Patent Application
20060238736
The present invention relates generally to a measuring apparatus, and more particularly to a measuring apparatus of an incident angular distribution or a light distribution characteristic (also called “an effective light source” and “a σ distribution”) on an illuminated plane. Here, a measurement of an effective light source distribution of an illumination optical system intends to cover not only a measurement of the σ distribution but also a measurement of the centroid (or center of gravity) of illumination light.
A projection exposure apparatus, which is conventionally used for the photolithography technique to manufacture fine semiconductor devices, projects a circuit pattern of a reticle (mask) onto a wafer and the like via a projection optical system, thus transferring the circuit pattern.
Along with the recent demand on finer and lower profile electronic devices, finer semiconductor devices mounted on these electronic devices are increasingly demanded. A higher quality exposure requires an effective light source optimal to a reticle pattern. An effective light source distribution depends upon an adjustment of a light intensity distribution near an exit plane of, e.g., a fly-eye lens to a desired shape such as a normal illumination shape, an annular illumination shape, and a quadrupole illumination shape. Moreover, a projection exposure apparatus is demanded to control its numerical aperture (“NA”), coherence factor σ (which is an illumination optical system's NA/a projection optical system's NA), and an effective light source, thus setting up a condition optimal to each of various characteristics.
Control over the coherence factor a needs a more accurate measurement of an effective light source distribution. A measurement of a light intensity distribution on a lens' incident pupil plane will provides the effective light source distribution. The measurement of the effective light source distribution also provides a measurement of the centroid of illumination light on the incident pupil plane, which causes an asymmetry of the imaging performance, i.e., the so-called telecentricity.
The conventional method of measuring the light intensity distribution on the pupil plane is seen, for example, in Japanese Patent No. 2,928,277, Japanese Patent Applications, Publication Nos. 2000-19012, 5-74687, and 2002-110540.
These conventional measurement methods require, in addition to a reticle, etc., a measuring apparatus such as a mirror, a condenser optical system, a relaying optical system, an image sensor and the like to be arranged on a plane or planes conjugate with reticle and wafer planes. However, an accommodation of these components in an exposure apparatus would increase the cost due to its limited capacity. In addition, such measurement methods should be applicable to an immersion exposure apparatus.
The present invention is directed to a measuring apparatus, as well as an exposure apparatus having the same, that can measure a light source distribution without an additional space.
A measuring apparatus according to one aspect of the present invention used for an exposure apparatus that includes an illumination optical system that illuminates a reticle using light from a light source, a reticle stage that supports and drives the reticle, and a projection optical system that projects a pattern of the reticle onto a plate measures an effective light source distribution as an incident angular distribution of the light on a reticle plane of the reticle includes a relaying optical system configured to guide the light that has passed the illumination optical system, and a housing that contains the relaying optical system and is mounted on the reticle stage in place of the reticle.
A measuring apparatus according to another aspect of the present invention in an exposure apparatus that includes an illumination optical system that illuminates a reticle by using a light from a light source, and a projection optical system that projects a pattern of the reticle onto a plate measures an effective light source distribution as an incident angular distribution on a reticle plane of the light, and includes a pinhole member configured to transmit the light that has passed the illumination optical system into a pinhole with a diameter determined by a blur of a geometrical optics and wave optics, and a deflector configured to deflect the light that has passed through the pinhole.
An exposure apparatus having the above measuring apparatus constitutes still another aspect of the present invention.
An exposure method according to another aspect of the present invention includes the steps of calculating a pupil transmittance distribution in a projection optical system based on a first effective light source distribution of the projection optical system acquired by the above measuring apparatus, and a second effective light source distribution derived from a pupil plane light intensity distribution measured on a plate plane using a light that has passed the projection optical system without a reticle, calculating an imaging performance by using a result of the calculating step and the first or second effective light source distributions, adjusting the effective light source distribution and/or the projection optical system by using the imaging performance, and exposing the plate based on the effective light source distribution and/or the projection optical system that have been adjusted.
A device manufacturing method including the steps of exposing a plate using the above exposure apparatus, and developing the plate exposed constitutes another aspect of the present invention.
Other objects and further features of the present invention will become readily apparent from the following description of the preferred embodiments with reference to the accompanying drawings.
Referring to the accompanying drawings, a description will be now given of exposure apparatuses according to various embodiments of the present invention. These exposure apparatuses are step-and-scan projection exposure apparatuses, but they are also applicable to a step-and-repeat type and other types.
A light that has passed the reticle goes through a reduction projection optical system 4, and is imaged onto a wafer 5. In order to adjust the positioning and focusing of the reticle 2, a device (alignment scope) 6 observes a mark formed on the reticle plane. The reticle plane observing device 6 includes an optical system 7 and an image sensor 8. The optical system 7 has an objective lens, a relay lens, and a magnification-variable erector lens, and the image sensor 8 uses a two-dimensional charge-coupled device (CCD).
The first embodiment measures an effective light source distribution using this configuration. In place of a usual reticle, a tool conjugate with the size of the reticle 2 and pellicle 3 is transported. The tool is located in a space from where the reticle 2 and the pellicle 3 are removed, thus needing no additional spaces.
This embodiment includes a collimator lens 11 explained later configured to collimate the light and a deflector 12 configured to bent the light, and reduces a thickness of the housing within the size of the reticle 2 and the pellicle 3. The tool uses a driving mechanism for the reticle stage 2a and the alignment scope 6, and requires no dedicated driving mechanism.
The light relayed from the light source by the illumination optical system 1 narrows down to a necessary area through a masking blade 9 that restricts the illumination area on the reticle plane, and enters a pinhole 10. The pinhole 10 may be slightly defocused from a plane conjugate with the reticle plane for spatial convenience. The collimator lens 11 substantially collimates the light that has passed through the pinhole 10. By using the deflector 12 such as a mirror or prism, the illumination light having been turned into collimated light enters a Fourier transforming optical system 13. An aerial pupil image is formed by the Fourier transforming optical system 13 at a position conjugate with the observation plane of the reticle plane observing element 6, and it is imaged at an approximately conjugate image sensor 8, thus measuring a light amount distribution through image processing. Based on this light amount and an observation diameter in the image sensor 8, a coordinate is calculated for a conversion into an effective light source distribution of an illumination optical system. The first embodiment observes an image using the image sensor 8, and the position of the pinhole 10 and the observation position of the image sensor 8 may be in an imaging relationship.
Stopping down the illumination light by the masking blade 9 reduces the influence by the stray light (unnecessary light). An aperture of the masking blade 9 should not be so small as to shut out the light incident upon the pinhole 10, and it should be, desirably, set as small as possible. The pinhole 10 is formed on a Cr surface through etching. Desirably, the size and density of the Cr surface should be so large that stray light does not leak from illumination light stopped down by the masking blade 9. For control over the largely remaining Cr surface transmission, a mechanical light shield is provided with a sheet metal tiger-den material with holes around the pinhole 10.
It is desirable that the diameter of the pinhole 10 be so large as to measure a light intensity distribution with sufficient resolving power. As a factor for determining the resolving power, a blur of the distribution and the image sensor's resolving power can be considered. Here, using
P=2×IllNA×S/L (1)
The illumination light that has passed the pinhole 17 with a discrete angular distribution characteristic has a divergent angle due to diffraction, i.e., an airy disk 22. The following equation is met with a wavelength λ and a pinhole's diameter φ, where A is a divergence angle 23 that corresponds to a first zero-point radius of this airy disk 22:
A=3.83×λ/φ/π (2)
Since respective light that has passed through the fly-eye lens 15 are each affected by the diffraction, use of a pinhole diameter satisfying the following formula enables a light from each fly-eye lens to be wave-optically separated and measured:
P>2×A (3)
When using a lens system for measuring an effective light source distribution, a blur due to an aberration needs to be considered. Assume that the image sensor 8's observation range radius corresponding to the illumination system's NA 19 is H, and an RMS spot diameter due to an aberration is D. D can geometrically-optically separate each light when satisfying the following formula:
D<2×H×S/L (4)
Besides these items, it is desirable to consider the dose incident on the image sensor, and the image sensor's resolving power.
These formulas define a pinhole diameter with a desired resolving power. An objective of setting a pinhole diameter is to separate light from a fly-eye lens for measurement, or to specify an amount of blurring at the beginning so as to set a pinhole diameter. In view of a dynamic range of the image sensor 8, light from the fly-eye lens may not be separated. The first embodiment sets the diameter of the pinhole 10 to φ50 μm. In order to control blurring caused by an aberration of a lens system, a tolerance for a lens system's distortion is relaxed, and the distortion is corrected by using image processing software.
The above pinhole diameter determination method can be applied across all the pupil plane light intensity distribution measurements for an exposure apparatus using a pinhole.
The illumination light on a reticle plane contemplates a maximum NA of 0.25 to 0.35 for an immersion exposure apparatus, although depending on an illumination condition. In order to keep the tool within the size of the reticle and pellicle, it is necessary to control a spread of the effective diameter of the illumination light that has passed the pinhole 10. For this purpose, the first embodiment has prepared a collimator lens 11 that collimates the illumination light that has passed through the pinhole 10. Besides using a collimator lens, it is possible to use a deflector having a curvature.
In this embodiment, the reticle stage supports the reticle and moves in a y direction or a scanning direction in
A method of moving the tool to a measuring position may be a scanning search for a light source, or may use an alignment mark. The scanning search for the light source takes time before measuring. On the other hand, the alignment mark has a high reproducibility, and enables a light intensity distribution to be measured by a designated coordinate.
The optical system 25 can move in the X direction, and provides an measurement of an illuminating system's effective light source distribution with a required image height. If each optical system 25 uses the same optical system, they are viable even if rotated by 180°. This configuration calibrates the optical system 7 of the reticle plane observing device 6 and the optical system 25 of the tool.
The first embodiment almost maintains telecentric an aerial pupil image of the Fourier transforming optical system 13. The collimated light at the observation position of the reticle plane observing device 6 controls a size measurement error caused by a defocus. A known aperture stop and the like calculate an absolute value of a σ distribution observable by the image sensor 8 and the central position of the image sensor 8, and correct an optical system's aberration such as a distortion.
The reticle plane observing device 6 can defocus the observation position. Defocusing the observation position, measuring two or more pupil plane illumination distributions, and finding its centroid shift can measure the inclination of illumination light. This result represents an inclination of the illumination light on the reticle plane, and the so-called telecentricity is obtained by subtracting telecentricity of the projection optical system from it. Of course, when a reference position of the image sensor is previously measured for each image height, an measurement at only the best focus position provides the inclination of the illumination light.
When the tool or reticle plane observing device 6 houses a beam splitter and analyzer, and they are switched, an effective light source distribution can be measured for each polarized light at the time of polarization illumination.
The configuration of the present embodiment can provide a measurement of an illumination system's effective light source distribution corresponding to a scan σ. In a step-and-scan exposure apparatus, the light intensity distribution at each position imaged on a wafer will become a scanning accumulation with the entire illumination slit. A scanning effective light source distribution is different from a stationary effective light source distribution. It is important to measure an actual effective light source distribution corresponding to the scan σ.
A scanning effective light source distribution can be measured with a slit that elongates in the scanning direction instead of a pinhole. An accumulation of obtained light intensity distributions in the scanning direction provides an actual scanning effective light source distribution. Use of a slit that is longer in the scanning direction than the illuminated area may provide a scanning effective light source distribution through only one measurement.
The effective light source distribution of an illumination optical system measured in this embodiment can be utilized to adjust and correct the illumination optical system. Since measurement results are outputted from the reticle plane observing device 6, its calibration with the exposure apparatus body is easy. The projection optical system's transmittance can be obtained when an effective light source distribution on a wafer plane is compared with an effective light source distribution on a reticle plane of this embodiment.
Next follows a second embodiment that uses an effective light source distribution measuring tool, which also includes an image sensor according to the present invention, to measure an effective light source distribution of an illumination system. The second embodiment is different from the first embodiment in that the tool also includes the image sensor, and measures an effective light source distribution without using the reticle plane observing device 6.
This embodiment transports a tool that is as large as the reticle 2 and pellicle 3 in place of the ordinary reticle, and has a pinhole, an optical system, and an image sensor.
This embodiment uses a wireless communication as a data communication means. Data acquired by the image sensor 28 is sent directly to the apparatus body using a wireless communication unit 29.
A third embodiment of the present invention uses the effective light source distribution obtained in the first and second embodiments to adjust an illumination system and evaluate the imaging performance.
Next, the light intensity distribution on the pupil plane is measured while the reticle 2 and the pellicle 3 are removed (Step 1010). For an immersion exposure apparatus, step 1010 provides two wafer stages—one for exposure and one for measurement. Thus, the stage for measurement can be driven for measurements during exposure, and the throughput improves. In addition, the wafer stage for exposure is for immersion in this embodiment, but the stage for measurement is not provided for immersion. Therefore, even if a pinhole that is formed by etching glass with Cr is used, no problems occur about the concentration and durability of Cr against the immersion liquid.
The transmittance distribution of the pupil in the projection optical system 4 is calculated from the results of steps 1006 and 1010 (Step 1012). In addition, an aberration of the projection optical system is measured in the exposure apparatus simultaneously (Step 1014).
The imaging performance is calculated based on the effective light source distribution is calculated on the reticle plane or the wafer plane, the transmittance distribution of the projection optical system's pupil, the projection optical system's aberration, the projection optical system's NA, the reticle pattern to be used, and the wavelength to be used (Step 1016). This imaging performance is compared with a target imaging performance (Step 1018), and if out of the permissible range, at least one of the effective light source distribution, the projection optical system's aberration, and its NA are adjusted (Step 1020). The comparing items with the target imaging performance include the critical dimension (“CD”) uniformity and the HV difference, OPE calculation, etc. The second and subsequent adjustments direct to the target values by calculating sensitivity. When the current imaging performance is permissible range from the target imaging performance, the adjustment ends (Step 1022). An automatic series of adjustments are preferable. The pupil's transmittance and aberration of the projection optical system need not be measured in the exposure apparatus body. However, they may be measured when the exposure apparatus is assembled, and the data may be used for calculation.
This embodiment can quickly adjust the effective light source distribution of an illumination system and aberration of a projection optical system. This optimization, which is not for each unit such as an illumination system or projection system, but for a whole exposure apparatus that is installed with an illumination system and projection system, is suitable for actual exposure of a wafer.
Next, referring to
The above embodiments can measure an effective light source distribution without an additional space. These embodiments can be applied to existing exposure apparatuses having no means for accurately measuring effective light distributions. These embodiments calculate a blur of a light that passes a pinhole, and correct it in accordance with the geometrical optics and wave optics, thus providing a measurement of the original light source distribution. Since the illumination light is maintained almost telecentric at the observation position of the image sensor, the incidence sensitivity characteristic on the image sensor is small, and the measurement can be more accurate than ever.
The illumination system's effective light source distribution acquired from these embodiments can be utilized as an index in adjusting a coherence factor a and telecentricity obtained from the effective light source distribution. Use of an acquired light intensity distribution for an imaging simulation enables the imaging performance to be evaluated. Actually measured data can be directly used as an illumination system's light intensity distribution in designing a diffraction optical element.
Further, the present invention is not limited to these preferred embodiments, and various variations and modifications may be made without departing from the scope of the present invention.
This application claims a foreign priority benefit based on Japanese Patent Applications No. 2005-123006, filed on Apr. 20, 2005, which is hereby incorporated by reference herein in its entirety as if fully set forth herein.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2005-123006 | Apr 2005 | JP | national |
