JPH0831405B2 - Ion projection lithographic apparatus and method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION This application is incorporated herein by reference, "Image Alignment Method and Apparatus in Ion Lithography".
Is a continuation-in-part application of United States Patent Application No. 201,959 filed June 2, 1988.

The present invention is an ion beam projection lithography for manufacturing a semiconductor device or the like.

Among various methods required for manufacturing a semiconductor device or the like, lithography is a very important step. Briefly, the lithographic method is photoresist,
Or it begins by coating a silicon wafer with a photosensitive material, which is simply called a "resist". The lithographic exposure means projects an image of the pattern contained on the mask or reticle on the resist coated wafer. The wafer is stepped through a series of exposure positions whereby the same pattern of mask is exposed multiple times on the wafer. Development removes the resist pattern that outlines the desired image on the wafer surface. The wafer is then subjected to any one of a number of possible methods such as etching, oxidation, ion implantation, diffusion and volume. After the wafer process is inspected, it is recoated with resist and the cycle is repeated 8-15 times, resulting in the same microcircuit checkerboard array on the wafer.

Most projection lithography to date has used a light beam to expose wafers, but the demands for smaller surface sizes and denser components have led to studies of higher precision. . Much effort has been expended in providing an x-ray lithographic system, while some other systems have been accepted, including ion beam projection lithography, but have received much less attention.

The main objective of the present invention is to overcome what appears to be the limitations and drawbacks of previous proposals for ion beam projection lithography. Another object is to provide the industry with an overall practical ion lithographic apparatus capable of forming very small surfaces in the pattern painted on the wafer and suitable for commercial use.

According to one aspect of the invention, means for supplying an ion beam, a mask in the path of the ion beam having an aperture for generating a desired beam pattern, an optical column behind the mask, the ion path. An optical column defined by first and second main lenses arranged along the first main lens crossover, which is an accelerating Einzel lens arranged to form a crossover in the optical column A second main lens, which is a gap lens positioned behind and arranged to project a reduced image of the mask, and a target station behind the gap lens for supporting a target receiving the image. An ion projection lithographic apparatus is provided.

Preferred embodiments of this aspect of the invention have the following features. The mask is located substantially at the first focal plane of the first primary lens and the target station is located substantially at the second focal plane of the second primary lens. The means for supplying the ion beam comprises an ion source and means for supplying a virtual image of the ion source remote from the light source column to the light source column, and (1) a second focal plane of the first main lens and a second focal plane. The distance between the first focal plane of the two primary lenses and (2) the distance between the mask and the point defining the virtual image of the ion source and mask is such that the target produces the image of the mask. The blurring and geometric distortions that are selected and that result from the first and second primary lenses are substantially minimized at the same time. The first and second main lenses are arranged to position the crossover of the first main lens substantially at the first focal plane of the second main lens so as to generate a substantially telecentric beam following the second main lens. Be positioned at.

In a particularly preferred embodiment of the invention, the first and second main lenses substantially simultaneously satisfy the following conditions, whereby the beam emerging from said second main lens is substantially telecentric and The image plane is substantially free of color blur and geometric distortion when is reached, that is, (1) qp = f ₁ f ₂ Where q is the distance between the second focal plane of the first main lens and the first focal plane of the second main lens; p is the distance between the point defining the virtual image of the ion source and the mask; f ₁ is the first focal length of the first main lens;
f ₁ ′ is the first focal length of the second main lens; f ₂ is the second focal length of the first main lens; δ ₁ and δ ₁ ′ are the first and second main focal lengths due to the energy change of the ions in the beam. First of the lens
Change in focal length, and G (p) = ap ³ + bp ² + cp
+ D. Here, a, b, c, and d are the first to the second focal plane of the first main lens in the presence of the third-order aberration.
A lens constant associated with the conversion function from the focal plane, said conversion function being Where r ₁ , θ ₁ are the abscissas at the first focal plane of the first main lens at the mask and θ ₂ is the constant ion radiation conversion at the second focal plane of the first main lens. It is the horizontal angle coordinate. The coefficient D'is a conversion coefficient from the first focal plane to the second focal plane of the second main lens, Here, r ₂ ′ is the radial coordinate of the radiation at the second focal plane (that is, the target position) of the second main lens.

In a preferred embodiment of the invention, the ions of the ion beam are provided at the target with an energy of between about 50 and 200 kV and the ions of the ion beam are provided at the mask with an energy of between about 1 and 10 kV. It The image of the mask at the target is at least
Reduced by a factor of 1.5, the image on the target is no less than 10 mm in width and height.

The preferred embodiment of the present invention also has the following features. The ion projection lithographic apparatus has a pair of electric field control apertures, one of which is generated from the second electrode in the area around the mask and in the area around the opening of the third electrode of the Einzel lens. It is arranged on each side of the second electrode of the Einzel lens to reduce the applied electric field strength. A voltage ratio in the range of 7-20 is applied to the first and second main lenses. The means for delivering the ion beam comprises an ion source and a lens disposed between the ion source and the mask, the lens finely tuning the apparatus to simultaneously minimize color blur and lens distortion. Positioned along the axis of the optical column to select the actual ion source point. The lens is a solenoid lens,
The solenoid lens contributes to the selection of a desired ion type from different types of different masses emitted from the ion source by different types of different deflections, preferably the apertures have different masses from the desired type. Positioned between the first and second main lenses to prevent the passage of ions.

According to another aspect of the invention, the multipole means is adapted for application of the quadrupole field or in the second main lens so as to vary the magnification of the image at the target in the X direction relative to the magnification of the image in the Y direction. And preferably the multipole means is located after the second main lens in a region substantially free of electric field.

According to another aspect, the multipole means is arranged to apply a dipole field for adjustable movement of the image in a plane perpendicular to the axis of the optical column between the first and second main lens. And preferably the multipole means are adapted to generate superimposed high order fields for control of distortion and blur in the image. Preferably, the multipole means is a pair of contiguous 16 pole circular arrays of arced electrodes adapted to introduce an offset parallel to the beam. Also in a preferred embodiment of the present invention, a voltage regulator is provided for changing the lens voltage to fine tune the distortion and blur balance in the image. Preferably the voltage regulator is further capable of non-proportional adjustment of the lens voltage to tune the magnification of the image at the target.

The preferred embodiment of the first-mentioned aspect of the invention also has the following features. The voltage of the first electrode of the Gap lens is substantially the same as the voltage of the first and third electrodes of the Einzel lens, and the voltage of the second electrode of the Gap lens is substantially the same as the voltage of the second electrode of the Shinzel lens. Identical, the ripple effect of the power supply is such that the voltage ratio of the lens column remains substantially unchanged, thus substantially preserving the image quality at the target station.

The device is arranged such that the diameter of the gap lens can be varied in order to change the magnification of the image at the target, when the beam leaving the second main lens remains substantially telecentric and substantially in the image plane. No chromatic aberration and geometric distortion. The lens column includes an outer, rigid metal shell that extends substantially the entire length of the optical column, the shell being at a constant potential and the first and third electrodes and the gap of the Einzel lens. It directly supports the first electrode of the lens. The central electrode of the Einzel lens is supported by an insulator engaged for support on the interior of the electrode shell and the second electrode of the gap lens.
The electrodes are supported via an insulating bush supported by the downstream end of a rigid shell.

Preferably, a tubular shield made of a magnetic material of high permeability extends around and is supported by a rigid metal shell. Preferably, the tubular shield comprises a series of removable longitudinal segments and each end of the tubular shield in such a way that an end plate, preferably of high magnetic permeability material, provides magnetic continuity with the tubular shield. Attached to.

In another aspect of the invention, an ion projection lithographic apparatus comprises a means for supplying an ion beam, a mask in the path of the ion beam having an aperture for producing a desired beam pattern, an optical column behind the mask. An optical column defined by first and second main lenses arranged along the ion path, a first main lens arranged to focus the ion beam, behind the first main lens A second main lens positioned to project an image of the mask that is positioned and scaled down, supports a target for receiving the image and a shield made of a magnetic material of high permeability extending around the optical column. A target station behind the second main lens for the shield, the shield comprising a plurality of plates defining a tubular shield. Composed of an end plate made of a magnetic material of the magnetic continuity of high permeability of the removable longitudinal segments and tubular shield that. Preferred embodiments of this aspect of the invention have the following features. A conductor is provided within the shield and is arranged to provide adjustment to the magnetic field to which the ion beam is exposed, preferably the conductor comprises an array of elongated loops within the tubular shield. An array of longitudinally extending conductive loops is associated with the tubular shield, the loops being arranged to establish a magnetic flux circumferentially within the tubular shield. Preferably, it includes means for applying an alternating current to the loop to demagnetize the tubular shield and means for applying a temporary small bias current to the loop to increase the permeability of the tubular shield. Each conductive loop comprises longitudinally extending conductive segments extending along the inside and outside of the tubular shield. The tubular shield is positioned to allow adjustments to the magnetic field to which the ion beam is exposed.

In another aspect of the invention, a lithographic apparatus comprises means for supplying an ion beam, a mask in the path of the ion beam having an aperture for producing a desired beam pattern, an optical column behind the mask, an ion The optical column defined by first and second main lenses arranged along a passage, a first main lens arranged to focus the ion beam, positioned behind the first main lens And a second main lens arranged to project the image of the reduced mask, and a target station behind the second main lens for supporting the image-receiving target. The means for supplying the ion beam comprises a solenoid lens disposed between the ion source and the mask for selecting an actual ion source point along the axis of the ion source and the optical column, the solenoid lens comprising The different deflections of the different types contribute to the selection of the desired ion type from the different types of mass emitted from the ion source. Preferably, the solenoid is a counter-wound dual solenoid lens intended to prevent rotation of the ion beam during passage through the solenoid lens, and preferably the aperture allows passage of ions of different mass from the desired type. It is placed between first and second main lenses that are sized to prevent.

In yet another aspect of the invention, an ion projection lithographic apparatus comprises means for supplying an ion beam, a mask in the path of the ion beam having an aperture for producing a desired beam pattern, and an optic behind the mask. A column, a first and a second arranged along the ion path
An optical column defined by a main lens, a first main lens arranged to focus the ion beam,
A second main lens positioned behind the first main lens and arranged to project a reduced image of the mask; a target station behind the gap lens for supporting a target receiving the image; And 4 to the multipole means for varying the magnification of the image at the target in the X direction relative to the magnification of the image in the Y direction and the multipole means arranged in or beyond the second main lens. It is composed of a voltage control device for applying a polar electric field. Preferably, the multipole means is located behind the second main lens in a region that is substantially free of electric field and preferably 16 of arc-shaped electrodes.
It is a circular array of poles.

Another aspect of the invention is a lithographic method for producing a surface of the order of 0.1 micron on a target, the dimensions of which are capable of imaging the desired surface of the lithographic mask in the target field. Providing a projection lithographic apparatus, the apparatus having a set of adjustable parameters,
A metrology mask having an array of metrology planes is used to empirically determine the effect of each parameter across the target field, while the other parameters are held constant to provide a set of wedge functions. Metrology across sets the parameters of the device by linear optimization, which is based on measurements, interrupts exposure periodically, replaces the lithographic mask with a metrology mask that defines an array of measurement planes across the field, Measure approximately the geometric distortion of the ion beam projected across the target field and
From the steps of determining an error value of the set, and by linear optimization, determining an adjustment to the parameter of the set to reduce the error value on the basis of the wedge function, making said adjustment, and restarting the exposure of the target. Become.

Preferably, the metrology mask defines an array of orthogonal slit pairs to generate a pair of metrology ion beamlets preceding the target and empirically determines the effect of each parameter across the target field while other parameters are empirically determined. The process in which is held constant consists of measuring the width and determining the position of the center of gravity of the beamlet at the target.

In another aspect of the invention, a lithographic device is
With a set of adjustable parameters, adapted to project an image of a lithographic mask onto a target mounted on an X, Y stage, the X, Y stage being adapted to index across multiple exposure positions. In the ion beam projection lithographic apparatus, a measurement mask designed to replace the lithographic mask, a precision measurement stage attached to the X and Y stages, and a one-way measurement mask that is indexed across the ion beam field A detector mounted on a metrology stage capable of determining an error value based on the blur and geometric distortion detected at points distributed across the field, the error value being a linear optimization technique. Generate corrections to the set of parameters to reduce blur and geometric distortion of the device. It is useful for that. Preferably, the metrology mask defines an array of orthogonal slit pairs for generating corresponding pairs of metrology ion beamlets preceding the target and a detector mounted on the precision metrology stage measures the width and at the detector. For determining the center of gravity of each beamlet and preferably the detector is arranged to block the measurement beamlet in such a way that only one beamlet of an orthogonal pair hits the corresponding slit at one time. Associated with a pair of orthogonal slits.

The described ion beam projection lithographic apparatus and method allows for the formation of low distortion, large fields, reduced images of mask patterns on wafers. The described form of ion optics allows for the simultaneous reduction of the energy spread (color blur) of the ion beam and the effect of the image aberration balance caused by lens distortion of the optical column of practical size. In this regard, it can be noted that the real image of the mask on the wafer is formed in such a way that the two lenses significantly reduce the image degradation caused by the inherent aberrations of these lenses themselves. It is to be. As is known, in forming a real image of a mask with a single electrostatic lens, barrel or pincushion distortion is experienced depending on whether the image of the mask is formed before or after crossover, in which case A crossover is an image of approximately a point object produced by an ion source.

By providing a second main lens that follows the first main lens, which itself has the same type of drawback, the image is a second one in which the distortion produced by the second lens corrects the distortion produced by the first main lens. Formed behind the main lens. An image that is substantially geometrically distortion-free can be formed at some point downstream from the second primary lens. In addition, the combination of the two lenses can be selected such that at certain points downstream of the second primary lens there is a region that is less sensitive to the energy spread of the ions, ie the presence of color blur. Due to the integrated selection of the parameters of the optical column, lens distortion and blurring are simultaneously minimized according to the invention.

By using an accelerating Einzel lens preceding the gap lens, the energy of the beam can be kept low while the ion source has sufficient ion energy for its operation and the desired ion energy at the target is desired. There is no extreme voltage along the length of the optical column.

As mentioned above, in the preferred embodiment, the mask is positioned near the first focal plane of the first main lens, which is an accelerating Einzel lens, and the wafer is near the second focal plane of the second main lens, which is a gap lens. Positioned. The beam crossover resulting from the focusing of the first main lens is located near the first focal plane of the second main lens. Under these conditions, the ion path from the mask image element is substantially parallel to each other between the two lenses and the beam exiting the optical column and forming the real image of the mask on the wafer is substantially telecentric.

As mentioned above, by positioning the solenoid lens directly behind the ion source and in front of the mask, the solenoid lens selects the mass of the ions impinging on the wafer and may therefore be generated by the mask thickness. Effective to reduce the angle of incidence of the ion beam on the mask, which reduces no shading. Solenoids are also conveniently used to adjust the position of the actual ion source point along the ion optical axis. This last function interacts with the balance of color blur from the energy spread of the ion source and thereby allows fine tuning without the need for mechanical shifting of the ion optics.

The present invention is particularly useful in methods and apparatus for accurately laying a die or pattern on a wafer. The marks are placed on the wafer, for example adjacent to the corners of the die being exposed. Each of the sets of microbeams traveling through the same optical system adjacent but distinct from the beam pattern imaged on the wafer is scanned by a small scanning plate which does not affect the main field position, but the wafer surface. Do not scan each of the microbeams over the limited area above. When each microbeam is scanned separately across the groove, the secondary particle signal is made up of the marks.
This signal, which is sensitive to the relative position of the microbeam and the reference map of the mark on the wafer, is used to determine when the microbeam position is associated with the mark.
In relation to other microbeams spaced around the perimeter of the chip, this provides a means for determining the position, magnification and orientation of the die field associated with the wafer. In this way, an accurate overlay of one field can be created on the pattern present on the wafer by processing the signal and applying a correction voltage to the field imaging optics.

This technique provides a real-time measurement of the image field location relative to the existing pattern throughout the exposure time of the new field. This technique also provides a means of aligning a new area with an existing pattern field on the wafer, even when the wafer is distorted by processing.

Referring to the plan view of FIG. 1 and the three-dimensional cutaway view of FIG. 2 of the preferred embodiment, a tubular shield 10 surrounds the ion beam lithography machine and provides protection and support for its components. These components are the ion source 12, solenoid 18, mask assembly 20, optical column 14 and end station 60. A magnetic shield 340 made from a highly permeable iron alloy surrounds the entire optical column 14 and the magnetic field of the earth,
It substantially eliminates external stray static and time-varying magnetic fields such as those generated by the materials of construction and the power supply. The magnetic continuity of the iron shield at each end of the column is provided by the iron plates 350,352.

Referring to Figures 1b and 1c, a cutaway side and cross-sectional view of the shield is shown. The shield comprises a plurality of plates 344 that overlap and are sequentially supported by aluminum support members 356. An array of longitudinally extending wires 345 extend along the folds 346 on the inside and then the outside of the magnetic field to form separate elongated loops. A controller 347 causes these wire loops to be excited by an alternating current of decreasing magnitude over time so as to demagnetize the shield. Then the device 34
7 applies a momentary small direct current so as to increase the magnetic permeability of the shield. An additional longitudinally extending current carrying wire 348 forming another set of separate elongated loops inside the shield is provided to allow a reduction of the stray field in the region of the ion beam. Through wire 348 and wire 34
The current in the loop defined by 5,346 is independently variable to obtain the minimum magnetic action for the ion beam in the optical column.

Referring back to FIGS. 1 and 2, ions are generated from the ion source 12 and pass through the dual coil solenoid 18, which analyzes the beam 244 to detect other ion beams generated from the ion source. From the type, the predetermined ion type, helium, which separates helium in this example, is separated. Following the ion source 12 are a suppression electrode 152 and an extraction electrode 150. The suppression electrode has a higher negative voltage in relation to the ion source plasma than the extraction electrode and thus prevents the downstream electrode from accelerating and thereby suppresses unwanted heating of the ion source 12. The potential and shape of the suppressor and extractor electrodes are chosen to form an ion beam with a small effective size (20 μm diameter or less).
The potential of the extraction electrode 150 defines the energy of the beam passing towards the mask, independent of the potential selected for the suppression electrode 152.

Following the extraction electrode 150 is an ion source X, Y alignment stage 154, which allows sliding movement of the entire ion source for proper alignment with respect to the axis of the optical column 14. Two coils 240, 2 of solenoid 18
42 is wound to provide opposite direction magnetic excitation that acts on the ions to prevent the beam from rotating about its axis as a result of penetrating the solenoid lens. The beam of the solenoid lens has a mask structure 20.
By reducing the angle of impact on the mask and thus the uniform angle of the ion beam, more flux can impinge on the mask 164 and enter the optical column through its aperture. This angular reduction is also important because the mask has openings of finite thickness (typically 1-5 μm) and typically very narrow patterns. The reduced angle of incidence minimizes the shadowing that occurs at the edges of the mask aperture. Within the solenoid's first winding 240, an electrical shutter 38, consisting of a multi-pole array, is provided to deflect the beam from the column axis using the dipole field. This multipole is X
A quadrupole magnetic field can be applied to adjust the magnification in the and Y directions (in a plane perpendicular to the beam axis and thereby correct the elliptical distortion in the ion source).

Also provided in this region is a dose monitor 156 positioned beyond the vacuum isolation valve 36 and solenoid that allows the optical column 14 to be sealed from the ion source during ion source replacement or repair. The outer circumference of the beam impinges on a known area of monitor 156 and the induced current is measured. In this way the flux of the beam through the optical column can be derived and the exposure time required for the applied resist can be determined.

Following the dose monitor 156 is shown in FIG.
The mask structure best shown clearly in the enlarged view of the figure
20 consecutive times. The mask assembly 20 includes a complementary set of masks 164, one of which is shown on a rotating disk 136, so that each of them is a complete pattern of successive exposures of the target through each mask. Reference is made to a set of different masks having apertures corresponding to respective portions of the desired exposure pattern, such that the target is exposed to the beam over. Each mask has a pusher rod 160 driven by a piezoelectric transducer 162 (Fig. 2a).
Mounted on a flex mount 158 that permits rotation of the die pattern in response to the linear motion of the. In the preferred embodiment, the rotation of the mask 164 and thus of the die pattern on the order of. +-. 500 microradians is controlled in response to a beam alignment apparatus described below. Mask structure 20
Precedes the mechanical exposure shutter 104 (shown in detail in FIG. 5) and serves to remove the heat generated by the portion of the beam that does not pass through the mask opening but is instead blocked by the mask. Mask cooling cylinder 16
8 are arranged. The cylinder 168 radiatively cools the mask and extends around the beam axis. Coolant is the inlet
It is cooled by being introduced through 174, circulated through concentric coil 170 and discharged through outlet 172. When the exposure shutter 104 is positioned to block the beam to the mask, the shutter simultaneously blocks the heat dissipation cooling action of the cylinder 168. The cooling device holds the mask at a substantially constant temperature.

A gap lens used to focus the beam trajectory and accelerate the crossover or Einzel lens 22 and the ions and image the mask on the target 26.
Transformed by the Einzel lens to form an image of the ion source between 24 and. As is well known, a gap lens is a two-electrode lens having a first electrode and a second electrode at a potential difference. In this case, the first electrode 182 of the gap lens 24 is part of the rigid shell and the second electrode 86 is supported by the insulator bushing 226 at the end of the rigid shell. The gap lens 24 forms an ion beam that travels substantially parallel to the axis of the column, as indicated by the beam trajectories shown symmetrically.
It forms an image of the mask as it strikes the wafer. This type of device is called telecentric. The substantially telecentric nature of the beam is advantageous to reduce image magnification errors caused by irregularities in the positioning of the wafer along the optical axis or imperfections in the wafer itself, such as a twisted wafer.

In the vicinity of the position where the Einzel lens 22 forms the crossover of the ion beam, two continuous multipoles 184 and 186
There is a multipolar structure 28 consisting of. The application of dipole fields of equal magnitude and opposite sign to the two multipoles allows the beam to deviate from its original path, but remains parallel to it. The magnitude of these deflections is controlled by a beam alignment device as described below and may be, for example, +/- 5 microns. The multipole is preferably closer to the cylindrical surface than the cylindrical rod, and an even multiple of the electric field up to the 16-pole electric field (e.g.,
16 bending arcs 25 as shown in FIG. 2d that can be used to generate dipoles, quadrupoles, octupoles etc.)
It is an array of 0s. In the preferred embodiment, the higher order electric fields are preset to compensate for distortions in the device, but the deflecting dipole fields are superposed.

As a practical feature in this embodiment, placing the multipole near the crossover allows the multipole aperture to be reduced as the beam diameter is focused towards its focal point. Furthermore, the length-to-diameter ratio of the multipoles is about 5: 1, preferably 10: 1 to avoid peripheral field effects. Due to the reduced multi-pole diameter due to the near crossover position, the length is correspondingly reduced so as not to affect the overall device length during the design stage.

Multipole assembly 28 closely follows a mass selective aperture 30 which, in conjunction with the solenoid, serves to substantially select ions of only the desired mass for imaging on the wafer. Opening 30
And in a field-free area downstream from the gap lens 24, a multipole structure 34 for generating a quadrupole field is provided with a relative magnification of the image in the X (Mx) and Y (My) directions in mutually opposite wafer planes. Is provided to deflect the. For example, if the magnification in the X direction is made smaller,
In that case, the magnification in the Y direction is larger. Therefore, this element is used to make fine adjustments to the magnification difference between X and Y to correct for errors such as slight tilting of the wafer in the image plane about the X or Y axis. The main lens voltage is used as described below for absolute adjustment of magnification. In the embodiment of FIG. 2, the quadrupole structure 34 is
It is an array of 16 bending arcs as shown in Figure 2d and is controlled by a beam alignment device as described to effect the variation of Mx and My by a factor of about ± 5 × 10 ^-4 . In addition, the 16 bending arcs allow any choice and therefore variation of the X and Y axis orientation.

Further, apart from the crossover, the placement of the quadrupole assembly 34 is an important factor in minimizing aberrations and distortions in the optical system. Therefore, the quadrupole is placed away from the crossover and preferably in a field-free area. Four
The selective position with respect to the polar element is within the second electrode of the gap lens but may be located away from the crossover.

Acceleration Einzel / Gap lens combination (Fig. 2)
Provides a sufficient space between the gap lens 24 and the target 26 for the arrangement of multipoles of a predetermined length. A constant diameter-to-length ratio is maintained for optimum performance, as previously described in the study of multipoles positioned between lenses. After the gap lens, the beam diameter is crossover (about 1mm)
A longer (about 15 mm) and therefore correspondingly larger multipole array must be used for quadrupole field applications. However, it will be appreciated by those skilled in the art that the differential control of Mx and My using a quadrupole field in the ion projection lithographic lens system is useful in other lens combinations than the embodiment shown in FIG. Is clear.

Immediately before the beam impinges on the wafer, it is necessary to monitor the position, magnification and orientation of the image formed on the wafer and to generate a signal that induces a corrective action of the optical element in the area where the error is detected. There is an alignment beam scanner and detector 32 used.

This alignment beam scanner and detector 32 is shown in detail in FIG. 2b. The beamlet 188 has a beamlet that passes through the outside of the insulating member 190, so that the main die field in the scanner device 32 is a field.
Separated from 246, while die field 246 passes through its center as shown. At this point the beamlet is scanning plate 19
Scanned separately from the die field by the application of a dipole field generated on 2. The beamlets are scanned across fiducial marks 194 on wafer 248. The backscatter electrons originating from the beamlets impinging on the mark may be a channeltron or an electron amplifier detector 196
Detected by. Although only a single detector 196 is shown, it should be understood that the detector is provided with each alignment mark. The signal from the detector is used to locate the position and size of the die field 246 on the wafer 248. In response to the signal, a correction field can be applied by the ion optics or the field rotation can be adjusted by the rotation of the mask 164.

Alignment blocks or alignment rings 300, which also have fiducial marks 302, can be positioned downstream from the beamlet scanner and detector and directly above the target (as schematically shown in FIG. 9;
In the embodiment described above, the mask is imaged on the wafer rather than the metrology step 306 shown in FIG. 9).

Positioning the alignment block downstream from the beamlet scanner and detector, which also has fiducial marks on the block, is described in European Patent Application No. 294,363.
And in particular in Figures 1, 2 and 16.

The position of the ring relative to the wafer can be detected by an interferometer and the die field positioning in this embodiment can correspond only to the alignment marks on the ring. In this case the die pattern can be placed on the wafer without the use of wafer alignment marks and the wafer can be easily stepped from position to position to form a repeating pattern in the so-called "blind stepping" mode. Wear. Alignment rings have additional utility in measuring device-to-device image errors in metrology modes as described below.

Returning to FIGS. 1 and 2, the wafer 248 has precision X, Y (in the wafer plane) and Z so that the wafer can be stepped from one position to the other.
Mounted on a stage 40 that allows movement (along the optical axis), replicas of the image pattern can be formed at various locations on the wafer. The position of the wafer along the optical axis can be adjusted. The wafer is fixed by the chuck 42. When the wafer is completely covered by the repeating pattern, i.e. the wafer is stepped through a selected sequence of beam exposures, its chuck 42 is removed from the back of the stage and vacuumed through vacuum locks 46 and 48. Removed from the device. The chuck with the new unexposed wafer is inserted into the proper position on the stage.

As shown in FIG. 1, various components of the peripheral device are used to scan the device. A suitable vacuum pump, shown as a turbo pump, deflates the optical column. The pump is electrically and mechanically isolated from the lens so that the lens assembly is not subject to any external vibration from the source. Flexible bellows 52 are used to receive this insulation. The bellows has a vacuum inside and an atmospheric pressure outside. The large lateral force present on the lens is matched by the opposing bellows 54 on the other side of the machine. Since the optical column is at high potential with respect to ground, an insulating bush 56 for high voltage insulation is provided. An equivalent bush 58 isolates the second bellows from ground on the other side of the machine. The other pump on the lens and the ion source are treated in the same way by a pump placed at ground potential.

There are various other power and auxiliary functions of the machine as shown in FIG. 1 at a ground potential in a position convenient for maintenance. These include a lens power supply 66, a power distribution panel 68 where power is distributed to the machine, components at high potential, and a service module 70 that includes an isolation transformer to power the controller for the vacuum system 72. There is. Cabinet 74 for X, Y, Z stage electronic controller
Is located near the stage. Within console 76 is a main touch screen control console containing the larger microprocessor or computing device used to operate the machine. Controller functions are transferred to an auxiliary controller mounted at the rear of the machine 78 which can be used for maintenance. The machine may be mounted with bulkheads in such a way that the wafer processing section of the machine is in a clean room and the rest of the machine is also in a via or maintenance room, as shown in FIG. However, it is not strictly clean as required in advanced semiconductor processing clean rooms. The wall is indicated by reference numeral 80, the clean room is indicated by reference numeral 82, and the maintenance room is indicated by reference numeral 84.

Surrounding the ion source is the enclosure 342 shown in the front view of the embodiment of FIG. 1a. The shield 342 has a cabinet in which a pump or power supply or other control device can be placed and which is insulated from ground potential by placing an insulator 62. The ion source is within the shield 64 as shown in FIG. 1 and provides a gas supply of the desired type, eg hydrogen, helium or neon.

The total length L ₁ of the device is 6.1 meters and the conjugate length L ₂ (distance from mask to target) is 2.1 meters. The maximum height H ₁ is 2.5 meters. The ion beam expands from a point at the ion source 12 to about 10 mm at the first winding of the solenoid 240 and about 60 mm at the mask 164. The Einzel lens 22 focuses a beam of about 70 mm on a crossover diameter of about 1 mm. After crossover, the beam at Gallens 24 expands to 15-24 mm (depending on the size chosen for the gearup lens) and its diameter is maintained substantially on target 26 due to the telecentric nature of the beam.

FIG. 2c shows a block diagram of the power supply used in the embodiment of FIG. For positive ions, the ion source 12 is held at a positive potential with respect to the extraction electrode 150 by the first power supply 198 at a voltage Vo that can change the accelerating potential of the extracted ions from, for example, 0-10 Kv. The suppression power supply 220 is the suppression electrode 152.
Is held at a negative potential with respect to the ion source due to the suppressing electrons.

The first device power supply 222 includes first and third electrodes 176 and 178.
On the other hand, a negative voltage is applied to the central electrode of the Einzel lens 180. In this shape of the electrode potential position, the Einzel lens is accelerating because the ions (positive ions in this example) are accelerated between the first and second electrodes and then decelerated between the second and third electrodes. Referred to as Ellens. In this particular embodiment, the net energy of the ions is unaffected by this lens when the first and third electrodes are at the same potential. The first and third electrodes 176, 178 of the Einzel lens and the first electrode of the gap lens 182 are an integral component of the rigid shell of the optical column 14 and receive a positive potential from the second power source 224 as shown. To do. (Other accelerating Einzel lens arrangements in which there is a voltage difference between the first and third electrodes are practical so that the energy reduction of the ions between the middle and third electrodes is not equal to the energy gain between the first and middle electrodes. Such an arrangement can be implemented here without the departure from many aspects of the invention by the addition of suitable insulators and power supplies.) The potential across the gap lens is from the optical column by insulator 226. It is provided by connecting the negative terminal of the second power supply 224 to the second electrode 86 of the gap lens to be insulated. Therefore gap lenses accelerate positive ions prior to their exit from the optical column. The first power source 222 is
For example, the voltage V ₁ can be set to the ground of this second electrode on an Einzel lens between 0 and 10 Kv.
The second power supply 224 is variable over a typical range V _{2 of} 50-200 Kv, which has a typical value of approximately 100 Kv. In practice, power supplies 222 and 224 are used to provide a typical lens voltage ratio for both lenses in the range of 7: 1 to 20: 1 depending on the implanted ion energy Vo set by power supply 198.

Supplemental power supplies 228 and 230 provide small voltage adjustments ΔV ₁ and ΔV ₂ to device power supplies 222 and 224, respectively. The arrangement shown in FIG. The voltage ratio of V _{E for} the Einzel lens for
And The ratio for the gap lens of V _G in the case of

In fact, the value of V ₁ is so small that the voltage ratio applied across the electrodes of the Einzel and Gallens lenses is approximately equal to avoid the detrimental effects of the inherent ripple of the power supply. The energy of the ions hitting the target is then in the range of about 50-200 Kv.

The focal length of the electrostatic lens changes as a function of the potential difference between their electrodes and therefore the magnification of the image at some distance beyond the lens can be changed. In the preferred embodiment of the two lens arrangement of FIG. 2, the magnification of the image at the target 26 is the ratio of the first focal length of the second lens over the second focal length of the first lens or the voltage ratio of each lens. It is equal to the ratio, ie V ₂ / V ₁ . First and second
The approximate magnification can be selected by selecting the voltage ratio of the lens. Furthermore, if the voltage ratio applied to the lens is increased proportionally, the focal length will decrease approximately proportionally but the magnification will not change. However, this adjustment affects the distortion of the two lenses as explained below. Therefore, changes in lens voltage provide a fine tuning means for distortion and magnification. Because of this fine tuning, supplemental power supplies 228 and 230 are preferably under computer control for lens voltage changes of up to a few percent (e.g. 0-300 volts). The supplemental power supply is adjusted in response to the signal generated by the alignment beam, preferably for changes in lens voltage of up to a few percent.

FIG. 2e is a block diagram of the alignment device. Alignment marks that may be on the wafer and / or alignment block as described above
Backscatter electrons generated from 194 are detected by a detector 196 positioned above the mark. No. 2
In the embodiment shown in e, the four orthogonal pair of alignment marks m ₁ ~m ₄ is attached is placed around the area of the wafer can be die field is imaged thereon. The distance between the pair m ₂ and m ₃ in this preferred embodiment
The distance a ₂ between a ₁ and the pair m ₁ and m ₄ is not substantially equal.

Referring to FIG. 2e, the signal S ₁ generated by the electrodes
Is supplied to the correction signal generator 460. The signal generator 460 also receives a signal S ₂ from the beamlet scanning device 192, which represents the scanning position of the beamlet scanned by the scanning plate 192 shown in FIG. 2b, for example. Signal generator 460 outputs signal S ₁
And S ₂ to detect the misalignment of the die field and to apply the appropriate correction signal S ₃ to the various image correction elements to apply the appropriate corrective action to the die field for proper alignment. To occur. The correction elements are a multipole 28 for the control of the x, y variations, a means 34 for the magnification correction and a rotation control, eg a mask rotation control and thereby a means for the rotational alignment of the image on the wafer. 162 can be included.

A practical advantage of the design with the Einzel lens preceding the gap lens is that there is no need to provide an insulator between the lenses because the outer envelope of the optical column is at the signal potential. This provides a very rigid mechanical structure that ensures stable column alignment.
Mechanical robustness also reduces vibration between the mask 20 and the wafer chuck 42 and allows for alignment of the ion source and mass filter with respect to the optical column and prevention of ions moving away from the axis in a robust manner. To

Yet another important feature of the present invention is the use of solenoid lenses in the ion lithography system for mass selection and changing the actual ion source point along the optical axis as shown in FIG. The beam from the ion source bends the lighter mass ions, such as hydrogen, at a greater angle to the axis and bends the ions toward the axis so that the desired ion, such as helium, moves approximately parallel to the axis. Penetrate the solenoid lens. Ions heavier than the desired ions, such as oxygen, are not significantly bent and are substantially continuous in branching straight from the ion source.
The first mass selection can occur in the plane of the mask. The heavier ions move away from the optical axis and strike the mask structure. Heavier ion species that pass through the mask appear to diverge from the ion source and are not significantly foreseen by the first main lens. Near the crossover, the mass selective aperture then only allows a small fraction of heavier ions to pass through and reach the target. The photoions become concentrated on the axis and can therefore pass through the mask towards the optical column. However, due to the proximity to the axis of the first main lens, these ions are out of focus and therefore also strike the mass selective aperture.

Solenoids are also used to adjust the actual ion source placement on the machine axis. Adjusting this position alters the position of the crossover and further interacts with the chromatic aberration balance resulting from energy diffusion in the ion source, which is discussed further below. In the top panel of FIG. 4, the actual source position of the helium ions is SV ₁ for the first current level i ₁ applied to the solenoid.
Under these conditions, the passages of helium ions will first form a crossover at the location of the mass selective aperture.
Focused by the main lens. In the lower panel of FIG. 4, the second current level i ₂ is applied to the solenoid winding resulting in a slightly different ion path that produces the actual source of helium ions at SV ₂ . In this case, the helium ions are focused on the crossover just above the mass selective aperture. Since the condition for generating the telecentric beam after the second main lens is to position the crossover in the vicinity of the first focal plane by the second main lens, adjustment of the crossover position using the solenoid lens is performed on the optical axis. Allows a degree of telecentricity that is varied without having to physically move the lens along.

As can be seen, at both locations of the actual ion source in FIG. 4, helium ions are selected for passage over the target and both lighter and heavier than helium are filtered. Therefore, the solenoid valve performs both functions at the same time. It will be apparent to those skilled in the art that operation of the solenoid valve for mass selection and adjustment of the actual ion source position can be accomplished by Einzel's single lens system or multiple lens system and various sequences along the beam path. It is useful in other electrostatic lens arrangements that can include gap lenses that produce Moreover, due to the use of the present optical column, the solenoid can be replaced by a single electric lens if the mass selection function is performed differently. Also to be realized is that the diameter of the aperture and its position along the optical axis can be chosen for optimal mass filtration.

Returning to FIG. 2, a double wound solenoid is used in the preferred embodiment. The opposing windings generate opposing magnetic fields that prevent the net rotation of the beam. In addition, shorter focal lengths up to about 50% can be obtained via a single winding device as the beam experiences four sets of peripheral fields such that the beams are exactly two opposed. .

The circular rotary disk 136 and the exposure shutter 104 are the fifth
It is shown in a view along the optical axis from the illustrated ion source. In this embodiment, the circular rotating disk has four complementary masks that create the entire pattern in addition to one measuring mask that is placed in place in a measuring mode that measures mechanical strain, as described below. Hold. The circular rotating disc rotates so that the mask enters the axis of the optical column as shown in FIGS. A separate mechanism (not shown) clamps the mask to the back wall of the first lens when in position on the optical column. Opposite the position of the mask, there is a vacuum lock chamber that holds it on another mask and a mechanism (not shown) for loading and unloading the mask on this circular rotating disk. Once one set of masks has been unloaded, a vacuum valve can then be used to insulate the set from the main chamber and that set can be removed and replaced with another set. The circular rotating disk 136 is consistent with the mask structure described above, which can be rotated in a precise manner to monitor mask rotation as required by beam alignment equipment.

The exposure shutter 104 is used to control the exposure time. There are two shutter positions 234 that cover the mask and prevent exposure and two shutter positions 236 where the mask is fully exposed. The peripheral slot 232 of the shutter 104 is shown in FIG. These allow the beam alignment feature to occur without exposing the central die field of the mask. For example, just prior to exposure of the die field, the shutter can be positioned so that the three pairs of alignment marks in the mask are exposed to the beam through the slots. This is sufficient to determine five parameters: X and Y translation in the wafer plane, rotation Φ and magnifications Mx and My, and the resulting signal from detector 196 controls each of the five parameters. The sixth beamlet is used to provide an error signal as a measure of the accuracy of the other beamlet signals. Once these parameters are established, exposure can begin. At some point during exposure, during rotation of the shutter, one of the pairs is replaced with another pair of alignment marks so that the alignment is properly held by the exposure when the exposure is finished. The exposure shutter can also be used in concert with an electrical shutter, for example positioned between the ion source and the mask to deflect the beam from the mask during exposure.

When the shutter covers the mask, it also prevents the mask from being exposed to the cooling surface of the mask cooling mechanism,
The mask then remains at ambient temperature and the beam hits the shutter, which cools in the same way as the mask. Therefore, no excessive cooling is required for the electronic shutter.

The energy spread of ions, called imperfections, caused by image imperfections produced in an ion projection lithographic apparatus,
And gives rise to an inherent distortion produced by the lens itself called lens distortion. Since the spherical aberration coefficient of an electrostatic lens is always positive, this lens distortion cannot be eliminated by lens combinations such as those done in optical properties where the coefficient may consist of any symbol. However, the distortion of the electrostatic lens can be balanced by adding to it a second main lens that follows the first main lens, which has the same type of drawback. The image can then be formed after the second main lens where the distortion produced by the second main lens balances the distortion produced by the first main lens.

As is known, in forming the actual image of the mask with a single electrostatic lens, barrel or pincushion distortion can occur before or after the crossover as the image of the mask is shown in FIG. Experienced, depending on whether formed. If a second main lens is added, the barrel distortion that may be introduced can be used to correct the pincushion distortion of the first main lens. At some point from the downstream of the second main lens, barrel distortion of the first main lens and pincushion distortion of the second main lens are balanced and an undistorted image is formed. Moreover, the effect of image blurring in a two-lens arrangement is a function of various optical parameters, including lens-to-lens distance and ion source-to-mask distance. With the proper combination of these distances, the image with minimal blurring occurs downstream from the second primary lens.

The two-lens system of the present invention allows the effects of color blur and lens inherent distortion to be minimized at the same image location in a practical size system. In addition, the actual image of the mask with minimized distortion and blur is formed in the telecentric beam at the target. The first lens is shaped to form a virtual image of the mask element at infinity, ie the beam paths between the two lenses from a particular mask element are substantially parallel to each other. The second lens forms its virtual real image at its exit focal plane. To achieve these above-mentioned conditions, the mask is placed near the first focal plane of the first lens F ₁ and the wafer is near the second focal plane of the second lens F ₂ ′ as shown in FIG. It is in. Therefore, with the mask and image near their respective focal planes, the distance between the two lenses can be selected at the design stage to minimize blurring without affecting the magnification of the image. Since the distance between the two lenses can be chosen, additional parameters are provided besides the ion source / mask distance,
Allows a minimum of coincidence as found in chromatic aberration and geometric or lens distortion.

For example, in FIG. 8, lens distortion ΔRo is plotted as a function of actual ion source / mask distance p for two different spacings q between two lenses. When the ion source distance is changed, the strain passes through at a minimum. If the image blur ΔR _E is also plotted, it is also a special ion source for each lens spacing.
/ It is found to pass with a minimum in the mask distance. Thus, there are source distances where distortion minima occur and generally other source distances where color blur minima occur. It is possible to have distortions for appropriately chosen lens spacings and color blur minima occur at the same actual source / mask distance.

Another feature of the optical system of the present invention relates to producing various magnification or reduction factors. The reduction factor is directly proportional to the focal length of the last lens and the focal length of the last lens is in turn directly proportional to the lens diameter. By simply adjusting the diameter of the last lens, a corresponding change in magnification can be obtained with great flexibility without affecting the balance of distortion and aberrations. For example, in the embodiment of FIG. 2, it is simply necessary to adjust the diameter of the final electrode of the gap lens to cover a range of different coefficients from 2: 1 to 10: 1. As a practical feature for the Einzel / Gap lens combination, removal of the gap lens final electrode 86 is facilitated because it resides at the end of the optical column as shown in FIG. In another embodiment, the first electrode of the gap lens is made removable from the optical column to allow the diameter of both electrodes to be charged.

Relationship between lens distortion, color blur, and telecentricity For mathematical processing, Fig. 3 shows under the focal plane.
F _1, F ₂ and F ₁ ', F _2' shows an ion optical system for the two said major lens L and L 'with. The mask is placed in the focal plane F ₁ and the wafer is placed in the final image plane F ₂ ′. Nominally point source is imaged by L in is arranged at a distance in front of the F ₁ and F ₁ '. This source may be a virtual or real image.

The distance p is between the source and the mask, and the drift distance q is the second focal plane F ₂ of the first main lens and the first focal plane of the second main lens.
Defined between F ₁ ′. The solid line B depicts the ideal first order path of radiation from the ion source to the image and the dashed line b shows the path of the radiation furnace that has been disturbed by lens distortion and / or chromatic aberration. In the case of such incomplete, the focal plane is shown in Figure 3 by which and the vertical dashed lines F ₁ and F ₂ are known with respect to the shift position, the respective F ₁ and F ₂ the first main lens The first and second focal planes,
Have the distances d ₁ = F _{1 −1} and d ₂ = ₂ −F ₂ , and F ₁ ′
And F ₂ 'is the first and the second focal plane of the second main lens, d ₁ = F _1' - having ₁ 'and d ₂ = _2' -F ₂ '.

Specific radiation (γ, θ) in the first rank optical characteristics
The abscissa of is linearly transformed into new coordinates (γ ', θ') at other locations along the axis of the device. This is conveniently denoted as a matrix transformation.

Here, A is a transformation matrix.

For the focal plane-to-focus transformation through the lens L, the transformation matrix T is Where f ₁ and f ₂ are _first and _second lenses L
Focal length, while for lens L'the transformation matrix T'is Where f ₁ ′ and f ₂ ′ are the first and second focal lengths of L ′ and the transformation matrix Q with respect to the drift from F ₂ to F ₁ ′ is Is.

The whole conversion mask / wafer Γ is thus Equation 5 here results from a straight-ahead matrix multiplication. So the special conversion formula is Here, the fact that γ ₁ = pθ ₁ (7) with respect to the nominal point source was used.

Since the (1,2) element of the matrix Γ is zero,
The final position along the axis corresponds to the Gaussian image. That is, γ ′ ₂ is independent of θ ₁ .

Furthermore, from equation (6), the magnification is Which shows the inverted real image due to the negative sign. Further, from the equation (6), the telecentric condition (θ ₂ ′ = O) requires qp = f ₁ f ₂ (9).

It should be noted that the image condition and the magnification value are q
And p are independent, while the telecentric condition defines a relationship between these two distances.

The simple linear transformation described above is disturbed by color effects (energy diffusion of the ion beam) and the inherent lens distortion of lenses L and L '. If the perturbations for T and T'are Δ and Δ ', then the perturbed transform is Γ' = T + P (T '+ Δ') Q (T + Δ) = T'QT + Δ'QΔ + T'QΔ + Δ'QT (10) Is.

Maintaining only the first order perturbation term results in a representation of the order for the perturbation matrix P: P = Γ′−Γ = T′QΔ + Δ′QT (11).

The condition under which the final beam traversal position γ ₂ ′ remains undisturbed is then checked.

According to equation (11), Δγ ₂ ′ = P ₁₁ γ ₁ + P ₁₂ θ ₁ = (pP ₁₁ + P ₁₂ ) θ ₁ (12) where P ₁₁ is the (1,1) element of the matrix P and P ₁₂
Is a (1,2) element. Thus Δγ ₂ ′ = 0 and the beam remains unperturbed for all θ ₁ if pP ₁₁ + P ₁₂ = 0 (13).

Regarding the color blur, the ion energy is changed from the nominal value E to the amount δ.
When E is changed, the focal length is f _i = f _i −δ _i ; f _i ′ = f _i ′ δ (14) and Then f _i changes to _i (i = 1,2). In addition, the focal planes change their position from F ₁ to Fi and from F ₁ ′ to F i ′, respectively, corresponding to displacements along the beam axis as shown in FIG. 3 and therefore di = F ₁ Fi; di ′ ＝ F ₁ ′ Fi ′ (16) The well-known Liouville theorem applies independently of these changes. ₁ / f ₂ = f ₁ / f ₂ ; ₁ ′ / ₂ ′ = f ₁ ′ / f ₂ (17) for the approximate zero rank and thus δ ₂ f ₁ = δ ₁ f ₂ ; δ ₂ ′ f ₁ A study that follows ′ = δ ₁ ′ f ₂ ′ (18) is sufficient.

For the lens, the conversion from F ₁ to F ₂ is Then, we can derive one by keeping only the first rank terms in d and δ.

Similarly for lens L ' Substituting these expressions for Δ and Δ ′ into equation (11) and using the unambiguously derived expressions for P ₁₁ and P ₁₂ in equation (13) In general, then q and p are chosen such that equations (9) and (22) are simultaneously satisfied, meaning that the blur is absent in the image plane and the final beam is telecentric. be able to. Equation (22) can be simplified without changing the nature of this general result. For Einzel and gap lenses operating at voltage ratios of 10: 1 or higher, incident ion energy changes do not result in significant changes in principal plane position. In other words, the focal plane shift is primarily the result of changes in focal length. That is, diδi; di′δi ′ (23) By this approximation and the approximation of (18) in the equation, the approximate result is qp = f ₁ f ₂ (24) achromatic condition with telecentricity with respect to telecentricity condition. Regarding In practice a solution can be found for p if δ ₁ / f ₁ <δ ₁ ′ / f ₁ ′. That is, the first lens must have a smaller color effect than the second. Note that the choice of p and q to satisfy equations (24) and (25) does not affect the image or magnification conditions described above in any way. Furthermore, the last lens is the formula (2
4) It can be easily estimated to a size that does not affect the conditions expressed by (25). This is because δ ₁ ′ is proportional to f ₁ ′ regardless of the size of the lens. Estimating only the size of the last lens (scaling) provides a convenient way of varying the magnification for a column.

The effect of lens distortion will be examined next. As the ions pass through the round lens, third order aberrations occur. The focal plane-to-focal plane transformation, which includes these aberrations, is given by Have. Where A, B, C, D and a, b, c, d
Is a constant for a particular lens geometry and voltage ratio. In addition to the point source condition γ ₁ = pθ ₁ in equation (7),
Applying this expression to lens L results in the following for the wobble matrix Δ: become. Here, F (p) = Ap ³ + Bp ² + Cp + D (28) G (p) = ap ³ + bp ² + cp + d Coordinates of radiation at F ₁ ′ of lens L ′ are Given by.

Neglecting the induced aberrations greater than the third order, equations (7), (26) and (29) yield γ ₁ ′ (f ₁ −qp / f ₂ ) θ ₁ (30) θ ₁ ′ − (p / f ₂ ) θ ₁ .

Assuming that the telecentricity condition is almost satisfied, γ ₁ ′ is small and magnifications larger than γ ₁ ′ ² are ignored. Applying the equation (26) to the lens L ′, Is given.

The perturbation transformation on L'follows immediately Here, F '(p) = C ' (qp / f 2 -f 1) (p / f 2) + D '(p / f 2) 2 G' (p) = c '(qp / f 2 -f ₁ ) (p / f ₂ ) + d '(p / f ₂ ) ²
(33) Substituting the above Δ and Δ ′ into the equation (11) means that the total fluctuation transformation matrix P, Is given.

Thus, the perturbation Δγ in the image plane that causes geometrical aberrations
₂ ' Is.

The condition for an undistorted image is thus (f ₂ / p) G (p) = F ′ (p) / f ₁ ′ (36) and for almost telecentric states, the equation (3
3) means F '(p) D' (p / f ₂ ) ² (37), which is a condition without distortion due to telecentricity, G (p) / (D '/ f') ~ (p / f ₂ ) ² (38) is added.

According to equation (31), D'has a length dimension. Therefore, when the last lens is estimated in size,
D '/ f _1' is constant while the are and once satisfied is not obtained equation (38) is always satisfied regardless of the actual size of the last lens. As mentioned above, this is also the situation for the achromatic state.

Since G (p) is a polynomial in p, the actual solution of Eq. (38) for p is not guaranteed based on logic. However, the choice of lenses and their associated parameters, voltage ratios and magnitudes is acceptable practice as found for p and q which simultaneously satisfy the three conditions represented by (24) and (38). The value of σ s can be selected and thus ensure the effects of lens distortion and blurring as minimized in the telecentric beam.

According to these principles, the lens arrangement according to the invention, in which the first main lens is an accelerating Einzel lens and the second is a gap lens, together with other important characteristics of the described lens arrangement, are associated with blurring and bricks. Allows for substantially simultaneous minimization of strain and acquisition of near telecentricity. It is a special feature that the ion beam of such a device can have the desired low level energy in the mask and very high energy in the wafer. The accelerating Einzel lens is slightly more aberrated than the preceding gap lens in the design of FIG. In order to satisfy the achromatic δ ₁ / f ₁ <δ ₁ ′ / f ₁ ′, the gap lens can be selected to operate at a lower voltage ratio than the accelerating Einzel lens. However, in general, G (p)> D ' / f 1' at and formula (38) is liable to satisfy fairly large values of p for this arrangement.

The condition of perfect telecentricity is not a basic lithographic requirement. Allowing the condition that the beam is nearly telecentric rather than completely telecentric allows a great deal of range and flexibility in simultaneously meeting the conditions for achromatic, distortion-free transport.

Using the ion optics overview of the preferred embodiment shown in FIG. 2, the device can simultaneously have the following performance characteristics with respect to the fabrication of high quality microchips with submicron features.

1. at least 10mm reduced at a rate of at least 1.5: 1 or 2: 1 relative to the mask and adapted in the image plane
The formation of an image that is.

2. Image distortion is less than 0.2 micron.

3. The blur of the image due to the energy diffusion of ions is about 50 nm.

4. The device is almost telecentric in the image plane.

5. Initial energy of ions (on the mask surface) is 1 to 10 KeV
Is.

6. The final energy of the ions hitting the target is 50-200K
eV.

7. Machine dimensions are compatible with typical integrated circuit manufacturing equipment requirements due to the short conjugate length of the lens system.

Metrology Linear error such as x, y translation, Mx or
The My and Φ errors are detected by the alignment beamlets and the signal from them is used to apply a supplemental voltage to the corresponding ion optics in real time.

However, inevitably, imperfections in the physical construction of the machine lead to non-linear geometric distortion and blurring of the image.
From time to time, to compensate for these errors as much as possible,
The effect of various ion optical harameters on such distortions and blurs must be determined empirically so that the optical elements of the device can be set to correct.

To measure the error, the instrument uses a metrology mask with slits (10a) to create a pattern of beamlets.
(Fig.) And the detector slit devices S ₁ , S ₂ (Figs. 9 and 11a) attached to the precision stage 306 are placed in the measurement mode. The beamlets generated by the slitting device of the metrology mask are in the X direction or Y
Slits on the measurement stage by moving the stage in the same direction (with the detector 304 attached to the stage)
Interrupted by S ₁ or S ₂ . The momentum required to intercept successive beamlets determines the actual position of each beamlet and compares this position with the nominal position of the corresponding slit in the mask (taking into account the projection scale). This determines the geometric distortion error in each region of the beam field. Similarly, a given beamlet can be compared to the nominal beamlet width to determine the amount of blur in the region of the beam field by determining the momentum of the slot that continues to generate the detection signal. The width of the beamlet is determined. The necessary correction parameters can be derived from these error functions.

A preferred embodiment of using this device is described in detail below. The pattern generated by the measurement mask is (2n + 1) of X and Y ribbon shape measurement beamlets in the projection area.
^It consists of ² pairs (n = an integer sufficient to provide the desired resolution). The pattern in FIG. 10a is for n = 2. First
As can be seen in FIG. 0b, the beamlets of each pair are incoherent but close to each other to define a unique X, Y field point due to the intersection of their longitudinal projections. These beamlets, which are made to a die field by a special metrology mask and are projected through the column, are attenuated by the optical system. In the case of a factor of 4, the initial mask openings are spaced across a field of 10 mm x 10 mm with a reduction of 4: 1 across a field of approximately 40 mm x 40 mm and on the wafer.

The position and width of the center of gravity of each ribbon beamlet are set such that the secondary electrons supported by the measurement stage 306 (FIGS. 9 and 11a) attached to the X and Y stages 118 (FIG. 9) in sequence are collected. It may be measured using an absolute precision of 0.01 micron using a small detector 304. Beamlet slits 330 on stage
(S ₁ or S _{2 in} FIGS. 9b and 11a) and strikes on metal surface 308. Secondary electrons 31 emitted from the surface 308
A 0 is detected by Channer Lutron 304.

Referring to FIGS. 9a and 9b, the metrology mask (9a
The two vertical slits in the figure) are a pair of orthogonal beamlets 0 ₁
And 0 ₂ are formed. On the measuring stage (see Fig. 9b) two orthogonal detection slits above the detector mounted on the stage for receiving the imaged opposite beamlets.
S ₁ and S ₂ are provided. The sensing slits are arranged as shown so that only one of the imaged orthogonal beamlets can pass through each slit on the stage at a time. In this embodiment, beamlet 0 ₁ is centered perpendicular to the negative axis X and beamlet 0 ₂ is perpendicular to the positive Y axis before imaging as shown in Figure 9a. Be centered. After imaging each beamlet inversion, 0 ₂ is centered perpendicularly to the negative Y axis and 0 ₁ is centered perpendicularly to the positive X axis. Detection slit
S ₂ can be positioned coincident with beamlet 0 ₂ reversal and detection. The detection slit S ₁ is positioned away from the impinging beam 0 ₁ when 0 ₂ is being measured. This arrangement ensures a signal measured by the detector that only originates from 0 ₂ . Similarly, stage 306 can be moved to detect 0 ₁ without interference from 0 ₂ .

The results for image quality, blurring and distortion of various adjustable parameters of the ion projector can be plotted in a quantitative way using the measuring device described above. Sequentially this optimizes the image quality and substantially corrects some imperfections inherent in the physical structure of the machine and some types of non-linear errors that may be present in a patterned manufacturing mask. Parameters can be set to prescribed values.

There are a set of typical n adjustable parameters for the machine.

1. Source X position relative to the axis of the first primary lens.

2. The Y position of the source with respect to the axis of the first main lens.

3. X position of the solenoid lens with respect to the axis of the first main lens.

4. The Y position of the solenoid lens relative to the axis of the first main lens.

5. The X position of the second main lens with respect to the axis of the first main lens.

6. The Y position of the second main lens with respect to the axis of the first main lens.

7. X position of the mask with respect to the axis of the first main lens.

8. Y position of the mask with respect to the axis of the first main lens.

9. The current passing through the solenoid lens.

10. The voltage on each individual electrode of the multipole.

11. Fine adjustment on the voltage of the two main lenses.

12. The current passing through the current element of the magnetic shield.

13. The axial position of the image.

14. Spacing between the first and second main lens.

15. Extraction voltage from the ion source.

16. Selected source parameter (s) affecting ion beam energy spread
(Eg magnetic field at ion source, filament heating power or gas pressure).

In order to calculate the optimal settings for the adjustable parameters, it is first necessary to measure the respective tilt or "wedge" effect as shown below. The n parameters are set with nominal initial values Pj (j = 1,2, --- n). Using a metrology mask and a detector mounted on the stage and a quadrature slit as described above, the error Qi (X, Y) results in an orthogonal pair of beamlets projected from the mask onto the image plane (see Figure 10b). Measured at each field point (X, Y) defined by (as shown). The index i is related to the type of error.

For example, Q ₁ (X, Y) = deviation of X from position X Q ₂ (X, Y) = deviation of Y from position Y Q ₃ (X, Y) = blur of X of image beamlet Q ₄ ( X, Y) = one of the Y blur parameters Pj of the image beamlet is then changed by the amount ΔPj and the set of unchanged error functions Qji (X, Y) is measured by the measuring device. The wedge function is simply the difference between the previous error Qi and the adjusted error Qji of the parameter Pj divided by the magnitude of the adjustment ΔPj.

That is, Sji (X, Y) = [Qji (X, Y) −Q ₁ (X, Y)] / ΔPj The parameter j is returned to its initial value Pj and the measurement process is performed by the wedge function Sji for all parameters j = 1, 2, ---
-Iterate until determined for n.

To the extent that the perturbation on the error function is linearly dependent on the adjustable parameter, usually the shift ΔPj of the value of Pj for small adjustments, the error function resulting from the general setting of the shift is Qi (X, Y) = It is given by Qi (X, Y) + ΣjΔPjSji (X, Y).

Known techniques in numerical analysis (eg, 1981, 10
033 New York, 5th Avenue 111 Academic
Using the practical optimization technology (practical optimization technics) by P.G.
It can be determined by the computer equipment of the machine that minimizes some function of Y). For example, a typical optimization sets a limit L ₁ for each type of error i and minimizes the difference between the maximum error QMAX observed over the image field and the limit L ₁ . That is, | QMAX ₁ −L ₁ | is the minimum, where QMAX ₁ = MAX [| Qi
(X, Y) |] (for all X, Y) could be optimized. In addition, the optimization is a parameter value
The actual constraints on Pj and / or error Q ₁ can be obeyed. Typical useful constraints can limit the range of each parameter value. That is, | Pj | PMAXj where Pj = Pj + ΔPj and PMAXj are the maximum allowable values of Pj. If it is desired to correct known mask errors simultaneously, QMASK
₁ (X, Y) then | QMAXi-L | is minimized.

Now, QMAXi = MAX [Qi (X, Y) + QMASK ₁ (X, Y)].

The great utility of the measuring device and the optimization method is that the wedge function Sji (X, Y) is generally quite stable over time and only occasionally measured. On the other hand, a special combination of the wedge functions (and hence the parameters Pj) is also commonly used to obtain optimum performance when the changes occur in the physical environment of the ion projector, eg changes in temperature or pressure. Change. With the aid of an appropriate computer program, the given change is the error Q ₁ (X,
Once Y) is measured, it is quickly calculated by the technique described above.

In the preferred embodiment, the metrology sensing device is permanently attached to the stage 40 shown in FIG. 2 or to the stage 118 as shown in FIGS. 11a and 11b. In that case it can be used to quickly measure any existing error Q ₁ (X, Y) whose tunable parameters can then be optimized.

Since the measurement needs to be taken over the size of the image field, it is typically no larger than 20 × 20 mm and the X, Y positions of slits S ₁ and S ₂ are laser interferometer 33
It can be measured using two and relatively small orthogonal mirrors 314. The X, Y positions of the measurement slits S ₁ and S ₂ are associated with the column via a laser mirror 316 and rigidly connected interferometers 116 and 332. The beam is also maintained in a stable condition at the end of the column by using the reference ring 300 shown in FIG. 9 and the scanning and sensing device described above. The ring is rigidly attached to the end of the column and has alignment marks 302.
The scanning and deflecting device senses these marks and maintains the position of the exposure in relation to them.

In the arrangement in which the alignment apparatus described above is used, the rough X, Y stage 118 shown in FIGS. 11a and 11b is not needed for laser interferometer position control to create an image on the wafer 248. The positional accuracy required for the measurement detection slit is achieved by mounting the slits S ₁ and S ₂ and the mirror 314 on a separate limited range stage 306.

Referring to FIGS. 11a and 11b, measurement stage 306
Can be mounted on a crude X, Y stage 118 and moved very quickly within plus or minus 5 microns of the beamlet position. The metrology stage itself, with the detector cage, is equipped with a transducer, such as a piezo-electric transducer that can move the detector in the X and Y directions by plus or minus 15 microns. Once the X, Y stage moves the metrology stage to one approximate position of the projection beamlet, the metrology stage is moved to find the exact center and width of the beamlet.

The crude X, Y stage 118 is mounted in a passage 370 that allows the use of a large block 374, eg, about 9 inches square, into which a wafer 248 can be inserted from the back side. The movements required to expose the wafer are achieved, for example, by a dc (direct current motor) under control, for example consisting of a glass ruler or two-axis laser interferometry. The stage can be stepped to any position within this 9x9 inch area with good accuracy within plus or minus 5 microns. When the mirror is attached to the stage, the position can then be set within a fraction of a micron in combination with the laser interferometer.

The usefulness of the metrology tool depends on the accuracy of the X, Y positioning of the detector slits and hence the accuracy of the stage 306. First
In the arrangement shown in FIGS. 1a and 11b, the stage can be precisely and calibrated by pressing the measurement fiducial markers on the wafer 248 at various X, Y values determined by the laser interferometer. .. The position of these markers on the wafer can be independently verified using off-line metrology tools such as Nikon 2I optics or electron beam tools.

[Brief description of drawings]

1 is a plan view of an ion beam lithography apparatus of a preferred embodiment, FIG. 1a is a front view of the apparatus of FIG. 1, FIG. 1b is a schematic side view of a magnetic shield of the apparatus of FIG. 1, and 1c. 1 is a cross-sectional view taken along the line 1c-1c in FIG. 1, FIG. 2 is a three-dimensional cutaway view of the apparatus of FIGS. 1 and 1a showing the ion beam trajectories and internal components of the apparatus, and FIG. 2a is 2 is an enlarged view of the column area shown in FIG. 2, FIG. 2b is an enlarged schematic view of the alignment beam scanner and detection device of FIG. 1, and FIG. 2c is a block schematic diagram of the power supply used by the embodiment of FIG. 2d is a schematic perspective view of the multipole array used in FIG. 1, FIG. 2e is a block diagram of the alignment device according to the present invention, and FIG. 3 is agitated by the presence of lens distortion and / or color blur. Two-lens optics showing active ion beam radiation Schematic diagram, FIG. 4 is a schematic diagram showing mass selection and actuation of a solenoid for locating a virtual source point, FIG. 5 is a schematic diagram of the mask carousel and beam shutter from the embodiment of FIG. 1, FIG. 6 is a schematic diagram of barrel and pincushion distortion correction achieved by a two lens system, and FIG. 7 shows the formation of parallel ion paths in the region between the lenses and the telecentric beam after the second lens according to the invention. FIG. 8 is a schematic diagram showing creation, FIG. 8 is a graphical illustration showing lens distortion and linewidth spread from color blur plotted as a function of distance between lenses for two source / mask distances, and FIG. 9 shows measurements. FIG. 9a is a schematic view of the target area of the present invention by means of an apparatus in mode, FIG. 9a is a schematic view of the opening of the metrology mask for forming the beamlets used in the metrology mode, 9b is a schematic view of a slit of a measuring stage for measuring a beamlet formed by a mask, FIG. 10a is an explanatory view of an image pattern used for measurement by an apparatus in a measuring mode, and FIG. 10b is a view of FIG. 10a. An enlarged view of the components of the image pattern, FIG. 11a is a top view of the target area and the metrology stage, and FIG. 11b is a sectional view taken along line AA of FIG. 11a. In the figure, reference numeral 10 is an enclosure, 12 is an ion source, 14 is an optical column, 18 is a solenoid, 20 is a mask structure, 22 is an Einzel lens, 24 is a gap lens, 26 is a target, 28 and 34 are multipole bodies, 30 is an aperture, 32 is an alignment beam scanner and detector, 38 is a shutter, 40 is a stage, 42 is a chuck, 46 and 48 are vacuum locks, 50 is a vacuum pump, 52 and 54 are bellows, 66 is a power supply, and 68 is power. Distribution panel, 70 service module, 72 vacuum device, 76 console, 150 extraction electrode, 152 suppression electrode, 164 mask, 168 cylinder, 170 coil, 176 first electrode, 178
Is a third electrode, 180 is a central electrode, 192 is a scanning plate, 194 is a mask, 196 is a detector, 210 is a control aperture, 244 is a beam, 248 is a wafer, 300 is an alignment ring, 302 is a mark,
306 is a measurement stage, and 460 is a correction signal generator.

 ─────────────────────────────────────────────────── ─── Continued Front Page (72) Inventor Gerhard Stengle Austria, Carinthia, A-9241 Bernberg, Umberg 41 (72) Inventor Hilton F. Gravish USA, Massachusetts 01970, Salem, 10 Norman Street (56) References JP-A-2-3062 (JP, A) JP-A-64-64216 (JP, A)

Claims (49)

[Claims] 1. A means for supplying an ion beam, a mask in the path of the ion beam having an opening for producing a desired beam pattern, an optical column arranged behind the mask, along the ion path. An optical column defined by first and second main lenses arranged in parallel, the first main lens being an acceleration Einzel lens arranged to form a crossover in the optical column, the crossover A second main lens, which is a gap lens positioned behind and arranged to project a reduced image of the mask, and arranged behind the gap lens for supporting a target receiving the image. A projection lithographic device, which comprises a target station 2. The mask is located substantially at a first focal plane of the first main lens and the target station is located substantially at a second focal plane of the second main lens. 1. The ion projection lithographic apparatus according to 1. 3. The means for supplying the ion beam comprises means for supplying a virtual image of the ion source to an ion source and an optical column remote from the ion source, and (1) the first
The distance between the second focal plane of the main lens and the first focal plane of the second main lens, and (2) the distance between the mask and the ion source and the point defining the virtual image of the mask are 7. The target selected to generate the image of the mask, wherein blurring and geometric distortions emanating from the first and second primary lenses are substantially minimized at the same time. 2. The ion projection lithographic apparatus according to 2. 4. The first and second main lenses cross the first main lens substantially at a first focal plane of the second main lens to generate a telecentric beam following the second main lens. An ion projection lithographic apparatus according to claim 3, wherein the ion projection lithographic apparatus is positioned to position the overcoat. 5. The first and second main lenses simultaneously satisfy substantially the following conditions such that the beam emerging from the second main lens is substantially telecentric and when it reaches the image plane is substantially: No blurring and geometric distortion, ie (1) qp = f ₁ f ₂ Here, q is the distance between the second focal plane of the first main lens and the first focal plane of the second main lens, and p is the distance between the point defining the virtual image of the ion source and the mask. , F ₁ is the first focal length of the first main lens, f ₁ ′ is the first focal length of the second main lens, f ₂ is the second focal length of the first main lens, and δ ₁ and δ ₁ ′ are , G (p) = ap ³ + bp ² + cp + d, where a, b, c, and d are the changes in the first focal lengths of the first and second main lenses due to changes in the energy of the ions in the beam. In the presence of third-order aberrations,
A lens constant associated with a conversion function from the first focal plane to the second focal plane of the first main lens, said conversion function being: Θ ₂ = −r ₁ / f ₂ + ar ³ ₁ + br ² ₁ Θ ₁ + cr ₁ Θ ² ₁ + d Θ ³ ₁ where Θ ₂ is the transformed lateral angular coordinate of the constant ion radiation at the second focal plane of the first main lens, r ₁ , Θ
₁ is the abscissa at the first focal plane of the first main lens in the mask and D'is the conversion factor from the first focal plane to the second focal plane of the second main lens, r _{2 ^{'~f 1' Θ 3 2 +}} D'Θ 2 where, r ₂ 'ion projection according to claim 1, characterized in that the radial coordinate of the radiation at the second focal plane of the second main lens Lithographic equipment. 6. Ion projection lithographic apparatus according to claim 1, characterized in that the ions of the ion beam are supplied at the target with an energy of between approximately 50 and 200 kV. 7. The ion projection lithographic apparatus according to claim 6, wherein the ions in the ion beam are supplied to the mask at an energy between about 1 and 10 kV. 8. The image of the mask at the target is reduced at a rate of at least 1.5, and the image at the target is not less than 10 mm in width and height. The ion projection lithographic apparatus according to item 1. 9. The ion projection lithographic apparatus further comprises a pair of electric field control apertures, one of the control apertures being in the area around the mask and around the opening of the third electrode of the Einzel lens. Ion projection according to any one of claims 1 to 5, characterized in that it is arranged on the side of the second electrode of the Einzel lens in order to reduce the electric field strength generated from the second electrode in the area of. Lithographic equipment. 10. The ion projection lithography apparatus according to claim 1, wherein a voltage ratio in the range of 7 to 20 is applied to the first and second main lenses. 11. The means for delivering the ion beam comprises an ion source and a lens disposed between the ion source and the mask, the lens simultaneously minimizing blurring and lens distortion. Ion projection according to any one of the preceding claims, characterized in that it is arranged to select the actual ion source point along the axis of the optical column for fine tuning of the device for this purpose. Lithographic equipment. 12. The lens is a solenoid lens,
The solenoid lens is adapted to contribute to selection of a desired ion type from different types of different masses emitted from the ion source by different types of different deflections. The described ion projection lithographic apparatus. 13. The ion projection lithographic apparatus further comprises a mass selective aperture located between the first and second main lenses to prevent passage of ions of different mass from the desired type. 13. The ion projection lithographic apparatus according to claim 12, which is characterized in that. 14. The ion projection lithographic apparatus comprises a Y
Multipole means arranged in or beyond the second main lens with respect to a quadrupole field for varying the magnification of the image at the target in the X direction relative to the magnification of the image in the X direction. The ion projection lithography apparatus according to any one of claims 1 to 5, further comprising: 15. The ion projection lithographic apparatus according to claim 14, wherein the multipole means is disposed behind the second main lens in a region substantially free of an electric field. 16. An arrangement between the first and second main lenses further adapted to apply a dipole electric field for adjustable movement of the image in a plane perpendicular to the axis of the optical column. 6. An ion projection lithographic apparatus according to any one of claims 1 to 5, characterized in that it comprises a multi-pole means. 17. The ion projection system according to claim 16, wherein said multipole means is further adapted to generate a superimposed high order electric field for controlling distortion and blurring in said image. Lithographic equipment. 18. The one of the arc-shaped electrodes, wherein said multipole means is adapted to introduce an offset parallel to said beam.
17. The ion projection lithographic apparatus of claim 16, which is a circular array of 16 continuous poles in pairs. 19. The ion projection lithographic apparatus further comprises a voltage adjusting device for changing a lens voltage ratio for finely tuning a balance of distortion and blurring in an image. And the ion projection lithographic apparatus according to claim 10. 20. The ion projection lithography of claim 19, wherein the voltage regulator is further capable of independent adjustment of the lens voltage ratio to tune the magnification of the image at the target. A device. 21. The voltage of the first electrode of the gap lens is substantially the same as the voltage of the first and third electrodes of the Einzel lens and the voltage of the second electrode of the Gap lens is of the Einzel lens. 2. The method according to claim 1, characterized in that it is substantially identical to the second electrode, so that the ripple effect of the power supply does not substantially change the voltage ratio of the optical column and thus substantially retains the image quality at the target station. 5. The ion projection lithographic apparatus according to any one of 5 above. 22. The diameter of the gap lens is configured to be variable to change the magnification of the image at the target, and the beam exiting the second main lens remains substantially telecentric during the change. The ion projection lithographic apparatus according to any one of claims 1 to 5, characterized in that there is substantially no chromatic aberration or geometric distortion in the image plane. 23. The optical column includes an outer, rigid metal shell substantially extending the entire length of the optical column, the metal shell being at a constant potential and the first and second Einzel lenses. Directly supporting the third electrode and the first electrode of the gap lens, the center electrode of the Einzel lens is supported by an insulator engaged for support on the inside of the metal shell and the first electrode of the gap lens. The ion projection lithography apparatus according to any one of claims 1 to 5, wherein the two electrodes are supported via an insulating bush supported by a downstream end of the rigid shell. 24. Ion projection according to claim 23, characterized in that a tubular shield made of a magnetic material of high permeability extends around and is supported by the rigid metal shell. Lithographic equipment. 25. The tubular shield comprises a plurality of plates (34
25. An ion projection lithographic apparatus according to claim 24, characterized in that it comprises a removable longitudinal segment consisting of 4). 26. The end plate of high permeability magnetic material is attached to each end of the tubular shield in such a manner as to provide magnetic continuity with the tubular shield. The described ion projection lithographic apparatus. 27. Means for supplying an ion beam, a mask in the path of the ion beam having an opening for producing a desired beam pattern, an optical column arranged behind the mask, along the ion path An optical column defined by first and second main lenses arranged in an array, the first main lens arranged to focus the ion beam, positioned behind the first main lens and The second arranged to project a reduced image of the mask;
A main lens, a target for receiving the image, and a target station behind the second main lens for supporting a shield made of a magnetic material of high permeability extending around the optical column, the shield comprising: A removable longitudinal segment comprising a plurality of plates (344) defining the tubular shield and an end plate (350, 352) of a magnetic material of high permeability for magnetic continuity with the tubular shield. An ion projection lithographic apparatus characterized in that 28. The tubular shield comprises a set of longitudinally-extending circumferentially spaced ribs mounted around the optical column, and a gap between and adjacent to the ribs. 28. An ion projection lithographic apparatus according to claim 27, characterized in that it is defined by a plurality of removable longitudinally extending plates (344) spread in a continuous sequence. 29. The tubular shield includes a conductor within the shield arranged to provide conditioning for the magnetic field to which the ion beam is exposed.
28. The ion projection lithographic apparatus according to any one of 28. 30. The ion projection lithographic apparatus of claim 29, wherein the conductor comprises an array of elongated loops within the tubular shield. 31. An array of longitudinally extending conductive loops is associated with the tubular shield, the loops being arranged to establish a magnetic flux circumferentially within the tubular shield. 24. The ion projection lithographic apparatus described in 24. 32. Means for applying an alternating current to the loop so as to demagnetize the tubular shield and for temporarily applying a small bias current to the loop so as to increase the permeability of the tubular shield. 32. An ion projection lithography apparatus according to claim 31, characterized in that it comprises means. 33. The ion projection lithographic apparatus of claim 31, wherein each of the conductive loops comprises a longitudinally extending conductive segment extending along the interior and exterior of the tubular shield. 34. The ion projection lithography apparatus of claim 31, wherein the tubular shield includes a conductor arranged to provide conditioning to the magnetic field to which the ion beam is exposed. 35. Means for supplying an ion beam, a mask in the path of the ion beam having an aperture for producing a desired beam pattern, an optical column disposed behind the mask, along the ion path. An optical column defined by first and second main lenses arranged in an array, the first main lens arranged to focus the ion beam, positioned and contracted behind the first main lens The second arranged to project an image of the mask
A main lens, a target station located behind the second main lens for supporting the image receiving target, wherein the means for supplying the ion beam is actually along the axis of the ion source and the optical column. A solenoid lens disposed between the ion source and the mask to select an ion source point of the solenoid lens, the solenoid lens having different masses of the different masses generated from the ion source by different types of different deflections. An ion projection lithography apparatus, characterized in that it contributes to the selection of a desired ion type from the types. 36. The dual solenoid lens of claim 35, wherein said solenoid lens is a counter-turn double solenoid lens intended to prevent rotation of said ion beam during passage of said solenoid lens. Ion projection lithographic equipment. 37. The ion projection lithographic apparatus comprises an aperture located between the first and second main lenses sized to prevent passage of ions of different mass from a desired type. The ion projection lithography apparatus according to claim 35 or 36. 38. Means for supplying an ion beam, a mask in the path of the ion beam having an aperture for generating a desired beam pattern, an optical column arranged behind the mask, along the ion path. An optical column defined by first and second main lenses arranged in an array, the first main lens arranged to focus the ion beam, positioned and contracted behind the first main lens The second arranged to project an image of the mask
A main lens, a target station located behind the gap lens for supporting the image-receiving target, and a multipole means arranged in or beyond the second main lens and in the Y direction. Ion projection lithographic apparatus comprising a voltage controller adapted to apply a quadrupole field to the multipole means to change the magnification of the image at the target in the X direction relative to the magnification of the image. . 39. An ion projection lithographic apparatus according to claim 38, wherein said multipole means is arranged in a region substantially free of electric field and behind said second main lens. 40. An ion projection lithography apparatus according to claim 38 or 39, wherein said multipole means is a 16-pole circular array of arc-shaped electrodes. 41. A lithographic method for producing a surface having a size of about 0.1 micron on a target, wherein an ion projection lithographic apparatus capable of forming desired features of a lithographic mask on a target field is provided. The instrument has a set of adjustable parameters and uses a metrology mask with an array of metrology planes to empirically determine the effect of each parameter across the target field, while the other parameters are set to a set of wedges. Set constant to provide a function, setting parameters of the device by linear optimization across the target field, which is based on metrology measurements, periodically interrupts the exposure, the lithographic mask to the target Substituting a metrology mask defining an array of metrology planes across the field, determining a set of error values from the measurements, and A lithographic method, characterized in that the optimization determines an adjustment to the set of parameters to reduce the error value on the basis of the wedge function, makes the adjustment, and restarts the exposure of the target. . 42. The metrology mask defines an array of orthogonal slit pairs to generate corresponding pairs of metrology ion beamlets prior to the target and is empirically based on the action of each parameter across the target field. Ion projection lithography method according to claim 41, characterized in that the other parameters are held constant while the width is measured and the center of gravity of the beamlet at the target is determined. . 43. Having a set of adjustable parameters,
An image of the lithographic mask is projected onto a target mounted on the stage in the X and Y directions.
In an ion beam projection lithographic apparatus in which a Y-direction stage is arranged to index across a large number of exposure positions, a measurement mask adapted to replace the lithographic mask, and a precision attached to the X- and Y-direction stages. Measuring stage, determining an error value based on the sensed blur and geometric distortion at the point where the metrology mask is adapted to index across the ion beam field while the metrology mask is distributed across the ion beam field. With a detector attached to the measurement stage, the error value being useful for performing a correction to the set of parameters by linear optimization to reduce blur and geometric distortion of the device. An ion beam projection lithographic apparatus characterized in that 44. The metrology mask defines an array of orthogonal slit pairs for generating a corresponding pair of metrology ion beamlets preceding the target and a detector mounted on the precision metrology stage measures the width and 44. The ion projection lithography apparatus according to claim 43, wherein the position of the center of gravity of the beamlet at the detector is determined. 45. The detector is arranged to block the measurement beamlet in such a way that only one beamlet of an orthogonal pair hits the corresponding slit at one time.
44. The ion projection lithographic apparatus according to claim 43, which is associated with a pair of orthogonal slits. 46. A means for supplying an ion beam, a mask in an ion beam passage having an opening for generating a desired beam pattern, an optical column arranged behind the mask, and an ion beam passage. An optical column defined by first and second main lenses disposed alongside, a target station behind the second main lens for supporting a target receiving the image, the mask being desired In addition to a mask structure for defining the image beam of the target beam, the target station brings the image beam in alignment with a desired position on the wafer. A reference mark imaged by the optical column on the reference beam to generate an alignment control signal. For example, it has been deflected means interruption from the image beam ion projection lithographic apparatus, characterized in that provided in order to repeatedly scanning said reference beam across a reference mark of each of the target station. 47. An ion projection lithography apparatus according to claim 46, wherein said deflecting means is associated with each reference beam. 48. The deflecting means comprises an electrostatic deflector and the image beam passes through a tubular shield while the reference beam passes outside the shield and the shield directs the image beam from the electric field of the deflector. 48. The ion projection lithographic apparatus according to claim 47, which is protected. 49. The first main lens is an acceleration Einzel lens arranged to form a crossover in the optical column, and the second main lens is positioned behind the crossover and 49. The ion projection lithographic apparatus according to claim 46, which is a gap lens arranged to project a reduced image of the mask.