JP2018519535A - Method for operating a microlithographic projection apparatus - Google Patents

Abstract

A method for operating a microlithographic projection apparatus includes: an image of an object field (14) illuminated on a mask (16) in a mask plane, an illumination system (12) and a mask (16) on a photosensitive surface (22). Providing a projection objective (20) configured to form on a positioned image field. Edge placement errors are determined at different field points within the image field. The mask (16) is then illuminated with projection light having an improved field dependence of the angular illumination distribution. The angular illumination distribution according to the improved field dependence varies across the object field so that the edge placement error determined in step b) is reduced at different field points.
[Selection] Figure 18

Description

  The present invention relates generally to the field of microlithography, and more specifically to an illumination system used in a projection exposure apparatus or mask inspection apparatus. The present invention is particularly relevant to correcting edge placement errors (EPE) that represent differences in desired and actual feature edge locations in the image plane of the objective at the wave level.

  Microlithography (also called photolithography or simply lithography) is a technique for the fabrication of integrated circuits, liquid crystal displays, and other microstructured devices. A microlithographic process is used in conjunction with an etching process to pattern features in a thin film stack formed on a substrate, eg, a silicon wafer. In each layer of fabrication, the wafer is first coated with a photoresist, a material that is sensitive to radiation, such as deep ultraviolet (DUV) light. Next, the wafer having the photoresist thereon is exposed to projection light in a projection exposure apparatus. The apparatus projects a mask containing the pattern onto the photoresist such that the photoresist is exposed only at certain locations determined by the mask pattern. After exposure, the photoresist is developed to produce an image corresponding to the mask pattern. An etching process then transfers the pattern into the thin film stack on the wafer. Finally, the photoresist is removed. Repeating this process with various masks results in a multi-layered microstructured component.

  A projection exposure apparatus is typically coated with a light source, an illumination system that illuminates the mask with the projection light generated thereby, a mask stage for aligning the mask, a projection objective, and a photoresist. And a wafer alignment table for aligning the wafers. The illumination system illuminates a field on the mask, which can have, for example, a rectangular or curved slit shape.

  In current projection exposure apparatus, a distinction can be made between two different types of apparatus. In one type, each target portion on the wafer is irradiated by exposing the entire mask pattern onto the target portion in one shot. Such an apparatus is commonly referred to as a wafer stepper. In the other type of device, commonly referred to as a step-and-scan device or scanner, each target portion gradually scans the mask pattern in the scan direction under the projection beam, while synchronizing the substrate in this direction. Irradiated by moving in parallel or antiparallel. The ratio of the speed of the wafer to the speed of the mask is usually less than 1 and equal to the magnification of the projection objective, for example 1: 4.

  It should be understood that the term “mask” (or reticle) must be interpreted broadly as a patterning means. Commonly used masks include opaque or reflective patterns and can be, for example, binary, alternating phase shift, attenuated phase shift, or various hybrid mask types. However, there are also active masks, for example masks achieved as programmable mirror arrays. In addition, a programmable LCD array can be used as an active mask.

  As technology for manufacturing microstructured devices advances, there is a constantly increasing demand for illumination systems. Ideally, the illumination system illuminates each point of the illumination field on the mask with projection light having a well-defined spatial illumination distribution and angular illumination distribution. The term angular illumination distribution represents how the total light energy of a light beam that converges towards a particular point in the mask plane is distributed among the different directions of the light rays that make up the light beam.

  The angular illumination distribution of the projection light incident on the mask is usually adapted to the type of pattern projected onto the photoresist. In many cases, the optimal angular illumination distribution will depend on the size, orientation, and pitch of the features contained in the pattern. The most commonly used projection light angle illumination distributions are referred to as conventional illumination settings, annular illumination settings, dipole illumination settings, and quadrupole illumination settings. These terms refer to the illumination distribution in the pupil plane of the illumination system. For example, using an annular illumination setting, only the annular region is illuminated in the pupil plane. Accordingly, there is only a small angle range in the angular illumination distribution of the projection light, and all light rays are incident obliquely on the mask at a similar angle.

  Various means are known in the art for modifying the angular illumination distribution of the projection light in the mask plane to achieve the desired illumination setting. In the simplest case, a diaphragm (diaphragm) including one or more apertures is arranged in the pupil plane of the illumination system. Since the location in the pupil plane is converted to an angle in a Fourier-related field plane such as the mask plane, the size, shape, and location of the aperture in the pupil plane determines the angular illumination distribution in the mask plane. However, any change in lighting settings requires a change of aperture. The replacement of the diaphragm will require a very large number of diaphragms with apertures having slightly different sizes, shapes, or locations, making it difficult to fine tune the lighting settings. In addition, the use of an aperture inevitably results in a loss of light and thus a reduction in device throughput.

  Accordingly, many common illumination systems include adjustable elements that allow the pupil plane illumination to be continuously changed to at least some extent. Many illumination systems use interchangeable diffractive optical elements to produce the desired spatial illumination distribution in the pupil plane. This spatial illumination distribution can be adjusted when a pair of zoom optics and axicon elements is positioned between the diffractive optical element and the pupil plane.

  Recently, it has been proposed to use a mirror array that illuminates the pupil plane. In EP 1 262 836 A1, the mirror array is achieved as a microelectromechanical system (MEMS) comprising more than 1000 micromirrors. Each of the mirrors can be tilted in two different planes perpendicular to each other. Thus, radiation incident on such a mirror device can be reflected in (substantially) any desired hemispherical direction. A condenser lens located between the mirror array and the pupil plane converts the reflection angle generated by the mirror into a location in the pupil plane. This known illumination system makes it possible to illuminate the pupil plane with a plurality of spots, each associated with one particular micromirror, which can be moved freely across the pupil plane by tilting the mirror.

  Similar illumination systems are known from US 2006/0087634 A1, US 7,061,582 B2, WO 2005/026843 A2, and WO 2010/006687 A1. US 2010/0157269 A1 discloses an illumination system in which a micromirror array is imaged directly onto a mask.

  As described above, it is usually desirable to illuminate all points on the mask so that they have the same illumination distribution and angular illumination distribution at least after scan integration. When multiple points on the mask are illuminated with different illumination, this typically results in an undesirable change in critical dimension (CD) on the wafer level. For example, in the presence of an illumination change, an image of a uniform line on the photoconductor on the mask may have an illumination change along its length. Because of the fixed exposure threshold of the resist, such illumination changes are directly translated into structure width changes that will be defined by the line image.

  If the angular illumination distribution changes unintentionally across the illumination field on the mask, it also has an adverse effect on the quality of the image produced on the photosensitive surface. For example, if the angular illumination distribution is not perfectly balanced, i.e., when more light is incident on one side of the mask than from the other side, the photosensitive surface is completely aligned with the projection objective. If it is not arranged on the focal plane, the conjugate image point on the photosensitive surface is shifted in the horizontal direction.

  US 6,404,499 A and US 2006/0244941 A1 are arranged in parallel and aligned parallel to the scanning direction to modify the spatial illumination distribution within the illumination field (ie, the field dependence of the illumination). A mechanical device is proposed that includes two opposing arrays of opaque finger diaphragm elements. Each pair of opposing diaphragm elements can be displaced along the scanning direction such that the distance between the opposing ends of these diaphragm elements is changed. When this device is placed in the field plane of the illumination system that is imaged on the mask by the objective system, it produces a slit illumination field whose width along the scan direction can vary along the scan orthogonal direction can do. Since irradiation is integrated during the scanning process, integrated irradiation (sometimes referred to as illumination dose) can be fine-tuned for a plurality of scan orthogonal positions within the illumination field.

  Unfortunately, these devices are mechanically very complex and more expensive. This is also due to the fact that these devices usually have to be placed in or close to the field plane where the blades of the movable field stop are located.

  It is more difficult to adjust the angular illumination distribution in a visual field dependent manner. This is mainly because the spatial illumination distribution is a function of only the spatial coordinates x and y, whereas the angular illumination distribution also depends on the angles α and β.

  WO 2012/100791 A1 discloses an illumination system in which a mirror array is used to generate a desired illumination distribution in the pupil plane of the illumination system. A fly-eye optical integrator having a plurality of light incident facets is disposed in the immediate vicinity of the pupil plane. Images of the light incident facets are superimposed on the mask. The light spot generated by the mirror array has an area that is at least five times smaller than the total area of the light incident facet. This makes it possible to generate a variable light pattern on the light entrance facet and thus to generate different angular illumination distributions in different parts of the illumination field. For example, an X dipole illumination setting may be generated for one part of the illumination field and a Y dipole illumination setting for another part of the illumination field.

  WO 2012/028158 A1 discloses an illumination system in which the illumination distribution on the light entrance facet of a fly-eye optical integrator is modified using a plurality of modulator units arranged in front of the optical integrator. Each modulator unit is associated with one of the light incident facets and redistributes the spatial and / or angular illumination distribution on the associated light incident facets variably without blocking any light. In this way, for example, two or more different parts can be illuminated with different illumination settings on a single die associated with different semiconductor devices.

  The unpublished patent application PCT / EP2014 / 003049 discloses a technique in which the illumination distribution on the light entrance facets of a fly-eye optical integrator is modified by imaging a digital mirror device (DMD) onto these light entrance facets. doing. This approach is advantageous in that it is not necessary to generate very small light spots using an analog micromirror array, as is the case with the illumination system known from WO 2012/100791 A1 mentioned above. The field dependence of the angular illumination distribution is adjusted so that the angular illumination distribution is completely uniform (ie, field independent) across the illumination field.

  However, it is sometimes mentioned that it may be desirable to intentionally introduce the field dependence of the angular illumination distribution. This introduction can be suitable, for example, when the projection objective has field-dependent properties. As far as masks are concerned, such field-dependent properties are usually the result of features having various orientations or dimensions. The adverse effects resulting from such visual field dependence can be successfully reduced by selectively introducing the visual field dependence of the angular illumination distribution.

  The industry that uses microlithographic projection apparatus for the manufacture of integrated circuits and other electronic or micromechanical devices is continually pursuing smaller feature sizes, higher output, and higher yields. One of the very important objectives is to reduce edge placement error (EPE). The edge placement error is between the actual (or simulated) contour location of the lithographically defined structure on the wafer (or similar support) on the one hand and the desired contour location on the other hand. It represents the difference. Edge placement error is a basic quantity that determines other common quantities such as critical dimension (CD) and overlay error. The reduction in edge placement errors directly results in higher yields and / or smaller feature sizes.

  Figures 16a, 16b and 16c show how edge placement errors are usually calculated. In the upper half of each figure, a target structure ST having a desired contour is shown. In the lower half, the rectangle indicated by the solid line represents the actual structure ST 'produced on the wafer in the microlithography process.

In the case shown in FIG. 16a, the actual structure ST ′ is wider than the target structure ST. It edges extending along the longitudinal direction of the structure ST ', only the positive edge placement error E = d _m -d _t is displaced, in this case, d _m is the measured distance from the symmetry line, d _t is the target distance from the symmetry line.

  As shown in FIG. 16b, when the target distance is equal to the measurement distance, the edge arrangement error E is zero.

If the measured distance d _m is smaller than the target distance d _t, the image placement error E becomes negative as shown it in Figure 16c.

EP 1 262 836 A1 US 2006/0087634 A1 US 7,061,582 B2 WO 2005/026843 A2 WO 2010/006687 A1 US 2010/0157269 A1 US 6,404,499 A US 2006/0244941 A1 WO 2012/100791 A1 WO 2012/028158 A1 PCT / EP2014 / 003049 US 2009/0116093 A1 WO 2009/100856 A1

E. Delano's paper "Primary Design and y-y Diagram (First-order Design and they, -y Diagram)", Applied Optics, 1963, Vol. 12, No. 1251, 1256

  An object of the present invention is to provide a method of operating a microlithographic projection apparatus that makes it possible to reduce edge placement errors.

  According to the invention, the above object is achieved by a method in which in step a) a mask, an illumination system configured to illuminate the mask, and a projection objective are provided. The projection objective is configured to form an image of the object field illuminated within the mask plane on the mask on an image field positioned on a photosensitive surface such as a resist or the like or, in the case of a mask inspection apparatus, a CCD sensor. The

  In the next step b), the edge placement error is determined at different field points within the image field. This stage can be achieved by measurement or simulation.

  In the last step c), the mask is illuminated with projection light having an improved field dependence of the angular illumination distribution. The angular illumination distribution according to the improved field dependence varies across the object field so that the edge placement error determined in step b) is reduced.

  It is well known in the art that angular illumination distribution has an effect on edge placement errors, but it determines the field dependence of edge placement errors and these edge placement errors are field dependent. It has never been proposed to generate a field-dependent angular illumination distribution in the illumination field that is determined to be reduced in a manner.

  The edge placement error determined in step b) may include at least one of the group consisting of CD variation and overlay variation.

  Once the edge placement error is determined in step b), the mask can be illuminated with projection light having an original field dependence of the angular illumination distribution. The edge placement error on the photosensitive surface is then simulated or measured at different field points within the image field. Then, in step c), the visual field dependence of the original angular illumination distribution can be changed in order to achieve an improved visual field dependence of the angular illumination distribution. These steps can be repeated once or several times. This means that the improved visual field dependency of the angular illumination distribution becomes the original visual field dependency of the angular illumination distribution in the next decision stage. In this way, the field dependence of the angular illumination distribution can be recursively improved until the edge placement error is very small or reaches a minimum value.

  When the edge placement error is first determined, the original angular illumination distribution may be constant, i.e., there is no field dependency. However, it is possible to start with an original angular illumination distribution that already has visual field dependence. This original visual field dependency can be calculated based on, for example, the feature size and orientation on the mask.

  Step c) may comprise illuminating the mask with projection light that has not only an improved field dependence of the angular illumination distribution but also an improved field dependence of the illumination. The illumination varies over the object field so that the edge placement error determined in step b) is reduced at different field points. In other words, in the common optimization process, the visual field dependency of the angular irradiation distribution and the visual field dependency of the irradiation are improved so that the edge arrangement error is reduced.

  In the above case, step b) additionally simulates or measures the edge placement error on the photosensitive surface at a different field point and illuminating the mask with projection light having an original field dependency of illumination. Stages. Furthermore, step c) includes an additional step of changing the original field dependency of the irradiation so that an improved field dependency of the irradiation is obtained.

  If the mask has a portion where the mask pattern is uniform (ie, the width, pitch, and orientation of the structure do not change), the conventional technique uses this angle-independent angular illumination distribution and uniform scan integral illumination. It was something to illuminate.

  However, according to the present invention, despite the illumination described above, the angular illumination distribution can be varied over the area of the object field that coincides with this portion of the mask having a uniform mask pattern at least at one point during step c). . In other words, the angular illumination distribution is deliberately changed across a uniform mask pattern to reduce edge placement errors that can be caused by defects in the projection objective.

  Of course, if the mask includes a non-uniform mask pattern with local variation characteristics, the improved angular illumination distribution can be adapted to the local variation characteristics of the mask pattern. The local variation characteristic of the mask pattern may include at least one of a group consisting of a structure width, a structure pitch, and a structure orientation.

  In one embodiment, at least in some field points, the angular illumination distribution according to the improved field dependence is non-telecentric. At least one of the mask and the photosensitive surface is displaced along the optical axis of the projection objective prior to step c). This results in a lateral shift of the image location. In this way, edge placement errors, especially overlay errors, can be reduced in a visual field dependent manner.

  If the mask moves continuously in step c) during the scanning cycle, the angular illumination distribution can be changed during the scanning cycle. In this case, the angular illumination distribution depends not only on the visual field coordinates but also on the time.

  An illumination system capable of generating a field-dependent angular illumination distribution and capable of generating a field-dependent illumination is preferably an optical integrator configured to generate a plurality of secondary light sources positioned in the pupil plane of the illumination system including. The optical integrator includes a plurality of light incident facets each associated with one of the secondary light sources. The images of the light incident facets overlap at least substantially in the mask plane. A spatial light modulator is provided that has a light exit surface and is configured to transmit or reflect incident projection light in a spatially resolved manner. The objective system images the light exit surface of the spatial light modulator on the light entrance facet of the optical integrator. In step c), the spatial light modulator is controlled so that an improved angular illumination distribution in the mask plane is obtained.

  The illumination system may further include an adjustable pupil forming unit that directs the projection light onto the spatial light modulator. The pupil forming unit may itself comprise a first beam deflection array or a reflective or transmissive first beam deflection element. Both beam deflection elements are configured to illuminate the spot at a location on the spatial light modulator that can be changed by changing the deflection angle produced by the beam deflection element.

  The spatial light modulator can include a second beam deflection array of reflective or transmissive second beam deflection elements. Each second beam deflection element may be capable of entering an “on” state that directs incident light toward the optical integrator and an “off” state that directs incident light toward other locations. The second beam deflection array can be realized, for example, as a digital mirror device (DMD).

  The subject of the invention is also an illumination system of a microlithographic projection apparatus comprising an optical integrator configured to generate a plurality of secondary light sources in the pupil plane of the illumination system. The optical integrator includes a plurality of light incident facets each associated with one of the secondary light sources. The spatial light modulator has a light exit surface and is configured to transmit or reflect incident projection light in a spatially resolved manner. The pupil forming unit is configured to direct projection light onto the spatial light modulator. The objective system images the light exit surface of the spatial light modulator on the light entrance facet of the optical integrator. The control unit is configured to control the pupil forming unit and the spatial light modulator so that the mask is illuminated with projection light having improved field dependence of the angular illumination distribution. The angular illumination distribution according to the improved field dependence varies across the object field so that edge placement errors that vary across the image field are reduced.

  The subject of the invention is also a microlithographic projection apparatus comprising a mask, an illumination system configured to illuminate the mask, and an image of the object field illuminated in the mask plane on the mask positioned on the photosensitive surface A projection objective configured to be formed on an image field. Means are provided for illuminating the mask with projection light having an improved field dependence of the angular illumination distribution, and the angular illumination distribution according to the improved field dependence reduces edge placement errors that vary across the image field. Varies over the object field.

Definitions The term “light” is used herein to denote any electromagnetic radiation, particularly visible light, UV light, DUV light, VUV light, and EUV light, and X-rays.

  The term “ray” is used herein to describe light having a propagation path that can be represented by a line.

  In this specification, the term “beam” is used to describe a plurality of rays having a common origin in the field plane.

  The term “light beam” is used herein to describe all light passing through a particular lens or another optical element.

  In this specification, the term “position” is used to indicate the location of the reference point of an object in a three-dimensional space. Typically, the position is indicated by a set of three orthogonal coordinates. Thus, orientation and position fully describe the placement of objects in three-dimensional space.

  In this specification, the term “surface” is used to represent any plane or curved surface in a three-dimensional space. The surface can be part of the object or it can be completely separated from the object as is usually the case in the field plane or pupil plane.

  The term “field plane” is used herein to describe the mask plane or any other plane that is optically conjugate to the mask plane.

  The term “pupil plane” refers to a plane in which a Fourier relationship is established (at least approximately) with respect to the field plane. Generally, in the pupil plane, peripheral rays passing through different points in the mask plane intersect, and the chief ray intersects with the optical axis. As is customary in the art, the pupil plane is actually not a plane in the mathematical sense, but is slightly curved, and thus a “pupil plane” even if it should be called the pupil plane in the strict sense. Use terminology.

  In this specification, the term “uniform” is used to describe a position-independent characteristic.

  As used herein, the term “optical raster element” refers to any optical element disposed with other identical or similar optical raster elements such that each optical raster element is associated with one of a plurality of adjacent optical channels. Used to describe an element, such as a lens, prism, or diffractive optical element.

  In this specification, the term “optical integrator” is used to describe an optical system that increases the product NA · a when NA is the numerical aperture and a is the illumination field area.

  The term “condenser” is used herein to describe an optical element or optical system that establishes (at least approximately) a Fourier relationship between two planes, eg, a field plane and a pupil plane.

図（Ｆｉｒｓｔ−ｏｒｄｅｒ Ｄｅｓｉｇｎ ａｎｄ ｔｈｅ

Ｄｉａｇｒａｍ）」、Ａｐｐｌｉｅｄ Ｏｐｔｉｃｓ、１９６３年、第２巻第１２号、１２５１〜１２５６ページに記載されている。 In this specification, the term “conjugate plane” is used to represent a plurality of planes between which imaging relationships are established. More information on the concept of conjugate planes can be found in E. Delano's paper “Primary Design and

Figure (First-order Design and the

Diagram), Applied Optics, 1963, Vol. 2, No. 12, pages 1251-1256.

  In this specification, the term “field dependency” is used to express any function dependency of a physical quantity to a position in the field plane.

In this specification, the term “angle irradiation distribution” is used to indicate how the irradiation of the light flux changes depending on the angle of the light beam constituting the light flux. Normally, the angular illumination distribution can be expressed by _a function I _a (α, β) when α, β are angular coordinates representing the direction of light rays. If the angular illumination distribution has a field dependence that varies at different field points, I _a can also be a function of field coordinates, ie, I _a = I _a (α, β, x, y) Become. The field dependence of the angular illumination distribution can be represented by a set of expansion coefficients a _ij of the Taylor (or other suitable) expansion of I _a (α, β, x, y) at x, y.

In this specification, the term “irradiation” is used to describe the total illumination that can be measured at a particular field point. Irradiation can be estimated from the angular illumination distribution by integrating over all angles α, β. Usually, the irradiation also has a visual field dependency, and therefore, I _s = I _s (x, y) holds when x and y are the spatial coordinates of the visual field point. The visual field dependency of irradiation is also called spatial irradiation distribution. In a scanner type projection apparatus, the amount of light irradiation at the field point is obtained by integrating the irradiation over time.

  The various features and advantages of the present invention may be more readily understood with reference to the following detailed description taken in conjunction with the accompanying drawings, in which:

1 is a schematic perspective view of a projection exposure apparatus according to an embodiment of the present invention. It is an expansion perspective view of the mask projected by the projection exposure apparatus shown in FIG. 1 which shows the local change of the angular irradiation distribution on a mask. It is meridional sectional drawing which passes along the illumination system which is a part of apparatus shown in FIG. FIG. 4 is a perspective view of a first mirror array included in the illumination system shown in FIG. 3. FIG. 4 is a perspective view of a second mirror array included in the illumination system shown in FIG. 3. FIG. 4 is a perspective view of an optical integrator included in the illumination system shown in FIG. 3. FIG. 6 is a schematic meridional section through the first and second mirror arrays shown in FIGS. 4 and 5. FIG. 6 is a perspective view of the second mirror array shown in FIG. 5 but illuminated by two poles. FIG. 7 is a perspective view of the optical integrator shown in FIG. 6 but illuminated by two poles. FIG. 4 is a schematic meridional section through a portion of the illumination system showing only the mirror array, the condenser, and the optical raster element array. 11a is a top view of the second mirror array shown in FIG. 3, and FIG. 11b is a top view of the optical integrator shown in FIG. It is a figure which shows the irradiation distribution on the light-incidence facet of an optical integrator. It is a graph which shows the scanning integral irradiation distribution along the X direction produced | generated by the light-incidence facet shown in FIG. It is a figure which shows another irradiation distribution on the light-incidence facet of an optical integrator. It is a graph which shows the scanning integral irradiation distribution along the X direction produced | generated by the light-incidence facet shown in FIG. 16a to 16c are diagrams illustrating the definition of the edge placement error. FIGS. 17a and 17b are diagrams illustrating how the edge placement error can be corrected by generating a telecentricity error and displacing the mask or wafer. FIG. 3 is an enlarged perspective view of a mask similar to FIG. 2 showing how different mask patterns are illuminated with different angular illumination distributions. 2 is a flow chart showing important method steps.

I. 1. General Configuration of Projection Exposure Apparatus FIG. 1 is a very simplified perspective view of a projection exposure apparatus 10 according to the present invention. The apparatus 10 includes a light source 11 that can be realized, for example, as an excimer laser. In this embodiment, the light source 11 generates projection light having a central wavelength of 193 nm. Other wavelengths, such as 157 nm or 248 nm are also envisioned.

  The apparatus 10 further includes an illumination system 12 that adjusts the projection light supplied by the light source 11 in a manner that will be described in more detail below. The projection light emitted from the illumination system 12 illuminates the illumination field 14 on the mask 16. The mask 16 includes a pattern 18 formed by a plurality of small features 19 shown as thin lines in FIG. In this embodiment, the illumination field 14 has a rectangular shape. However, other shapes are also conceivable, for example a ring segment illumination field 14.

  The projection objective 20 including lenses L1 to L6 images the pattern 18 in the range of the illumination field 14 onto a photosensitive layer 22, for example a photoresist, supported by a substrate 24. The substrate 24, which can be formed by a silicon wafer, is placed on a wafer stage (not shown) so that the upper surface of the photosensitive layer 22 is accurately positioned in the image plane of the projection objective 20. The mask 16 is placed on the object plane of the projection objective 20 using a mask stage (not shown). Since the projection objective 20 has a magnification β that satisfies | β | <1, a reduced image 18 ′ of the pattern 18 in the range of the illumination field 14 is projected onto the photosensitive layer 22.

  During projection, the mask 16 and the substrate 24 move along a scanning direction corresponding to the Y direction shown in FIG. The illuminated field 14 is then scanned across the mask 16 so that a larger patterned area can be continuously imaged. The ratio between the speed of the substrate 24 and the speed of the mask 16 is equal to the magnification β of the projection objective 20. When the projection objective 20 does not invert the image (β> 0), the mask 16 and the substrate 24 move along the same direction as indicated by arrows A1 and A2 in FIG. However, the present invention can be used for stepper tools where the mask 16 and substrate 24 do not move during mask projection.

II. Field Dependent Angle Irradiation Distribution FIG. 2 is an enlarged perspective view of a mask 16 that includes another exemplary pattern 18. For the sake of simplicity, it is assumed that the pattern 18 includes only features 19 that extend along the Y direction, ie, includes only identical features 19 that extend along the Y direction and are separated by the same distance. It is further assumed that the feature 19 extending along the Y direction is optimally imaged on the photosensitive layer 22 using an X dipole illumination setting.

  In FIG. 2, the exit pupil 26a associated with the luminous flux is illustrated by a circle. The light beam converges towards a field point located at a certain X position of the illumination field 14 at the first point in the scan cycle. Two poles 27a separated along the X direction within the exit pupil 26a represent the direction in which the projection light propagates toward this field point. Assume that the light energy concentrated in each pole 27a is equal. Accordingly, the projection light incident from the + X direction has the same energy as the projection light incident from the −X direction. Assuming that the features 19 are uniformly distributed across the pattern 18, it should be expected that this X dipole illumination setting will be generated on each point on the mask 16.

  However, if this X dipole illumination setting is maintained throughout the scan cycle and over the entire length of the illumination field 14, the resulting structure after the exposure and subsequent etching steps will not result in the intended location. There is. More specifically, the edges of the structure may be displaced along the X direction in the manner described above with reference to FIGS. 16a and 16c. In other words, the same feature 19 is illuminated with the same illumination and the same angular illumination distribution, but edge placement errors may occur. Typically, edge placement errors can adversely affect critical dimension (CD) quotas and / or lead to severe overlay problems.

  There are various causes that can contribute to such edge placement errors. For example, certain proximity effects such as scattered light associated with the feature 19 may be different when the feature 19 located on the circumference of the mask 16 is compared to the feature 19 located at the center. May have the result that it may be imaged. Other sources of edge placement errors include lens heating effects within the projection objective 20. For example, an optical element positioned in the projection objective 20 near the field plane is illuminated in a non-rotational manner. A small fraction of the projected light (albeit small) may be absorbed by each optical element resulting in a non-rotationally symmetric heat distribution, resulting in a non-rotationally symmetric deformation of these optical elements. . If the optical element is positioned near the field plane (but not in the field plane), such deformation may result in field related aberrations such as distortion.

  The present invention, in accordance with one of its various aspects, relates to eliminating or at least reducing edge placement errors that may occur as a result of these and similar causes. Surprisingly, it has been found that edge placement errors caused by a number of different effects can be reduced significantly significantly by slightly correcting the angular illumination distribution and preferably further illumination in a field-dependent manner. ing. In principle, it is even possible to illuminate each point on the mask 16 with a different combination of illumination and angular illumination distribution. Typically, this need for correction is even stronger when the pattern 18 is not uniform as shown in FIG. However, even in the case of a uniform pattern 19 as shown in FIG. 2, field-dependent edge placement errors are often found and require at least partial correction.

  In FIG. 2, different illumination conditions at different field points are represented by two further exit points 26b, 26c that are generated at different X positions and different times during the scan cycle. At the exit pupil 26b, the light energy concentrated in each pole 27b is still equal. However, the light cone associated with the pole 27b is tilted compared to the light cone of light associated with the exit pupil 26a.

  In the exit pupil 26c, the pole 27c is positioned at the same position as the pole 27a. Therefore, the direction in which the projection light is incident on each field point is the same. However, the pole 27c is not balanced, i.e., the light energies concentrated in the pole 27c are different from each other. In this case, the projection light incident from the + X direction has less energy than the projection light incident from the −X direction.

  Both exit pupils 26b, 26c introduce a telecentricity error. This means that the energy centerline of the light cone is incident on the mask 16 from an oblique direction rather than perpendicularly. This oblique incidence can be used to axially displace the mask 16 and / or substrate 24 and to affect the location of the edge at the substrate level in a manner described in more detail below.

  The visual field dependence of the angular illumination distribution may be required not only in the X direction but also in the Y direction within the illumination visual field 14. Thus, one point on the mask 16 encounters a different angular illumination distribution while passing through the illumination field 14 during the scanning cycle. It must be taken into account that if a field dependence along the Y direction (ie the scanning direction) occurs, the overall effect for a particular field point is obtained by time integrating different angular illumination distributions.

  There are a wide variety of other field-dependent variations in angular illumination distribution that may be necessary to reduce edge placement errors. For example, the poles in the exit pupil associated with some field points can be deformed, blurred, or have a desired non-uniform illumination distribution.

  As mentioned above, it may be necessary to change not only the angular illumination distribution over the illumination field 14, but also the illumination obtained by integrating the angular illumination distribution over all possible angles. The following two sections III and IV describe in more detail how the desired changes in illumination and angular illumination distribution can be achieved.

III. General Configuration of Illumination System FIG. 3 is a meridional section through the illumination system 12 shown in FIG. For purposes of clarity, the diagram of FIG. 3 is significantly simplified and not to scale. In particular, this means that different optical units are represented by only one or very few optical elements. In practice, these units can include significantly more lenses and other optical elements.

  In the illustrated embodiment, the projection light emitted by the light source 11 is incident on the beam expansion unit 32, which outputs an expanded and substantially collimated light beam 34. For this purpose, the beam expansion unit 32 can include several lenses or can be realized, for example, as a mirror arrangement.

  The projection light beam 34 then enters a pupil forming unit 36 that is used to generate a variable spatial illumination distribution in subsequent planes. For this purpose, pupil forming unit 36 includes a first mirror array 38 of very small mirrors 40 that can be individually tilted about two orthogonal axes using actuators. FIG. 4 is a perspective view of the first mirror array 38 showing how the two parallel light beams 42, 44 are reflected in different directions depending on the tilt angle of the mirror 40 upon which they enter. FIG. 3 and 4, the first mirror array 38 includes only 6 × 6 mirrors 40, and actually the first mirror array 38 includes hundreds or even thousands of mirrors 40. Can do.

  The pupil forming unit 36 further includes a prism 46 having a first plane 48 a and a second plane 48 b that are both inclined with respect to the optical axis OA of the illumination system 12. At these inclined surfaces 48a and 48b, incident light is reflected by total internal reflection. The first surface 48 a reflects incident light toward the mirror 40 of the first mirror array 38, and the second surface 48 b guides light reflected from the mirror 40 toward the exit surface 49 of the prism 46. To do. Therefore, the angular illumination distribution of the light emitted from the exit surface 49 can be changed by individually tilting the mirrors 40 of the first mirror array 38. More details regarding the pupil formation unit 36 can be gathered from US 2009/0116093 A1.

  The angular illumination distribution generated by the pupil forming unit 36 is converted into a spatial illumination distribution using the first condenser 50. The capacitor 50, which can be omitted in other embodiments, directs the incident light toward a digital spatial light modulator 52 that is configured to reflect the incident light in a spatially resolved manner. For this purpose, the digital spatial light modulator 52 includes a second mirror array 54 of micromirrors 56 arranged in a mirror plane 57, and an enlarged excerpt C in FIG. 3 and an enlarged excerpt C in FIG. 'Can be seen most clearly. However, in contrast to the mirror 40 of the first mirror array 38, each micromirror 56 of the second mirror array 54 simply passes two stable operating states, ie incident light through the first objective 58. It has an “on” state directed toward the optical integrator 60 and an “off” state that directs incident light toward the light absorbing surface 62.

  The second mirror array 54 can be implemented as a digital mirror device (DMD), for example, as is commonly used in projectors. Such devices can include up to millions of micromirrors that can be switched thousands of times per second between two operating states.

  Similar to the pupil forming unit 36, the spatial light modulator 52 includes an incident surface 65 arranged perpendicular to the optical axis OA, and a first plane 66a and a first plane 66a both inclined with respect to the optical axis OA of the illumination system 12. It further includes a prism 64 having two planes 66b. At these inclined surfaces 66a and 66b, incident light is reflected by total internal reflection. The first surface 66 a reflects incident light toward the micromirror 56 of the second mirror array 54, and the second surface 66 b directs the light reflected from the micromirror 56 toward the surface 68 of the prism 64. Induce.

  When all the micromirrors 56 of the second mirror array 54 are in the “on” state, the second mirror array 54 has a substantially planar beam folding mirror effect. However, when one or more micromirrors 56 are switched to the “off” state, the spatial illumination distribution of the light emitted from the mirror plane 57 is corrected. This modification can be used in the manner described in more detail below to provide a field-dependent modification of the angular light distribution on the mask 16.

  As already described above, the light emitted from the prism 64 enters the optical integrator 60 through the first objective system 58. Since the light passing through the first objective 58 is substantially collimated, the first objective 58 can have a very low numerical aperture (eg, 0.01 or even lower) and thus a small number This can be realized by using a spherical lens having a small size. The first objective system 58 images the mirror plane 57 of the spatial light modulator 52 on the optical integrator 60.

  The optical integrator 60 includes a first array 70 and a second array 72 of optical raster elements 74 in the illustrated embodiment. FIG. 6 is a perspective view of the two arrays 70, 72. Each array 70, 72 includes a parallel array of cylindrical lenses extending along each of the X and Y directions on each side of the support plate. The spatial region where the two cylindrical lenses intersect forms an optical raster element 74. Accordingly, each optical raster element 74 can be regarded as a microlens having a cylindrical curved surface. The use of a cylindrical lens is particularly advantageous when the optical power of the optical raster element 74 should be different along the X direction and along the Y direction. Different refractive powers are required when the square illumination distribution on the optical integrator 60 is converted into the slit illumination field 14 as would normally be the case. Hereinafter, the surface of the optical raster element 74 that faces the spatial light modulator 52 is referred to as a light incident facet 75.

  The optical raster elements 74 in each of the first and second arrays 70, 72 have a one-to-one correspondence with one optical raster element 74 in the first array 70 and one optical raster element 74 in the second array 72. Are arranged one after the other so as to be associated with each other. Two optical raster elements 74 associated with each other are aligned along a common axis to define an optical channel. Within the optical integrator 60, a light beam that propagates in one optical channel does not intersect or overlap with a light beam that propagates in another optical channel. Accordingly, the optical channels associated with the optical raster elements 74 are optically separated from each other.

  In this embodiment, the pupil plane 76 of the illumination system 12 is positioned behind the second array 72, but can be equally positioned in front of it. The second condenser 78 establishes a Fourier relationship between the pupil plane 76 and a field stop plane 80 having an adjustable field stop 82 disposed therein.

  The field stop plane 80 is optically conjugate with the raster field plane 84 positioned in or near the light entrance facet 75 of the optical integrator 60. This is because each light incident facet 75 in the raster field plane 84 is imaged over the entire field stop plane 80 by the optical raster element 74 and the second condenser 78 of the second array 72 associated therewith. Means that. Images of the illumination distribution on the light entrance facets 75 in all optical channels overlap in the field stop plane 80, thereby providing a very uniform illumination of the mask 16. Another approach to account for uniform illumination of the mask 16 is based on the illumination distribution generated by each optical channel in the pupil plane 76. This illumination distribution is often called a secondary light source. All the secondary light sources illuminate the field stop plane 80 in common with projection light from different directions. When the secondary light source is “dark”, no light is incident on the mask 16 from the (small) directional range associated with this particular light source. Therefore, a desired angular light distribution can be set on the mask 16 by simply switching the secondary light source formed in the pupil plane 76 on and off. This switching is achieved by changing the illumination distribution on the optical integrator 60 using the pupil forming unit 36.

  The field stop plane 80 is imaged by the second objective system 86 on the mask plane 88 in which the squeeze 16 is disposed, using a mask base (not shown). An adjustable field stop 82 is also imaged on the mask plane 88 and defines at least the short side of the illumination field 14 extending along the scanning direction Y.

  The pupil forming unit 36 and the spatial light modulator 52 are connected to a control unit 90, and the control unit 90 is further connected to an overall system controller 92 shown as a personal computer. The control unit 90 turns the mirror 40 of the pupil forming unit 36 and the micromirror 56 of the spatial light modulator 52 so that the angular illumination distribution in the mask plane 88 changes in the intended manner within the illumination field 14 during the scanning cycle. Configured to control. In the next section, the function and control of the illumination system will be described.

IV. Functions and control of lighting system FIG. 7 shows how the pupil formation unit 36 generates an illumination distribution on the micromirror 56 of the spatial light modulator 52. The prisms 46 and 64 are not shown for the sake of simplicity.

  Each mirror 40 of the first mirror array 38 illuminates a spot 94 on the mirror plane 57 of the spatial light modulator 52 at a position that can be changed by changing the deflection angle produced by the respective mirror 40. Configured as follows. Thus, by tilting the mirror 40 about its tilt axis, the spot 94 can move freely across the mirror plane 57. In this way, a wide variety of different illumination distributions can be generated on the mirror plane 57. It is also possible for the spots 94 to partially or completely overlap, as shown at 95 locations. In this case, a stepwise irradiation distribution can be generated.

  FIG. 8 is a perspective view similar to FIG. 5 of the second mirror array 54 included in the spatial light modulator 52. In this figure, it is assumed that the pupil forming unit 36 has produced an illumination distribution on the second mirror array 54 of two square poles 27 each extending exactly over 6 × 6 micromirrors 56. The poles 27 are arranged point-symmetrically along the X direction.

  As shown in FIG. 9, the objective system 58 forms an image of the above-described irradiation distribution on the light incident facet 75 of the optical integrator 60. In this case, the illumination distribution formed on the second mirror array 54 is reproduced identically on the light incident facet 75 of the optical integrator 60 (except possible scaling due to the magnification of the objective 58). It is assumed that all the micromirrors 56 are in the “on” state. The regular grid shown on the light incident facet 75 represents an image of the boundary of the micromirror 56, but this image does not appear outside the pole 27 and is shown in FIG. 9 for illustrative reasons only. Yes.

2. Field Dependence Since the light incident facet 75 is positioned in the raster field plane 84, the illumination distribution on the light incident facet 75 passes through the optical raster elements 74 and the second condenser 78 of the second array 72 and the field stop plane. 80 is imaged.

  This imaging will now be described with reference to FIG. 10 which is an enlarged excerpt from FIG. 3 and not to scale. In this figure, only two pairs of optical raster elements 74 of the optical integrator 60, a second condenser 78, and an intermediate field stop plane 80 are shown.

  In the following, the two optical raster elements 74 associated with a single optical channel are referred to as a first microlens 101 and a second microlens 102. The microlenses 101 and 102 are sometimes called a field-of-view honeycomb lens and a pupil honeycomb lens. Each pair of microlenses 101, 102 associated with a particular optical channel generates a secondary light source 106 in the pupil plane 76. In the upper half of FIG. 10, it is assumed that converging light beams L1a, L2a, and L3a indicated by solid lines, dotted lines, and broken lines are incident on different points of the light incident facet 75 of the first microlens 101. . After passing through the two microlenses 101 and 102 and the condenser 78, the light beams L1a, L2a, and L3a converge at the focal points F1, F2, and F3, respectively. From the upper half of FIG. 10, the points where the light rays are incident on the light entrance facet 75 and the points where they pass through the field stop plane 80 (or any other conjugate field plane) are optically conjugate. It becomes clear.

  The lower half of FIG. 10 shows a case where the collimated light beams L1b, L2b, and L3b are incident on different regions of the light incident facet 75 of the first microlens 101. Since light incident on the optical integrator 60 is typically substantially collimated, this is a fairly realistic case. The light beams L1b, L2b, and L3b are focused on a common focal point F positioned on the second microlens 102, and then pass through the field stop plane 80 again in a collimated state. Again, as a result of optical conjugation, it can be seen that the regions where the light beams L1b, L2b, and L3b are incident on the light incident facet 75 correspond to the regions illuminated in the field stop plane 80. it can. Of course, if the microlenses 101, 102 have refractive power along both the X and Y directions, these considerations apply separately in the X and Y directions.

  Thus, each point on the light incident facet 75 directly corresponds to a conjugate point in the intermediate field stop plane 80 (and thus in the illumination field 14 on the mask 16). That is, depending on the position of the light incident facet 75 relative to the optical axis OA of the illumination system on the conjugate point in the illumination field 14 when the illumination on the point on the light incident facet 75 can be selectively influenced. It can affect the irradiation of light rays incident from the direction. The greater the distance between the optical axis OA and the light incident facet 75, the greater the angle at which the light ray is incident on a point on the mask 16.

3. Correction of illumination on light incident facets In the illumination system 12, a spatial light modulator 52 is used to correct illumination on points on the light incident facets 75. In FIG. 9, it can be seen that each pole 27 extends over a plurality of subregions that are images of the micromirror 56. When the micromirror is placed in the “off” state, the conjugate area on the light incident facet 75 is no longer illuminated, so that from the (small) directional range associated with this particular light incident facet 75, the projection Light is not incident on the conjugate area on the mask.

  This will be described in more detail below with reference to FIGS. 11a and 11b, which are top views of the micromirror 56 of the spatial light modulator 52 and the light entrance facet 75 of the optical integrator 60, respectively.

  A thick dotted line on the second mirror array 54 divides the mirror plane 57 of this array into a plurality of object areas 110 each containing 3 × 3 micromirrors 56. The objective system 58 forms an image of each object area 110 on the optical integrator 60. In the following, this image is referred to as image area 110 '. Each image area 110 ′ is completely coincident with the light incident facet 75, ie, the image area 110 ′ has the same shape, size, and orientation as the light incident facet 75 and is completely superimposed on the light incident facet 75. Is done. Since each object area 110 includes 3 × 3 micromirrors 56, image area 110 ′ also includes 3 × 3 images 56 ′ of micromirrors 56.

  In FIG. 11a, there are eight object areas 110 that are completely illuminated by the projection light by the pupil forming unit. These eight object areas 110 form two poles 27. Within a portion of the object area 110, one, two, three or more micromirrors 56d, represented as black squares, are in an “off” state in which incident projection light is directed toward the absorber 62 rather than the objective system 58. It can be seen that it is controlled by the control unit 90. Thus, by switching the micromirror between the “on” state and the “off” state, the projected light is incident on the corresponding region of the image area 110 ′ on the light incident facet 75, as shown in FIG. Can be variably blocked. Hereinafter, these areas are referred to as dark spots 56d '.

  As described above with reference to FIG. 10, the illumination distribution on the light incident facet 75 is imaged on the field stop plane 80. As shown in the upper portion of FIG. 12, when the light incident facet 75 includes one or more dark spots 56d ′, the illumination distribution generated in the mask plane 88 by the associated optical channel is also a certain constant. It will have a dark spot at the X position. Therefore, when a point on the mask passes through the illumination field 14, the total scanning integral irradiation depends on the X position of the point in the illumination field 14 as shown in the graph of FIG. The point at the center of the illumination field 14 will receive the highest scan integral illumination by not passing through the dark spot, and the point at the longitudinal end of the illumination field 14 will have reduced overall illumination to a different extent. Will accept. Accordingly, by selectively switching one or more micromirrors 56 of the spatial light modulator 52 from the “on” state to the “off” state, the angular light distribution on the mask 16 and further the visual field dependency of the spatial illumination distribution. Can be corrected.

  In the above, it had to be assumed that each object area 110 imaged on one of the light incident facets 75 includes only 3 × 3 micromirrors 56. Therefore, the resolution along the scan orthogonal direction X that can be used to correct the field dependence of the angular light distribution is relatively coarse. Increasing the number of micromirrors 56 in each object area 110 can improve this resolution.

  FIG. 14 shows a top view of one of the light incident facets 75 for an embodiment in which 20 × 20 micromirrors 56 are included in each object area 110. In this case, as shown in the graph shown in FIG. 15, a more complicated scanning integrated irradiation distribution can be achieved on the mask 16 along the X direction.

V. Reduction of edge placement error CD uniformity In the first stage, an attempt is made to improve CD uniformity by carefully defining the illumination along the scan orthogonal direction X in the illumination field 14. This approach is known per se in the art and will not be described in more detail here. Next, the micromirror 56 is controlled so that the target field dependency of irradiation is obtained in the illumination field 14. Since the influence of the projection objective 20 on the field-of-view dependence of irradiation at the wafer level cannot be easily predicted, this process may need to be repeated several times. Usually, after several iterations, the CD field-dependent variation reaches a minimum.

  After determining the target field dependence of the illumination, the target field dependence of the angular illumination distribution must be determined.

  It is known how various defects in the illumination setting affect the critical dimension at the wafer level, so to reduce the pattern dependent and field dependent variations of the critical dimension, It can be determined which corrections of the original visual field dependence must be caused. In general, only a small visual field dependence of the angular illumination distribution is required to reduce the critical dimension change that typically occurs.

  The micromirror 56 is then controlled so that the target field dependence of the angular illumination distribution is generated in the illumination field 14. The setting of each micromirror between the “on” and “off” states has an effect not only on the field dependence of the angular illumination distribution, but also on the field dependence of the illumination, so the illumination and angle The visual field dependence of the irradiation distribution can be implemented using a single process.

  Experiments show that critical dimension variation can be reduced by a factor of almost 2 with respect to dense line pitch when not only the illumination but also the angular illumination distribution is optimized separately at different field of view positions (almost by a factor of 2). It has been revealed.

2. Overlay control A similar approach to that outlined above can be used when the view-dependent overlay error is to be corrected.

  In the first stage, the visual field dependence of overlay error at the wafer level is measured or simulated. As described above with reference to FIG. 2, the asymmetry of the angular illumination distribution results in telecentricity errors. In this case, the energy center of the projection light is obliquely incident on the image point. This oblique incidence can be used to accurately shift the image point by displacing the wafer surface from its ideal position in the image plane in the axial direction.

  This is illustrated in FIGS. 17a and 17b. The upper part of FIG. 17 a shows how the telecentric beam 120 passes through the image plane 122 of the projection objective 20. In the middle portion of FIG. 17B, it can be seen that the image point 124 has the smallest diameter in the image plane 122. In the parallel plane 126 displaced in the axial direction with respect to the image plane 122, the diameter of the image point 128 is large, but the X and Y coordinates are not affected by the displacement (see the lower part of FIG. 17a). I want to be)

  FIG. 17b shows the same arrangement for the case of a non-telecentric beam 120 '. This non-telecentricity does not affect the size and position of the image point 124 ′ in the image plane 122. However, in the parallel plane 26, the image point is not only large, but also displaced laterally along the X direction.

  Thus, by carefully introducing asymmetry in the angular illumination distribution and slightly displacing the wafer along the optical axis, the field dependence of the image can be used to correct the field dependence of edge placement errors. It is possible to generate a lateral shift. Since any defocus arrangement of the wafer has a reduced image contrast, a compromise must be found between correction of field-dependent edge placement errors on the one hand and contrast reduction on the other hand.

VI. EUV
In the above, the present invention has been described with reference to a projection exposure apparatus 10 that uses VUV projection light. However, the concepts outlined above can be used for EUV projection apparatus.

  WO 2009/100856 A1 describes an EUV illumination system that makes it possible to generate the desired field dependence of the illumination and angular illumination distribution. Again, the small mirrors must be individually controlled to achieve the desired field dependency.

VII. Multi-Device Die FIG. 18 is a schematic diagram similar to FIG. 2 of a mask 16 that can be used to fabricate different integrated circuits or other devices on a single die. For this purpose, the mask 16 includes three first pattern areas 181a, 181b, 181c and three second pattern areas 182a, 182b, 182c arranged one after the other along the scanning direction Y. . In the simplified embodiment shown, the first pattern area and the second pattern area differ from each other depending on the density of line features 19 extending along the Y direction.

  Here, it is assumed that the first pattern areas 181a, 181b, 181c are illuminated with an angular illumination distribution corresponding to the dipole setting. Accordingly, the pupil 26 a includes two poles 27 a that are separated along the scanning orthogonal direction X.

  The second patterns 182a, 182b, 182c are illuminated with projection light having an angular illumination distribution corresponding to a combination of a dipole setting and a conventional setting. Accordingly, the exit pupil 26b associated with the light beam incident on the second pattern area includes not only the two poles 27b but also the center pole 27b '. Accordingly, the illumination setting associated with exit pupil 26b fully incorporates the illumination setting associated with exit pupil 26a.

  The micromirror 56 of the illumination system 12 can be controlled not to produce only the exit pupils 26a, 26b at each field point in the illumination field 14. In addition, the control technique corrects field-dependent edge placement errors by slightly modifying the exit pupils 26a, 26b within the bisector of the illumination field 14 during the scan cycle. This complex task is possible due to the ability to control the micromirror 56 very quickly and reliably, even in the case of a very large number of micromirrors 56.

  In other embodiments, the illumination settings do not change abruptly and are continuously transformed so that the most recent illumination setting is generated at the intermediate field point.

VII. Important Method Steps The important method steps of the present invention are then summarized with reference to the flowchart shown in FIG.

  In a first stage S1, a mask, an illumination system, and a projection objective are provided. The projection objective is configured to form an image of the object field illuminated in the mask plane on the mask on the image field positioned on the photosensitive surface.

  In a second step S2, edge placement errors are determined at different field points within the image field.

  In the third step S3, the mask is illuminated with projection light having an improved field dependence of the angular illumination distribution so that the edge placement error determined in step S2 is reduced.

14 Illumination field 16 Mask 18 Pattern 19 Feature 27a, 27b Pole

Claims (11)

A method of operating a microlithographic projection apparatus, comprising:
a) mask (16),
An illumination system (12) configured to illuminate the mask, and an image of an object field (14) illuminated on the mask (16) in a mask plane positioned on the photosensitive surface (22) A projection objective (20) configured to form on a field of view;
And a stage of giving
b) determining edge placement errors at different field points within the image field;
c) illuminating the mask (16) with projection light having an improved field dependence of the angular illumination distribution, wherein the angular illumination distribution according to the improved field dependence is determined in step b) Illuminating varying across the object field (14) such that the edge placement error is reduced at the different field points;
A method comprising the steps of: Step b)
Illuminating the mask (16) with projection light having an original field dependence of the angular illumination distribution;
Simulating or measuring the edge placement error on the photosensitive surface at the different field points;
Including
Step c) comprises modifying the original field dependence of the angular illumination distribution so that the improved field dependence of the angular illumination distribution is obtained.
The method according to claim 1. Stage c) comprises illuminating the mask (16) with projection light having improved field dependence of the illumination;
The illumination varies across the object field such that the edge placement error determined in step b) is reduced at the different field points;
The method according to any one of claims 1 to 2, characterized in that: Step b)
Illuminating the mask (16) with projection light having original field dependence of the irradiation;
Simulating or measuring the edge placement error on the photosensitive surface at the different field points;
Including
Step c) comprises modifying the original field dependence of the illumination such that the improved field dependence of the illumination is obtained;
The method according to claim 3. The mask (16) has a portion including a uniform mask pattern;
The angular illumination distribution according to the improved field dependence of the angular illumination distribution varies over an area of the object field that coincides with the portion at least at one point during step c);
The method according to any one of claims 1 to 4, characterized in that: The mask includes a non-uniform mask pattern having local variation characteristics;
The improved angular illumination distribution adapts to the local variation characteristics of the mask pattern;
The method according to any one of claims 1 to 5, characterized in that: At least in some field points, the angular illumination distribution according to the improved field dependence is non-telecentric;
At least one of the mask and the photosensitive surface is displaced along the optical axis of the projection objective prior to step c),
The method according to any one of claims 1 to 6, characterized in that: The illumination system given in step a) is
An optical integrator (60) configured to generate a plurality of secondary light sources (106) positioned in a pupil plane (76) of the illumination system, the optical integrator (60) comprising the secondary light sources The optical integrator (60) including a plurality of light incident facets (75) each associated with one of (106), wherein an image of the light incident facets at least substantially overlaps in the mask plane; ,
A spatial light modulator (52) having a light exit surface (57) and configured to transmit or reflect incident projection light in a spatially resolved manner;
An objective system (58) for imaging the light exit surface (57) of the spatial light modulator (52) on the light incident facet (75) of the optical integrator (60);
Including
In step c), the spatial light modulator is controlled such that the improved angular illumination distribution is obtained in the mask plane;
The method according to any one of claims 1 to 7, characterized in that: An illumination system of a microlithographic projection apparatus (10),
a) pupil plane (76);
b) A plurality of light incident facets (75) configured to generate a plurality of secondary light sources (106) in the pupil plane (76), each associated with one of the secondary light sources (106). ) Including an optical integrator (60),
c) a spatial light modulator (52) having a light exit surface (57) and configured to transmit or reflect incident projection light in a spatially resolved manner;
d) a pupil forming unit (36) configured to direct projection light onto the spatial light modulator;
e) an objective system (58) for imaging the light exit surface (57) of the spatial light modulator (52) on the light entrance facet (75) of the optical integrator (60);
f) configured to control the pupil forming unit (36) and the spatial light modulator (52) such that the mask is illuminated with projection light having improved field dependence of the angular illumination distribution A control unit (90), wherein the angular illumination distribution according to the improved field dependence varies across the object field such that edge placement errors varying across the image field are reduced; ,
An illumination system comprising:   The control unit is configured to control the pupil forming unit (36) and the spatial light modulator (52) such that a method according to any one of claims 2 to 8 is implemented. The illumination system according to claim 9. a) a mask (16);
b) an illumination system (12) configured to illuminate the mask;
c) a projection objective (20) configured to form an image of the object field (14) illuminated on the mask in a mask plane on an image field positioned on the photosensitive surface;
d) means (38, 54, 60, 90) for illuminating the mask with projection light having an improved field dependence of the angular illumination distribution, the angular illumination distribution according to the improved field dependence Means for illuminating (38, 54, 60, 90) varying across the object field such that edge placement errors varying across the image field are reduced;
A microlithographic projection apparatus comprising:

Images