KR100950736B1 - Ion implanter - Google Patents

Abstract

The ion implanter of the present invention includes an ion source for generating an ion beam, an electron beam source for radiating an electron beam scanned in the Y direction to the ion source, a power source for the electron beam source, and an ion beam in the vicinity of the injection position. An ion beam monitor for measuring the ion beam current density distribution in the Y direction, and a control device. The control device increases the scanning speed of the electron beam at a position corresponding to the monitor point at which the ion beam current density measured by the ion beam monitor is relatively large, while controlling the electron-beam power supply based on the measurement data of the ion beam monitor. Ion beam current density in the Y direction measured by the ion beam monitor by reducing the scanning speed of the electron beam at a position corresponding to a monitor point where the ion beam current density measured by the ion beam monitor is relatively small. Has the function of homogenizing the distribution.

Description

Ion implanter {ION IMPLANTER}

The present invention relates to an ion implanter which implants an ion into a substrate by using a combination of injecting a ribbon-shaped ion beam into the substrate and moving the substrate in a direction intersecting with the main surface of the ion beam.

In this type of ion implanter, in order to improve the homogenization of the ion beam on the substrate, the length of the ion beam in the ribbon shape (which is also referred to as sheet or band shape and also applies to the following description) (in the Y direction in the specification) It is important to improve the homogenization of the ion beam current density distribution.

As a technique for improving the homogenization of the ion beam current density distribution in the longitudinal direction of the ribbon-shaped ion beam, for example, Patent Document 1 controls the filament current of an ion source having a plurality of filaments so that the ion beam is incident on the substrate. Techniques for improving the homogenization of the ion beam current density distribution in the vicinity of the implantation location are disclosed.

Patent document 2 discloses that an electron beam scanned in one dimension is incident into a plasma vessel of an ion source, and gas is ionized by the electron beam to generate a plasma, thereby adjusting the ion beam current density distribution of the ion beam drawn from the ion source. Disclosed is a technique for improving.

[Patent Document 1] JP-A-2000-315473 (paragraphs 0012 to 0015, Fig. 1)

[Patent Document 2] JP-A-2005-38689 (paragraphs 0006 and 0008, Fig. 1)

In the technique disclosed in Patent Document 1, even when a plurality of filaments are arranged in the longitudinal direction of the ion beam, a space is inevitably present between the filaments, so that the plasma density and the ion beam current density are inevitably lowered. Therefore, there is a limit to improving the homogenization of the ion beam current density distribution.

In the technique disclosed in Patent Document 2, even when the homogenization of the ion beam drawn out from the ion source can be outlined, the homogenization may be damaged during the progress of the ion beam. As a result, the homogenization of the ion beam current density distribution at the implantation position may not necessarily be good.

An exemplary embodiment of the present invention provides an ion implanter capable of improving the homogenization of the ion beam current density distribution in the longitudinal direction (Y direction) at the implantation location on the substrate.

According to the first aspect of the present invention, the first ion implanter of the present invention

Ribbon traveling direction of the ion beam is set as the Z direction, two directions substantially orthogonal to each other in a plane substantially orthogonal to the Z direction are set as the X and Y directions, respectively, and the dimension in the Y direction is larger than the dimension in the X direction A type ion beam is transmitted to irradiate a substrate, and is an ion implanter which performs ion implantation. Such an ion implanter,

An ion source having at least one filament for generating arc discharge in a plasma vessel into which gas is introduced, wherein the ion source generates a ribbon ion beam having a dimension in the Y direction greater than the dimension in the Y direction of the substrate;

A substrate driving device which moves the substrate in a direction intersecting with a main surface of the ion beam at an implantation position where the ion beam is to enter the substrate;

One or more electron beam sources for generating an electron beam, ionizing the gas by radiating the electron beam into a plasma vessel of the ion source to generate a plasma, and scanning the electron beam into the plasma vessel in the Y direction;

A drawing voltage for controlling the generation amount of the electron beam, at least one electron-beam power supply for supplying a scan voltage for scanning to the electron beam source;

An ion beam monitor measuring an ion beam current density distribution in the Y direction of the ion beam at a plurality of monitor points in the Y direction, at or near the injection position;

Monitor points with a relatively high ion beam current density measured by the ion beam monitor while controlling the electron-beam power source based on the measurement data of the ion beam monitor to maintain a substantially constant value of the electron beam generated from the electron beam source. Increasing the scanning speed of the electron beam at the position of the ion source corresponding to the scanning speed of the electron beam at the position of the ion source corresponding to the monitor point where the ion beam current density measured by the ion beam monitor is relatively small. And a control device having a function of homogenizing the ion beam current density distribution in the Y direction measured by the ion beam monitor by performing at least one of relatively decreasing.

In the first ion implanter, the ion beam current density distribution in the Y direction of the ion beam is measured by the ion beam monitor at or near the implantation position. The control device then controls the electron-beam power source based on the measurement data of the ion beam monitor, and controls the density of the plasma generated by the electron beam by controlling the scan rate of the electron beam in the plasma vessel of the ion source. do. Specifically, the scan rate of the electron beam at the position of the ion source corresponding to the monitor point at which the ion beam current density measured by the ion beam monitor is relatively high while maintaining the amount of the electron beam generated from the electron beam source at a substantially constant value. And relatively reducing the scanning speed of the electron beam at the position of the ion source corresponding to the monitor point where the ion beam current density measured by the ion beam monitor is relatively small, and As a result, control for homogenizing the ion beam current density distribution in the Y direction measured by the ion beam monitor can be executed. According to this structure, the homogenization of the ion beam current density distribution in the Y direction at the injection position can be improved.

According to the second and third aspects of the present invention, (a) the control device includes: a function of supplying a scan signal which is a source of a scan voltage supplied from an electron-beam power source to an electron beam source, to the electron-beam power source; Calculating an average value of the ion beam current densities of the Y-direction distribution measured by the ion beam monitor; A function of uniformly controlling the filament current flowing through the filament of the ion source such that the calculated average value is substantially equal to a predetermined specific ion beam current density; Calculating an error of the Y direction distribution, which is a difference between the ion beam current density of the Y direction distribution and the preset ion beam current density measured by the ion beam monitor; Determining a monitor point at which the calculated error is larger than a predetermined allowable error and a sign of the error at this monitor point; Determining a scan voltage corresponding to a predetermined monitor point; Based on the determined sine of the error, the scan rate of the electron beam is increased in proportion to the degree of error at the scan voltage corresponding to the monitor point where the measured ion beam current density is large, and the measured ion beam current density is At the scan voltage corresponding to the small monitor point, the scan speed of the electron beam is reduced in proportion to the extent of the error, so that at all substantially the monitor points at which the ion beam collides, the waveform of the scan signal is equal to or less than the allowable error. Shaping; (B) the electron-beam power source may include an amplifier for amplifying the scan signal supplied from the control device to generate a scan voltage; have.

In the specification, "substantially all monitor points" means that all monitor points are preferred, but some non-critical monitor points may be excluded.

According to the fourth aspect of the present invention, the ion implanter of the present invention has an advancing direction of the ion beam set as the Z direction, and two directions substantially perpendicular to each other in a plane substantially orthogonal to the Z direction are respectively the X and Y directions. And a ribbon-shaped ion beam whose dimension in the Y direction is larger than the dimension in the X direction is transmitted to irradiate the substrate, thereby performing ion implantation. The ion implanter includes an ion source having one or more filaments for generating arc discharge in a plasma vessel into which gas is introduced, and for generating a ribbon-shaped ion beam having a dimension in the Y direction greater than a dimension in the Y direction of the substrate; A substrate driving device for moving the substrate in a direction intersecting with a main surface of the ion beam at an implantation position where the ion beam is to enter the substrate; One or more electron beam sources for generating an electron beam, ionizing a gas by radiating the electron beam into a plasma vessel of the ion source to generate a plasma, and scanning the electron beam into the plasma vessel in a Y direction; At least one electron-beam power supply for supplying a drawing voltage for controlling an amount of generation of an electron beam and a scan voltage for scanning to the electron beam source; An ion beam monitor measuring an ion beam current density distribution in the Y direction of the ion beam at a plurality of monitor points in the Y direction, at or near the injection position; While controlling the electron-beam power source based on the measurement data of the ion beam monitor to maintain the scan rate of the electron beam generated by the electron beam source at a substantially constant value, the ion beam current density measured by the ion beam monitor is Relatively reducing the amount of generation of the electron beam at the location of the ion source corresponding to the relatively large monitor point, and electrons at the location of the ion source corresponding to the monitor point having a relatively small ion beam current density measured by the ion beam monitor. And a control device having a function of homogenizing the ion beam current density distribution in the Y direction measured by the ion beam monitor by performing at least one of relatively increasing the amount of generation of the beam.

In the second ion implanter, the ion beam current density distribution in the Y direction of the ion beam is measured by the ion beam monitor at or near the implantation position. Then, the control device controls the electron-beam power supply based on the measurement data of the ion beam monitor, and controls the generation amount of the electron beam from the electron beam source, so that the density of the plasma generated by the electron beam in the plasma vessel To control. Specifically, the electron at the position of the ion source corresponding to the monitor point where the ion beam current density measured by the ion beam monitor is relatively high while maintaining the Y-direction scan speed of the electron beam generated by the electron beam source at a substantially constant value. At least one of relatively reducing the amount of generation of the beam and relatively increasing the amount of generation of the electron beam at the location of the ion source corresponding to the monitor point where the ion beam current density measured by the ion beam monitor is relatively small; Thus, control to homogenize the ion beam current density distribution in the Y direction measured by the ion beam monitor can be performed. According to this structure, the homogenization of the ion beam current density distribution in the Y direction at the injection position can be improved.

According to the fifth and sixth aspects of the present invention, (a) the control device comprises: a function of supplying a drawing signal which is a source of a drawing voltage supplied from an electron-beam power source to the electron beam source; Calculating an average value of the ion beam current densities of the Y-direction distribution measured by the ion beam monitor; Uniformly controlling the filament current flowing through the filament of the ion source such that the calculated average value is substantially equal to a predetermined specific ion beam current density; Calculating an error of the Y direction distribution, which is a difference between the ion beam current density of the Y direction distribution and a preset ion beam current density measured by the ion beam monitor; Determining a monitor point at which the calculated error is larger than a predetermined allowable error and a sign of the error at this monitor point; Determining a scan voltage corresponding to a predetermined monitor point;

Based on the determined sine of the error, the draw voltage is reduced in proportion to the degree of error at the scan voltage corresponding to the monitor point with the measured ion beam current density high, and the monitor point with the measured ion beam current density small Increasing the draw voltage in proportion to the degree of error at a scan voltage corresponding to to shape the waveform of the draw signal such that the error is less than or equal to an acceptable error at substantially all monitor points at which the ion beam collides; And may store the data of the shaped outgoing signal and the data of the filament current, and (b) the electron-beam power source may comprise an amplifier for amplifying the outgoing signal supplied from the control device to generate the outgoing voltage.

The ion implanter may further comprise an analytical electromagnet disposed between the ion source and the implantation position and analyzing the momentum by bending the ion beam from the ion source in the X direction.

The ion implanter may further include an acceleration / deceleration device disposed between the analysis electromagnet and the injection position, causing the ion beam to be bent in the X direction by an electrostatic field, and accelerating or decelerating the ion beam.

The present invention described in the fifth and sixth aspects has the above-described structure. Therefore, it is possible to improve the homogenization of the ion beam current density distribution in the Y direction at the injection position on the substrate. As a result, the homogenization of ion implantation on the substrate can be improved.

In addition, plasma generation using filament and homogenization of ion beam current density distribution due to plasma density distribution control using an electron beam source are used in combination. Therefore, high-density ion implantation can be easily performed while irradiating the substrate with an ion beam having a large current.

In addition, according to the seventh aspect of the present invention, the ion implanter may further include an analysis electromagnet disposed between the ion source and the implantation position and analyzing the momentum by bending the ion beam from the ion source in the X direction. There is,

Analysis electromagnet,

A set of body parts opposed to each other in the X direction across the beam path through which the ion beam passes, and at least one set of connecting portions connecting the ends of the body parts to each other in the Z direction while avoiding the beam path, the ion A coil that generates a magnetic field that bends the beam in the X direction,

A yoke collectively surrounding the outside of the body portion of the coil,

The coil has a structure in which a notch portion is arranged in a fan-shaped stacked coil while leaving the main body portion and the connecting portion, and the stack coil has a main surface on an outer circumferential surface of the laminated insulator. Winding the lamination with a plurality of turns, stacking the lamination of the conductor sheet and the insulating sheet in which the main sheet extends along the Y direction; It is comprised by forming the insulator laminated | stacked on the outer peripheral surface of a stack.

According to the eighth aspect of the present invention, the ion implanter further comprises an analysis electromagnet disposed between the ion source and the implantation position, the analyte electromagnet analyzing the momentum by bending the ion beam from the ion source in the X direction,

Analysis electromagnet,

A set of body portions opposed to each other in the X direction across the beam path through which the ion beam passes and covering approximately half or more of one side of the ion beam in the Y direction, and the body in the Z direction while avoiding the beam path A saddle-shaped coil having a set of connecting portions connecting the ends of the portions to each other, the first coil generating a magnetic field in cooperation with the second coil to bend the ion beam in the X direction;

A set of body portions that oppose each other in the X direction across the beam path and cover approximately half or more of the other side of the ion beam in the Y direction, and connect the ends of the body portions in the Z direction while avoiding the beam path A second saddle-shaped coil having a set of connecting portions, the second coil being disposed to overlap the first coil in the Y direction, and generating a magnetic field to cooperate with the first coil to bend the ion beam in the X direction; Coils,

A yoke collectively surrounding the outside of the body portion of the first coil and the second coil,

Each of the first and second coils has a structure in which a notch portion is arranged in a fan-shaped cylindrical stack coil while leaving the main body portion and the connecting portion, and the stack coil has a main surface Y on its outer circumferential surface. Winding the lamination with a plurality of windings, stacking the lamination of the conductive sheet and the insulating sheet extending along the direction; It is comprised by forming the insulator laminated | stacked on the outer peripheral surface of a stack.

According to the ninth aspect of the present invention, the ion implanter may further include an analytical electromagnet disposed between the ion source and the implantation position and analyzing the momentum by bending the ion beam from the ion source in the X direction,

Analysis electromagnet,

A set of body parts opposed to each other in the X direction across the beam path through which the ion beam passes, and a connecting portion connecting the ends of the body parts to each other in the Z direction while avoiding the beam path, wherein the ion beam is X An internal coil that generates a main magnetic field that bends in a direction;

Saddle shape with a set of body parts outside the inner coil and opposed to each other in the X direction across the beam path and a set of connections connecting the ends of the body parts to each other in the Z direction while avoiding the beam path. At least one first external coil, the coil of which generates a sub-magnetic field that supports or corrects the main magnetic field;

Saddle having a set of body parts outside the inner coil and facing each other in the X direction across the beam path and a set of connections connecting the ends of the body part to each other in the Z direction while avoiding the beam path. At least one second outer coil, wherein the coil is shaped to overlap with the first outer coil in a Y direction and generates a sub magnetic field that supports or corrects the main magnetic field;

A yoke collectively surrounding an outer side of said body portion of said inner coil and said first and second outer coils,

Each of the inner coil, the first and the second outer coils has a structure in which notches are disposed in a fan-shaped cylindrical stack coil while leaving the main body portion and the connecting portion, and the stack coil has a main surface on an outer circumferential surface of the laminated insulator. Winding the lamination with a plurality of windings, stacking the lamination of the conductor sheet and the insulating sheet extending along the Y direction; Forming laminated insulators on the outer circumferential surface of the stack; Winding the lamination with a plurality of windings, stacking the lamination of the conductor sheet and the insulating sheet, the main surface of which extends along the Y direction, on the outer circumferential surface of the stack; It is comprised by forming the insulator laminated | stacked on the outer peripheral surface of a stack.

According to a tenth aspect of the present invention, the ion implanter may further include an analytical electromagnet disposed between the ion source and the implantation position and analyzing the momentum by bending the ion beam from the ion source in the X direction,

Analysis electromagnet,

A set of body portions opposed to each other in the X direction across the beam path through which the ion beam passes and covering approximately half or more of one side of the ion beam in the Y direction, and the body in the Z direction while avoiding the beam path A saddle-shaped coil having a set of connecting portions connecting the ends of the portions to each other, the first inner coil cooperating with the second inner coil to generate a main magnetic field to bend the ion beam in the X direction;

A set of body portions that oppose each other in the X direction across the beam path and cover approximately half or more of the other side of the ion beam in the Y direction, and connect the ends of the body portions in the Z direction while avoiding the beam path The second internal coil having a set of connecting portions, the second internal coil being disposed to overlap the first internal coil in the Y direction, and generating a main magnetic field to cooperate with the first internal coil to bend the ion beam in the X direction. and,

A set of main body portions that are outside of the first inner coil and opposed to each other in the X direction across the beam path, and a set of connections that connect the ends of the main body portions to each other in the Z direction while avoiding the beam path; One or more first external coils that are saddle shaped coils and generate a sub magnetic field that supports or corrects the main magnetic field,

A set of main body portions that are outside of the second inner coil and opposed to each other in the X direction across the beam path, and a set of connections that connect the ends of the main body portions to each other in the Z direction while avoiding the beam path; At least one second outer coil disposed to overlap with the first outer coil in a Y-direction, and generating a sub magnetic field that supports or corrects the main magnetic field;

A yoke collectively surrounding the outside of the body portion of the first and second inner coils and the first and second outer coils,

Each of the first internal coil and the first external coil has a structure in which a notch part is disposed in a fan-shaped cylindrical stack coil while leaving the main body part and the connecting part, and the stack coil is formed on an outer circumferential surface of the laminated insulator. Winding the lamination with a plurality of windings, stacking the lamination of the conductor sheet and the insulating sheet whose main surface extends along the Y direction; Forming laminated insulators on the outer circumferential surface of the stack; Winding the lamination with a plurality of windings, stacking the lamination of the conductor sheet and the insulating sheet, the main surface of which extends along the Y direction, on the outer circumferential surface of the stack; Is formed by forming an insulator laminated on the outer circumferential surface of the stack,

Each of the second internal coil and the second external coil has a structure in which a notch portion is disposed in a fan-shaped cylindrical stack coil while leaving the main body portion and the connecting portion, and the stack coil is formed on an outer circumferential surface of the laminated insulator, Winding the lamination with a plurality of windings, stacking the lamination of the conductor sheet and the insulating sheet whose main surface extends along the Y direction; Forming laminated insulators on the outer circumferential surface of the stack; Winding the lamination with a plurality of windings, stacking the lamination of the conductor sheet and the insulating sheet, the main surface of which extends along the Y direction, on the outer circumferential surface of the stack; It is comprised by forming the insulator laminated | stacked on the outer peripheral surface of a stack.

According to an eleventh aspect of the invention, the analysis electromagnet may further comprise a set of magnetic poles projecting inwardly from the yoke such that they face each other in the Y direction across the beam path.

Since the invention according to the seventh to eleventh aspects of the present invention includes the above-described analytical electromagnet, the following effects are obtained.

Each coil of the analysis electromagnet is configured such that the notch portion is arranged in the fan-shaped cylindrical stack coil as described above with the main body portion and the connecting portion left, so that the portion extends in the Y direction substantially parallel to the end of the main body portion. I am in a state to do it. In the case where the dimension in the Y direction of the main body portion increases, it is coped by correspondingly increasing the dimension in the Y direction of the connecting portion. As a result, the protruding distance of the connection portion in the incident and radial directions of the beam does not increase. According to this structure, it is possible to reduce the distance that the connecting portion of the coil protrudes from the yoke in the incident and radial directions of the beam.

Therefore, the size of the analysis electromagnet can be reduced, thereby reducing the area required for installing the analysis electromagnet. In addition, the weight of the analytical electromagnet can be reduced. In addition, the possibility of the magnetic field generated by the connection of the coils shaking the shape of the ion beam is reduced.

Accordingly, the protruding distance of the connecting portion of each coil can be reduced, and the length of the connecting portion can be shortened, thereby reducing unnecessary power consumption at the connecting portion. In addition, each coil has a structure in which a conductor sheet is stacked with an insulating sheet interposed therebetween. Thus, when compared with a multi-wound coil in which the sheathed conductor is wound multiple times, the area ratio of the conductor is high and the power consumption is correspondingly low. As a result, power consumption can be reduced.

As a result, as the analytical electromagnet becomes smaller, the size of the ion implanter can be reduced, thereby reducing the area required to install the ion implanter. In addition, the weight of the ion implanter can be reduced. In addition, as the power consumption of the analysis electromagnet is reduced, the power consumption of the ion implanter can be reduced.

The invention according to the eighth aspect of the present invention can obtain the following additional effects. That is, since the analysis electromagnet is provided with the 1st and 2nd coil, it can cope easily with the ion beam which has a large Y direction dimension.

The present invention according to the ninth aspect can obtain the following additional effects. That is, since the analysis electromagnet includes the first and second coils in addition to the internal coils, it is possible to generate a magnetic field with high homogeneity of magnetic flux density distribution in the Y direction in the beam path of the ion beam. As a result, fluctuations in the form of ion beams can be suppressed to a low level during radiation. This effect is more pronounced when the dimension in the Y direction of the ion beam is large.

The present invention according to the tenth aspect can obtain the following additional effects. That is, since the analysis electromagnet includes the first and second external coils in addition to the first and second internal coils, it is possible to easily cope with the ion beam having a large dimension in the Y direction, and to the beam path of the ion beam, It is also possible to generate a magnetic field with high homogeneity of magnetic flux density distribution in the direction. As a result, fluctuations in the form of ion beams can be suppressed to a low level during radiation. This effect is more pronounced when the dimension in the Y direction of the ion beam is large.

The present invention according to the eleventh aspect can obtain the following additional effects. That is, since the analytical electromagnet further includes a stimulus, the magnetic field can be easily concentrated in the gap between the stimuli, thereby easily generating a high magnetic flux density magnetic field in the beam path.

According to a twelfth aspect of the present invention, an ion implanter is disposed between an analytical electromagnet which analyzes a momentum by bending an ion beam from the ion source in the X direction, and distributes the ion beam to an electrostatic field. It may further include an acceleration / deceleration device to be bent in the X direction by the acceleration or deceleration of the ion beam,

The acceleration / deceleration device includes first to third electrodes arranged in the order of the first electrode, the second electrode, and the third electrode in the ion beam traveling direction starting from the upstream side, and the ion beam is connected to the first electrode. Accelerate or decelerate at two stages between the second electrode and between the second and third electrodes,

The second electrode is constituted by two electrode members which are opposed to each other in the X direction across the path of the ion beam and are applied with different potentials to deflect the ion beam in the X direction, the third electrode being specified after deflection. It is arranged along the trajectory of the ion beam with specific energy.

The invention according to the twelfth aspect has the acceleration / deceleration device described above, and thus the following additional effects are obtained.

That is, in the acceleration / deceleration apparatus, the ion beam can be deflected by the part of the second electrode which is configured to be separated into two electrode members, whereby the effect of energy distribution can be obtained. Due to the presence of the third electrode, an ion beam having a specific energy can be induced efficiently, and ions and neutral particles other than the ion beam can be efficiently blocked by the third electrode. Therefore, energy pollution can be suppressed efficiently. In particular, it is empirically known that in the deceleration mode, neutral particles easily occur by charge conversion during deceleration of the ion beam between the first electrode and the second electrode. Although many neutral particles occur, these particles travel in a straight line and impinge on the electrode and are blocked. Thus, neutral particles can be effectively removed from the acceleration / deceleration device.

In addition, the ion beam may be accelerated in two stages and may be deflected before acceleration in a dual subsequent stage. Therefore, deflection becomes easy. In addition, electrons generated by the collision of unwanted ions are bent by the second electrode to prevent electrons from reaching the first electrode. Therefore, the energy of X-rays generated by the collision of electrons may be lowered.

Other features and advantages of the invention will be apparent from the following detailed description, the accompanying drawings and the claims.

(1) About all ion implanters

1 is a schematic plan view showing an embodiment of the ion implanter of the present invention. In the specification and the drawings, the advancing direction of the ion beam 50 is always set in the Z direction, and two directions substantially orthogonal to each other in a plane substantially perpendicular to the Z direction are set as the X direction and the Y direction, respectively. For example, the X and Z directions are horizontal and the Y direction is vertical. The Y direction is a constant direction, but the X direction is not an absolute direction and changes depending on the position of the ion beam 50 on the path (see, eg, FIG. 1). In this specification, the case where the ion which comprises the ion beam 50 is positive ion is demonstrated as an example.

The ion implanter is an apparatus for ion implantation by irradiating the substrate 60 with the ribbon ion beam 50, the ion source 100 generating the ribbon ion beam 50; The ion beam 50 from the ion source 100 is bent in the X direction to analyze the momentum (e.g., mass spectrometry, the same applies to the following), and focus the focus 56 of the ion beam 50 at a predetermined momentum; An analysis electromagnet 200 which is a focal point in the X direction and is applied equally below); At the implantation position where the ion beam passing through the analysis electromagnet 200 is incident on the substrate 60, the direction in which the substrate 60 intersects with the main surface 52 of the ion beam 50 (see FIGS. 2 and 3). The board | substrate drive apparatus 500 which moves to this is provided. For example, the movement is a reciprocating linear movement. The substrate driving apparatus 500 includes a holder 502 for holding the substrate 60.

The path of the ion beam 50 from the ion source 100 to the substrate 60 is in a vacuum vessel (not shown) maintained in a vacuum atmosphere.

In the specification, "main surface" means an end face of a ribbon-like or sheet-like member (e.g., ion beam 50, insulating sheets 266, 267 and conductor sheets 268, 269, described later). Rather, it means the broad side of the member. "Downstream" or "upstream side" means the downstream side or the upstream side in the advancing direction Z of the ion beam 50. The ion beam 50 generated from the ion source 100 and the ion beam 50 derived from the analysis electromagnet 200 are different in concept. In other words, the former is the ion beam before the momentum analysis and the latter is the ion beam after the momentum analysis. The difference between the ion beams is evident. Therefore, in the specification, the ion beams are not distinguished from each other, and both are represented as the ion beams 50.

The ion beam 50 generated from the ion source 100 and delivered to the substrate 60 is, for example, as shown in FIG. 2, for example, the dimension Wy in the Y direction is larger than the dimension Wx in the X direction, that is, It has a ribbon shape where Wy> Wx. Although the ion beam 50 has a ribbon shape, this does not mean that the dimension Wx in the X direction is as thin as paper or cloth. For example, although depending on the dimensions of the substrate 60, the dimension Wx in the X direction of the ion beam 50 is about 30 to 80 mm, and the dimension Wy in the Y direction is about 300 to 500 mm. The plane along which the ion beam 50 is larger, the plane along the YZ plane, is the major plane 52.

The ion source 100 generates a ribbon ion beam 50 in which the dimension Wy in the Y direction is larger than the thickness Ty in the Y direction of the substrate 60, as in the example shown in FIG. 3. For example, when the thickness Ty is 300 to 400 mm, the dimension Wy in the Y direction is about 400 to 500 mm. Because of such a dimensional relationship and the above-described movement of the substrate 60, the ion implantation can be performed by irradiating the substrate 60 with the ion beam 50.

For example, the substrate 60 is a semiconductor substrate, a glass substrate, or another substrate. The planar shape of the substrate is circular or rectangular in shape.

Near the focal point 56 of the ion beam 50 radiated from the analysis electromagnet 200, a slit 70 is arranged to analyze the momentum of the ion beam 50 in cooperation with the analysis electromagnet 200. The analysis slit 70 has a slit 72 extending substantially parallel to the Y direction. The reason for arranging the analysis slit 70 near the focus 56 of the ion beam 50 is that both the transfer efficiency and the momentum analysis capability of the ion beam 50 are improved.

The analysis electromagnet 200, the analysis slit 70, and the acceleration / deceleration device to be described later may be disposed as necessary.

As will be described in detail below, a plurality of electron beam sources Gn are disposed in the ion source 100 (specifically, the plasma vessel 118 constituting the ion source). A drawing voltage for controlling the generation amount of the electron beam and a scan voltage for scanning in the Y direction are supplied from the corresponding electron-beam power supply 114 to each electron beam source Gn. According to an embodiment, the number of electron beam sources Gn and electron-beam power sources 114 is two. The number is not limited to two. Each number may be one or more than two. That is, their number is all one or more arbitrary.

An ion beam monitor 80 for measuring the ion beam current density distribution in the Y direction of the ion beam 50 includes a plurality of monitors in the Y direction at or near the injection position where the ion beam 50 is incident on the substrate 60. Is placed on the point. Measurement data D ₁ representing the beam current density distribution is output from the ion beam monitor 80 and provided to the control device 90.

For example, as in the example shown in FIG. 1, the ion beam monitor 80 may be disposed near the back side (ie, downstream) of the implantation position. Alternatively, the monitor may be arranged near the front side (ie upstream side) of the injection position or configured to be able to move to the injection position. The ion beam monitor 80, the substrate 60 and the holder 502 need to be disposed so as not to interfere with each other. When the ion beam monitor 80 is placed near the rear side of the implantation position during the measurement, the substrate 60 and the holder 502 can be moved to a position that does not cause interference to the measurement. When the ion beam monitor 80 is disposed near the front side of the injection position during the measurement, the ion beam monitor 80 can be moved to a position that does not cause interference to the measurement.

The ion implanter also includes a control device 90 that controls the electron beam power source 114 based on the measurement data D ₁ provided from the ion beam monitor 80. In this embodiment, the control device 90 can also control the filament current If, which will be described below.

(2) Ion Source 100, Electron Beam Source Gn, etc. and Their Control

As shown in FIG. 4, the ion source 100 includes a gas (also including steam) 120 for generating a plasma, through a gas introduction port 119, and one or more filaments 122 (as seen in FIG. 4). In the embodiment, three) are disposed within the plasma vessel 118 having, for example, a cuboid shape, and an arc discharge is generated between the filament 122 and the plasma vessel 118 which also functions as an anode, so that the gas 120 is ionized. As a result, the plasma 124 is generated, and the ribbon type ion beam 50 as described above is drawn out of the plasma 124 by the extraction electrode system 126.

Gas 120 is a gas containing the desired elements (eg, dopants such as B, P, and As). Specific examples of the gas include a gas containing a source gas such as BF ₃ , PH ₃ , AsH ₃ or B ₂ H ₆ .

If necessary, a plurality of gas introduction ports 119 may be disposed in the Y direction. According to this structure, the gas concentration distribution can be easily homogeneous in the plasma container 118, thereby promoting the homogenization of the plasma density distribution.

The lead electrode system 126 has one or more electrodes (three in the example shown). These electrodes have ion extracting holes 128 at respective positions. The shape, arrangement, and the like of the ion extraction holes 128 of the extraction electrode system 126 are appropriately determined according to the selected shape of the extracted ion beam 50. As in the example shown in FIG. 5, the ion extracting holes 128 may be a plurality of small holes arranged in the Y direction or slits extending in the Y direction. Depending on the dimension W _X of the ion beam 50 in the X direction, a plurality of rows (eg, two or three columns) of a plurality of such ion extracting holes 128 may be arranged in the X direction.

The number of filaments 122 is any number of one or more. It may be desirable to arrange a plurality of filaments in the Y direction to produce an ion beam 50 having a large Y direction dimension W _Y and which is highly homogeneous.

The filament 122 may have a U shape as in the examples shown in FIGS. 4 and 5, or may have a straight shape extending in the Y direction as in the example shown in FIG. 6. The filaments may have other shapes.

The U-shaped filament 122 may be bent in the YZ plane as shown in FIG. 4, or may be bent in the XZ plane as shown in FIG. 5.

As shown in FIG. 4, each filament 122 is heated to receive a filament current If from a filament power source 134 of variable voltage and emit hot electrons. A DC arc power source 136 for generating arc discharge is connected between one end of each filament 122 and the plasma vessel 118. In this embodiment, the filament power source 134 may change (increase or decrease) the filament current If in accordance with the filament current control signal Sf provided from the control device 90.

In this example, one filament power source 134 is disposed in each filament 122. However, it is not necessary to arrange the plurality of filament power sources 134 individually. The filament power source may be assembled into one unit or may be configured as one filament power source that allows the filament current If to flow independently through each filament 122. In this example, the arc power source 136 is shared by all filaments 122. Alternatively, one arc power source may be disposed for each filament 122. In the case of shared power supplies, the structure can be simplified.

Magnets for forming a multi-pole magnetic field (multi-cusp magnetic fields) used to generate and maintain the plasma 124 may be arranged around the plasma vessel 118. Ion sources having the structures described above are also called bucket type ion sources (or multipolar magnetic field ion sources).

Between the plurality of filaments 122, specifically, the electron beam source Gn is disposed at an intermediate point, respectively. In this embodiment, as shown in FIG. 7, each electron beam source Gn includes a filament 140 for emitting electrons (hot electrons), an anode 144 for drawing electrons as an electron beam 138, and the like. An extraction electrode 142 disposed between these two components 140 and 144 to control the generation amount of the electron beam without changing the energy of the electron beam 138, and an electron beam 138 to be drawn out Is provided with a pair of scanning electrodes 146 for scanning in the Y direction.

According to this configuration, each electron beam source Gn generates an electron beam 138 and radiates the electron beam into the plasma vessel 118 of the ion source 100, thereby allowing the gas ( 120 is ionized to produce a plasma 124. In addition, the electron beam 138 may be scanned in the Y direction one-dimensionally in the ion source 100 (specifically, the plasma vessel 118). Examples of scan trajectories are shown in FIGS. 5 and 6. In short, the electron beam source Gn is used to calibrate the density distribution of the plasma 124 generated by the filament 122. In this embodiment, two electron beam sources Gn are provided. However, the number of electron beam sources is not limited to two. The number may be one or a plurality of other than two. That is, the number can be any number of one or more.

In the example shown in FIG. 7, each electron beam power source 114 includes a filament power source 150 that heats the filament 140, and a filament 140 and an extraction electrode (eg, to control the amount of the electron beam 138). A drawing power supply 152 for applying a DC drawing voltage Ve between the 142, an energy control power supply 154 for applying a DC anode voltage Va between the filament 140 and the anode 144, and a pair An amplifier 156 is applied between the scan electrodes 146 to apply a scan voltage Vy for the Y-direction scanning. In this embodiment, the filament power source 150 is a DC power source. Alternatively, the filament power source may be an AC power source.

For example, the control device 90 functions to provide a scan signal Sy which is a source of the scan voltage Vy, and the amplifier 156 amplifies (voltage amplifies) the scan signal Sy provided from the control device 90. To generate (output) the scan voltage Vy. In this example, the scan voltage Vy varies in the ± direction depending on the potential of the anode 144. According to this configuration, the electron beam power supply 114 may supply a drawing voltage Ve for controlling the amount of the electron beam 138, a scan voltage Vy for Y-direction scanning, and the like to the electron beam source Gn. Can be.

In short, the energy of the electron beam 138 withdrawn from each electron beam source Gn is determined at the level of the anode voltage Va, resulting in Va [eV]. The energy of the electron beam 138 is set at a level capable of ionizing the gas 120 by electron collision in the plasma vessel 118. For example, if gas 120 is a gas of the type described above, its energy may be set to about 500 eV to 3 keV, specifically about 1 keV.

The ion beam monitor 80 measures the Y-direction ion beam current density distribution of the ion beam 50 at a plurality of points arranged in the Y-direction. For example, as in the example shown in FIG. 9, the ion beam monitor 80 includes a plurality of beam current measuring devices 82 (for example, Faraday cups) arranged in the Y direction. The arrangement length of the plurality of beam current measuring devices 82 may be slightly larger than the Y direction dimension W _Y of the ion beam 50. According to this structure, the whole ion beam 50 in a Y direction can be measured. The beam current measuring devices 82 correspond to the monitor points, respectively. 9 is a schematic view. The number, shape, arrangement, and the like of the beam current measuring device 82 are not limited to those shown in FIG.

For example, each beam current measuring device 82 may be formed in a rectangular shape extending in the X direction instead of the circular shape as in the example shown in FIG. In this case, each beam current measuring device 82 has the X direction dimension larger than the X direction dimension W _X of the ion beam 50 incident on the device, so that the device is totally ion beam 50 in the X direction. It can be configured to accept. According to such a structure, the influence by the X direction ion beam current density distribution of the ion beam 50 can be eliminated. In other words, the average ion beam current density in the X direction can be measured. As described above, the substrate 60 moves along the X direction (not limited to the movement parallel to the X direction). When the beam current measuring device 82 is configured as described above, the ion beam current density distribution of the ion beam 50 can be measured in a state more similar to the actual ion implantation into the substrate 60.

The ion beam monitor 80 may be configured such that one beam current measuring device 82 is moved in the Y direction by a moving mechanism.

In this specification, the monitor point is not a mathematical point that does not have a predetermined area, but is a small measuring place having a predetermined area in which the _Y -direction dimension is sufficiently small compared to the _Y -direction dimension W _Y of the ion beam 50.

The area of each monitor point is already known. Thus, the measured value of the beam current of the ion beam 50 at each monitor point is substantially the same as the measured value of the beam current density at the monitor point. The beam current density at the monitor point can be obtained by dividing the beam current obtained at the monitor point by the area.

FIG. 8 shows a simplified view of the system shown in FIG. 1, from the ion source 100 to the ion beam monitor 80. Reference numeral 170 denotes the entire ion beam transport system. The Y-direction of the ion source 100 may be set substantially parallel to the Y-direction of the ion beam monitor 80. However, the ion beam transport system 170 is not always linear (see FIG. 1), and the X-direction of the ion source 100 and the X-direction of the ion beam monitor 80 are always as shown in FIG. 8. It is not parallel to each other. This does not cause a problem.

In an embodiment, the control device 90 consists of a computer having a CPU, a storage device, an input AD converter, an output DA converter and the like. The control device 90 has a function of performing one of the control A and the control B, which will be described later, and does not perform both the control A and the control B at the same time.

(A) Scanning speed control of the electron beam

In this case, the electron-beam power source based on the measurement data D ₁ of the ion beam monitor 80 to keep the amount of the ion beam 138 generated at the electron beam source Gn at a substantially constant value. While controlling the 114, (a) the scanning speed of the electron beam 138 at a position in the ion source corresponding to the monitor point where the ion beam current density measured by the ion beam monitor 80 is relatively large is relatively high. By performing both the raising and (b) relatively lowering the scanning speed of the electron beam 138 at a position in the ion source corresponding to the monitor point where the measured ion beam current density is relatively small, the control device 90 The ion beam monitor 80 has a function of homogenizing the ion beam current density distribution in the Y-direction measured.

(B) the amount of electron beam control

In this case, electrons are based on the measurement data D ₁ of the ion beam monitor 80 to maintain the Y-direction scanning speed of the ion beam 138 generated at the electron beam source Gn at a substantially constant value. The amount of generation of the electron beam 138 at a position in the ion source that controls the beam power source 114 and at the same time corresponds to the monitor point where the ion beam current density measured with the ion beam monitor 80 is relatively large. , And (b) relatively increasing the amount of generation of the electron beam 138 at a position in the ion source corresponding to the monitor point where the measured ion beam current density is relatively small, thereby controlling the controller 90 ) Has the function of homogenizing the ion beam current density distribution in the Y-direction measured by the ion beam monitor 80.

In any of the above-mentioned control (A) and control (B), the control device 90 may perform at least one of the above-mentioned control (a) and control (b). However, since the homogenization control of the distribution of the ion beam current density is faster, it is preferable that the control device performs both of the above controls. The term "function" can be used interchangeably with "means". This will apply equally to other functions described later.

Specific examples of the control A and the control B will be described later.

(A) Scanning speed control of the electron beam

In this case, the power supply shown in FIG. 7 is used as the electron beam power supply 114. In this example, the withdrawal voltage Ve output from the withdrawal power supply 152 is set constant, and the amount of the ion beam generated at the electron beam supply source Gn is set constant. In addition, the anode voltage Va output from the energy control power supply 154 is preferably set constant, and the energy of the electron beam 138 is set constant. In an embodiment, these values are set constant. 10 to 12 show a flow chart of the control performed using the control device 90 in this case.

Prior to this control, the monitor point Py on the ion beam monitor 80, the electron beam source Gn which shares the increase / decrease in the ion beam current density at this monitor point Py, and the electron beam source ( The correspondence between the scanning voltages Vy supplied to Gn is checked in advance and stored in the control device 90. In the case of using only one electron beam source Gn, the electron beam source Gn is uniquely defined, so there is no need to check and store the electron beam source Gn. In addition, the electron beam supply Gn does not need to be included in the correspondence relationship mentioned later.

Focusing attention on any monitor point Py on the ion beam monitor 80, the correspondence relationship indicates that the electron beam source Gn is increasing / decreasing the ion beam current density at the monitor point Py. And a value of the scan voltage Vy supplied to the electron beam source Gn, and this correspondence can be expressed by Equation 1 below. Subscripts i, j, and k represent predetermined positions and are integers. For example, the correspondence can be determined by checking the electron beam source Gn, the scan voltage Vy, and the monitor point Py at which the ion beam current density is increased or decreased at a predetermined voltage. The correspondence relationship is determined only by the configuration of the ion implanter. Thus, once determined, it does not change unless the configuration of the ion implanter is changed. The data representing the corresponding relationship may be stored in the control device 90 (specifically, the storage device).

[Equation 1]

Py _i ↔ (Gn _j , Vy _k )

The following procedure is described with reference to FIG. 10 and the like. The preferred ion beam current density Iset of the ion beam 50 striking the substrate 60 and the tolerance ε of this density are set in the control device 90 (step 900). The set ion beam current density Iset is referred to as the preset ion beam current density. The tolerance ε represents the degree to which the actual ion beam current density, specifically the ion beam current density Imon measured by the ion beam monitor 80, can deviate from the preset ion beam current density Iset.

Then, the filament condition is approximately set (step 901). This means that upon generation of the plasma 124 the ion beam 50 is withdrawn from the ion source 100 without using the electron beam source Gn and using only the filament 122, the ion beam monitor 80 The ion beam current density (Imon) measured by) is approximately set manually. Specifically, the filament power source 134 is adjusted, and the filament current If, which will flow through the filament 122 of the ion source 100, is set approximately. At this time, the adjustment of the arc current supplied from the arc power source 136 may be additionally performed. Usually, if the configuration of the ion source 100 or the ion implanter does not change, this rough setup only needs to be performed once.

If the coarse setting of the filament conditions is performed more precisely, subsequent control (eg, control after step 905) can be completed more quickly. This may also apply to the case of the example shown in FIG. 16.

For example, it is preferable to make a coarse setting so that the measured ion beam current density Imon at all the monitor points Py is similar to the preset ion beam current density Iset and the distribution is homogenized to some extent. 13 and 14A schematically show an example in which this setting is performed. FIG. 13 shows an example in which the measured ion beam current density Imon is set slightly smaller than the preset ion beam current density Iset, and FIG. 14A shows that the measured ion beam current density Imon is the preset ion beam. An example is shown that is set slightly larger than the current density (Iset). Any of these settings may be employed.

As shown in FIG. 13, the peak position of the measured ion beam current density Imon corresponds almost to the position of the filament 1222. In the figure, AG ₁ represents an area affected by one electron beam source Gn, and AG ₂ represents an area affected by another electron beam source Gn. However, this drawing is only a schematic drawing.

Subsequently, the control device 90 supplies the scan signal Sy having a predetermined initial waveform to the electron beam power supply 114 (specifically, the amplifier 156 of the electron beam power supply) to have the scan voltage having the same waveform. (Vy) is output (step 902). For example, the initial waveform is a triangular waveform. The frequency is for example 10 kHz. The frequency is not limited to this.

The electron beam source Gn generates an electron beam 138 which is scanned with an initial waveform in the Y-direction. Using this electron beam source and filament 122, plasma 124 is generated at ion source 100 and ion beam 50 is extracted (step 903). The ion beam monitor 80 receives the ion beam 50 and measures the ion beam current density Imon (step 904). An example of this is shown in FIGS. 13 and 14A. Hereinafter, the example of FIG. 14A will be described.

In addition, the control device 90 performs subsequent processes such as computation. Based on the measurement data D ₁ supplied from the ion beam monitor 80, the average value Iave of the ion beam current density Imon in the Y-direction distribution measured by the ion beam monitor 80 is calculated. (Step 905). This is a predetermined value.

Then, the average value Iave is compared with the preset beam current density Iset, and it is determined whether the two values are substantially equal to each other (step 905). If the two values are substantially equal to each other, the process proceeds to step 908, if not the process proceeds to step 907. The expression "substantially the same" means that the values are equal to each other or within a certain small error range. This expression can be used interchangeably with the expression "almost the same".

Step 907 is a subroutine of filament current control. 11 shows the contents of the subroutine. In this subroutine, first, it is determined whether the average value Iave is a large value compared to the preset ion beam current density Iset (step 920). If larger, the process proceeds to step 921, otherwise the process proceeds to step 922.

In step 921, the filament power source 134 is controlled by the filament current control signal Sf supplied from the control device 90, and the filament current If to be flowed through all the filaments of the ion source 100 is It is uniformly reduced by a predetermined amount (in other words, equally or by the same amount, applied equally below). In step 922, in contrast to the above, the filament current If to be flowed through all the filaments 122 is increased uniformly by a predetermined amount. For example, the predetermined amount is about 1% to 2% of the filament current If at the timing at which the rough setting of the filament state is finished (step 901). In the case where the predetermined amount is large, control is performed rapidly, but the control is likely not to converge. In contrast, when the predetermined amount is small, the control is slow, but the possibility is eliminated. Therefore, the predetermined amount can be determined in consideration of both cases.

However, the control of steps 900 to 911 including the filament current control subroutine (step 907) and the electron beam scanning speed control subroutine (step 910) described below is performed in real time in the ion beam implantation process on the substrate 60. It is not performed, for example, at an appropriate timing before processing of the substrate 60 at the time of interruption. Control speed is hardly an issue. Therefore, control can be performed in which speed is not important and stability and reliability are important. For example, control may require time in minutes. This may also apply to the control shown in FIGS. 16 and 17.

After the filament current control subroutine in step 907, the process returns to step 905 and the above control is repeated until the determination in step 906 is YES. As a result, the average value Iave becomes almost equal to the preset ion beam current density Iset. 14B shows a schematic example of this state. The process then proceeds to step 908.

In step 908, the error Ierr in the Y-direction distribution, which is the difference between the measured ion beam current density Imon in the Y-direction distribution and the preset ion beam current density Iset, is expressed in the following equation, for example. Is calculated accordingly.

[Equation 2]

Ierr = Imon-Iset

Then, it is determined whether or not the magnitude (absolute value) Ierr | of the error at every monitor point Py where the ion beam 50 collides is less than the tolerance error [epsilon]. If there is one point whose magnitude is less than the tolerance, the process proceeds to step 910. Otherwise, the process proceeds to step 911.

As in the embodiment, the determination is preferably performed for all monitor points Py where the ion beam 50 impinges. However, you may omit the decision for some non-critical monitor points. It is not necessary to make a determination for the monitor point Py at which the ion beam 50 does not collide. That is, the determination may be performed for virtually all monitor points Py at which the ion beam 50 collides.

Step 910 is an electron beam scan rate control subroutine. 12 shows the contents of the subroutine. In the subroutine, first the monitor point Py whose magnitude of error | Ierr | is greater than the allowable error [epsilon] is determined (in other words, checked and applied the same below), and at the monitor point Py where the error is large A signal of an error Ierr is determined (step 930). As can be seen from Equation 2, in the embodiment, the measured ion beam current density Imon is greater than the preset ion beam current density Iset, and the measured ion beam current density Imon Is negative when the ion beam current density Iset is smaller than the preset ion beam current density Iset. See also FIG. 14B. The number of monitor points Py determined as described above is generally large at the initial stage of control and decreases as the control of steps 905 to 908 further proceeds.

Subsequently, the electron beam source Gn corresponding to the determined monitor point Py and the scanning voltage Vy of the point are determined (step 931). This step can be performed using the above-described correspondence (see Equation 1 and its description). However, in the case of using only one electron beam source Gn, the electron beam source Gn is uniquely determined, and it becomes unnecessary to determine the electron beam source Gn.

Subsequently, the waveform of the scanning signal Sy is such that the scanning speed of the electron beam 138 at the timing at which the error Ierr is the scanning voltage Vy corresponding to the positive monitor point Py is the magnitude of the error. So that the scanning speed of the electron beam 138 at a timing where the error Ierr is the scan voltage Vy corresponding to the negative monitor point Py is reduced in proportion to the magnitude of the error Ierr. It is formed (step 932). As a result, the waveform of the scan signal Sy changes from the initial triangular waveform to a slightly distorted waveform. In short, a waveform is obtained in which the slope at the point where the scanning speed is increased or decreased is increased or decreased from the initial waveform or the triangular waveform.

In order to perform control more precisely, it is preferable to set the scanning speed between two points of different scanning speeds to the scanning speed obtained by interpolating the scanning speeds of two points.

In the case where the scanning speed of the electron beam 138 is increased, the generation of the plasma 124 is reduced (rare) due to the electron beam 138 at the point where the speed is increased, and the ion beam drawn from that point ( Beam current density of 50) is increased. In the case where the scanning speed of the electron beam 138 is increased, the generation of the plasma 124 is increased due to the electron beam 138 at the point where the speed is increased, and the ion beam drawn out from the point ( Beam current density of 50) is increased.

Increasing the scanning speed of the electron beam 138 means that the time-deviation ratio dSy / dt of the scanning signal Sy increases, and thus the time-deviation ratio dVy / dt of the scanning voltage Vy is increased. The decrease in scanning speed means that the time-deviation rate (dSy / dt) and thus the time-deviation rate (dVy / dt) is reduced.

The proportional constant when the scanning speed of the electron beam 138 is increased or decreased in proportion to the magnitude of the error | Ierr | may be appropriately determined. If the proportional constant is increased, control is performed rapidly, but there is a high possibility that the control will not converge. Conversely, if the proportional constant is reduced, control is slowed down, but this possibility is eliminated. Therefore, the proportional constant can be determined considering both cases.

Subsequently, by using the scan signal Sy in which the above-described waveform is formed, the electron beam 138 generated from the electron beam source Gn is scanned (step 933). That is, the electron beam 138 is scanned by using the scan voltage Vy obtained by amplifying the scan signal Sy whose waveform is formed by the amplifier 156. As a result, the error Ierr is reduced and the number of monitor points Py in which the error is larger than the allowable error? Is reduced. However, there may occur a case where the average voltage Iave of the ion beam current density Imon is changed depending on the waveform shape. 14C shows a schematic example of this state.

Thus, after the electron beam scan rate control subroutine of step 910, the process returns to step 905. The above control is repeated until the determination in step 909 is YES. As a result, at all (almost all) monitor points Py where the ion beams 50 collide, the magnitude of the error | Ierr | becomes less than the allowable error ε, and the average value Iave is a preset ion beam current density. (Iset) is almost the same (see step 906). 14D shows a schematic example of this state.

If the determination in step 909 is YES, the data of the waveform-formed scan signal Sy, the data of the filament current If, and other data as necessary, are stored in the control device 90 (specifically, the storage device). . As a result, the control to uniformize the Y-direction ion beam current density distribution by using the control device 90 ends.

After the equalization control is finished, by using the stored data as necessary, the ion beam 50 is drawn out from the ion source 100, and ion implantation is performed on the substrate 60.

As described above, according to the ion implanter, the uniformity of the ion beam current density distribution in the Y direction at the implantation position on the substrate 60 can be improved. As a result, the uniformity of ion implantation on the substrate 60 can be improved.

Furthermore, plasma generation using the filament 122 and uniformization of the ion beam current density distribution due to plasma density distribution control using the electron beam power source Gn are used in combination. Therefore, ion implantation can be easily performed by irradiating the substrate 60 with the ion beam 50 having a large current and high uniformity. This may also apply to the following embodiments.

(B) the amount of electron beam control

An example of this case will be described with reference to Figs. 15 to 17 mainly. In these figures, parts that are the same as or correspond to the control part of (A) are indicated by the same reference numerals. In the following description, the difference with the control of said (A) is emphasized.

In this case, like the electron beam power source 114, the power source shown in Fig. 15 is used. The electron beam power source 114 applies an extraction voltage Ve for controlling the generation amount of the electron beam 138 between the filament 140 and the extraction electrode 142 instead of the DC extraction power source 152. Has In the case of this embodiment, the control device 90 has a function of supplying a drawing signal Se which is a circular shape of the drawing voltage Ve. The amplifier 162 amplifies (voltage amplifies) the drawing signal Se supplied from the control device 90 to generate (output) the drawing voltage Ve. Instead of the amplifier 156, the electron beam power supply has a scan power supply 166 that simply outputs a triangular scan voltage Vy.

In this embodiment, that is, the waveform and magnitude of the scanning voltage Vy are set constant, and the scanning speed of the electron beam 138 generated from the electron beam source Gn is set constant. In addition, the anode voltage Va output from the energy control power supply 154 is set constant, and the energy of the electron beam 138 is set constant. Thus, in this embodiment, the values are set constant. For example, the frequency of the scan voltage Vy is 10 kHz. The frequency is not limited to this value.

16 and 17 show flowcharts of the control performed using the control device 90 in this case. In FIG. 16, step 902 shown in FIG. 10 is replaced by step 912 and step 910 is replaced by step 913. Since the content of the filament current control subroutine (step 907) is the same as that shown in FIG. 1, reference is made to FIG.

In step 912, the control device 90 supplies the scanning signal Sy having the initial waveform to the electron beam power supply 114 (specifically, its amplifier 162), and the drawing voltage having the same waveform. Outputs (Ve). For example, the initial waveform is a DC voltage with a constant voltage.

Step 913 is a quantity control subroutine of the electron beam. 17 shows the contents of the subroutine. Steps 930 and 931 are the same as those shown in FIG. 12, and thus redundant descriptions are omitted.

In step 934 subsequent to step 931, the waveform of the drawing signal Se is such that the drawing voltage Ve is at the timing when the error Ierr is the scan voltage Vy corresponding to the positive monitor point Py. It decreases in proportion to Ierr and is shaped such that the withdrawal voltage Ve increases in proportion to the size of the error | Ierr | at a timing where the error Ierr is the scan voltage Vy corresponding to the negative monitor point Py. As a result, the waveform of the outgoing signal Se is changed from the initial constant value to a slightly distorted waveform. In summary, a waveform is obtained in which the voltage value at a position where the amount of the electron beam increases or decreases increases or decreases from an initial constant wave.

When the extraction signal Se and the extraction voltage Ve increase, the amount of the electron beam at the position where the signal increases increases, and the generation of the plasma 124 increases due to the electron beam 138 at that position. (Thick), the beam current density of the ion beam 50 drawn therefrom increases. When the extraction signal Se and the extraction voltage Ve decrease, the amount of electron beam at the location where the signal decreases decreases, and the generation of the plasma 124 is reduced due to the electron beam 138 at that location. (Thinner), the beam current density of the ion beam 50 drawn therefrom decreases.

The proportionality constant when the draw voltage Ve increases or decreases in proportion to the error size | Ierr | may be appropriately determined. When the proportional constant increases, control is executed quickly, but the control is likely not to converge. In contrast, when the proportional constant decreases, the control is slow, but such a possibility is excluded. Therefore, the proportional constant can be determined by considering both cases.

Thereafter, the electron beam 138 is generated from the electron beam source 138 by using the drawing signal Se shaped into the waveform as described above (step 935). As a result, the error Ierr is reduced and the number of monitor points where the error is larger than the allowable error [epsilon] is reduced. In this case, the average value Iave of the measured ion beam current density Imon may change depending on the waveform shaping. A schematic example of this condition is the same as that shown in FIG. 14C.

Thus, after the electron beam quantity control subroutine of step 913, the process returns to step 905. The above control is repeated until the determination at step 909 is YES. As a result, for all (substantially all) monitor points Py with which the ion beam 50 collides, the magnitude | Ierr | of the error is equal to or less than the allowable error [epsilon], and the average value Iave is previously It is substantially equal to the set ion beam current density Iset (see step 906). A schematic example of this condition is the same as that shown in FIG. 14D.

If the determination in step 909 is YES, the data of the waveform-shaped outgoing signal Se and the data of the filament current If, and other data as necessary, are stored in the control device 90 (specifically, the storage device). (Step 911). As a result, the control for equalizing the ion beam current density distribution in the Y direction is terminated by using the control device 90.

In addition, according to the present embodiment, it is possible to improve the homogeneity of the ion beam current density distribution in the Y direction at the injection position on the substrate 60. As a result, the homogeneity of ion implantation on the substrate 60 can be improved.

As in the example shown in FIG. 18, alternatively, the electron beam source Gn may be embedded in a cylinder 172 that is individually exhausted from the interior of the plasma vessel 118, and the electron beam source Gn may be an arrow. It may be exhausted differently as indicated by Q. According to this structure, since the vacuum degree of the electron beam source Gn can be improved, the performance of the electron beam source Gn can be prevented from being lowered by the gas 120 introduced into the plasma vessel 118 ( See FIG. 4).

As in the example shown in FIG. 18, the mesh electrode 174 may be disposed near the front surface of the cylinder 172. According to this structure, the plasma 124 may be shielded by the mesh electrode 174. Therefore, the plasma 124 can be prevented from entering the electron beam source Gn and degrading the performance of the electron beam source Gn.

Regardless of the arrangement of the cylinder 172 or the mesh electrode 174, the electron beam source Gn may be disposed externally near the plasma vessel 118, and the electron beam 138 may be placed from the source to the plasma vessel ( 118).

As described above, the number of filaments 122, electron beam sources Gn, and the like are not limited to the numbers in the above-described embodiments, and are appropriately selected according to the required Y-direction dimension Wy of the ion beam 50, and the like. Can be. The manner of disposing the filament 122 and the electron beam source Gn is not limited to the manner of the above-described embodiment, and may be appropriately selected according to the required Y-direction dimension Wy of the ion beam 50 and the like.

(3) About the whole analysis electromagnet

The analysis electromagnet 200 will be described. Prior to the description, a conventional analytical electromagnet is described for comparison.

(3-1) Conventional Analysis Electromagnets

For example, Patent Document 3 discloses an example of an analytical electromagnet related to the momentum analysis of a ribbon ion beam.

Patent document 3: JP-A-2004-152557 (paragraph 0006-0022, FIG. 1 and FIG. 21)

The conventional analytical electromagnet disclosed in Patent Document 3 will be described with reference to FIG. 43. In the figure, the yoke 26 is indicated by a dashed-dotted line to help understand the shapes of the coils 12 and 18. The traveling direction of the ion beam 2 is set in the Z direction, and two directions substantially perpendicular to each other in a plane substantially perpendicular to the Z direction are set as the X and Y directions, respectively. Thereafter, a ribbon ion beam 2 extending in the Y direction is incident at the inlet 24 of the analysis electromagnet 40 and radiated from the outlet 26.

The analysis electromagnet 40 includes a yoke 36 in which the upper and lower coils as shown in FIG. 1 of Patent Document 3, that is, the two coils 12 and 18 correspond to the yokes shown in FIG. It has a structure combined with.

The coil is a saddle-shaped coil (referred to as a banana-shaped coil in Patent Document 3), and a set of main body portions (coil mains in Patent Document 3) opposed to each other across the path (beam path) of the ion beam 2. 14); It includes a set of connecting portions (referred to in patent document 3 as end rising portions) 16 which rise inclined to avoid the beam path and connect the end portions of the body portion 14 to each other in the Y direction. The connecting portion 16 rises inclined at the inlet 24 and the outlet 26 so as to prevent the ion beam 2 from colliding with the portion, thereby securing a beam passing area.

In addition, the coil 18 is a saddle-shaped coil having a structure similar to that of the coil 12 (but having a surface symmetrical shape with respect to the coil 12), and the set of main body 20 and the set of connecting portions ( 22).

Each of the coils 12 and 18 is a multi-turn coil in which a conductor (coated conductor) whose periphery is covered by an insulator is wound a plurality of times, and a fan-shaped coil in plan view is bent near both ends. It is formed by the method of forming the connecting portions 16 and 22. As the conductor, generally, a hollow conductor through which a cooling medium (for example, cooling water) can flow is used. In the specification, "insulation" means electrical insulation.

The yoke 36 collectively surrounds the outside of the body portions 14, 20 of the coils 12, 18.

The analysis electromagnet 40 has the following problem.

(A) At the inlet 24 and the outlet 26, the protruding length L ₁ from which the connecting portions 16, 22 protrude from the yoke 36 in the direction of incidence and radiation of the beam is large. This is mainly due to the following reasons.

(a) In order to be able to deflect the ribbon-shaped ion beam 2 elongated in the Y direction as uniformly as possible, the main body portions 14 and 20 of the coils 12 and 18 are vertical by increasing the dimension a in the Y direction. Long (in the vertical direction longer than the example shown in FIG. 43). As described above, in the coils 12 and 18, a bending process is applied to the fan coils to form the connecting portions 16 and 22. The dimension a is thus reflected substantially directly in the protrusion length L ₁ . As the dimension a increases further, the protrusion length L ₁ also increases.

(b) In the coils 12 and 18, the connecting portions 16 and 22 are formed by applying a bending process to the fan coil as described above. Due to limitations on the bending process, relatively large bends 30, 32 inevitably occur near the boundary between the body portions 14, 20 and the connecting portions 16, 22. Due to the presence of these bends 30, 32, the distance L ₂ between the end portion of the yoke 36 and the end portions of the connecting portions 16, 22 is increased. Since the distance L ₂ is included in the protruding distance L1, the protruding distance L ₁ increases. Because of the limitation on the bending process, as dimension a further increases, the radius of curvature of the bends 30, 32 must be further increased, and the distance L ₂ and the protrusion distance L ₁ become longer.

The protruding distance L1 can be expressed by the following equation.

[Equation 3]

L ₁ = a + L ₂

(c) The connecting portions 16 and 22 rise inclined. This too also increases the protrusion distance L ₁ .

As described above, when the protruding length L ₁ of the connecting portions 16 and 22 from the yoke 36 is large, the analysis electromagnet 40 becomes correspondingly larger, and the area required for installing the analysis electromagnet 40 also increases. do. Therefore, the ion implanter becomes large, and the area required for installing the ion implanter also increases. In addition, the weight of the analysis electromagnet 40 also increases. The possibility of a magnetic field (also referred to as a fringe field) generated by the connections 16, 22 on the outside of the yoke 36 shaking the shape of the ion beam 2 (shape and posture, the same applies below) Will grow.

(B) The power consumption of the coils 12 and 18 becomes large. This is mainly caused by the following reason.

(a) The connecting portions 16 and 22 do not generate a magnetic field which deflects the ion beam 2. As described above, the protruding distance L ₁ of the connecting portions 16 and 22 is large. Thus, the lengths of the connections 16, 22 are correspondingly long, and the power consumption at the connections 16, 22 is unnecessarily large. This increases the power consumption of the coils 12, 18.

(b) As described above, the coils 12 and 18 are multi-turn coils of the sheath conductor. Therefore, it is difficult to increase the ratio of the conductor region (ie, the space factor of the conductor) in the cross sections of the coils 12, 18. Accordingly, power loss is correspondingly large, and power consumption is increased. In the case where the sheathed conductor is a hollow conductor, the drop rate of the conductor is further decelerated, so that the power loss is further increased. Therefore, power consumption is further increased.

As described above, when the power consumption of the coils 12 and 18 is large, the power consumption of the analysis electromagnet 40 is increased, thereby increasing the power consumption of the ion implanter.

The above-described problem of the conventional analysis electromagnet 40 can be solved by the analysis electromagnet 200 described below. Hereinafter, the overall configuration of the analysis electromagnet 200, the structure of the coil, the method of the coil, the characteristics of the analysis electromagnet 200, the control method and other examples will be described in order.

(3-2) Overall structure of the analysis electromagnet 200

Examples of analytical electromagnets 200 are shown in FIGS. 19-21 and the like. 21 shows an analysis electromagnet with the vacuum vessel 246 omitted. The analysis electromagnet 200 has a magnetic field along the Y direction in the beam path 202 through which the ribbon-shaped ion beam 50 impinges on the electromagnet, and the ion beam 50 passes, and the ion beam 50 is X It is configured to bend in the direction to perform momentum analysis. The magnetic field is schematically illustrated by magnetic lines of force 204 in FIG. 20 and the like. When the ion beam 50 impinges on the analytical electromagnet 200, the ongoing ion beam 50 receives a Lorentz force Fx directed to the right by the magnetic field, as viewed in the travel direction Z, and thus the right side. Is biased. As a result, momentum analysis is performed. The center trajectory of the ion beam 50 is shown by the dashed-dotted line in FIG. 19, and the radius of curvature is indicated by R. In FIG. The angle (deflection angle) at which the ion beam 50 is deflected by the analysis electromagnet 200 is represented by α.

For example, the radius of curvature R is 300 to 1500 mm and the deflection angle α is 60 to 90 degrees. 19 exemplarily illustrates a case where the deflection angle α is 90 degrees.

Referring also to FIG. 22, the analysis electromagnet 200 includes a first inner coil 206, a second inner coil 212, one or more (three in this embodiment) first outer coil 218, one or more (Three in this embodiment) a second external coil 224, a yoke 230, and a set of magnetic poles 232. The beam path 202 is surrounded by a vacuum vessel 236 made of nonmagnetic material and maintained in a vacuum atmosphere. Vacuum vessel 236 is also referred to as an analyzer tube.

The first and second internal coils 206, 212 are extracted and shown in FIG. 23. Referring to the drawings, the coil can be more easily understood.

In this example, the coils 206, 212, 218, 224 seal in the Y direction about a symmetry plane 234 (see FIG. 20, etc.) passing through the center in the Y direction of the beam path 202 and parallel to the XZ plane. It has a qualitative face symmetry. The coil 320 (see FIGS. 37 and 39, etc.), the first coil 326 and the second coil 328 (see FIG. 25) described later are configured in a similar manner. In the case of employing such a surface symmetric structure, a magnetic field having high symmetry in the Y direction can be easily generated in the beam path 202. This contributes to suppressing the shaking of the shape of the ion beam 50 upon radiation from the analysis electromagnet 200.

Hereinafter, as illustrated in FIGS. 20, 24, 28, and the like, when the plurality of first external coils 218 and the plurality of second external coils 224 need to be distinguished from each other, the first external coils 218 may be used. Is referred to as the first outer coils 218a, 218b and 218c starting from the top in the Y direction and the second outer coil 224 as the second outer coils 224a, 224b and 224c starting from the bottom in the Y direction. This is because the second outer coil is face symmetrical with respect to the first outer coil 218 as described above.

In the case where a sign indicating a component, for example coil 206, is underlined in the figure, such a sign refers to the entire component, such as a coil.

Referring primarily to FIGS. 23 and 27, the first internal coils 206 oppose each other in the X direction across the beam path 202 and on one side (upper side in the embodiment) of the ion beam 50 in the Y direction. A set of body portions 208 covering approximately half or more (in other words, substantially half or more) and the ends of the body portions 208 in the Z direction while avoiding the beam path 202 [ In other words, the saddle with a set of connections 210 connecting each other to an end on the inlet 238 side of the analysis electromagnet 200 and an end on the outlet 240, which may also be applied to other coils. It is a coil of shape. The first internal coil cooperates with the second internal coil 212 to generate a main magnetic field that bends the ion beam 50 in the X direction. The main magnetic field is a magnetic field that causes the ion beam 50 to bend to a substantially predetermined radius of curvature R.

The first inner coil 206 is referred to as a saddle shaped coil because, as a whole, the coil has a shape similar to the saddle. The same applies to the other coils 212, 218, 224 and the coils 326, 328 described later.

In order to prevent the ion beam 50 from colliding with the connecting portion 210 and to reduce the influence of the magnetic field generated by the connecting portion on the ion beam 50, the connecting portion is directed upward in the Y direction. It is separated from. For the same purpose as above, the connections of the other coils are separated from the beam path 202 toward the top or the bottom in the Y direction.

Mainly referring to FIG. 23, the second internal coils 212 oppose each other in the X direction across the beam path 202 and approximately half of the other side (lower side in the embodiment) of the ion beam 50 in the Y direction or A set of body portions 214 covering more (in other words, substantially half or more) and the ends of the body portions 214 in the Z direction while avoiding the beam path 202. A saddle shaped coil having a set of connections 216. The second inner coil is disposed to overlap the first inner coil 206 in the Y direction, and generates a main magnetic field that cooperates with the first inner coil 206 to bend the ion beam 50 in the X direction. . That is, the second internal coil 212 generates a magnetic force line 204 in the same direction as the first internal coil 206.

The second inner coil 212 has a similar dimension and structure to the first inner coil 206. In general, the number of turns of the conductor (specifically, the conductor sheet 268, FIG. 25, etc.) is equal to the number of turns of the first internal coil 206. However, as described above, the second internal coil has a shape that is plane symmetrical about the plane of symmetry 234 with respect to the first internal coil 206. The connector 216 is disposed on the opposite side (ie, lower side) in the Y direction with respect to the connector 210 across the beam path 202.

Although indicated by lines in FIG. 23, a small (eg, about 20 mm) gap 242 is formed between the first internal coil 206 and the second internal coil 212. In the gap, a total of two cooling plates 312 (see FIG. 34), which will be described later, may be disposed, that is, one cooling plate is on the side of the first internal coil 206, and one cooling plate is on the second internal coil 212. ) Side.

Referring primarily to FIG. 22, each of the first outer coils 218 is outside of the first inner coils 206 and a set of body portions 220 facing each other in the X direction across the beam path 202. And a set of connecting portions 222 connecting the ends of the body portion 220 to each other in the Z direction while avoiding the beam path 202. The first external coil generates a sub magnetic field that supports or corrects the main magnetic field. The first external coils 218 are arranged to overlap each other in the Y direction.

Specifically, the body portion 220 and the lateral portions (parts corresponding to the lateral portions 284 shown in FIG. 27) of each of the first external coils 218 overlap each other in the Y direction. It is arranged. Strictly speaking, it is difficult to say that the vertical portions (parts corresponding to the vertical portions 282 shown in FIG. 27) of the connecting portion 222 are overlapped as described above, but when viewed as a whole, the first external coil It can be said that 218 are arranged to overlap each other in the Y direction. The second outer coil 224 is similarly constructed.

The first outer coil 218 has a structure substantially similar to the first inner coil 206. However, the dimension in the Y direction is smaller than the dimension of the first inner coil 206 and the number of windings of the conductor is generally smaller than the number of the first inner coil 206. The first outer coil 218 has the same number of conductors (specifically, see conductor sheet 269, FIG. 25, etc.) turns. In an embodiment, the first outer coil 218 has different Y direction dimensions. Alternatively, the first outer coils have the same Y direction dimension. The second outer coil 224 is similarly configured.

For example, in the first inner coil 206 and the second inner coil 212, the Y direction dimension of the main body portion and the connecting portion is about 230 mm, and in the first outer coil 218a and the second outer coil 224a. The dimension of is about 50 mm, the dimension in the first outer coil 218b and the second outer coil 224b is about 60 mm, and in the first outer coil 218c and the second outer coil 224c. The dimension of is about 100 mm.

Although indicated by lines in FIG. 22, between the first outer coils 218, between the second outer coils 224, and the lowermost first outer coils 218 and 218c and the uppermost second outer coil 224. Small gaps 244, 246 and 248 are formed in 224c (see also FIG. 24). In the gap, a cooling plate 312 (see FIG. 34) described later may be disposed. For example, the dimensions of the gaps 244 and 246 are about 10 mm, and the dimensions of the gap 248 correspond to the dimensions of the gap 242 or about 20 mm. The gaps 244 and 246 are disposed around their respective circumferences along the respective outer coils 218 and 224.

The first outer coil 218 may generate a magnetic field in the same direction or in the opposite direction as the magnetic field generated by the first inner coil 206 and the second inner coil 212. Alternatively, the direction of the magnetic field can be reversed by control. The second outer coil 224 is configured in a similar manner. A portion of the magnetic force line (magnetic field) generated by the body portion 220 of the first external coil 218 spreads (in other words, leaks) toward the beam path 202, whereby the main magnetic field is affected. Accordingly, the first external coil 218 may generate a sub magnetic field that supports or corrects the main magnetic field. In this case, each first outer coil 218 has the effect of supporting or correcting a magnetic field in an area near the inside of the coil. The second outer coil 224 is configured in a similar manner.

Referring primarily to FIG. 22, each second outer coil 224 is outside the second inner coil 212 and has a set of body portions 226 facing each other in the X direction across the beam path 202. ) And a set of connecting portions 228 connecting the ends of the main body portion 226 to each other in the Z direction while avoiding the beam path 202. The second external coil generates a sub magnetic field that supports or corrects the main magnetic field. The second outer coils 224 are disposed to overlap each other in the Y direction and to overlap the first outer coils 218 in the Y direction.

The second outer coil 224 has a structure substantially similar to the second inner coil 212. However, the dimension in the Y direction is smaller than that of the second inner coil 212, and the number of turns of the conductor is generally smaller than the number of turns of the second inner coil 212. The number of turns of the conductor (specifically the conductor sheet) and the dimension in the Y direction of the second outer coil 224 are as described above.

An example of the number of turns of each conductor will be described. The number of turns of the first inner coil 206 and the second inner coil 212 is about 110 turns, and the number of turns of the first outer coil 218 and the second outer coil 224 is about 85 turns. It is Turon.

Since the main body portions 208, 214, 220, and 226 of the coil are disposed on the yoke 230 substantially as a whole, the main body portion is a portion that generates a predetermined magnetic field (main or sub magnetic field) in the beam path 202. I can speak. The main body portion 322 of the coil 320 described later is similarly configured.

The connecting portions 210, 216, 222, and 228 of the coil may be said to be portions which electrically connect the ends of each of the set of main body portions to each other in the Z direction and cooperate with the main body portions to form a loop-shaped conductive path. The connecting portions 324 and 325 of the coil 320 to be described later are similarly configured.

FIG. 20 is a longitudinal cross-sectional view taken along line A-A in FIG. 19 showing the body portions 208, 214, 220, 226 of the coils 206, 212, 218, 224. 39 to 41 which will be described later also show the main body of the coil.

The yoke 230 is made of ferromagnetic material and collectively surrounds the outside of the body portions 208, 214, 220, 226 of the coils 206, 212, 218, 224. The yoke 230 thus configured also gives the effect of reducing magnetic field leakage to the outside. The yoke 230 has a plan view shape of a so-called fan type as shown in FIG. 19. The shape of the cross section (cross section along the XY plane) of the yoke 230 is a shape of a rectangular frame. The yoke 230 configured as described above is also referred to as a window frame type yoke.

In an embodiment, the upper yoke 231 constituting the yoke 230 is removable. A method of using the upper yoke 231 will be described later.

The set of magnetic poles 232 is made of ferromagnetic material and protrudes inwardly from the yoke 230, for example about 15 mm, to face each other in the Y direction across the beam path 202. The top view shape of each magnetic pole 232 is an arch shape extended along the center track 54 of the ion beam 50 shown in FIG. This shape is also referred to as a fan-like shape. The gap length G between the magnetic poles 232 is somewhat larger (eg, by 100 to 150 mm) than the dimension Wy of the ion beam 50 in the Y direction. Stimulus 232 is not essential. However, if the magnetic poles are arranged, the magnetic field lines 204 can be easily concentrated in the gaps between the magnetic poles 232, so that it is easy to generate a magnetic field having a high magnetic flux density in the beam path 202.

For example, the gap length G between the poles 232 has a size equal to or less than one half of the radius of curvature R. FIG. Specifically, if the radius of curvature R is 800 mm, the gap length G is for example 500 mm. In general, the gap length _G is greater than the width W _G of the magnetic pole 232. That is, G≥W _G. According to such a dimensional relationship, it is possible to prevent the magnetic pole 232 and the yoke 230 from expanding unnecessarily.

20-22, it appears that a gap exists between the first inner coil 206 and the first outer coil 218 and between the second inner coil 212 and the second outer coil 224. . In an embodiment, the stack insulator 262 shown in FIGS. 24 and 25 is interposed in the gap.

(3-3) coil structure

Next, the structure of the coil and the like will be described in detail. 24 is an enlarged schematic view of a cross section of the first inner and outer coils taken along the line DD of FIG. 22, and FIG. 25 is an exploded view illustrating the first inner coil and the uppermost first outer coil shown in FIG. 24. It is a cross section.

The first inner coil 206 and the first outer coil 218 are fan-shaped cylindrical stacked coils 290 (FIG. 29) with the main body parts 208 and 220 and the connecting parts 210 and 222 remaining. (Not shown) (not shown). In the fan-shaped cylindrical stack coil, the lamination (set 264) of the insulating sheet 266 in which the main surface 266a extends along the Y direction and the conductor sheet 268 in which the main surface 268a extends along the Y direction are made. The first stack insulator 261 (stacked in the direction of arrow 270 that intersects the Y direction, which applies equally below) is stacked in a state of being wound with several windings, and the second stack insulator 262 is formed of the lamination. The lamination (set 265) of the insulating sheet 267 formed on the outer circumferential surface, the main surface 267a extending along the Y direction, and the conductor sheet 269 on which the main surface 269a extends along the Y direction, has several The stack is wound in two windings, and a third stack insulator 263 is formed outside the lamination.

In order to facilitate understanding of the notches 272-275, notches 272-275 of the first internal coil 206 are shown in FIG. 27. Similar notches 272-275 are also disposed in the first outer coil 218.

The yoke 230 fits into two notches 272 and 273 located in the outward and inward direction of the radius of curvature R. As shown in FIG. That is, these notches have a shape corresponding to the shape of the yoke 230. The notches 276 to 279 of the coil 320 described later are similarly configured. Two notches 274 and 275 on the advancing direction Z side of the ion beam 50 form an upper half of the inlet 238 and the outlet 240, respectively.

The second stack insulator 262 may constitute a first inner coil 206 (FIG. 25 shows this case), or may constitute a first outer coil 218, and coils 206 and 218. May be shared by

30 shows a cross-sectional structure of the stack coil 290 shown in FIG. As shown in FIG. 30, the stack coil is constituted by inner and outer coils 292 and 294 having the same cross-sectional structure as in FIG. Even in this case, the second stack insulator 262 may constitute an internal coil 292 (FIG. 30 shows this case), or may constitute an external coil 294, and coils 292 and 294. May be shared by

In the stack coil 290, the portions 272a to 275a respectively corresponding to the notches 272 to 275 are cut out by a cutting process or the like to form the notches 272 to 275. FIG. Thereafter, the inner coil 292 is configured as the first inner coil 206, and the outer coil 294 is configured as the first outer coil 218.

In addition, the present embodiment has a structure in which a gap is arranged in the external coil 294 of the stack coil 290 by a cutting process or the like in order to separate the first external coil 218 into three parts (three steps). Has

Each stack insulator 261, 262, 263 of the stack coil 290 is formed by, for example, winding a prepreg sheet with a plurality of windings. The prepreg sheet 300 of FIG. 31 is a prepreg sheet. A prepreg sheet is a sheet | seat processed in the semi-hardened state by immersing insulating resin in the support member which has insulation and heat resistance.

The support member is constituted by, for example, glass fibers or carbon fibers. The resin is composed of, for example, an epoxy resin or a polyimide resin. Stack insulators 261-263 formed using such a prepreg sheet may be referred to as fiber reinforced plastic (FRP). The thickness of the stack insulators 261 to 263 may be appropriately selected depending on the strength required as the structural member.

Each insulating sheet 266, 267 is, for example, a sheet composed of Nomex (registered trademark), Lumilar (registered trademark), or Kapton (registered trademark), or other insulating sheet. The thickness of the insulating sheets 266 and 267 may be appropriately selected depending on the required insulating strength and the like. For example, the thickness may be about 75 μm or smaller.

Each conductor sheet 268, 269 is constituted by, for example, a copper sheet or an aluminum sheet. The thickness may be appropriately selected depending on the current passing through. For example, in the case of copper sheets, the thickness is about 0.4 mm, and in the case of aluminum sheets, the thickness is about 0.5 mm. The width in the direction corresponding to the Y direction of these sheets may be appropriately selected according to the Y direction dimension of the coil required, for example, about 230 mm (eg, the width before the process described later is about 234 mm). Further, the widths of the stack insulators 261 to 263 and the insulating sheets 266 and 267 can also be appropriately set in accordance with the above values.

Insulating sheet 266 and conductor sheet 268 may overlap in a manner opposite to FIG. 25 as described below. The conductor sheet 268 may be disposed on the inner side (the left side of FIG. 25, that is, the stack insulator 261 side) of the first internal coil 206, and the insulating sheet 266 may be disposed to overlap the outer side. As needed, the insulating sheet 266 can be arrange | positioned so that it may overlap with both sides of the conductor sheet 268, respectively. The insulating sheet 267 and the conductor sheet 269 of the first external coil 218 are similarly configured.

As shown in a plan view, the conductor sheet 268 of the first internal coil 206 has a structure similar to that of a fan, with a plurality of windings, as shown in FIG. 26, and the terminal 340 of the sheet. It is connected to the end. However, the number of windings is not limited to that shown. When the current IM flows through the conductor sheet 268, a magnetic force line 204 that forms a main magnetic field may be generated. The same current IM and magnetic field lines 204 are also shown in FIG. 27.

As shown in plan view, the conductor sheet 269 of the first outer coil 218 also has a structure similar to that in FIG.

The second inner coil 212 and the second outer coil 224 are configured similarly to the first inner coil 206 and the first outer coil 218. As described above, the coil has a plane symmetrical shape about the symmetry plane 234 with respect to the first inner coil 206 and the first outer coil 218.

If necessary, a member or the like for reinforcing the coil may be further disposed on the outer circumference of the outer stack insulator 263 (in the case of the coil shown in FIG. 38, the stack insulator 262).

Taking the first internal coil 206 as an example, an example of the structure of the connecting portion of the coil will be described in more detail with reference to FIG. 27.

Each connecting portion 210 of the first inner coil 206 is connected with two vertical portions 282 substantially perpendicular to the end of the body portion 208 in the Z direction and extending substantially parallel to the Y direction. And a lateral portion 284 connected substantially perpendicular to the vertical portion 282 and extending substantially parallel to the XZ plane. That is, the vertical portions 282 are connected to each other by the lateral portions 284. Thus, the first internal coil 206 includes a lateral conductive path 286 substantially perpendicular to the Y direction and a vertical conductive path 288 substantially parallel to the Y direction. That is, most of the conductive path of the first internal coil 206 is composed of a combination of the conductive paths 286 and 288 excluding the edge portion. The current densities at all positions of the conductive paths 286 and 288 are set equal to each other.

The connections 216, 222, 228 of the other coils 212, 218, 224 are configured similarly to the connections 210. Thus, each of the other coils 212, 218, 224 have a lateral conduction path substantially perpendicular to the Y direction and a vertical conduction path substantially parallel to the Y direction. That is, most of the conductive path of the coil is composed of a combination of the lateral conductive path and the vertical conductive path excluding the edge portion. The current densities at all positions of the lateral and vertical conducting paths are set equal to each other. The coil 320 described later is similarly configured.

The connection portion of the coil is preferably configured as described above. According to the structure, the protruding distance of the connection portion from the analysis electromagnet 200 can be reliably shortened in the direction of incidence and radiation of the beam. The protrusion distance will be described in detail below.

A structural example of the power supply of the coil is shown in FIG. In the example, DC main power source 250 is connected to first internal coil 206 and second internal coil 212, respectively. The main power supply 250 may supply current IM having substantially the same level to each other to the first internal coil 206 and the second internal coil 212, respectively. The two main power supplies 250 need not be disposed separately, but can be configured as a single combined main power supply.

Also in this example, the DC sub power source 252 is connected to the first external coils 218 (218a to 218c) and the second external coils 224 (224a to 224c), respectively. The sub power source 252 may supply current IS to the first external coil 218 and the second external coil 224, respectively, and flows through the first external coil 218 and the second external coil 224. The current IS can be controlled independently. The plurality of sub power supplies 252 need not be individually disposed, but a single combined power supply capable of independently controlling the current IS flowing through the first external coil 218 and the second external coil 224, respectively. It can be configured as a sub power supply.

(2-3) coil manufacturing method

Next, an example of a coil manufacturing method will be described using the first inner coil 206 and the first outer coil 218 as an example.

First, a fan-shaped cylindrical stack coil 290 shown in FIG. 29 is manufactured. This manufacturing method is carried out in the following manner.

First, as shown in FIG. 31, the mandrel 296 having a sharp portion 297 protruding outwardly as opposed to the sharp portion 291 of the stack coil 290 shown in FIG. 29 is indicated by arrow 299 . By rotating in a predetermined direction about the axis 298, the prepreg sheet 300 as described above is wound by a plurality of windings. As a result, the stack insulator 261 shown in FIGS. 30 and 32 is formed.

Next, as shown in FIG. 32, the mandrel 296 is rotated in the same manner as described above, so that the insulating sheet 266 and the conductive sheet 268 are overlapped with each other on the outer circumferential surface of the stack insulator 261. It is wound up with a plurality of windings and laminated. As a result, the lamination of the insulating sheet 266 and the conductor sheet 268 shown in FIG. 30 is formed.

Next, as in the case of FIG. 31, the prepreg sheet 300 is wound around the outer circumferential surface of the lamination of the insulating sheet 266 and the conductor sheet 268 with a plurality of windings, and the stack insulator 262 shown in FIG. 30 is shown. To form.

Next, as in the case of FIG. 32, the insulation sheet 267 and the conductor sheet 269 are wound with a plurality of windings on the outer circumferential surface of the stack insulator 262 in a state where they overlap each other, and the insulation sheet shown in FIG. 30 is obtained. Laminations of the 267 and the conductor sheet 269 are formed.

Next, as in the case of FIG. 31, the prepreg sheet 300 is wound around the outer circumferential surface of the lamination of the insulating sheet 267 and the conductor sheet 269 with a plurality of windings, and the stack insulator 263 shown in FIG. To form.

After the step, after removing the mandrel 296, the stack coil is constituted by the inner coil 292 and the outer coil 294, but the sharp portion 291a protrudes outwardly, ie, outwardly, to the sharp portion 291. (290a) is obtained.

When a lead plate is disposed at the start and end portions of the conductor sheet 268, the conductor sheet 268 may be connected to the terminal 340 using the lead plate. The conductor sheet 269 is similarly constructed.

Before the winding step, preferably, abrasive particles (shots) such as metal grains are blown to the main surfaces 268a and 269a on the front and rear sides of the conductor sheets 268 and 269 to roughen the surface (that is, the shot blasting step is applied). ). According to this structure, the surface area can be increased, and intimate contact with the insulating sheets 266 and 267 can be improved. Even when the shot blasting process is applied only to one main surface of each of the conductor sheets 268 and 269, the effect can be obtained. However, it is desirable to apply the process to both major surfaces. This may also be applied to the insulating sheets 266 and 267.

Similarly, it is preferable to apply a shot blasting process to the main surfaces 266a and 267a on the front and rear sides of the insulating sheets 266 and 267 to roughen the surface. According to this structure, the surface area can be increased, and the intimate contact with the conductor sheets 268, 269 and the like can be further improved.

Next, after winding a heat shrinkable tape (not shown) around the outer circumference of the stack coil 290a, the sharp portion 291a is pressed to form a sharp portion 291 as indicated by arrow 302 in FIG. 33. Perform the molding process. The resulting article is heat cured. As a result, a stack coil 290b that forms the stack coil 290 shown in FIG. 26 is obtained. The strength of the structure is improved due to the winding of the heat shrinkable tape. Instead of the heat shrinkable tape, a prepreg tape configured similar to the prepreg sheet described above can be wound.

Next, after vacuum infiltrating resin into the stack coil 290b, it heat-cures under a pressurized state. In short, this means carrying out the resin molding process. As a result, the stack coil 290 shown in FIG. 29 is obtained. The resin molding process can increase the adhesive strength between the layers of the stack coil 290 to improve the strength and electrical insulation of the coil.

Next, a cutting process is performed to both end surfaces of the stack coil 290 in the axial direction (in other words, the height direction) to form a flat surface. Thereafter, a cut process is performed on the portions 272a to 275a corresponding to the notches to form the notches 272 to 275.

In the case where the external coil 294 is configured as the plurality of first external coils 218, a groove processing is applied to a portion of the external coil 294 corresponding to the gap 244 to form the gap 244.

Next, the stack coil 290c to which the cutting process and the grooving process are applied is immersed in an etching solution for etching the material of the conductor sheets 268 and 269 (as described above, copper or aluminum) to perform an etching process. do. As a result, burrs and the like of the conductor sheets 268 and 269 generated on the machining surface are removed during the cutting process and the grooving process, thereby preventing short circuits (layer short circuits) between the layers between the conductor sheets 268 and 269. And the end faces of the conductor sheets 268 and 269 become more concave than the end faces of the insulating sheets 266 and 267, thereby increasing the creepage distance of the layer insulation in the conductor sheets 268 and 269. Insulation performance can be improved.

After winding a heat shrinkable tape around the whole stack coil 290d to which the above-mentioned etching process was applied, it heat-hardens. As a result, it is possible to obtain a fan-shaped cylindrical stack coil in which the first internal coil 206 and the first external coil 218 shown in Figs. The strength of the structure is improved due to the winding of the heat shrinkable tape. When the coil has a forced cooling structure described later, the cooling plate 312 can be attached in the following manner before winding the heat shrinkable tape. Instead of the heat shrinkable tape, it is possible to wind up a prepreg tape that is configured similarly to the prepreg sheet described above.

As shown in FIG. 34, a cooling plate 312 having a refrigerant passage 314 is provided with an upper end 306 and a lower end 307, a first internal coil 206 and a first through an insulator 316. It is attached to the gap 244 of the outer coil 218 by pressure contact, respectively. Preferably, the cooling plate 312 is not only the upper and lower end surfaces of the main body portions 208 and 220 of the coils 206 and 218 in the Y direction, but also the upper and lower end surfaces of the coupling portions 210 and 222 in the Y direction. It is also arranged on the face. That is, it is preferable that a cooling plate is arrange | positioned as wide as possible. For example, coolant flows through the refrigerant passage 314. In an example, an insulator 316 is wound around the cooling plate 312. However, winding the insulator is not essential.

Coils 206 and 218 may be forcedly cooled through the end face by cold plate 312. This cooling structure is also referred to as an end cooling system.

In the above case, a high thermal conductivity thermal diffusion component (such as silicon grease) is interposed between the cold plate 312 and the insulator 316 and between the end surfaces of the insulator 316 and the coils 206 and 218 (eg, application). Is preferred. According to such a structure, an air space can be removed as much as possible, and heat conductivity performance and cooling performance can be improved.

Each gap 244 may be configured as a wedge-like shape that becomes narrower as it travels toward the inner side (left side of FIG. 34) of the coil 218. In addition, since the cooling plate 312 attached to the gap may be configured in a similar shape to the wedge, the cooling plate 312 is press-inserted into the gap. According to such a structure, the clearance gap formed between the end surface of the coil 218 and the cooling plate 312 can be made small, and a close contact can be improved. Therefore, the cooling performance can be further improved.

When the cooling plate 312 is arrange | positioned as mentioned above, after heat-shrinkable tape or a prepreg tape is wound around the whole coil in the state shown in FIG. 34, it heat-cures. Due to this, the fixing and intimate contact of the cooling plate 312 can be realized.

Finally, as needed, in both the case where the cooling plate 312 is disposed and in the case where the cooling plate is not disposed, the entire coil including the first inner coil 206 and the first outer coil 218 is It may be molded into a resin. According to such a structure, the moisture resistance, insulation, mechanical strength, etc. of a coil can be improved further. In this case, 5 to 30 wt% of filler (filler) may be mixed with the resin. According to such a structure, the crack resistance of a resin, etc. can be improved.

In a manner similar to that described above, the second inner coil 212 and the second outer coil 224 can be manufactured as coils in which these coils are incorporated. Coils described later, that is, the coil 320 shown in FIGS. 37 to 39, the first and second coils 326 and 328 shown in FIG. 40, and the internal coil 330 and the first and first shown in FIG. 41. 2 The outer coils 218 and 224 are manufactured in a similar manner as described above. The inner and outer coils can be made integral with each other.

Using the coils 206, 212, 218, and 224, the analysis electromagnet 200 and the like shown in FIGS. 19 and 20 can be assembled, for example, according to the following procedure. That is, in a state where the upper yoke 231 of the yoke 230 is removed and separated, the member in which the second internal coil 212 is integrated with the second external coil 224 is inserted into the yoke 230 from the upper side. Thereafter, the vacuum container 236 is inserted from above, and the member in which the first internal coil 206 is integrated with the first external coil 218 is inserted from above. Finally, the upper yoke 231 is attached.

(3-5) Characteristics of the analysis electromagnet 200 and the like

In the analysis electromagnet 200, the first inner coil 206 and the first outer coil 218 are fan-shaped cylindrical stack coils (left with main body parts 208 and 220 and connecting parts 210 and 222). Since the notches 272 to 275 are disposed in the 290, the connecting portions 210 and 222 extend substantially parallel in the Y direction from the ends of the main body portions 208 and 220. Therefore, even in the case where the dimension in the Y direction of the main body portions 208 and 220 increases, the case in which the dimension in the Y direction of the connecting portions 210 and 222 increases correspondingly is addressed. As a result, the protruding distance of the connecting portions 210 and 222 in the beam incident and radial directions does not increase.

The above items will be described with reference to FIG. 23 by using the first internal coil 206 as an example. When the dimension a in the Y direction of the body portion 208 increases, it is coped by changing the dimension b in the Y direction of the connecting portion 210 correspondingly. Specifically, dimensions a and c are substantially identical to each other. Therefore, even when the dimension a increases, the protruding distance L ₃ (see FIG. 19) of the connecting portion 210 does not increase in the incident and radial directions of the ion beam 50. The protruding distance L ₃ is determined by the distance L ₅ between the end face of the yoke 230 and the end face of the connection portion 210 and the thickness b of the connection portion 210. That is, the protruding distance L3 may be expressed as follows. As can be seen from the description of the structure of the first internal coil 206, the body portion 208 also has a thickness b .

[Equation 4]

L ₃ = b + L ₅

Unlike Equation 1 described above, which represents the protruding distance L1 of the conventional analysis electromagnet 40, Equation 2 does not include the dimension a in the Y direction. This is a feature that is significantly different from the conventional analytical electromagnet 40.

In addition, the distance L ₅ may be smaller than the length L ₂ of the conventional analysis electromagnet 40. The reason for this is as follows. Unlike the conventional coil 12, the connecting portion 210 is not formed by inclining the connecting portion 16 by a bending process, and as described above, the notch portion 272 is formed in the fan-shaped cylindrical stack coil 290. 275 to 275, and the connection portion 210 extends substantially parallel in the Y direction. In addition, the edge portion 254 between the body portion 208 and the connection portion 210 may be in a state where these portions are less rounded or substantially vertical by a cutting process or the like.

For the same reason as above, the protruding distance L ₃ of the connection portion 210 from the yoke 230 in the direction of incidence and radiation of the beam may be reduced.

The second inner coil 212 and the second outer coil 224 are similarly configured.

When the dimension a in the Y direction is set to the same value, that is, 250 mm, the protrusion distance L ₁ of the conventional analysis electromagnet 40 is about 300 mm, whereas the protrusion distance L of the analysis electromagnet 200 is different. ₃ ) is about 110 mm.

For the same reasons as described above, even when the inner coils 206 and 212 and the outer coils 218 and 224 are arranged in duplicate, such as in the analysis electromagnet 200, the beams from the yoke 230 in the direction of incidence and radiation of the beam The protruding distance L ₄ of the outer coil 218 can be reduced. In the conventional analysis electromagnet 40, when the coils are arranged in double inside and outside, the protruding distance of the connection part greatly increases.

For the same reason as above, since the analysis electromagnet 200 can be miniaturized, the area required for installing the analysis electromagnet 200 can be reduced. In addition, the weight of the analysis electromagnet 200 may be reduced. In addition, the possibility that the magnetic field generated by the connections of the coils 206, 212, 218, 224 interfere with the shape of the ion beam 50 is reduced.

According to this configuration, the protruding distance of the connecting portion of the coils 206, 212, 218, and 224 can be reduced, and the length of the connecting portion can be shortened, thereby reducing the waste of power at the connecting portion.

As described above, the coils 206, 212, 218, and 224 have a structure in which the conductor sheets 268 and 269 are stacked with the insulating sheets 266 and 267 interposed therebetween. Thus, when compared to a multi-turn coil in which the sheath conductor is wound several times, the space factor of the conductor is high and the power loss is correspondingly low. As a result, power consumption can be reduced.

For example, consider the case where the dimension a in the Y direction of each coil is set to 250 mm. In the prior art, the dripping rate of the conductor of the multi-winding coil of the sheathed conductor is about 60 to 70% even when the conductor is not hollow (not the hollow conductor), and further decreases in the case of the hollow conductor. Alternatively, the droplet rate of the conductors of the coils 206, 212, 218, 224 may be set to about 84 to 85%.

As a result, in the analysis electromagnet 200, it is possible to generate a magnetic field of the required intensity with less power consumption than in the conventional analysis electromagnet 40. At the same power consumption, it is possible to generate a stronger magnetic field than that generated by the conventional analytical electromagnet 40. In the latter case, since the radius of curvature R of the ion beam deflection can be reduced, the analysis electromagnet 200 can be further miniaturized.

The magnetic field of 0.2 Tesla is generated by two coils 206 and 212 (coils 218 and 224 are not used) in the same manner as the conventional analysis electromagnet 40 with the dimension a in the Y direction of each coil being 250 mm. In that case, the power consumption of the analysis electromagnet 40 is about 67 kW, whereas the power consumption of the analysis electromagnet 200 is only about 24 kW.

The ion implanter shown in FIG. 1 has an analytical electromagnet 200 having the features described above. Therefore, as the size of the analysis electromagnet 200 becomes smaller, the entire ion implanter can be downsized, so that the necessary area for installing the ion implanter can be reduced. The weight of the ion implanter can also be reduced. In addition, as the power consumption of the analysis electromagnet 200 decreases, the power consumption of the entire ion implanter may also be reduced.

In addition, since the analysis electromagnet 200 includes the first internal coil 206 and the second internal coil 212 as described above, compared to the case where one coil is used for each of the upper side and the lower side, The ion beam 50 with the large dimension Wy can be easily coped.

In addition, the first external coil 218 and the second external coil 224 may generate a sub magnetic field that supports or corrects the main magnetic field. Because of the sub magnetic field, the main magnetic field can be corrected, and the homogeneity of the magnetic flux density distribution in the Y direction can be improved. The sub magnetic fields 218 and 224 generated by the external coils 218 and 224 may be weaker than the main magnetic field and thus can be easily controlled.

Due to the above-described main and sub magnetic fields, it is possible to generate a magnetic field with high homogeneity in the magnetic flux density distribution in the Y direction in the beam path 202. As a result, fluctuations in the form of the ion beam 50 (bending, tilting, etc., which are equally applicable to the following) at the time of radiation from the analysis electromagnet 200 can be suppressed to a low level. This effect is more remarkable when the dimension Wy in the Y direction of the ion beam 50 is large.

Even when one first external coil 218 and one second external coil 224 are used, the effect of correcting the main magnetic field can be obtained. However, it is preferable to arrange the plurality of first outer coils 218 and the plurality of second outer coils 224 as in the example. In this case, the magnetic flux density distribution in the Y direction of the magnetic field generated in the beam path 202 can be more finely corrected by these external coils 218 and 224. Therefore, a magnetic field with higher homogeneity in the Y direction can be generated. As a result, fluctuations in the form of the ion beam 50 at the time of radiation can be suppressed to a lower level.

(2-5) Control Method of Analysis Electromagnet 200

An example of a control method of the analysis electromagnet 200 will be described. The current flowing through the first external coil 218 and the second external coil 224 allows the shape of the ion beam 50 radiated from the analysis electromagnet 200 to approach the shape of the ion beam 50 at the time of incidence. Can be controlled.

Specifically, the shape of the ion beam 50 radiated from the analysis electromagnet 200 is excessively inward of the radius of curvature R in the ion beam 50 radiated from the analysis electromagnet 200 with respect to the central axis. Reducing the current flowing through the first outer coil 218 and the second outer coil 224 corresponding to the bent portion, and the first outer coil 218 and the second outer portion corresponding to the portion lacking inward bending. By implementing at least one of increasing the current flowing through the coil 224, a form parallel to a predetermined central axis (center axis 318 shown in FIGS. 35 and 36) substantially parallel to the Y direction is approached. It is supposed to. For this reason, the ion beam 50 radiated | emitted from the analysis electromagnet 200 becomes a straight line which is not inclined, and has a form near the form at the time of incidence.

35 and 36 show examples of the shape of the ion beam 50 emitted from the analysis electromagnet 200, respectively. In the figure, the predetermined central axis substantially parallel to the Y direction is indicated by 318 , the plane of symmetry is indicated by 234 , the central trajectory of the ion beam 50 is denoted by 54 and the radius of curvature is denoted by R have.

In the case of the form shown in FIG. 35, the shape of the ion beam 50 is not shaken when viewed in the advancing direction of the ion beam 50, so that the first outer coils 218a to 218c and the second outer coils 224a to The value of the current flowing through 224c may be maintained.

In the case of the form shown in FIG. 36, the ion beam 50 is distorted (curved) in an arch shape similar to the L shape when viewed in the travel direction Z, that is, the curvature as it proceeds further upward in the Y direction. It bends more excessively toward the inside of the radius R and more excessively bent toward the inside as it proceeds further downward. Therefore, the current flowing through the first outer coil 218a is greatly reduced, the current flowing through the first outer coil 218b is slightly reduced, and through the first outer coil 218c and the second outer coil 224c. The current flowing at the current value is maintained, the current flowing through the second outer coil 224b decreases slightly, and the current flowing through the second outer coil 224a decreases greatly. As a result, while maintaining the position of the center trajectory 54 of the ion beam 50 radiated from the analysis electromagnet 200, the shape of the ion beam can be approximated to the form parallel to the central axis 318. That is, the shape can be made close to the shape shown in FIG.

If the shape of the ion beam 50 radiated from the analysis electromagnet 200 is shaken in a different form from that shown in FIG. 36, the correction is performed in the same concept as that transmitted, and the shape is close to the shape shown in FIG. 35. It can be done.

When the shape of the ion beam 50 radiated from the analysis electromagnet 200 is shaken, the following problem mainly occurs. According to the control method, such a problem can be prevented from occurring.

In general, the analysis slit 71 shown in FIG. 1 is disposed downstream of the analysis electromagnet 200. The slit 72 of the analysis slit 70 is linear. Therefore, when the shape of the ion beam 50 is shaken, a portion cut by the analysis slit 70 occurs, and the amount of the ion beam 50 of a predetermined ion species passing through the analysis slit 70 decreases. Since the cleavage occurs, the homogenization of the ion beam 50 is impaired. Increasing the width of the slit 72 in the X direction to prevent such cutting occurs, the resolution is lowered.

In addition to the above-described problem of the analysis slit 70, when ion implantation is performed to the substrate 60 using the ion beam 50 whose shape is shaken, a problem arises that the homogeneous implantation is damaged.

(2-6) Another Example of Analysis Electromagnet 200

Next, another example of the analysis electromagnet 200 will be described. The same reference numerals are given to the same or corresponding parts as the previous examples shown in FIGS. 19 to 22 and the like, and overlapping descriptions are avoided. The following description focuses on the differences from the previous examples.

Referring also to FIG. 37, the analysis electromagnet 200 shown in FIG. 39 avoids the set of body portions 322 and the beam path 202 facing each other in the X direction across the beam path 202. And a coil 320 having two sets of connecting portions 324 and 325 connecting the ends of the body portion 322 to each other in the Z direction, which bends the ion beam 50 in the X direction. Generates a magnetic field that causes The two connecting portions 324 on the upper side of FIG. 37 are one set of connecting portions, and the two connecting portions 325 on the lower side are other sets of connecting portions.

As shown in FIG. 38 showing the cross-sectional structure of the coil 320, the coil has the same cross-section as the first internal coil 206 (see FIG. 25) and the internal coil 292 (see FIG. 30) of the stack coil 290. Has a structure. That is, the coil 320 has notches 276 to 281 disposed on a fan-shaped cylindrical stack coil having the same structure as the internal coil 292 while leaving the main body 322 and the connecting portions 324 and 325. Has a structure. In addition, the coil 320 may be manufactured by the same manufacturing method as described above.

The coil 320 is configured as one coil in which the aforementioned first and second internal coils 206 and 212 (see FIG. 23) are vertically integrated with each other.

The shape of the notches 276 and 277 is similar to the shape of the notches 272 and 273 described above. The notches 278 and 279 have a surface symmetry with respect to the notches 276 and 277 about the symmetry plane (see FIG. 39). In detail, the notches 280 and 281 are through-holes, which form the inlet 238 and the outlet 240, respectively, and the ion beam 50 may pass through the notch. More specifically, ion beam 50 may pass through the notch through vacuum vessel 236.

The vacuum container 236 may pass through the coil 320 by inserting the vacuum container 236 in the Z direction through the notches 280 and 281. In this case, when a flange or the like is placed in the vacuum container 236 to interfere, the flange or the like is detached once. The analytical electromagnet 200 can be assembled by a similar method.

The connection part 324 has a structure similar to the connection part 210 of the first internal coil 206. The connection part 325 has a surface symmetry with respect to the connection part 324 about the symmetry plane 234.

The dimension a1 in the Y direction of the main body portion is substantially equal to the sum of the dimension c1 in the Y direction of the connecting portion 324 and the dimension c1 in the Y direction of the connecting portion 325 (that is, 2c1).

In the example analysis electromagnet 200, the coil 320 is also configured as one coil in which the above-described first and second internal coils 206 and 212 are integrated with each other. Therefore, for the same reason as described above, the protruding distance of the connecting portions 324 and 325 of the coil 320 from the yoke 230 is reduced, thereby obtaining the effect of miniaturizing the analysis electromagnet 200 and reducing power consumption.

The analysis electromagnet 200 shown in FIG. 40 includes first and second coils 326 and 328 that, in cooperation with each other, generate a magnetic field that bends the ion beam 50 in the X direction. The coils 326 and 328 are similar in structure to the first and second internal coils 206 and 212 (see FIG. 23), respectively. Accordingly, the first and second coils 326 and 328 may also be manufactured by the same manufacturing method as described above.

In the example analysis electromagnet 200, the first and second coils 326 and 328 have a similar structure to the first and second internal coils 206 and 212. Therefore, for the same reason as described above, the protruding distance of the connection portion of the coil from the yoke 230 is reduced, thereby obtaining the effect of miniaturizing the analysis electromagnet 200 and reducing power consumption.

Since the analysis electromagnet includes the first and second coils 326 and 328, it is possible to easily cope with the ion beam 50 having a large dimension Wy in the Y direction.

The analysis electromagnet 200 illustrated in FIG. 41 has a structure similar to that of the coil 320 and an internal coil 330 that generates a main magnetic field that causes the ion beam 50 to be bent in the X direction, and configured as described above. And external to the inner coil 330 and including first and second outer coils 218 and 224 for generating a sub magnetic field that supports or corrects the main magnetic field. That is, instead of the first and second internal coils 206 and 212 shown in FIG. 20 and the like, the analysis electromagnet includes an internal coil 330. Therefore, the inner coil 330 and the first and second outer coils 218 and 224 may be manufactured by the same manufacturing method as described above.

The characteristic matters at the time of manufacturing these coils are demonstrated. Using the stack coil 290 (see FIG. 29) in which the axial dimension (height) is set to a predetermined value, a notch similar to the notches 276 to 281 in FIG. 292, 294). In the external coil 294, a gap similar to the gap 248 shown in FIG. 22 is provided by a cutting process or the like, thereby forming the first and second external coils 218 and 224. In a manner similar to that of FIG. 22, each of the first and second outer coils 218, 224 is composed of a plurality of coils.

In the example shown in FIG. 41, the number of the first external coils 218 is two, but is not limited to two. The number of first outer coils is any number of one or more. The second outer coil 224 is similarly constructed.

The example analysis electromagnet 200 also includes an inner coil 330 and first and second outer coils 218, 224 configured as described above. Therefore, for the same reason as described above, the protruding distance of the connection portion of the coil from the yoke 230 is reduced, thereby obtaining the effect of miniaturizing the analysis electromagnet 200 and reducing power consumption.

The analysis electromagnet, in addition to the inner coil 330, includes first and second outer coils 218, 224 configured as described above. Therefore, a magnetic field with high homogeneity of the magnetic flux density distribution in the Y direction can be generated in the beam path 202 of the ion beam 50. As a result, fluctuations in the form of the ion beam 50 at the time of radiation can be suppressed to a low level. This effect is more remarkable when the dimension Wy in the Y direction of the target ion beam 50 is large.

Since the plurality of first outer coils 218 and the plurality of second outer coils 224 are arranged, the magnetic flux in the Y direction of the magnetic field generated in the beam path 202 by these outer coils 218 and 224. The density distribution can be calibrated more finely. Therefore, a magnetic field with higher homogeneity in the Y direction can be generated. As a result, fluctuations in the form of the ion beam 50 at the time of radiation can be suppressed to a lower level.

In addition, in the case where the ion implanter shown in FIG. 1 includes the analytical electromagnet 200 of each example, the entire ion implanter can be miniaturized in accordance with the miniaturization of the analytical electromagnet 200. Can be reduced. The weight of the analytical electromagnet can also be reduced. In addition, as the power consumption of the analysis electromagnet 200 is reduced, the power consumption of the entire ion implanter may be reduced.

(4) About the acceleration / deceleration device 400

The acceleration / deceleration device 400 shown in FIG. 1 deflects the ion beam 50 passing through the analysis slit 70 in the X direction by the electrostatic field, and accelerates or decelerates the ion beam 50 by the electrostatic field. Let's do it. Preferably, the acceleration / deceleration device 400 is disposed on the downstream side as far as possible to effectively exert the effect of suppressing the energy pollution described later. In the example shown in FIG. 1, the acceleration / deceleration device is arranged between the analysis slit 70 and the injection position, ie between the analysis slit 70 and the substrate drive device 500.

In providing the acceleration / deceleration device 400, the acceleration / deceleration device 400 may not only accelerate / decelerate the ion beam 50 but may also deflect the ion beam 50 in the X direction. Therefore, the ion beam 50 of predetermined energy can be selectively induced, and energy contamination (mixing of unwanted energy ions) can be suppressed. In addition, these effects can be realized by one acceleration / deceleration device 400. Therefore, the delivery path of the ion beam 50 can be shortened as compared with the case where the energy analyzer is arranged separately. Therefore, the transmission efficiency of the ion beam 50 can be improved. Especially when the ion beam 50 has a low energy and a large current, the ion beam 50 during transmission is easily diverged by the space charge effect. Therefore, the effect of shortening the transmission distance becomes remarkable.

40 shows a more specific example of an acceleration / deceleration device 400. The acceleration / deceleration device 400 includes first to third electrodes 402 arranged in the order of the first electrode 402, the second electrode 404, and the third electrode 406 in the traveling direction starting from the upstream side. , 404, 406). In an example, each electrode has openings 412 and 416 extending in the Y direction, through which ion beam 50 flows. In an example, electrode 402 consists of one electrode. Alternatively, the electrode may consist of two electrodes with the same potential interposed between them and in the X direction. The same can be applied to the electrode 406. The electrode 404 has a gap 414 extending in the Y direction, and the ion beam 50 flows through the gap.

A potential V1 with respect to the ground potential is applied to the first electrode 402. In general, the potential V1 is a high potential of positive (acceleration mode) or negative (deceleration mode).

When the potential is applied to the electrodes 402, 404, 406 or the electrode members 404a, 404b described below, if the potential is a value other than OV, voltage application means corresponding to the electrode (for example, from a DC power supply or a DC power supply) Potential is supplied from a voltage distribution resistor or the like (not shown) for distributing the voltage of the same. When the potential is 0 V, the corresponding electrode is grounded.

In general, the second electrode 404 is set to a potential that is at a level between the first electrode 402 and the third electrode 406. In the case of known electrostatic acceleration tubes, the second electrode 404 consists of a single electrode. In this example, the second electrode is configured to be separated into two electrode members 404a and 404b facing each other in the X direction across the path of the ion beam 50. Further, different potentials V2a and V2b; V2a? V2b are applied to the electrode members 404a and 404b, respectively, so that the ion beam 50 is deflected in the X direction. Specifically, a potential V2b lower than the potential V2a of the counter electrode 404a (counter electrode) is applied to the electrode member 404b on the side where the ion beam 50 is deflected, or is set to V2b <V2a. . Means for applying such a potential are as described above.

A gap 414 flowing through the ion beam 50 is disposed between the two electrode members 404a and 404b constituting the electrode 404. Preferably, the gap 414 bends in the deflection direction of the ion beam 50 as in this example. Specifically, the gap bends along the trajectory of ions 418 having a particular energy, ie specifically the desired energy after deflection. According to this structure, the ion beam 50 composed of the ions 418 having the desired energy can be efficiently induced.

A potential V3, which is generally 0 V, is applied to the third electrode 406. That is, the third electrode is grounded.

Preferably, the third electrode 406 downstream of the second electrode 404 is disposed along the trajectory of ions 418 having a particular energy, specifically the desired energy after deflection by the electrode 404. . According to this structure, the ions 418 having the desired energy can be efficiently induced, and the ions 420 and 422 having energy other than the desired energy and the neutral particles 424 are efficiently carried out by the electrode 406. Can be blocked. Therefore, energy pollution can be suppressed more efficiently.

The difference between the potentials V2a and V2b applied to the electrode members 404a and 404b constituting the electrode 404 is that the ion 418 having the desired (target) energy is the central trajectory of the acceleration / deceleration device 400. In particular, it is set to pass through a central trajectory (more specifically, the gap 414 and the opening 416) of the electrode 404, 406, which includes a second electrode 404 having a deflection function.

Table 1 collectively shows an example of the electrode and the potential applied to the electrode. Examples 1 and 2 are in an acceleration mode in which the ion beam 50 is accelerated by the acceleration / deceleration device 400, and example 3 is in a deceleration mode in which the ion beam 50 is decelerated. In the case of Example 1, acceleration energy of 30 keV can be realized, and in case of Example 2, acceleration energy of 130 keV can be realized. In the case of Example 3, acceleration energy of 8 keV can be realized. In any case, the potential V2b of the electrode member 404b, which is one electrode constituting the second electrode 404, is set lower than the potential V2a of the counter electrode 404a.

Potential V1 [kW] Potential V2a [kW] Potential V2b [kW] Potential V3 [kW] Example 1 30 0 -48 0 Example 2 130 100 52 0 Example 3 -8 0 -One 0

According to the acceleration / deceleration device 400, the ion beam 50 may be deflected by a second electrode 404 composed of two electrode members 404a and 404b and to which different potentials V2a and V2b are applied. . At this time, since the amount of deflection depends on the energy of the ion beam 50 at the time of deflection, the ions 418 having the desired energy can be separated from the ions 420 and 422 having different energies. The ions 420 are ions having energy lower than the desired energy, and the deflection amounts of these ions are larger than the deflection amounts of the ions 418. The ions 422 are ions having energy higher than the desired energy, and the deflection amounts of these ions are smaller than the deflection amounts of the ions 418. Neutral particles 424 may be separated because they travel in a straight line without being deflected. That is, since the acceleration / deceleration apparatus 400 exhibits an energy separation function, the acceleration / deceleration apparatus 400 can selectively induce the ion beam 50 composed of the ions 418 having the desired energy, and can suppress energy contamination. In an example, ions 420 and 422 other than ions 418 having the desired energy and neutral particles 424 impinge on electrode 406 downstream of second electrode 404, thereby blocking these ions. And removed.

In addition, the acceleration / deceleration device 400 also exhibits the original function of accelerating or decelerating the ion beam 50 in addition to the energy separation function described above. Since these functions can be realized by a single acceleration / deceleration device 400, it is not necessary to arrange the energy separator separately. When compared with the case where the energy separator is installed separately, the delivery path of the ion beam 50 can be shortened. Therefore, the transmission efficiency of the ion beam 50 can be improved.

In addition, the ion beam 50 may be accelerated between two stages, ie between the electrodes 402 and 404, and between the electrodes 404 and 406. Example 2 of Table 1 shows an example of such a case. Prior to acceleration in subsequent stages (ie, during low energy periods), ion beam 50 may be deflected by electrode 404. In comparison with the case where deflection is performed after full acceleration, the ion beam 50 can be easily deflected. Specifically, the difference between the potentials V2a and V2b applied to the two electrode members 404A and 404B constituting the electrode 404 can be made small. As a result, there is an advantage that electrical insulation is facilitated in the vicinity of the electrode 404.

Ions and neutral particles other than ions 418 having the desired energy may be blocked and removed by electrode 406 downstream of electrode 404. Therefore, energy pollution can be suppressed more efficiently. In particular, in the deceleration mode (see Example 3 in Table 1), it is easy for the neutral particles 424 to be generated by the charge conversion during the deceleration of the ion beam 50 between the electrode 402 and the electrode 404. It is known empirically. Even when many neutral particles 424 occur, these particles travel in a straight line and impinge on the electrode 406 to be blocked. Thus, the neutral particles 424 can be effectively removed from the acceleration / deceleration device 400.

In the acceleration mode, electrons are generally radiated, ions with energies other than the desired energy are accelerated from the location of the colliding electrode to the higher potential side, and X-rays with high energy corresponding to those accelerated electrons are accelerated. The electrons arise from the part of the electrode that collides. Known electrostatic acceleration tubes do not have a deflection function. Thus, the accelerated electrons can reach the high potential electrode (the electrode corresponding to the electrode 404) without being bent, and are accelerated by the large energy corresponding to the potential of the higher potential electrode to the higher potential electrode. Since they collide, X-rays having high energy are generated therefrom.

Alternatively, as in the acceleration / deceleration device 400, the second electrode 404 is composed of two electrode members 404a and 404b, and different potentials are applied to the electrode member, whereby an electrode having a deflection function is provided. Is provided. According to this structure, electrons emitted from the location where the ions with unwanted energy collide with each other are bent by the electrode 404 and thus cannot reach the higher potential electrode 402. Specifically, the electrons are bent toward the electrode member 404a having the higher potential among the two electrode members 404a and 404b constituting the electrode 404 and then collide with the electrode member 404a. At this time, the acceleration energy of the electron is the energy corresponding to the potential of the electrode member 404a, which is lower than when the electron collides with the electrode 402 of higher potential. For example, as in Example 1 of Table 1, the energy of the colliding electrons is about 0 eV and X-rays are substantially not generated. In the case of Example 2, the energy is about 100 keV, which is lower than about 130 keV when the electrons strike the electrode 402. In any case, therefore, the energy of the generated X-rays may be lower than the energy in known electrostatic acceleration tubes.

If necessary, other electrodes can be arranged upstream of the electrode 402 or downstream of the electrode 406. For example, a high potential electrode for accelerating or decelerating the ion beam 50 may be disposed upstream of the electrode 402. An electrode of negative potential for suppressing reversed electrons may be disposed downstream of the electrode 406.

1 is a schematic plan view showing an embodiment of the ion implanter of the present invention,

2 is a schematic perspective view partially showing an example of a ribbon ion beam,

3 is a diagram showing an example of the dimensional relationship in the Y direction between the ion beam and the substrate,

4 is a schematic cross-sectional view showing an example of the structure of the ion source shown in FIG. 1,

FIG. 5 is a schematic plan view showing an example of the arrangement of the filament and the electron beam source, the scanning position of the electron beam, and the like in the ion source shown in FIG. 4;

6 is a schematic plan view showing another example of the arrangement of the filaments,

FIG. 7 is a diagram showing an example of the structure of the electron beam source and the electron-beam power source shown in FIG. 1,

FIG. 8 is a simplified illustration of a system from the ion source shown in FIG. 1 to the ion beam monitor,

FIG. 9 is a schematic front view showing an example of the ion beam monitor shown in FIG. 1;

FIG. 10 is a flowchart showing a specific example of the content for controlling homogenizing the ion beam current density distribution in the Y direction using the control device shown in FIG. 1,

FIG. 11 is a flowchart showing an example of the filament current control subroutine shown in FIG. 10;

FIG. 12 is a flowchart illustrating an example of the electron beam scan rate control subroutine shown in FIG. 10;

FIG. 13 is a schematic diagram showing an example of an ion beam current density distribution after carrying out the rough setting for the filament state shown in FIG. 10;

14A to 14D are schematic diagrams showing a process of homogenizing the ion beam current density distribution in the Y direction by executing the filament current control and the electron beam scan speed control shown in FIG.

FIG. 15 is a diagram showing another example of the structure of the electron beam source and the electron-beam power source shown in FIG. 1;

FIG. 16 is a flowchart showing another specific example of the content of controlling homogenizing the ion beam current density distribution in the Y direction using the control device shown in FIG. 1;

17 is a flowchart showing an example of the amount control subroutine of the electron beam shown in FIG. 16;

18 is a schematic cross-sectional view showing another example of a method of disposing an electron beam source with respect to a plasma vessel of an ion source,

19 is a plan view illustrating an example of the analysis electromagnet shown in FIG. 1;

20 is a cross-sectional view taken along the line A-A of FIG. 19,

21 is a perspective view illustrating the analysis electromagnet shown in FIG. 19 with the vacuum container omitted;

FIG. 22 is a perspective view illustrating an analysis electromagnet shown in FIG. 19;

FIG. 23 is a perspective view illustrating the first and second internal coils shown in FIG. 22;

FIG. 24 is a schematic diagram showing an enlarged cross section of the first inner and outer coils taken along the line D-D of FIG. 22;

FIG. 25 is a cross-sectional view illustrating an exploded view of the first inner coil and the uppermost first outer coil shown in FIG. 24;

FIG. 26 is a schematic plan view showing a manner of winding the conductor sheet shown in FIG. 25;

FIG. 27 is a perspective view illustrating the first internal coil illustrated in FIG. 23;

FIG. 28 is a diagram showing an example of a power supply structure for the coil of the analysis electromagnet shown in FIG. 19;

FIG. 29 is a perspective view showing an example of a stack coil that is the original first and second internal coils shown in FIG. 22;

30 is an exploded view illustrating cross sections of the inner and outer coils taken along the line F-F of FIG. 29;

FIG. 31 is a plan view illustrating an example of a method of winding a prepreg sheet using a mandrel;

32 is a plan view illustrating an example of a method of winding an insulating sheet and a conductor sheet using a mandrel,

33 is a plan view illustrating an example of a stacked coil wound using a mandrel,

34 is a cross-sectional view illustrating an example of attaching a cooling plate to first and second internal coils;

35 is a diagram illustrating an example of an ion beam having a conventional form immediately after being emitted from an analytical electromagnet,

36 is a diagram showing an example of an ion beam having a distorted shape immediately after being emitted from an analytical electromagnet,

37 is a perspective view illustrating another example of a coil of an analysis electromagnet,

FIG. 38 is an exploded view illustrating a cross-sectional view of the coil taken along the line J-J of FIG. 37;

39 is a cross-sectional view showing another example of the analysis electromagnet corresponding to FIG. 20,

40 is a cross-sectional view showing still another example of the analysis electromagnet corresponding to FIG. 20;

41 is a cross-sectional view showing still another example of the analysis electromagnet corresponding to FIG. 20,

42 is a cross-sectional view showing an example of the acceleration / deceleration device shown in FIG. 1,

Fig. 43 is a perspective view showing an example of a conventional analytical electromagnet in which yokes are indicated by dashed-dotted lines to help understand the shape of the coil.

Claims (15)

The traveling direction of the ion beam is set as the Z direction, two directions orthogonal to each other in a plane orthogonal to the Z direction are set as the X and Y directions, respectively, and a ribbon type ion beam having a dimension in the Y direction larger than An ion implanter that is delivered to irradiate a substrate to perform ion implantation, An ion source having at least one filament for generating arc discharge in a plasma vessel into which gas is introduced, wherein the ion source generates a ribbon ion beam having a dimension in the Y direction greater than the dimension in the Y direction of the substrate; A substrate driving device which moves the substrate in a direction intersecting with a main surface of the ion beam at an implantation position where the ion beam is to enter the substrate; At least one electron beam source generating an electron beam, radiating an electron beam into the plasma vessel of the ion source to ionize a gas to generate a plasma, and scanning the electron beam into the plasma vessel in the Y direction; A drawing voltage for controlling the generation amount of the electron beam, one or more electron-beam power supplies for supplying the scan voltage for scanning to the electron beam source; An ion beam monitor measuring an ion beam current density distribution in the Y direction of the ion beam at a plurality of monitor points in the Y direction, at or near the injection position; The ion beam current density measured by the ion beam monitor is preset by controlling the electron-beam power source based on the measurement data of the ion beam monitor to maintain the amount of the electron beam generated from the electron beam source at a constant value. Relatively increasing the scan speed of the electron beam at the location of the ion source corresponding to the monitor point greater than a particular ion beam current density, and wherein the ion beam current density measured by the ion beam monitor is such that The ion beam current density distribution in the Y direction measured by the ion beam monitor by performing at least one of relatively decreasing the scan speed of the electron beam at the position of the ion source corresponding to the monitor point smaller than the beam current density. Control device having the function of homogenizing Ion implanter comprising a. The method of claim 1, (a) the number of the electron beam source and the electron-beam power source are both one, (b) the control device, Supplying a scan signal which is a source of scan voltage supplied from the electron-beam power source to the electron beam source, to the electron-beam power source; Calculating an average value of the ion beam current densities of the Y-direction distribution measured by the ion beam monitor; A function of uniformly controlling the filament current flowing through the filament of the ion source so that the calculated average value is equal to a predetermined specific ion beam current density within a preset error range; A function of calculating an error of the Y direction distribution, which is a difference between the ion beam current density of the Y direction distribution and a preset ion beam current density measured by the ion beam monitor, A monitor point whose calculated error is larger than the intended allowable error, the ability to determine the sign of the error at this monitor point, A function of determining the scan voltage corresponding to the scheduled monitor point, Based on the determined sine of the error, the scan rate of the electron beam is proportional to the degree of error at a scan voltage corresponding to a monitor point whose measured ion beam current density is greater than the predetermined specific ion beam current density. Increase and decrease the scan rate of the electron beam in proportion to the degree of error at a scan voltage corresponding to a monitor point whose measured ion beam current density is less than the predetermined specific ion beam current density, so that the ion beam collides with all The ability to shape the waveform of the scan signal so that the error at the monitor point is less than or equal to the acceptable error, Has the function of storing the data of the shaped scan signal and the data of the filament current, (c) the electron-beam power source comprises an amplifier for amplifying the scan signal supplied from the control device to generate a scan voltage. The method of claim 1, (a) the number of both the electron beam source and the electron-beam power source is plural, (b) the control device, Supplying a scan signal which is a source of scan voltage supplied from the electron-beam power source to the electron beam source, to the electron-beam power source; Calculating an average value of the ion beam current densities of the Y-direction distribution measured by the ion beam monitor; A function of uniformly controlling the filament current flowing through the filament of the ion source so that the calculated average value is equal to a predetermined specific ion beam current density within a preset error range; A function of calculating an error of the Y direction distribution, which is a difference between the ion beam current density of the Y direction distribution and a preset ion beam current density measured by the ion beam monitor, A monitor point whose calculated error is larger than the intended allowable error, the ability to determine the sign of the error at this monitor point, Determining the electron beam source and scan voltage corresponding to the scheduled monitor point; Based on the determined sine of the error, the scan rate of the electron beam is proportional to the degree of error at a scan voltage corresponding to a monitor point whose measured ion beam current density is greater than the predetermined specific ion beam current density. Increase and decrease the scan rate of the electron beam in proportion to the degree of error at a scan voltage corresponding to a monitor point whose measured ion beam current density is less than the predetermined specific ion beam current density, so that the ion beam collides with all The ability to shape the waveform of the scan signal so that the error at the monitor point is less than or equal to the acceptable error, Has the function of storing the data of the shaped scan signal and the data of the filament current, (c) each of the electron-beam power sources includes an amplifier for amplifying the scan signal supplied from the control device to generate a scan voltage.  The traveling direction of the ion beam is set as the Z direction, two directions orthogonal to each other in a plane orthogonal to the Z direction are set as the X and Y directions, respectively, and a ribbon type ion beam having a dimension in the Y direction larger than An ion implanter that is delivered to irradiate a substrate to perform ion implantation, An ion source having at least one filament for generating arc discharge in a plasma vessel into which gas is introduced, wherein the ion source generates a ribbon ion beam having a dimension in the Y direction greater than the dimension in the Y direction of the substrate; A substrate driving device for moving the substrate in a direction intersecting with a main surface of the ion beam at an implantation position where the ion beam is to enter the substrate; At least one electron beam source generating an electron beam, radiating an electron beam into the plasma vessel of the ion source to ionize a gas to generate a plasma, and scanning the electron beam into the plasma vessel in the Y direction; A drawing voltage for controlling the generation amount of the electron beam, one or more electron-beam power supplies for supplying the scan voltage for scanning to the electron beam source; An ion beam monitor measuring an ion beam current density distribution in the Y direction of the ion beam at a plurality of monitor points in the Y direction, at or near the injection position; The ion beam current density measured by the ion beam monitor is controlled while controlling the electron-beam power source based on the measurement data of the ion beam monitor to maintain the scan speed of the electron beam generated by the electron beam source at a constant value. Relatively reducing the generation amount of the electron beam at the position of the ion source corresponding to the monitor point greater than a predetermined specific ion beam current density, and the ion beam current density measured by the ion beam monitor is The ion beam current density distribution in the Y direction measured by the ion beam monitor by performing at least one of relatively increasing the generation amount of the electron beam at the position of the ion source corresponding to the monitor point smaller than the ion beam current density. Control device having the function of homogenizing Ion implanter comprising a. The method of claim 4, wherein (a) the number of the electron beam source and the electron-beam power source are both one, (b) the control device, Supplying a drawing signal which is a source of a drawing voltage supplied from the electron-beam power source to the electron beam source, to the electron-beam power source; Calculating an average value of the ion beam current densities of the Y-direction distribution measured by the ion beam monitor; A function of uniformly controlling the filament current flowing through the filament of the ion source so that the calculated average value is equal to a predetermined specific ion beam current density within a preset error range; A function of calculating an error of the Y direction distribution, which is a difference between the ion beam current density of the Y direction distribution and a preset ion beam current density measured by the ion beam monitor, A monitor point whose calculated error is larger than the intended allowable error, the ability to determine the sign of the error at this monitor point, A function of determining the scan voltage corresponding to the scheduled monitor point, Based on the determined sine of the error, the draw voltage is reduced in proportion to the degree of error at the scan voltage corresponding to the monitor point whose measured ion beam current density is greater than the predetermined specific ion beam current density, At a scan voltage corresponding to a monitor point where the measured ion beam current density is less than the predetermined specific ion beam current density, the draw voltage is increased in proportion to the degree of error, thereby allowing error at all monitor points where the ion beam collides. The ability to shape the waveform of the outgoing signal to be equal to or less than the possible error, Has the function of storing the data of the shaped outgoing signal and the data of the filament current, (c) the electron-beam power source includes an amplifier for amplifying the drawing signal supplied from the control device to generate the drawing voltage. The method of claim 4, wherein (a) the number of both the electron beam source and the electron-beam power source is plural, (b) the control device, Supplying a drawing signal which is a source of a drawing voltage supplied from the electron-beam power source to the electron beam source, to the electron-beam power source; Calculating an average value of the ion beam current densities of the Y-direction distribution measured by the ion beam monitor; A function of uniformly controlling the filament current flowing through the filament of the ion source so that the calculated average value is equal to a predetermined specific ion beam current density within a preset error range; A function of calculating an error of the Y direction distribution, which is a difference between the ion beam current density of the Y direction distribution and a preset ion beam current density measured by the ion beam monitor, A monitor point whose calculated error is larger than the intended allowable error, the ability to determine the sign of the error at this monitor point, Determining the electron beam source and scan voltage corresponding to the scheduled monitor point; Based on the determined sine of the error, the draw voltage is reduced in proportion to the degree of error at the scan voltage corresponding to the monitor point whose measured ion beam current density is greater than the predetermined specific ion beam current density, At a scan voltage corresponding to a monitor point where the measured ion beam current density is less than the predetermined specific ion beam current density, the draw voltage is increased in proportion to the degree of error, thereby allowing error at all monitor points where the ion beam collides. The ability to shape the waveform of the outgoing signal to be equal to or less than the possible error, Has the function of storing the data of the shaped outgoing signal and the data of the filament current, (c) each of the electron-beam power sources includes an amplifier for amplifying a drawing signal supplied from the control device to generate a drawing voltage. The method according to claim 1 or 4, The ion beam from the ion source is bent in the X direction to be disposed between the analysis electromagnet for analyzing the momentum and the injection position, the ion beam is bent in the X direction by an electrostatic field, and the ion beam is accelerated or decelerated. Further includes an acceleration / deceleration device, The acceleration / deceleration device includes first to third electrodes arranged in the order of the first electrode, the second electrode, and the third electrode in the ion beam traveling direction starting from the upstream side, and the ion beam is connected to the first electrode. Accelerate or decelerate at two stages between the second electrode and between the second and third electrodes, The second electrode is constituted by two electrode members which are opposed to each other in the X direction across the path of the ion beam and are applied with different potentials to deflect the ion beam in the X direction, the third electrode being specified after deflection. An ion implanter disposed along the trajectory of an ion beam with specific energy. The device according to any one of claims 1 to 6, further comprising an analysis electromagnet disposed between the ion source and the injection position, and analyzing the momentum by bending the ion beam from the ion source in the X direction. and, The analysis electromagnet, A set of body parts opposed to each other in the X direction across the beam path through which the ion beam passes, and at least one set of connecting portions connecting the ends of the body parts to each other in the Z direction while avoiding the beam path, the ion A coil that generates a magnetic field that bends the beam in the X direction, A yoke collectively surrounding the outside of the body portion of the coil, The coil has a structure in which a notch portion is arranged in a fan-shaped stacked coil while leaving the main body portion and the connecting portion, and the stack coil is formed on an outer circumferential surface of the laminated insulator, Winding the lamination with a plurality of turns, stacking the lamination of the conductor sheet and the insulating sheet, the main surface of which extends along the Y direction; An ion implanter configured to form an insulator laminated on an outer circumferential surface of the stack. The device according to any one of claims 1 to 6, further comprising an analysis electromagnet disposed between the ion source and the injection position, and analyzing the momentum by bending the ion beam from the ion source in the X direction. and, The analysis electromagnet, A set of body portions opposed to each other in the X direction across the beam path through which the ion beam passes and covering half or more of one side of the ion beam in the Y direction, and the body portion in the Z direction while avoiding the beam path A first coil having a saddle shape having a set of connecting portions connecting the ends to each other, the first coil generating a magnetic field in cooperation with the second coil to bend the ion beam in the X direction; A set of main body portions opposed to each other in the X direction across the beam path and covering half or more of the other side of the ion beam in the Y direction, and connecting the ends of the main body portions in the Z direction while avoiding the beam path; A saddle-shaped coil having a set of connecting portions, the second coil being disposed to overlap the first coil in the Y direction and generating a magnetic field in cooperation with the first coil to bend the ion beam in the X direction. and, A yoke collectively surrounding the outside of the main body portion of the first coil and the second coil, Each of the first and second coils has a structure in which a notch portion is disposed in a fan-shaped cylindrical stack coil while leaving the main body portion and the connecting portion, and the stack coil has a main surface on an outer circumferential surface of the laminated insulator. Winding the lamination with a plurality of windings, stacking the lamination of the conductive sheet and the insulating sheet extending along the Y direction; An ion implanter configured to form an insulator laminated on an outer circumferential surface of the stack. The device according to any one of claims 1 to 6, further comprising an analysis electromagnet disposed between the ion source and the injection position, and analyzing the momentum by bending the ion beam from the ion source in the X direction. and, The analysis electromagnet, A set of body parts opposed to each other in the X direction across the beam path through which the ion beam passes, and a connecting portion connecting the ends of the body parts to each other in the Z direction while avoiding the beam path, wherein the ion beam is X An internal coil that generates a main magnetic field that bends in a direction; A set of body parts outside the inner coil and opposite each other in the X direction across the beam path, and a set of connections connecting the ends of the body parts to each other in the Z direction while avoiding the beam path; One or more first external coils that are saddle shaped coils and generate a sub magnetic field that supports or corrects the main magnetic field; A set of body parts outside the inner coil and opposite each other in the X direction across the beam path, and a set of connections connecting the ends of the body parts to each other in the Z direction while avoiding the beam path; At least one second outer coil, the saddle-shaped coil, disposed in the Y direction to overlap with the first outer coil, and generating a sub magnetic field to support or correct the main magnetic field; A yoke collectively surrounding the outer side of the main body of the inner coil and the first and second outer coils, Each of the internal coil and the first and second external coils has a structure in which notches are disposed in a fan-shaped cylindrical stack coil in a state in which the main body part and the connection part are left, and the stack coils are stacked insulators. Winding the lamination with a plurality of windings, stacking the lamination of the conductor sheet and the insulating sheet, the main surface of which extends along the Y-direction, on the outer circumferential surface of; Forming laminated insulators on the outer circumferential surface of the stack; Winding the lamination with a plurality of windings, stacking the lamination of the conductor sheet and the insulating sheet, the main surface of which extends along the Y direction, on the outer circumferential surface of the stack; An ion implanter configured to form an insulator laminated on an outer circumferential surface of the stack. The device according to any one of claims 1 to 6, further comprising an analysis electromagnet disposed between the ion source and the injection position, and analyzing the momentum by bending the ion beam from the ion source in the X direction. and, The analysis electromagnet, A set of body portions opposed to each other in the X direction across the beam path through which the ion beam passes and covering half or more of one side of the ion beam in the Y direction, and the body portion in the Z direction while avoiding the beam path A saddle-shaped coil having a set of connecting portions connecting the ends to each other, the first inner coil generating a main magnetic field in cooperation with the second inner coil to bend the ion beam in the X direction; A set of main body portions opposed to each other in the X direction across the beam path and covering half or more of the other side of the ion beam in the Y direction, and connecting the ends of the main body portions in the Z direction while avoiding the beam path; The second internal coil having a set of connecting portions and disposed to overlap the first internal coil in the Y direction, and generating a main magnetic field to cooperate with the first internal coil to bend the ion beam in the X direction; , A set of body parts that are outside of the first inner coil and that face each other in the X direction across the beam path, and a set of connections that connect the ends of the body part to each other in the Z direction while avoiding the beam path At least one first external coil having a saddle-shaped coil, wherein said at least one first external coil generates a sub-magnetic field that supports or corrects said main magnetic field; A set of body parts that are outside of the second inner coil and that face each other in the X direction across the beam path, and a set of connections that connect the ends of the body part to each other in the Z direction while avoiding the beam path; At least one second external coil having a saddle-shaped coil disposed in the Y direction and overlapping with the first external coil and generating a sub magnetic field for supporting or correcting the main magnetic field; A yoke collectively surrounding the outer side of the body portion of the first and second inner coils and the first and second outer coils, Each of the first internal coil and the first external coil has a structure in which a notch part is disposed in a fan-shaped cylindrical stack coil while leaving the main body part and the connecting part, and the stack coil has an outer circumferential surface of the laminated insulator. Winding the laminations with a plurality of windings, stacking the laminations of the conductor sheet and the insulating sheet, the main surface of which extends along the Y direction; Forming laminated insulators on the outer circumferential surface of the stack; Winding the lamination with a plurality of windings, stacking the lamination of the conductor sheet and the insulating sheet, the main surface of which extends along the Y direction, on the outer circumferential surface of the stack; Is formed by forming an insulator laminated on the outer circumferential surface of the stack, Each of the second internal coil and the second external coil has a structure in which a notch portion is disposed in a fan-shaped cylindrical stack coil while leaving the main body portion and the connecting portion, and the stack coil has an outer circumferential surface of the laminated insulator. Winding the laminations with a plurality of windings, stacking the laminations of the conductor sheet and the insulating sheet, the main surface of which extends along the Y direction; Forming laminated insulators on the outer circumferential surface of the stack; Winding the lamination with a plurality of windings, stacking the lamination of the conductor sheet and the insulating sheet, the main surface of which extends along the Y direction, on the outer circumferential surface of the stack; An ion implanter configured to form an insulator laminated on an outer circumferential surface of the stack. The ion implanter of claim 8, wherein the analysis electromagnet further comprises a set of magnetic poles protruding inwardly from the yoke to face each other in the Y direction across the beam path. 10. The ion implanter of claim 9, wherein the analysis electromagnet further comprises a set of magnetic poles protruding inwardly from the yoke to face each other in the Y direction across the beam path. The ion implanter of claim 10, wherein the analysis electromagnet further comprises a set of magnetic poles protruding inwardly from the yoke to face each other in the Y direction across the beam path. 12. The ion implanter of claim 11, wherein the analysis electromagnet further comprises a set of magnetic poles protruding inwardly from the yoke to face each other in the Y direction across the beam path.

Images