AM-LCD having a structure as shown in FIG. 19C is completed.
[0211]
Through the above steps, a channel formation region 709, impurity regions 720 and 721, and an LDD region 728 are formed in the NTFT of the driver circuit. The impurity region 720 is a source region, and the impurity region 721 is a drain region. In addition, a channel formation region 712, impurity regions 722 to 725, and an LDD region 729 are formed in the NTFT of the pixel matrix circuit. Here, in the LDD regions 728 and 729, a region overlapping with the gate electrode (GOLD region) and a region not overlapping with the gate electrode (LDD region) were formed.
[0212]
On the other hand, a channel formation region 717 and impurity regions 715 and 716 are formed in the p-channel TFT. The impurity region 715 is a source region, and the impurity region 716 is a drain region.
[0213]
For example, when a liquid crystal display is manufactured using a TFT manufactured using a semiconductor film formed by the above method, a laser processing after the laser processing is not conspicuous compared to the conventional case. This is because variation in TFT characteristics, particularly mobility variation, is suppressed by the present invention.
[0214]
FIG. 20A illustrates an example of a circuit configuration of an active matrix liquid crystal display device. The active matrix liquid crystal display device of this embodiment includes a source signal line side driver circuit 501, a gate signal line side driver circuit (A) 507, a gate signal line side driver circuit (B) 511, a precharge circuit 512, and a pixel matrix circuit. 506.
[0215]
The source signal line side driver circuit 501 includes a shift register circuit 502, a level shifter circuit 503, a buffer circuit 504, and a sampling circuit 505.
[0216]
The gate signal line driver circuit (A) 507 includes a shift register circuit 508, a level shifter circuit 509, and a buffer circuit 510. The gate signal line side driver circuit (B) 511 has a similar configuration.
[0217]
Further, in the present invention, it is easy to make the length of the LDD region different on the same substrate in consideration of the driving voltage of the NTFT, and the optimum shape is set for the TFTs constituting each circuit in the same process. It can also be built in.
[0218]
FIG. 20B is a top view of the pixel matrix circuit, and the AA ′ cross-sectional structure of the TFT portion and the BB ′ cross-sectional structure of the wiring portion correspond to FIG. 19C. A part is shown with the same code | symbol. In FIG. 20B, reference numeral 601 denotes a semiconductor layer, 602 denotes a gate electrode, and 603 denotes a capacitor line. In this embodiment, the gate electrode and the gate wiring are formed of a first conductive layer and a third conductive layer, and the gate bus line is formed of the first conductive layer, the second conductive layer, and the third conductive layer. And a clad structure formed of
[0219]
FIG. 21A shows a top view of a CMOS circuit which is a part of the driver circuit and corresponds to FIG. 19C. 610 is a source electrode of PTFT, 611 is a drain electrode, 612 is a source electrode of NTFT, and 613 and 614 are gate wirings. Further, in this embodiment, the NTFT and PTFT active layers are in direct contact with each other and share the drain electrode. However, the present invention is not particularly limited to this structure, and the structure (active layer is completely separated) as shown in FIG. Structure). In FIG. 21, reference numeral 620 denotes a PTFT source electrode, 621 denotes a drain electrode, 622 denotes an NTFT source electrode, and 623 and 624 denote gate wirings.
[0220]
Although the process for manufacturing a top gate type TFT has been described in this embodiment, it is needless to say that the laser irradiation apparatus of the present invention can be used to manufacture an inverted stagger type TFT.
[0221]
The laser irradiation apparatus of the present invention can be used for laser annealing for improving crystallization and crystallinity of a silicon film, and for laser activation for adding a dopant to activate a dopant.
[0222]
[Embodiment 8] The present invention can also be used when an interlayer insulating film is formed on a conventional MOSFET and a TFT is formed thereon. That is, it is also possible to realize a three-dimensional semiconductor device in which a reflective AM-LCD is formed on a semiconductor circuit.
[0223]
The semiconductor circuit may be formed on an SOI substrate such as SIMOX, Smart-Cut (registered trademark of SOITEC), ELTRAN (registered trademark of Canon Inc.), or the like.
[0224]
In addition, when implementing a present Example, you may combine any structure of Examples 1-7.
[0225]
[Embodiment 9] In this embodiment, a case will be described in which a TFT is formed on a substrate in the manufacturing process shown in Embodiment 7 and an AM-LCD is actually manufactured.
[0226]
When the state of FIG. 19C is obtained, an alignment film is formed to a thickness of 80 nm on the pixel electrode 738. Next, a glass substrate with a color filter, a transparent electrode (counter electrode), and an alignment film formed thereon is prepared as a counter substrate, and each alignment film is rubbed and a sealing material (sealing material) is used. The substrate on which the TFT is formed is bonded to the counter substrate. In the meantime, the liquid crystal material is held. Since this cell assembling process may use a known means, a detailed description thereof will be omitted.
[0227]
Examples of the liquid crystal material include TN liquid crystal, PDLC, ferroelectric liquid crystal, antiferroelectric liquid crystal, and a mixture of ferroelectric liquid crystal and antiferroelectric liquid crystal. For example, 1998, SID, “Characteristics and Driving Scheme of Polymer-Stabilized Monostable FLCD Exhibiting Fast Response Time and High Contrast Ratio with Gray-Scale Capability” by H. Furue et al., 1997, SID DIGEST, 841, “A Full -Color Thresholdless Antiferroelectric LCD Exhibiting Wide Viewing Angle with Fast Response Time "by T. Yoshida et al., 1996, J. Mater. Chem. 6 (4), 671-673," Thresholdless antiferroelectricity in liquid crystals and its application to The liquid crystal disclosed in "displays" by S. Inui et al. or US Pat. No. 5,945,569 can be used.
[0228]
A liquid crystal exhibiting an antiferroelectric phase in a certain temperature range is called an antiferroelectric liquid crystal. Among mixed liquid crystals having antiferroelectric liquid crystals, there is a so-called thresholdless antiferroelectric mixed liquid crystal that exhibits electro-optic response characteristics in which transmittance continuously changes with respect to an electric field. This thresholdless antiferroelectric mixed liquid crystal has a V-shaped electro-optic response characteristic, and a drive voltage of about ± 2.5 V (cell thickness of about 1 μm to 2 μm) is also found. ing.
[0229]
Here, FIG. 23 shows an example of the light transmittance characteristics of the thresholdless antiferroelectric mixed liquid crystal exhibiting a V-shaped electro-optic response with respect to the applied voltage. The vertical axis of the graph shown in FIG. 23 is the transmittance (arbitrary unit), and the horizontal axis is the applied voltage. The transmission axis of the polarizing plate on the incident side of the liquid crystal display device is set to be substantially parallel to the normal direction of the smectic layer of the thresholdless antiferroelectric mixed liquid crystal that substantially coincides with the rubbing direction of the liquid crystal display device. . Further, the transmission axis of the output-side polarizing plate is set to be substantially perpendicular (crossed Nicols) to the transmission axis of the incident-side polarizing plate.
[0230]
As shown in FIG. 23, it can be seen that when such a thresholdless antiferroelectric mixed liquid crystal is used, low voltage driving and gradation display are possible.
[0231]
When such a low-voltage thresholdless antiferroelectric mixed liquid crystal is used in a liquid crystal display device having an analog driver, the power supply voltage of the image signal sampling circuit is suppressed to about 5V to 8V, for example. Is possible. Therefore, the operating power supply voltage of the driver can be lowered, and low power consumption and high reliability of the liquid crystal display device can be realized.
[0232]
Further, even when such a low-voltage thresholdless antiferroelectric mixed liquid crystal is used in a liquid crystal display device having a digital driver, the output voltage of the D / A conversion circuit can be lowered. The operating power supply voltage of the A conversion circuit can be lowered, and the operating power supply voltage of the driver can be lowered. Therefore, low power consumption and high reliability of the liquid crystal display device can be realized.
[0233]
Therefore, using such a thresholdless antiferroelectric mixed liquid crystal driven by a low voltage uses a TFT (for example, 0 nm to 500 nm or 0 nm to 200 nm) having a relatively small LDD region (low concentration impurity region). It is also effective in some cases.
[0234]
In general, the thresholdless antiferroelectric mixed liquid crystal has a large spontaneous polarization, and the dielectric constant of the liquid crystal itself is high. For this reason, when a thresholdless antiferroelectric mixed liquid crystal is used in a liquid crystal display device, a relatively large storage capacitor is required for the pixel. Therefore, it is preferable to use a thresholdless antiferroelectric mixed liquid crystal having a small spontaneous polarization. Further, the driving method of the liquid crystal display device may be line-sequential driving, so that the period of writing the gradation voltage to the pixel (pixel feed period) may be lengthened to compensate for the small storage capacity. .
[0235]
In addition, since low voltage drive is implement | achieved by using such a thresholdless antiferroelectric mixed liquid crystal, the low power consumption of a liquid crystal display device is implement | achieved.
[0236]
Note that any liquid crystal having electro-optical characteristics as shown in FIG. 23 can be used as the display medium of the liquid crystal display device described in this specification.
[0237]
Next, the external appearance of the AM-LCD manufactured as described above is shown in FIG. As shown in FIG. 22, the active matrix substrate and the counter substrate face each other, and liquid crystal is sandwiched between these substrates. The active matrix substrate includes a pixel matrix circuit 631, a scanning line driver circuit 632, and a signal line driver circuit 633 formed on the substrate 630.
[0238]
The scanning line driving circuit 632 and the signal line driving circuit 633 are connected to the pixel matrix circuit 631 by scanning lines 641 and signal lines 642, respectively. These drive circuits 632 and 633 are mainly composed of CMOS circuits.
[0239]
A scanning line 641 is formed for each row of the pixel matrix circuit 631, and a signal line 642 is formed for each column. A TFT 640 of a pixel matrix circuit is formed in the vicinity of the intersection of the scanning line 641 and the signal line 642. The gate electrode of the TFT 640 of the pixel matrix circuit is connected to the scanning line 641 and the source is connected to the signal line 642. Further, a pixel electrode 643 and a storage capacitor 644 are connected to the drain.
[0240]
The counter substrate 650 has a transparent conductive film such as an ITO film formed on the entire surface of the substrate. The transparent conductive film is a counter electrode for the pixel electrode 643 of the pixel matrix circuit 631, and the liquid crystal material is driven by an electric field formed between the pixel electrode and the counter electrode. On the counter substrate 650, an alignment film, a black mask, and a color filter are formed as necessary.
[0241]
IC chips 635 and 636 are attached to the substrate on the active matrix substrate side using the surface to which the FPC 634 is attached. These IC chips 635 and 636 are configured by forming circuits such as a video signal processing circuit, a timing pulse generation circuit, a γ correction circuit, a memory circuit, and an arithmetic circuit on a silicon substrate.
[0242]
Further, in this embodiment, the liquid crystal display device is described as an example, but the present invention is applied to an EL (electroluminescence) display device and an EC (electrochromic) display device as long as it is an active matrix display device. It is also possible to do.
[0243]
Example 10
In this example, an example in which an EL (electroluminescence) display device is manufactured using the present invention will be described. 30A is a top view of the EL display device of the present invention, and FIG. 30B is a cross-sectional view thereof.
[0244]
In FIG. 30A, reference numeral 3001 denotes a substrate, 3002 denotes a pixel portion, 3003 denotes a source side driver circuit, 3004 denotes a gate side driver circuit, and each driver circuit reaches an FPC (flexible printed circuit) 3006 through a wiring 3005. Connected to an external device.
[0245]
At this time, the first sealant 3101, the cover material 3102, the filler 3103, and the second sealant 3104 are provided so as to surround the pixel portion 3002, the source side driver circuit 3003, and the gate side driver circuit 3004.
[0246]
FIG. 30B corresponds to a cross-sectional view taken along line AA ′ of FIG. 30A. A driving TFT included in the source side driver circuit 3003 on the substrate 3001 (here, an n-channel type is used here). TFTs and p-channel TFTs are shown.) 3201 and pixel TFTs included in the pixel portion 3002 (however, here, TFTs for controlling current to EL elements are shown) 3202 are formed. .
[0247]
An interlayer insulating film (planarization film) 3301 made of a resin material is formed on the driving TFT 3201 and the pixel TFT 3202, and a pixel electrode (cathode) 3302 electrically connected to the drain of the pixel TFT 3202 is formed thereon. As the pixel electrode 3302, a conductive film having a light-blocking property (typically, a conductive film containing aluminum, copper, or silver as its main component or a stacked film of such a conductive film and another conductive film) can be used. In this embodiment, an aluminum alloy is used as the pixel electrode.
[0248]
An insulating film 3303 is formed over the pixel electrode 3302, and an opening is formed over the pixel electrode 3302 in the insulating film 3303. In this opening, an EL (electroluminescence) layer 3304 is formed on the pixel electrode 3302. A known organic EL material or inorganic EL material can be used for the EL layer 3304. The organic EL material includes a low molecular (monomer) material and a high molecular (polymer) material, either of which may be used.
[0249]
A known technique may be used for forming the EL layer 3304. The EL layer may have a stacked structure or a single layer structure by freely combining a hole injection layer, a hole transport layer, a light emitting layer, an electron transport layer, or an electron injection layer.
[0250]
An anode 3305 made of a transparent conductive film is formed on the EL layer 3304. As the transparent conductive film, a compound of indium oxide and tin oxide or a compound of indium oxide and zinc oxide can be used. In addition, it is desirable to remove moisture and oxygen present at the interface between the anode 3305 and the EL layer 3304 as much as possible. Therefore, it is necessary to devise such that the both are continuously formed in a vacuum, or the EL layer 3304 is formed in a nitrogen or rare gas atmosphere, and the anode 3305 is formed without being exposed to oxygen or moisture. In this embodiment, the above-described film formation is possible by using a multi-chamber type (cluster tool type) film formation apparatus.
[0251]
The anode 3305 is electrically connected to the wiring 3005 in a region indicated by 3306. A wiring 3005 is a wiring for applying a predetermined voltage to the anode 3305 and is electrically connected to the FPC 3006 through a conductive material 3307.
[0252]
As described above, an EL element including the pixel electrode (cathode) 3302, the EL layer 3304, and the anode 3305 is formed. This EL element is surrounded by a first sealing material 3101 and a cover material 3102 bonded to the substrate 3001 by the first sealing material 3101 and enclosed by a filler 3103.
[0253]
As the cover material 3102, a glass plate, an FRP (Fiberglass-Reinforced Plastics) plate, a PVF (polyvinyl fluoride) film, a mylar film, a polyester film, or an acrylic film can be used. In the case of this embodiment, a light-transmitting material is used because the radiation direction of light from the EL element is directed toward the cover material 3102.
[0254]
However, it is not necessary to use a light-transmitting material when the light emission direction from the EL element is opposite to the cover material, and a metal plate (typically a stainless steel plate), a ceramic plate, or an aluminum foil is used. A sheet having a structure sandwiched between PVF films or mylar films can be used.
[0255]
As the filler 3103, an ultraviolet curable resin or a thermosetting resin can be used, and PVC (polyvinyl chloride), acrylic, polyimide, epoxy resin, silicone resin, PVB (polyvinyl butyral), or EVA (ethylene vinyl acetate) is used. Can be used. When a hygroscopic substance (preferably barium oxide) is provided inside the filler 3103, deterioration of the EL element can be suppressed. In this embodiment, a transparent material is used so that light from the EL element can pass through the filler 3103.
[0256]
Further, a spacer may be contained in the filler 3103. At this time, if the spacer is formed of barium oxide, the spacer itself can be hygroscopic. In the case where a spacer is provided, it is also effective to provide a resin film on the anode 3305 as a buffer layer that relieves pressure from the spacer.
[0257]
The wiring 3005 is electrically connected to the FPC 3006 through a conductive material 3307. The wiring 3005 transmits a signal transmitted to the pixel portion 3002, the source side driver circuit 3003, and the gate side driver circuit 3004 to the FPC 3006, and is electrically connected to an external device by the FPC 3006.
[0258]
In this embodiment, the second sealing material 3104 is provided so as to cover the exposed portion of the first sealing material 3101 and a part of the FPC 3006, and the EL element is thoroughly shielded from the outside air. Thus, an EL display device having the cross-sectional structure of FIG.
[0259]
Example 11
In this example, examples of a pixel structure that can be used for the pixel portion of the EL display device shown in Example 10 are shown in FIGS. In this embodiment, 3401 is a source wiring of the switching TFT 3402, 3403 is a gate wiring of the switching TFT 3402, 3404 is a current control TFT, 3405 is a capacitor, 3406 and 3408 are current supply lines, and 3407 is an EL element. .
[0260]
FIG. 31A shows an example in which the current supply line 3406 is shared between two pixels. That is, there is a feature in that the two pixels are formed so as to be symmetrical with respect to the current supply line 3406. In this case, since the number of power supply lines can be reduced, the pixel portion can be further refined.
[0261]
FIG. 31B illustrates an example in which the current supply line 3408 is provided in parallel with the gate wiring 3403. In FIG. 31B, the current supply line 3408 and the gate wiring 3403 are provided so as not to overlap with each other. However, if the wirings are formed in different layers, they overlap with each other through an insulating film. It can also be provided. In this case, since the exclusive area can be shared by the power supply line 3408 and the gate wiring 3403, the pixel portion can be further refined.
[0262]
In FIG. 31C, a current supply line 3408 is provided in parallel with the gate wiring 3403 in the same manner as the structure of FIG. 31B, and two pixels are symmetrical with respect to the current supply line 3408. It is characterized in that it is formed. It is also effective to provide the current supply line 3408 so as to overlap one of the gate wirings 3403a and 3403b. In this case, since the number of power supply lines can be reduced, the pixel portion can be further refined.
[0263]
[Embodiment 12] A CMOS circuit and a pixel matrix circuit manufactured by implementing the present invention can be used for various electro-optical devices (active matrix liquid crystal display, active matrix EL display, active matrix EC display). . That is, the present invention can be implemented in the manufacturing steps of all electronic devices in which these electro-optical devices are incorporated as display media.
[0264]
Such electronic devices include video cameras, digital cameras, projectors (rear or front type), head mounted displays (goggles type displays), car navigation systems, car stereos, personal computers, personal digital assistants (mobile computers, mobile phones) Or an electronic book). Examples of these are shown in FIGS. 32, 33 and 34. FIG.
[0265]
FIG. 32A shows a personal computer, which includes a main body 2001, an image input portion 2002, a display portion 2003, a keyboard 2004, and the like. The present invention can be applied to the image input unit 2002, the display unit 2003, and other signal control circuits.
[0266]
FIG. 32B shows a video camera, which includes a main body 2101, a display portion 2102, an audio input portion 2103, operation switches 2104, a battery 2105, an image receiving portion 2106, and the like. The present invention can be applied to the display portion 2102 and other signal control circuits.
[0267]
FIG. 32C illustrates a mobile computer, which includes a main body 2201, a camera unit 2202, an image receiving unit 2203, operation switches 2204, a display unit 2205, and the like. The present invention can be applied to the display portion 2205 and other signal control circuits.
[0268]
FIG. 32D shows a goggle type display, which includes a main body 2301, a display portion 2302, an arm portion 2303, and the like. The present invention can be applied to the display portion 2302 and other signal control circuits.
[0269]
FIG. 32E shows a player using a recording medium (hereinafter referred to as a recording medium) on which a program is recorded, and includes a main body 2401, a display portion 2402, a speaker portion 2403, a recording medium 2404, an operation switch 2405, and the like. This player uses a DVD (Digital Versatile Disc), CD, or the like as a recording medium, and can perform music appreciation, movie appreciation, games, and the Internet.
The present invention can be applied to the display portion 2402 and other signal control circuits.
[0270]
FIG. 32F illustrates a digital camera, which includes a main body 2501, a display portion 2502, an eyepiece portion 2503, operation switches 2504, an image receiving portion (not shown), and the like. The present invention can be applied to the display portion 2502 and other signal control circuits.
[0271]
FIG. 33A illustrates a front type projector, which includes a projection device 2601, a screen 2602, and the like. The present invention can be applied to the liquid crystal display device 2808 constituting a part of the projection device 2601 and other signal control circuits.
[0272]
FIG. 33B shows a rear projector, which includes a main body 2701, a projection device 2702, a mirror 2703, a screen 2704, and the like. The present invention can be applied to the liquid crystal display device 2808 constituting a part of the projection device 2702 and other signal control circuits.
[0273]
Note that FIG. 33C is a diagram illustrating an example of the structure of the projection devices 2601 and 2702 in FIGS. 33A and 33B. The projection devices 2601 and 2702 include a light source optical system 2801, mirrors 2802, 2804 to 2806, a dichroic mirror 2803, a prism 2807, a liquid crystal display device 2808, a phase difference plate 2809, and a projection optical system 2810. Projection optical system 2810 includes an optical system including a projection lens. Although the present embodiment shows a three-plate type example, it is not particularly limited, and for example, a single-plate type may be used. Further, the practitioner may appropriately provide an optical system such as an optical lens, a film having a polarization function, a film for adjusting a phase difference, or an IR film in the optical path indicated by an arrow in FIG. Good.
[0274]
FIG. 33D illustrates an example of the structure of the light source optical system 2801 in FIG. In this embodiment, the light source optical system 2801 includes a reflector 2811, a light source 2812, lens arrays 2813 and 2814, a polarization conversion element 2815, and a condenser lens 2816. Note that the light source optical system illustrated in FIG. 33D is an example and is not particularly limited. For example, the practitioner may appropriately provide an optical system such as an optical lens, a film having a polarization function, a film for adjusting a phase difference, or an IR film in the light source optical system.
[0275]
However, the projector shown in FIG. 33 shows a case where a transmissive electro-optical device is used, and an application example in a reflective electro-optical device and an EL display device is not shown.
[0276]
FIG. 34A shows a cellular phone, which includes a main body 2901, an audio output portion 2902, an audio input portion 2903, a display portion 2904, operation switches 2905, an antenna 2906, and the like. The present invention can be applied to the audio output unit 2902, the audio input unit 2903, the display unit 2904, and other signal control circuits.
[0277]
FIG. 34B illustrates a portable book (electronic book), which includes a main body 3001, display portions 3002 and 3003, a storage medium 3004, operation switches 3005, an antenna 3006, and the like. The present invention can be applied to the display portions 3002 and 3003 and other signal circuits.
[0278]
FIG. 34C illustrates a display, which includes a main body 3101, a support base 3102, a display portion 3103, and the like. The present invention can be applied to the display portion 3103. The display of the present invention is particularly advantageous when the screen is enlarged, and is advantageous for displays having a diagonal of 10 inches or more (particularly 30 inches or more).
[0279]
【The invention's effect】
According to the present invention, in a laser irradiation apparatus that processes coherent light into a linear beam, the light intensity on the irradiation surface of the linear beam is dispersed, and the in-plane homogeneity of the laser annealing effect by the laser beam can be improved. it can.
[Brief description of the drawings]
FIG. 1 is a configuration diagram of a beam homogenizer of the present invention.
FIG. 2 is a configuration diagram of a beam homogenizer according to the present invention.
FIG. 3 is a configuration diagram of a cylindrical lens group of the present invention.
FIG. 4 is a configuration diagram of a cylindrical lens group of the present invention.
FIG. 5 is a diagram showing a laser irradiation system of Example 1.
6 is a top view of the laser annealing apparatus of Example 1. FIG.
FIG. 7 shows a modification of the cylindrical lens group of the present invention.
FIG. 8 shows a modification of the cylindrical lens group of the present invention.
FIG. 9 is a configuration diagram of a beam homogenizer using the multi-phase lens of the present invention.
FIG. 10 is a diagram showing a difference between an optical system arrangement for producing a plane wave and an optical system arrangement for producing a spherical wave according to the present invention.
FIG. 11 is a diagram showing parameters necessary for obtaining the interference fringe pitch d of Example 3 by calculation;
FIG. 12 is a diagram showing parameters necessary for obtaining the interference fringe pitch d of Example 3 by calculation;
13 is an optical system for forming the linear laser of Example 4. FIG.
14 is a diagram illustrating a scanning method of a linear laser according to Embodiment 4. FIG.
15 is a diagram showing a state of laser irradiation on the multi-sided substrate of Example 5. FIG.
16 is a view showing a state of laser irradiation on a multi-sided substrate of Example 6. FIG.
FIG. 17 shows a manufacturing process of the AM-LCD of Example 7.
18 is a view showing a manufacturing process of an AM-LCD of Example 7. FIG.
19 is a diagram showing manufacturing steps of an AM-LCD according to Example 7. FIG.
20 is a top view of a pixel matrix circuit and a circuit arrangement according to Embodiment 7. FIG.
FIG. 21 is a top view of the CMOS circuit according to the seventh embodiment.
FIG. 22 is a diagram showing an external appearance of an AM-LCD according to Example 9.
23 is a characteristic diagram of light transmittance with respect to applied voltage of a thresholdless antiferroelectric mixed liquid crystal showing a V-shaped electro-optic response of Example 9. FIG.
FIG. 24 is a schematic view of a silicon film irradiated with a conventional linear beam.
FIG. 25 is a configuration diagram of a beam homogenizer.
FIG. 26 shows the intensity distribution of a linear beam.
FIG. 27 shows the simulation result of the intensity distribution in the longitudinal direction of the linear beam.
FIG. 28 shows the simulation result of the intensity distribution in the longitudinal direction of the linear beam.
FIG. 29 shows the simulation result of the intensity distribution in the longitudinal direction of the linear beam.
30A is a top view and FIG. 30B is a cross-sectional view of an EL display device according to Example 10;
31 is an equivalent circuit diagram of a pixel portion of an EL display device according to Example 11. FIG.
FIG. 32 is an explanatory diagram of an electronic apparatus according to a twelfth embodiment.
FIG. 33 is an explanatory diagram of a projector according to a twelfth embodiment.
FIG. 34 is an explanatory diagram of an electronic apparatus according to a twelfth embodiment.
[Explanation of symbols]
201 Laser equipment
202 Cylindrical lens group
203 Cylindrical lens group
204 Cylindrical lens
205 slit
206 Cylindrical lens group
207 Mirror
208 Cylindrical lens
210 Linear laser beam

Claims (3)

A laser generator for generating a laser beam;
A beam homogenizer,
A laser irradiation apparatus having a stage movable in one direction,
The beam homogenizer is
A first cylindrical lens group comprising (2n + 1) first cylindrical lenses;
A second cylindrical lens group composed of N (n-1) second cylindrical lenses;
A third cylindrical lens;
A third cylindrical lens group composed of (n′−1) fourth cylindrical lenses, and sequentially arranged on the optical path from the emission side of the laser device,
In the third cylindrical lens group,
The principal points of the (n′−1) fourth cylindrical lenses are shifted by an interval of d / (n′−1),
The principal planes of the (n′−1) fourth cylindrical lenses form the same plane,
Images obtained by orthogonal projection of principal points of the (n′−1) fourth cylindrical lenses on a plane perpendicular to the plane are arranged on the same straight line at an interval of d / (n′−1) (n '-1) points,
The laser beams that have passed through the (n′−1) fourth cylindrical lenses are out of phase with each other,
By synthesizing the laser beams out of phase, the intensity of interference fringes on the irradiated surface is made uniform,
The λ the wavelength of the laser beam, the focal length of one cylindrical lens constituting the third cylindrical lens group and f, and the width of one cylindrical lens constituting the first cylindrical lens group and L The d is denoted by d = λf / L,
N is a natural number, n is an integer of 3 or more, and n ′ is an integer satisfying 3 ≦ n ′ ≦ n. A laser generator for generating a laser beam;
A beam homogenizer,
A laser irradiation apparatus having a stage movable in one direction,
The beam homogenizer is
A first cylindrical lens group comprising (2n) first cylindrical lenses;
A second cylindrical lens group composed of N (n-1) second cylindrical lenses;
A third cylindrical lens;
A third cylindrical lens group composed of (n′−1) fourth cylindrical lenses;
Are sequentially arranged on the optical path from the emission side of the laser device,
In the third cylindrical lens group,
The principal points of the (n′−1) fourth cylindrical lenses are shifted by an interval of d / (n′−1),
The principal planes of the (n′−1) fourth cylindrical lenses form the same plane,
Images obtained by orthogonal projection of principal points of the (n′−1) fourth cylindrical lenses on a plane perpendicular to the plane are arranged on the same straight line at an interval of d / (n′−1) (n '-1) points,
The laser beams that have passed through the (n′−1) fourth cylindrical lenses are out of phase with each other,
By synthesizing the laser beams out of phase, the intensity of interference fringes on the irradiated surface is made uniform,
The λ the wavelength of the laser beam, the focal length of one cylindrical lens constituting the third cylindrical lens group and f, and the width of one cylindrical lens constituting the first cylindrical lens group and L The d is denoted by d = λf / L,
N is a natural number, n is an integer of 3 or more, and n ′ is an integer satisfying 3 ≦ n ′ ≦ n. In claim 1 or 2,
The laser generator is Nd: YAG laser beam or Nd: YVO ₄ A laser irradiation apparatus characterized by generating a laser beam.

Images