JP7044888B2 - Lithography method and equipment - Google Patents

Description

[Cross-reference to related applications]
This application claims the interests of European Application No. 181511235.1 filed on January 11, 2018 and European Application No. 17202511.6 filed on November 20, 2017, in whole by reference. Incorporated in this book.

[Technical field]
The present invention relates to a lithography method and also a lithography apparatus.

A lithography device is a machine that gives a desired pattern to a target portion of a substrate. Lithographic devices can be used, for example, in the manufacture of integrated circuits (ICs). In that situation, a mask, also called a mask or reticle, can be used to generate a circuit pattern corresponding to each layer of the IC, which pattern is a substrate with a layer of radiosensitive material (resist) (eg, silicon). It can be imaged on a target portion (eg, including a portion of one or several dies) on a wafer). Generally, a single substrate contains a network of adjacent target portions that are continuously exposed. Known lithographic devices include so-called steppers that illuminate each target area by exposing the entire pattern to the target area at once, while scanning the pattern through a beam in a given direction (the "scanning" direction). It includes a so-called scanner that illuminates each target portion by scanning the substrate in parallel or antiparallel in this direction.

Attaching the pellicle to the mask has traditionally been done in DUV lithography equipment. The pellicle is a transmissive film a few mm (eg, 5 mm) away from the mask pattern. The contaminant particles received by the pellicle are far from the mask pattern and, as a result, do not significantly affect the quality of the image projected onto the substrate by the lithography appliance. In the absence of pellicle, the contaminated particles are on the pattern of the mask, obscuring part of the pattern and preventing the pattern from being projected correctly onto the substrate. Therefore, the pellicle plays an important role in preventing contaminated particles from adversely affecting the projection of the pattern onto the substrate by the lithography appliance.

While pellicle provides useful and valuable functionality, pellicle itself causes unwanted side effects in that it affects the image projected onto the substrate by the lithography appliance. The pellicle has a finite thickness and a refractive index greater than that of air. As a result, if the pellicle is not perpendicular to the optical axis of the lithography appliance, the radiation passing through the pellicle will be displaced. This causes distortion in the pattern projected onto the substrate by the lithography equipment. Whether specified herein or elsewhere, it is desirable to provide, for example, a method of preventing or mitigating one or more of the prior art problems.

According to a first aspect of the invention, a support structure configured to support a mask and associated pellicle, wherein the mask gives a pattern to the cross section of the radiated beam to form a patterned radiated beam. It comprises a support structure and a projection system configured to project a patterned radiated beam onto the target portion of the substrate, with a wall extending between the support structure and the projection system. The wall includes an opening that allows a patterned radiated beam to pass from the mask and pellicle to the projection system, and the wall is provided with a lithography system equipped with a two-dimensional array of pressure sensors. ..

The signal output from the two-dimensional array of pressure sensors makes it possible to advantageously calculate the shape formed by the pellicle during scan exposure (eg, using shape reconstruction by near-field acoustic holography). This makes it possible to correct the distortion of the image due to the pellicle. A two-dimensional array of pressure sensors is still useful during manufacturing exposure of the substrate as it does not interfere with the patterned radiation beam.

The two-dimensional array of pressure sensors may extend to either side of the wall opening.

The pressure sensor may be placed in a recess formed in the wall.

The upper surface of the pressure sensor may be flush with the upper surface of the wall. If the pressure sensor is flush with the top surface of the wall, this provides a smooth continuous surface and prevents the pressure sensor from creating large turbulence in the gas flowing over the top surface of the wall.

The pressure sensor may be provided at a pitch of up to 3 cm.

The mask and pellicle may be present in the lithography appliance. The pitch between the pressure sensors may generally correspond to the distance between the pressure sensor and the pellicle.

Further, a processor configured to receive an output signal from the array of pressure sensors and calculate the shape formed by the pellicle during the scanning operation of the mask and pellicle may be provided.

The processor may be configured to reconstruct the shape formed by the pellicle using near-field acoustic holography.

Further controllers may be configured to apply adjustments to the lens elements of the projection system during scan exposure to compensate for the distortion caused by the geometry formed by the pellicle.

According to the second aspect of the present invention, there is provided a method for measuring pellicle deflection in a lithography apparatus. This method loads a mask assembly comprising a mask and a pellicle into the lithography apparatus of the first aspect of the present invention, performs a scanning operation of the mask assembly, receives a signal output from a pressure sensor, and performs pressure. It comprises using the signal output from the sensor to calculate the shape formed by the pellicle during the scanning operation.

Computing the shape formed by the pellicle during the scanning operation may comprise reconstructing the shape formed by the pellicle using near-field acoustic holography.

This method may further comprise determining the natural frequency of the pellicle when calculating the shape formed by the pellicle, followed by consideration of the natural frequency and the harmonics of the natural frequency.

This method may further comprise filtering the signal output from the pressure sensor to remove frequencies above the known maximum frequency of pellicle motion.

The scanning motion may be a set of scanning motions corresponding to the scanning motions used during the manufacturing exposure performed using the mask and pellicle.

A set of scanning operations may include a scanning operation used to expose a field located at the edge of the substrate and a scanning operation used to expose a field located away from the edge of the substrate. ..

The shape formed by the pellicle during the scanning operation may be calculated, or the correction applied to the lithography appliance may be calculated before the manufacturing exposure is performed using the mask and pellicle.

The calculated corrections may be applied during manufacturing exposure of the substrate using masks and pellicle.

The output signal from the pressure sensor may continue to be received during the manufacturing exposure performed using the mask and pellicle. The output signal may be used to adjust the calculated shape formed by the pellicle.

The correction applied to the lithographic appliance may be adjusted to take into account the adjusted calculated shape formed by the pellicle.

The shape formed by the pellicle during the scanning operation may be calculated, or the correction applied to the lithography apparatus may be calculated while the manufacturing exposure is being performed using the mask and pellicle.

According to a third aspect of the present invention, there is provided a computer program comprising computer-readable instructions configured to cause a computer to perform the method according to the second aspect of the present invention.

According to a fourth aspect of the present invention, there is provided a computer comprising a memory for storing processor-readable instructions and a processor configured to read and execute the instructions stored in the memory. Processor-readable instructions include instructions configured to control the computer to perform the method according to a second aspect of the invention.

The features of one aspect of the invention can be combined with the features of different aspects of the invention.

Embodiments of the present invention will be described only by way of reference to the accompanying drawings showing the corresponding portions of the corresponding reference numerals.
It is a figure which shows schematicly about the lithography apparatus which concerns on one Embodiment of this invention. It is a figure which shows the influence of the non-planar pellicle on the radiated beam schematicly. 3A and 3B are diagrams illustrating a portion of the lithography apparatus of FIG. 1 in more detail and schematically. It is a figure which shows schematic the pellicle deformation during the scanning movement of a pellicle. It is a figure which shows the x, y distortion of the image projected using the lithography apparatus caused by the pellicle deformation. It is a flowchart of the method which concerns on embodiment of this invention.

Although this document may specifically refer to the use of lithographic equipment in the manufacture of ICs, the lithographic equipment described herein may be used in the manufacture of integrated optical systems, magnetic domain memories, liquid crystal displays (LCDs), thin film magnetic heads, and the like. May have. Those skilled in the art can consider the term "wafer" or "die" to be synonymous with the more general terms "board" or "target portion", respectively, in the context of such alternative applications. The substrate referred to herein may be treated, for example, with a track (usually a tool that applies a resist layer to the substrate and develops the exposed resist) or a measurement or inspection tool before and after exposure. Where applicable, the disclosures herein may apply to such and other substrate processing tools. Further, the substrate can be processed multiple times, for example to create a multilayer IC, and therefore the term substrate as used herein may refer to a substrate that already contains a plurality of processing layers.

The terms "radiation" and "beam" as used herein include any kind of electromagnetic radiation, ultraviolet (UV) radiation (eg, having wavelengths of 365 nm, 248 nm, 193 nm, 157 nm or 126 nm) and extremes. Includes ultraviolet (EUV) radiation (eg, having a wavelength in the range of 5-20 nm).

The support structure holds the mask (which may also be called the reticle). It holds the mask in a way that depends on the orientation of the mask, the design of the lithographic device, and other conditions. The support structure can use mechanical clamping, vacuum, or other clamping techniques, such as electrostatic clamping under vacuum conditions. The support structure can be, for example, movable as needed and the mask can be, for example, a frame or table that can ensure that it is in the desired position with respect to the projection system.

As used herein, the term "projection system" is used as appropriate according to other factors, such as the exposure radiation used, or the use of immersion fluid or vacuum, as appropriate, eg, refraction optics, reflected optics, reflections. It should be broadly interpreted as covering any type of projection system, including refraction optical systems. When the term "projection lens" is used herein, it can be considered synonymous with the more general term "projection system".

Lighting systems may also include various types of optical components, including refraction, reflection, and catadioptric components for directing, shaping, or controlling a beam of radiation, such components. , Collectively or alone below, may be referred to as a "lens".

The lithography apparatus may be of a type having two (dual stage) or more substrate tables (and / or two or more support structures). In such a "multi-stage" machine, additional tables can be used in parallel, or one or more other tables can be used for exposure while performing preparatory steps on one or more tables.

The lithography device may be a device in which the substrate is immersed in a liquid (for example, water) having a relatively high refractive index so as to fill the gap between the final element of the projection system and the substrate. Immersion technology is well known as a technique for increasing the numerical aperture of a projection system.

FIG. 1 schematically shows a lithography apparatus according to a specific embodiment of the present invention. This device
A lighting system (illuminator) IL configured to regulate the beam PB of radiation (eg UV radiation or DUV radiation), and
A support structure MT that supports the mask MA, and a support structure MT that is connected to a positioning device (not shown) to accurately position the mask with respect to the item PL.
A substrate table (eg, wafer table) WT that holds the substrate (eg, resist-coated wafer) W and is connected to a positioning device PW for accurately positioning the substrate with respect to the item PL.
With a projection system (eg, a refraction projection lens) PL configured to image the pattern given to the radiation beam PB by the mask MT onto the target portion C (eg, including one or more dies) of the substrate W. , Equipped with.

As shown, the device is transmissive (eg, using a transmissive mask). Alternatively, the device may be reflective (eg, using a programmable mirror array of the form described above).

The pellicle P is attached to the frame, which in turn is attached to the mask MA. The pellicle P is a permeable membrane away from the pattern on the mask. The pellicle P prevents contaminated particles from entering the mask pattern and keeps such contaminated particles away from the mask pattern. The pellicle P may be separated from the mask pattern by, for example, several mm, for example, about 5 mm. The mask MA, frame F and pellicle P are all placed in the environment defined by housing 2. The two-dimensional array 30 of the pressure sensor is arranged on the wall 33 of the housing 2. The array 30 of the pressure sensor monitors the pressure of gas (air or the like) in the housing during the scanning operation of the mask MA, the frame F, and the pellicle P.

The illuminator IL receives a beam of radiation from the source SO. The radiation source and the lithography apparatus may be separate bodies, for example, when the radiation source is an excimer laser. In such cases, the source is not considered to form part of the lithography equipment and the emitted beam is directed from the source SO towards the illuminator IL, eg, a beam delivery system containing a suitable oriented mirror and / or beam expander. Pass with the help of BD. In other cases, for example if the source is a mercury lamp, the source may be an integral part of the device. The radiation source SO and the illuminator IL may be referred to as a radiation system together with the beam delivery system BD, if necessary.

The illuminator IL may include adjusting means AM for adjusting the angle and / or spatial intensity distribution of the beam. In general, at least the outer and / or inner radius ranges (usually called σ outer and σ inner) of the intensity distribution in the pupil plane of the illuminator can be adjusted. The illuminator IL may also generally include various other elements such as an integrator IN and a capacitor CO. The illuminator provides a regulated beam of radiation with the desired uniformity and intensity distribution in the beam cross section.

The radiated beam PB is incident on the mask MA held by the support structure MT. After passing through the mask MA, the beam PB passes through the pellicle P and then enters the projection system PS. The projection system focuses the beam PB on the target portion C of the substrate W. With the help of a positioning device PW and a position pressure sensor IF (eg, an interfering device), the substrate table WT can be accurately moved to position different target portions C, for example, in the path of the beam PB. Similarly, the support structure MT can be used to accurately position the mask MA with respect to the path of the beam PB, for example during scan exposure. The mask MA and the substrate W can be aligned using the mask alignment marks M1 and M2 and the substrate alignment marks P1 and P2.

Scanning exposure can be performed using lithographic equipment. In scanning exposure, the support structure MT and the substrate table WT are scanned synchronously (ie, single dynamic exposure) while the pattern given to the beam PB is projected onto the target portion C. The velocity and orientation of the substrate table WT with respect to the support structure MT is determined by the reduction and image inversion characteristics of the projection system PS.

The lithography apparatus further includes a processor PR. The processor receives the signals output from the array of pressure sensors and is configured to use these signals to calculate the shape formed by the pellicle P during the scanning movement of the mask assembly MS. The processor can calculate the adjustments applied to the lithography appliance during scan exposure to reduce the distortion caused by the pellicle. The lithography apparatus further includes a controller CT. The controller CT is configured to apply adjustments to the lithography equipment during scan exposure. The adjustment may include the adjustment of the lens element of the projection system PS.

The processor PR and the controller CT may be provided as a single unit. The processor PR and / or controller CT may include a computer. The computer may include memory for storing processor-readable instructions. The processor PR may be configured to read and execute instructions stored in memory.

It has been understood for some time that the pellicle P affects the patterned radiated beam PB passing through it. However, consideration of the effects of pellicle has been limited to treating the pellicle as if it had the form of a flat sheet traversing the radiated beam PB. Currently, pellicle P has been found to be dynamically distorted during scan exposure. This dynamic deformation causes distortion in the image projected on the substrate W by the lithography apparatus. Embodiments of the present invention address this issue and make it possible to reduce strain. This is achieved without the need to make changes to the pellicle P, frame F, or mask M.

FIG. 2 schematically shows the offset of the radiated beam caused by the pellicle P when the pellicle (or part of the pellicle) is at an angle with respect to the optical axis of the lithography appliance. It is included in FIG. 2 to help explain the offset Cartesian coordinates. The Cartesian coordinates used in the other figures follow the conventions of scanning lithography equipment. The y direction is the scanning direction (that is, the moving direction during scanning exposure), the x direction is in the mask plane in the non-scanning direction, and the z direction is the optical axis of the lithography apparatus.

The pellicle P has a refractive index _np larger than the refractive index n ₁ and n ₂ of the gas (for example, air) on both sides of the pellicle. The thickness of the pellicle is d. The offset introduced by the pellicle P follows Snell's law and is to some extent determined by the thickness of the pellicle and the index of refraction of the pellicle. Further, since the pellicle is at an angle with respect to the XY plane, the XY offset is further determined by the angle of the pellicle with respect to the XY plane. The main ray R _p of the system is shown by the alternate long and short dash line, and the ray R at an angle θ _a with respect to the main ray R _p is also shown. The dashed line _R1 shows how the ray R propagates in the absence of the pellicle. The solid line R ₂ shows how the line propagates in the presence of the pellicle P. As can be seen, there is a significant shift Δyp in the y direction of the ray _R2 compared to the ray _R1 that would be seen in the absence of the pellicle _P. As can be understood from FIG. 2, the displacement of the ray R depends to some extent on the angle of the pellicle P with respect to the XY plane. The main ray R _p is shifted by a smaller amount than the ray R. Rays perpendicular to the pellicle P (not shown) are not shifted.

FIG. 3A is a schematic view showing a part of the lithography apparatus of FIG. 1 in more detail. Similar to FIG. 1, the pellicle P is fixed to the pellicle frame F, and the pellicle frame F is attached to the mask MA. The mask MA is attached to the support structure MT. The pellicle P, pellicle frame F, and mask MA are sometimes referred to as mask assembly MS.

The wall 33 extends between the mask assembly MS and the first lens element 24 of the projection system PS of the lithography apparatus. The wall 33 is provided with an opening 22, through which the patterned radiating beam can travel to the projection system PS. The opening 22 may be referred to as an exposure slit. The wall 33 is shown in FIG. 3B as viewed from above.

The mask assembly and support structure MT are placed in the environment 18 defined by the housing 20. The environment 18 defined by the housing may be referred to as the mask assembly environment 18. The housing 20 has an additional opening 21 at the opposite upper end of the mask MA to receive the radiated beam PB (see FIG. 1).

A two-dimensional array of pressure sensors 30 is provided on the wall. The wall 33 faces the support structure MT and thus faces the pellicle P of the mask assembly MS when the mask assembly is held by the support structure MT. The processor PR receives an output from the array of pressure sensors 30. The two-dimensional array of pressure sensors 30 extends to either side of the opening 22 of the wall 33. The 2D array shown contains 60 sensors, but the 2D array may have a number of other sensors.

Gas, eg air, is present in the mask assembly environment 18. Gas may be provided at a pressure higher than the pressure in the projection system PS to prevent contaminated particles from moving from the projection system to the mask assembly environment 18.

The volume 26 is surrounded by a pellicle P, a mask MA and a frame F. The gas is contained in the volume 26. The volumes are connected to the mask assembly environment 18 by a leak path (not shown) that allows gas (eg, air) to flow between them. The leak path is limited so that the speed at which the gas can travel between the volume 26 and the mask assembly environment 18 is limited. Since the flow rate is low enough, it can be considered that the amount of gas in the volume 26 is fixed during the scan exposure.

During scan exposure, the support structure MT and the mask assembly MS move rapidly from one side of the housing 20 (as indicated by the arrow in FIG. 3A) to the other side in the y direction. Scanning exposure may be performed, for example, within about 100 ms.

As schematically shown in FIG. 3A, during the scan movement from left to right of the mask assembly MS, the pressure of the gas on the right side of the mask assembly MS and the support structure MT reduces the volume containing that gas. It will increase because it is. At the same time, the pressure on the left side of the mask assembly MS and the support structure MT decreases as the volume containing the gas increases. As a result, the gas flows around the mask assembly MS and the support structure MT until the gas pressures are equal in the mask assembly environment 18. This gas flow can cause dynamic deformation of the pellicle P (ie, deformation that changes during the scanning movement of the pellicle).

The inertia of the gas (eg, air) in the volume 26 surrounded by the pellicle P, mask MA and frame F can also cause dynamic deformation of the pellicle. Referring to FIG. 3A, when the scanning movement of the mask assembly MS in the positive y direction begins, the inertia of the gas in the volume 26 tends to keep the gas in its original position. As a result, gas accumulates at the left end of the mask assembly during acceleration of the mask assembly MS. This causes the pellicle P to bulge outward at the left end of the mask assembly MS. The mask assembly MS slows down when it reaches the right edge of the scanning motion. Here, the gas is moving in the positive y direction within the volume 26 and tends to continue to move as the mask assembly MS slows down. As a result, gas accumulates at the right end of the mask assembly during deceleration of the mark assembly MS. As a result, the pellicle P bulges outward at the right end of the mask assembly MS. As mentioned above, the scan exposure may be performed within about 100 ms. During this time, the mask assembly accelerates from rest, moves 100 mm or more, then decelerates and stands still. It will be appreciated that the inertia of the gas in the volume 26 of the mask assembly MS causes significant deformation of the pellicle 26.

The dynamic deformation consists of bending of the pellicle P, which introduces distortion into the image projected onto the substrate W by the lithography apparatus LA. As described above in connection with FIG. 2, when the pellicle P is at an angle with respect to the mask MA, this introduces an offset into the projected image. Because the pellicle is curved and therefore has a range of angles with respect to the mask, the pellicle introduces distortion into the projected image rather than introducing a simple offset. In addition, the strain introduced by the pellicle changes during scan exposure. This is because the patterned radiated beam RB passes along the pellicle P during the scanning movement of the mask assembly MS and different parts of the pellicle are bent in different ways.

FIG. 4 schematically shows an example of pellicle deformation that may occur during the scanning operation of the mask assembly. As schematically shown in FIG. 4, the scanning operation of the pellicle P is in the Y direction in this example.

As mentioned above, the amount of gas in the volume 26 between the pellicle P and the mask MA is effectively fixed during the scan exposure. In addition, the gas in the volume G tends to resist compression or expansion. As a result, the total volume surrounded by the pellicle P remains substantially constant, and the outward expansion of one part of the pellicle P tends to coincide with the corresponding inward movement of another part of the pellicle P. There is. An example of this modification of the pellicle P is shown in FIG. The portion of the pellicle P toward the lower left end bulges inward, and the portion of the pellicle P toward the upper right end bulges outward by a corresponding amount. Therefore, the volume enclosed by the pellicle P remains substantially constant. This deformation of the pellicle can be regarded as similar to the movement of the surface of the waterbed. That is, it resembles the movement of a flexible membrane that surrounds a volume of incompressible fluid.

The distortion caused by the pellicle in the image projected during scan exposure is relatively complex. Distortion can be regarded as aberrations that can be expressed as Zernike, which include several orders of Zernike. However, the distortion is relatively consistent. That is, when performing a scan exposure on a given lithography device using a given mask assembly MS with a particular pellicle, the distortion caused by the pellicle is about the same as the distortion caused during the previous exposure. Become. This is true if the speed and direction of the scan exposure is the same and the background pressure of the gas in the mask assembly environment is the same (ie, the pressure of the gas changes when the mask assembly MS is not moving). Not done). For scan exposures at the same speed but in opposite directions, the distortion caused by the pellicle is reversed.

The dynamic deformation of the pellicle P during scan exposure is determined using the pressure measurements output from the array of pressure sensors 30. The dynamic deformation can be thought of as the shape formed by the pellicle as a function of the position of the mask assembly MS during the scanning operation. The dynamic deformation of the pellicle during scan exposure can be pre-calculated by the processor using the measurements obtained prior to the scan exposure of the substrate (discussed further below). The calculated dynamic deformation of the pellicle during scan exposure is used to determine the adjustments applied to the projection system during scan exposure of the substrate. These adjustments reduce the distortion caused by the pellicle.

Referring again to FIG. 3, the pressure sensor 30 in the wall 33 of the housing 20 is provided as a two-dimensional array. The pressure sensor 30 may be, for example, a movable diaphragm attached to a magnet (eg, in the form of a microphone). The pressure sensor may be, for example, a MEMS microphone. In one embodiment, the pressure sensor may be an AKU242 digital silicon MEMS microphone available from Akustica, Philadelphia, USA. Other MEMS microphones (pressure sensors), such as the VM101 microphone available from Vespa Technologies, Massachusetts, USA, may be used.

The pressure sensor 30 can be arranged in the recess formed in the wall 33. Placing the pressure sensor 30 in the recess of the wall 33 is advantageous because it prevents the pressure sensor from protruding from the wall and causing large turbulence in the gas (eg, air) flowing across the wall.
The upper surface of the pressure sensor may be flush with the upper surface of the wall. If the pressure sensor is flush with the top surface of the wall, this provides a smooth continuous surface, which further helps to avoid the occurrence of turbulence.

An electrical connection (not shown) from the pressure sensor 30 can pass through the wall 33 and exit from the bottom of the wall, or through the wall and exit from the side wall. Alternatively, you can use a wireless connection. The processor PR receives an output signal from the pressure sensor 30.

In FIG. 3, the two-dimensional array of pressure sensors consists of 60 pressure sensors. However, this is just a schematic and any suitable number of pressure sensors can be used.

The pressure sensor 30 of the pressure sensor array may be located at a distance of about 2 cm from the pellicle P. The pressure sensors can be separated from each other at a pitch of about 2 cm (eg, a pitch of up to 3 cm). In general, close field acoustic holography is effectively used to determine the pellicle shape by generally corresponding to the distance between the pressure sensor 30 and the pellicle P, or by providing the pressure sensor 30 with a smaller spacing. be able to.

The pellicle P can measure, for example, about 110 mm × 150 mm. The deflection of the pellicle during scan exposure can have a relatively low spatial frequency (eg, 3 cm or more). As a result, a pitch of about 2 cm (eg, pitches up to 3 cm) of the array pressure sensor 30 can provide pressure measurements at a spatial frequency high enough to allow accurate measurements of dynamic pellicle deformation. can. Pellicle deflection on the order of millimeters occurs. The detection system 40 may be capable of determining pellicle deflection with an accuracy on the order of microns. This is sufficient to provide the exact properties of the millimeter-order deflection of the pellicle.

The pressure sensor 30 can have a sampling frequency higher than the frequency of movement of the pellicle during the scanning operation. The frequency of motion of the pellicle may be, for example, in the range of 25 Hz to 40 Hz. The pressure sensor 30 can provide, for example, an output measurement having a frequency up to about 100 Hz (eg, up to about 200 Hz) and can accurately detect frequencies as low as about 10 Hz. The pressure sensor 30 may be able to detect frequencies below 10 Hz, but the accuracy of such measurements may be reduced.

In general, the spatial pitch of the pressure sensor 30 and the frequency of the output from the pressure sensor can be selected high enough to effectively sample and determine the deflection of the pellicle P.

Proximity field acoustic holography (NAH) may be used by the processor PR to reconstruct the dynamic deformation of the pellicle P during the scanning operation of the mask assembly MS. In other words, to reconstruct the shape formed by the pellicle P as a function of the position of the mask assembly MS during the scanning operation. The reconstruction is accomplished using the computation performed by the processor PR. Reconstruction of the shape formed by the pellicle during scan exposure can determine the distortion caused by the pellicle. Once the distortions have been determined, the lens model can be used to determine the corrections (eg, adjustment of the lens elements of the projection system) applied to the lithography equipment to reduce these distortions. This advantageously improves the accuracy with which the pattern is projected onto the substrate. Close-field acoustic holography will be described further below.

The dynamic deformation of the pellicle P may be determined using the pressure sensor array 30 and the processor PR before the exposure of the substrate is started. It is expected that the behavior of the pellicle will be consistent and that the deformation measured before the exposure will be repeated during the exposure if the perimeter (perimeter) that determines the deformation remains the same. Because it is done. Therefore, the mask assembly MS can move in a scanning operation corresponding to the scanning exposure, but the exposure radiation does not enter the substrate. The dynamic pellicle deformation that occurs is determined by the processor PR based on the output signal from the array of pressure sensors 30. During subsequent scan exposure of the substrate, the dynamic deformation of the pellicle P is assumed to be the same and corrections are applied to the lithography appliance accordingly.

In one example, a lithographic appliance can be used to expose a substrate with a mask MA and associated pellicle P that were not previously used in the lithographic appliance. The desired scan exposure length is used during the exposure of the substrate and the desired scan speed can be used. However, this may vary depending on the exposure at different locations on the substrate. For example, when exposing a field located towards the center of the substrate, the full exposure scan length and maximum scan speed can be used. However, when exposing a field at the edge of a substrate, a partial field may be exposed. Therefore, shorter exposure scan lengths can be used. The speed of scan exposure may also decrease. Different locations around the edges of the substrate can be exposed using different scan lengths and / or scan speeds.

Prior to exposing the substrate, measurements can be obtained using the pressure sensor array 30 for a set of scan operations containing different scan lengths and velocities used during subsequent exposure of the substrate. For each scan length and / or velocity of a set of scan movements, the measurements obtained using the pressure sensor array 30 are used to determine the dynamic deformation of the pellicle P that occurs during scan exposure. The lens model is then used to determine the adjustments applied to the lens element during these scan exposures to reduce the distortion caused by the deformation of the pellicle.

In one example, the scan operation of the mask assembly MS for each scan speed and scan length may be performed in the positive Y direction and the negative Y direction to allow the acquisition of two sets of data. As further mentioned above, the behavior of the pellicle is expected to be symmetric with respect to the positive and negative Y-direction scans, so a single scan motion measurement is required to characterize the dynamic deflection of the pellicle. It is enough. However, performing the two scan operations provides additional data that allows the pellicle deformation to be more accurately determined (eg, by improving the signal-to-noise ratio). More than two scan operations can be used for a given scan length to obtain more data. This allows the deformation of the pellicle to be obtained more accurately (eg, by further improving the signal-to-noise ratio).

In one example, simulated exposure of the entire substrate can be performed without directing the exposure radiation through the mask MA. In this example, a set of scanning operations involves a simulated exposure of the entire substrate. Data can be collected from the pressure sensor 30 for each scanning operation of the mask assembly MS. Performing a simulated exposure of the entire substrate performs all scan speeds and scan times that occur during the manufacturing exposure of the substrate and produces data that can be used to determine the shape formed by the pellicle during the exposure. To. When the measurement is performed, the correction applied to the lens element is calculated. The correction can then be applied when the substrate is exposed. As mentioned elsewhere herein, pressure measurements can be performed during exposure of the substrate.

The computational power required to determine the pellicle deformation and the correction applied to the lens element may be such that they cannot be calculated and applied in real time during substrate exposure. Therefore, in the above embodiments, they are calculated prior to the exposure of the substrate. However, if sufficiently high processing power is available, the determination of pellicle deformation and associated lens correction can be determined in real time. Therefore, the shape formed by the pellicle during the scanning operation can be calculated, and the correction applied to the lithography appliance can be calculated during the manufacturing exposure.

Proximity acoustic holography is used to computationally reconstruct the shape formed by the pellicle at each moment during the scanning operation of the mask assembly MS. The calculations used in this reconstruction can require a large amount of calculations, so it is beneficial to remove noise when possible. Filtering of the data output from the pressure sensor can be performed using frequency. Close-field acoustic holography is described in WO2009 / 130243A2, US2013 / 0094678A1 and US2013 / 0128703A1, each of which is incorporated herein by reference.

One parameter that can be used to apply data filtering is the natural frequency (resonance frequency) of the pellicle P. The natural frequency of the pellicle is in the range of 20 to 50 Hz and depends on the tension of the pellicle. In practice, the natural frequency is about 25 Hz (eg, plus or minus 5 Hz). The natural frequency of the pellicle can be determined, for example, by vibrating the mask assembly MS using an actuator used to provide scanning operation for the mask assembly during substrate exposure. The output from the pressure sensor 30 is monitored when vibration is applied. The vibration is initially applied at a frequency lower than the expected natural frequency. Next, the frequency is increased, for example, in units of 0.1 Hz, until a spike in the signal output from the pressure sensor 30 is seen. This spike indicates the natural frequency of the pellicle P. The natural frequency and the harmonics of the natural frequency are present when the pellicle is deformed during the scanning operation.

Another frequency that will be present during the scanning operation of the mask assembly MS is the frequency of exposure performed by the lithography appliance. This exposure frequency can be, for example, between 2 and 10 Hz. In this case as well, harmonics of the exposure frequency may be present in the pellicle deformation. The two scans of the mask assembly MS may be sufficient to make it possible to measure the pellicle deformation caused by the exposure frequency. In this context, the term "two scans" is intended to mean a one-way scan followed by a return scan in the opposite direction. It may be possible to use a single scanning motion to be able to measure the deformation of the pellicle caused by the exposure frequency.

In general, the behavior of the pellicle is highly reproducible, so a single scanning motion may be sufficient to allow the pellicle deformation during the scanning motion to be determined. The determined pellicle deformation is repeated with subsequent scanning movements at the same speed and duration.

The dynamic deflection of the pellicle may have a maximum frequency limit well below about 200 Hz. Low-pass filtering of the signal output from the pressure sensor 30 may be applied to exclude signals with frequencies above about 200 Hz when the pellicle deflection is calculated. This further improves the signal-to-noise ratio.

Knowledge of the frequencies present in the pellicle deformation can be used to filter the signal output from the pressure sensor 30 thereby improving the signal-to-noise ratio. For example, frequencies present in the signal received from the pressure sensor 30 that deviate from the expected frequency of pellicle motion may be filtered (eg, by the processor PR).

When reconstructing the shape formed by the pellicle during scan exposure using close field acoustic holography, the phase difference between the signals output from the pressure sensor 30 of the pressure sensor array is used. Since the array of pressure sensors 30 is two-dimensional (as opposed to just a line of sensors), the information received from the sensors is sufficient to allow the shape formed by the pellicle to be reconstructed. be. The processor PR performs the correlation between the signals received from the different pressure sensors 30. If a strong signal is seen for a given correlation, this indicates that the object caused a pressure wave incident on these pressure sensors at different times. The source of the pressure wave can be identified using the phase difference (time delay) that produced the strongly correlated signal. The position of the mask assembly (and thus the pellicle P) in the Y direction is known at any given moment when the signal is being output from the pressure sensor 30. Therefore, since the location of the pellicle P is known (although its deflection is unknown), this information can be used to determine whether the source of the pressure wave corresponds to the pellicle or to another device. If the pellicle P is not the source of the pressure wave, the pressure wave is negligible. When the pellicle P is the origin of a pressure wave, the pressure wave is used as part of the reconstruction of the shape formed by the pellicle. This is done for the pressure sensor 30 across the pressure sensor array. Many of the origins of pressure waves are determined. Taken together, these show the shape formed by the pellicle P.

ここで、ｚ_ｓは圧力波の発生源の１次元の位置であり、ｚ_ｈは圧力センサの１次元の位置である。本発明の実施形態では、圧力センサ３０で受信される音波は既知である。逆伝播関数を使用して、音波を逆伝播させ、圧力波を引き起こしたペリクルのたわみを決定することができる。逆解法は、レイリーの伝搬カーネルを使用した測定平面（圧力センサ３０が配置されている平面）のデコンボリューションであってよい。 More specifically, when a pressure wave is generated, the pressure wave propagates according to the propagator function G.

Here, z _s is the one-dimensional position of the source of the pressure wave, and z _h is the one-dimensional position of the pressure sensor. In the embodiment of the present invention, the sound wave received by the pressure sensor 30 is known. A backpropagator function can be used to backpropagate a sound wave to determine the deflection of the pellicle that caused the pressure wave. The inverse method may be deconvolution of the measurement plane (the plane on which the pressure sensor 30 is located) using Rayleigh's propagation kernel.

As further mentioned above, once the deformation of the pellicle is determined with respect to the pellicle in the lithography equipment at a particular scan length and scan rate, this known variant of the pellicle is a subsequent use of the same scan length and scan rate. Expected to occur during substrate exposure.

As further mentioned above, the behavior of the pellicle P is consistent for a given scan speed and scan length, unless other parameters are changed. In practice, there may be some differences between the interiors of the housing 20 where the mask assembly MS is located in different lithographic devices. As a result, even if the same scan length and scan speed are used, the deformation of the pellicle can be different within different lithography appliances. Therefore, previously determined pellicle deformations may be used for a particular pellicle in a particular lithographic appliance, but should not be used for that pellicle in a different lithographic appliance. The previously determined deformation of the pellicle, if present, may be used for subsequent scan exposures. This is the case when the pellicle is attached to the same mask MA, or when the pellicle is attached to another mask MA (changing the mask does not significantly affect the deformation of the pellicle).

The pressure sensor 30 can be installed in all lithography equipment in the manufacturing plant. This allows dynamic pellicle deformation to be determined within each lithography appliance. As mentioned above, this is advantageous because different dynamic deformations of the same pellicle can occur in different lithographic devices.

The pressure sensor 30 is present during the manufacturing exposure of the substrate. Since the pressure sensors 30 are passive (ie, they do not affect the pellicle or mask), they may be used to collect data during the manufacturing exposure of the substrate. The processor PR can continue to determine the shape formed by the pellicle using the data obtained from the pressure sensor 30 during the manufacturing exposure of the substrate. This can, for example, make it possible to improve the calculated dynamic deformation of the pellicle. In other words, the accuracy with which pellicle deformation is determined can improve over time. Similarly, the adjustments applied to the lithographic device to reduce the distortion caused by the pellicle deformation can be improved over time.

As further mentioned above, the deformation of the pellicle is at least partially determined by the tension of the pellicle. This tension gradually decreases over time. This is because the pellicle absorbs some of the radiation from the patterned radiation beam, which over time causes aging of the pellicle. Due to this aging, the pellicle loses tension. As the tension decreases, the deformation of the pellicle during scan exposure becomes the same shape. However, its shape is amplified. In other words, the maximum deflection of the pellicle from a plane passing through the edge of the pellicle increases.

Since aging of the pellicle is gradual and predictable, a simple model is used to adjust the calculated dynamic deformation of the pellicle to take into account the aging of the pellicle that occurs after the pellicle deformation has been measured. Can be put in. Alternatively, if the output signal from the pressure sensor 30 is monitored during manufacturing exposure, the dynamic deformation of the pellicle may be calculated periodically. This includes changes in dynamic deformation due to aging.

When reconstructing the shape formed by the pellicle P using the signal output from the pressure sensor 30, the processor PR can take into account the known constraints of the pellicle. For example, as described above, the position of the pellicle is known during the scanning operation, and the pressure signal received from a starting point other than the pellicle can be ignored by the processor PR. In another example, it is known that the edge of the pellicle P is fixed to the frame F and therefore does not move in the z direction.

When reconstructing the geometry formed by the pellicle P using the signal output from the pressure sensor 30, the processor PR can take into account the geometry previously seen for other pellicle. This can be done, for example, using the shapes previously seen in other pelliclees of the same type. The same type of pellicle may be a pellicle of the same thickness. The same type of pellicle may have the same initial tension as the pellicle P at the time of manufacture. However, the tension of the same type of pellicle can decrease over time due to aging of the pellicle. The shape found in the pellicle is sized according to the aging of the pellicle. Aging of the pellicle is caused by the cumulative radiation dose of the radiation experienced by the pellicle. This radiation dose is calculated by the processor. Therefore, it is possible to measure the aging deterioration and the decrease in tension of the pellicle caused by the radiation dose. The processor PR can apply adjustments to previously seen shapes for the same type of pellicle to take into account the aging of that pellicle. The resulting shape may be taken into account by the processor PR when reconstructing the shape formed by the pellicle P. The embodiments of the present invention do not affect the manufacturing exposure performed using the lithography apparatus. As mentioned above, this means that measurement of pellicle deflection may be performed during manufacturing exposure.

The array of pressure sensors 30 can be retrofitted to existing lithography equipment. This can be achieved, for example, by replacing the existing wall 33 of the mask assembly housing 20 with a new wall provided with an array of pressure sensors 30.

The flowchart of FIG. 6 shows a method according to an embodiment of the invention that can be used to compensate for the distortion of the projected image caused by the shape formed by the pellicle P during the scan exposure performed by the lithography apparatus. The correction may be applied, for example, during the manufacturing exposure of the substrate, for example during the exposure of a series of wafers with dies forming integrated circuits. In general, this method involves using the output from the pressure sensor array and other information to calculate the shape formed by the pellicle during the scanning operation. The calculation may be a reconstruction of the shape formed by the pellicle using near-field acoustic holography. This method also uses a radiation beam aberration model to determine how the radiation beam PB is distorted by the pellicle and applies a rolling Gaussian slit exposure-model to scan the exposure. Including determining the effect of the pellicle on the exposure. The effect of the pellicle on the exposure is sometimes referred to as the fingerprint of the pellicle. The method further comprises using a lens model to determine the adjustment of the projection system PS applied to compensate for the pellicle fingerprint. The correction is then applied to the projection system during the manufacturing exposure of the substrate.

As described above, reversing the scanning exposure direction reverses the distortion of the pellicle. Therefore, two sets of adjustments, one for each direction of scan exposure, can be stored in memory. If scan exposures are performed at the edges of the wafer, they may be shorter and / or slower than scan exposures performed away from the edges of the wafer. As a result, the deformation of the pellicle is different when these exposures are performed. As a result, an additional set of adjustments for the projection system PS is stored in memory and used when exposure is done at the edges of the wafer.

If heating of the pellicle P by the radiated beam PB is expected to have a significant effect (eg, reduction of the pellicle tension due to thermal expansion), calibration can be performed while the radiated beam is incident on the pellicle. Alternatively, the effect of heating caused by the radiated beam on the tension of the pellicle P (using the coefficient of thermal expansion of the pellicle) can be calculated and added to the model. The temperature of the pellicle P can be expected to rise in a manner known as a function of time, for example, it may rise at the same rate as the temperature of the mask MA. The heating of the mask may be the subject of a separate existing model, and the temperature of the pellicle P may be derived from that model. The pellicle P may be heated directly by absorption of incident radiation, or indirectly by heat conduction from the mask MA to the pellicle P via the frame F.

With reference to FIG. 6 in more detail, a mask assembly with a new pellicle is loaded into the lithographic apparatus. New in this context may mean that the pellicle was not previously used in this lithographic device (it may be used in other lithographic devices). The natural frequency of the pellicle can be determined (further described above). This is an optional step that can reduce the amount of computation required to reconstruct the shape formed by the pellicle. However, the shape formed by the pellicle may be determined without first determining the natural frequency of the pellicle. The scanning operation of the mask assembly used during the manufacturing exposure is performed. An array of pressure sensors is used to measure the pressure during the scanning operation.

Close-field acoustic holography is used to reconstruct the shape formed by the pellicle during the scanning operation. The shape formed by the pellicle changes as a function of the pellicle position during the scanning operation. The signal can be filtered using the natural frequency of the pellicle and the exposure scan frequency (along with the harmonics of these frequencies). Filtering can be done before or as part of the calculation.

The radiation beam aberration model receives the shape of the pellicle as an input during the scanning operation and also the illumination mode used during the manufacturing exposure as an input. The radiation beam aberration model may be, for example, a ray deflection model, which may be a model that implements Snell's law (discussed above in connection with FIG. 2). Alternatively, the radiated beam aberration model may be a more sophisticated model that models the Zernike aberration of the radiated beam caused by the deformation of the pellicle (this type of model treats the pellicle as a lens element).

The output from the radiation beam aberration model is input to the rotating Gaussian slit exposure model. This model addresses the movement of the pellicle and mask with respect to the radiated beam during scan exposure (eg as a convolution) and outputs the pellicle fingerprint caused by the deformation of the pellicle. An example of a pellicle fingerprint is shown in FIG. The pellicle fingerprint shows how points in the image are displaced due to the effects of pellicle distortion.

Finally, the lens model is used to determine the correction applied to the lens element of the projection system PS to compensate for the pellicle fingerprint. Such lens models are well known in the art and therefore lens models are not described herein. The correction may be, for example, a fourth-order polynomial correction in the Y direction.

As mentioned above, the correction applied to the lens element may be determined prior to the manufacturing exposure. The correction is then applied during the manufacturing exposure to compensate for the pellicle fingerprint during that exposure.

The pellicle may be stationary before the exposure of the new substrate begins. When the exposure of the substrate begins, vibrations of the mask assembly MS occur, and these vibrations stabilize after about two or three scan exposures. The effect of these vibrations on the shape formed by the pellicle can be measured using embodiments of the present invention. This may then be taken into account when applying corrections to the lithography equipment during manufacturing scan exposure.

Adjustments applied by the lithographic device during scan exposure can be stored in the lithographic device. Alternatively, the adjustments may be stored remotely and may be transmitted to the lithographic device when they are needed.

Adjustments that compensate for pellicle fingerprints are combined with adjustments that compensate for sources of aberration elsewhere in the lithographic appliance (eg, adjustments that compensate for aberrations caused by the heating of the lens elements of the projection system during exposure). Can be done.

Although the described embodiments of the present invention refer to a model of a particular form, any suitable form of the model can be used.

Although the adjustments that compensate for the fingerprint of the pellicle have been described for the adjustment of the lens element, other adjustments can be used for the lithography apparatus. For example, the position of the substrate during scan exposure can be adjusted by a lithographic device (eg, some movement in the z direction can be used to compensate for the change in focus).

The Z-direction movement of the pellicle during scan exposure may be determined as part of the calculation of the shape formed by the pellicle during scan exposure. This output can be used to determine how much the effect of the pellicle is reduced by movement in the z direction. When the dust particles are present on the pellicle, the z-direction movement towards the mask causes the dust particles to approach the focal plane of the lithography appliance. The extent to which this occurs can be determined using the calculated pellicle shape. Next, the influence of the movement of the dust particles in the z direction can be determined.

Aspects of the invention can be implemented in any convenient way, including suitable hardware and / or software. For example, a programmable device that can form part of a controller CT can be programmed to implement embodiments of the invention. Accordingly, the invention also provides suitable computer programs for implementing aspects of the invention. Such computer programs can be delivered on suitable carrier media including tangible carrier media (eg, hard disks, CD ROMs, etc.) and intangible carrier media such as communication signals.

Although specific embodiments of the invention have been described above, it will be appreciated that the invention may be practiced in ways other than those described. This description is not intended to limit the invention.

Claims (12)

A support structure configured to support a mask and associated pellicle, wherein the mask can give a pattern to the cross section of the radiated beam to form a patterned radiated beam.
A projection system configured to project the patterned radiated beam onto the target portion of the substrate.
Equipped with
A wall extends between the support structure and the projection system, which has an opening that allows the patterned radiation beam to pass from the mask and the pellicle to the projection system. Including, the wall is provided with a two-dimensional array of pressure sensors .
The pressure sensor is placed in a recess formed in the wall and
A lithography device in which the upper surface of the pressure sensor is flush with the upper surface of the wall . The lithography apparatus according to claim 1, wherein the two-dimensional array of the pressure sensor extends to any side of the opening of the wall. The lithography according to claim 1 or 2 , further comprising a processor configured to receive an output signal from the array of pressure sensors and calculate the shape formed by the pellicle during the scanning operation of the mask and the pellicle. Device. The lithography apparatus according to claim 3 , wherein the processor is configured to reconstruct the shape formed by the pellicle using near-field acoustic holography. 30. 4 Lithography equipment. A method of measuring pellicle deflection in a lithography appliance.
Loading a mask assembly comprising a mask and a pellicle into the lithography apparatus according to claim 1.
Performing the scanning operation of the mask assembly and receiving the signal output from the pressure sensor,
Using the signal output from the pressure sensor, the shape formed by the pellicle during the scanning operation can be calculated.
How to prepare. The method of claim 6 , wherein calculating the shape formed by the pellicle during a scanning operation comprises reconstructing the shape formed by the pellicle using near-field acoustic holography. Claim 6 or 7 , further comprising determining the natural frequency of the pellicle when calculating the shape formed by the pellicle, and then considering the natural frequency and the harmonics of the natural frequency. The method described in. The output signal from the pressure sensor continues to be received during the manufacturing exposure performed using the mask and the pellicle, and the output signal is used to adjust the calculated shape formed by the pellicle. The method according to any one of claims 6 to 8 . 9. The method of claim 9 , wherein the correction applied to the lithographic apparatus is adjusted to take into account the adjusted calculated geometry formed by the pellicle. A computer program comprising computer - readable instructions configured to cause a computer to perform the method according to any one of claims 6-10 . Memory for storing processor-readable instructions and
A processor configured to read and execute instructions stored in the memory, and
Is a computer equipped with
The processor - readable instruction comprises an instruction configured to control the computer to perform the method of any of claims 6-10 .

Images