TWI437225B - Contamination monitoring and control techniques for use with an optical metrology instrument - Google Patents

Description

Pollution vision and control technology using optical metrology equipment

This application is in the field of optical metrology, and in particular, the present invention relates to optical metrology that can be performed in vacuum ultraviolet (VUV).

In one embodiment, a method of performing accurate and repeatable optical metrology in vacuum ultraviolet (VUV) is provided. In one embodiment, the techniques disclosed herein can be used to ensure that a vacuum ultraviolet reflectometer produces highly stable and repeatable results in the presence of gases and surface contaminants. In another embodiment, the techniques disclosed herein provide a method for obtaining accurate reflectance data from a sample whose surface can be contaminated.

Optical metrology techniques have long been used in process control applications in the semiconductor manufacturing industry due to their contactless, non-destructive, and generally high throughput properties. Most of these tools operate in a portion of the spectral region that spans deep ultraviolet to near infrared wavelengths (DUV-NIR, typically 200 to 1000 nm). The continual advancement of the development of smaller devices containing thinner layers has challenged the sensitivity of this test equipment. Efforts to develop optical metrology devices using shorter wavelengths (below 200 nm) have been considered, with greater sensitivity to small changes in processing conditions. A method for performing optical measurements at shorter wavelengths, such as systems and methods for vacuum ultraviolet (VUV) reflectometers, is described in U.S. Application Serial No. 10/668,642, filed on Sep. 23, 2003, In U.S. Patent No. 7,067,818, the disclosure of which is hereby incorporated by reference in its entirety in its entirety in its entirety in the the the the the the the the the the

Contamination of optical surfaces such as windows and mirrors is a serious obstacle to the operation of optical devices in VUV. Moisture and residual molecules (especially hydrocarbon compounds) can deposit on such surfaces over time to significantly reduce their effectiveness. These effects have been attributed to previous studies due to their impact on the design, development and performance of 193 and 157 nm lithography exposure tools.

In order to ensure that the extreme sensitivity provided by the VUV optical metrology equipment theory is realized in practice, it will be highly desirable to develop an apparatus having the inherent ability to reduce, remove or completely eliminate fouling on its optical surface. Moreover, if this self-cleaning capability can be achieved without the need to add potentially expensive and complex components, it would provide a great benefit to the tool holder.

When a contaminated layer appears on the surface of the sample under investigation, it can significantly contribute to the measured optical response in VUV resulting in inaccurate and/or erroneous results. These effects include ultra-thin films in the sample (<100 It has a specific meaning that the thickness of the film itself can be comparable to the thickness of the contaminated layer.

One contemplated technique for improving the measurement of a semiconductor wafer by removing a contaminated layer in a cleaning step includes using microwave radiation and/or radiant heating prior to measurement. Although enhanced measurement repeatability using this approach has been reported, this approach requires coupling a separate cleaning system to an existing measurement system resulting in increased system cost and design complexity.

In view of these shortcomings, it is desirable to develop a metrology system that is capable of removing contaminants from the surface of the sample to ensure accurate and highly repeatable results. This equipment will be able to simultaneously clean and measure specific locations on the sample without exceeding the extra components required for normal measurement, reducing system cost and design complexity. In addition, this device will not require alignment of the independent cleaning subsystem with the independent measurement subsystem. In addition, this device will avoid unnecessarily "cleaning" the entire sample while ensuring consistent cleaning results at all measurement locations.

One embodiment of the disclosed technology provides a method of generating a controlled environment within a VUV optical metrology device and subsequently monitoring it by minimizing or completely eliminating the fouling of contaminants on the surface of the optical component that may result in performance degradation. technology.

Another embodiment discloses a technique for reducing surface contaminants of optical components contained within an optical path (or sub-path) of an optical metrology device. This technique can be utilized in one embodiment without the need to couple or integrate additional components and/or test equipment to existing metrology devices.

Another embodiment discloses a technique whereby surface contamination on an optical component within an optical metrology device is monitored such that a cleaning procedure can be performed as deemed necessary. This technique can further enable independent optical paths of the device to be monitored independently of each other and subsequently cleaned.

In another embodiment, a technique for removing contaminants from the surface of the sample prior to the optical response of the record to ensure accurate results is disclosed. In an alternate embodiment, the technique can be implemented in a manner that does not require coupling or integrating additional components and/or test equipment to existing metrology devices.

In another embodiment, a technique for characterizing contaminants on the same surface is disclosed. In addition to providing an understanding of the nature of the contaminant itself, the technique also provides a means of performing accurate sample measurements, considering that a contaminated layer may be present on the surface.

In another embodiment, a method of controlling an atmosphere in an optical metrology tool is provided. The method can include providing at least a first environmentally controlled chamber and a second environmentally controlled chamber configured to deliver a beam of light having a wavelength below the DUV wavelength. The method can further include reducing the concentration of the optically absorbing material in at least one of the first and second environmentally controlled chambers by utilizing a vacuum evacuation technique, the at least one of the first and second environmentally controlled chambers One can be a controlled atmosphere chamber. Additionally, the method can include backfilling the controlled atmosphere chamber with a non-absorbent gas to improve optical performance by increasing the pressure within the controlled atmosphere chamber above a vacuum evacuation pressure level; and at a transmission wavelength lower than the DUV wavelength The beam simultaneously controls the atmosphere chamber to be backfilled.

Another embodiment includes a method for controlling an atmosphere in an optical metrology tool, which can include providing at least one environmentally controlled sample chamber and an environmentally controlled optics chamber, the sample chamber and optics chamber Each is configured to deliver a beam of light having a wavelength below the DUV wavelength. The method can further include reducing the concentration of moisture or oxygen in at least one of the sample chamber and the optics chamber from an ambience state by utilizing a vacuum evacuation technique, the reduced sample chamber and the optics The at least one of the chambers is a controlled atmosphere chamber. The method can also include: backfilling the controlled atmosphere chamber with a VUV non-absorbent gas to reduce contaminant migration by increasing a pressure within the controlled atmosphere chamber above a vacuum evacuation pressure level; and at a transmission wavelength A beam of light below the DUV wavelength is simultaneously in the backfill state of the controlled atmosphere chamber.

In another embodiment, a method for controlling an atmosphere in an optical metrology tool can include providing at least one environmentally controlled sample chamber and an environmentally controlled optics chamber, the sample chamber and optics chamber Each is configured to deliver a beam of light having a wavelength below the DUV wavelength. The method can further include providing an iso-beam optical path and a reference beam optical path that matches an optical path length of the reference beam optical path. The method can also include reducing the concentration of moisture or oxygen in at least one of the sample chamber and the optics chamber from an ambience state by utilizing a vacuum evacuation technique, the reduced sample chamber and the optics The at least one of the chambers is a controlled atmosphere chamber. The method can further include: backfilling the controlled atmosphere chamber with a VUV non-absorbent gas to improve optical performance by increasing a pressure within the controlled atmosphere chamber to a vacuum evacuation pressure level; and at a low transmission wavelength The beam at the DUV wavelength is simultaneously in the backfill state of the controlled atmosphere chamber.

In another embodiment, a method of determining an environmental pollution condition in an optical metrology tool is provided. The method can include obtaining a first intensity measurement from a reference sample at a first time and obtaining a second intensity measurement from the reference sample at a second time. Additionally, the method can include: analyzing the first and second intensity measurements; and determining the optical metrology tool from the analysis of the first intensity measurement and the second intensity measurement based on the variation between the first intensity and the second intensity Whether the environmental pollution status is suitable for further use.

In another embodiment, a method of determining an environmental pollution state in an optical metrology tool operating at least at a wavelength below a DUV wavelength is provided. The method can include obtaining, at a first time, a first intensity spectrum measurement from a reference sample, the first intensity spectrum measurement comprising at least a plurality of wavelengths below a DUV wavelength; and obtaining a first from a reference sample at a second time A two intensity measurement, the first intensity spectrum measurement comprising at least a plurality of wavelengths below the DUV wavelength. The method may further include: analyzing the first intensity measurement and the second intensity measurement; and determining an environment of the optical metrology tool from the analysis of the first and second intensity measurements based on the variation between the first intensity and the second intensity Whether the pollution status is suitable for further use.

A further understanding of the nature of the advantages of the concepts disclosed herein can be realized upon examination of the following description and associated drawings.

In order to enhance the sensitivity of optical metrology devices for challenging applications, it is necessary to extend the wavelength range in which such measurements are performed. In particular, it is advantageous to utilize shorter wavelength (higher energy) photonic systems that extend to and beyond the electromagnetic spectral region known as vacuum ultraviolet (VUV). In general, the vacuum ultraviolet (VUV) wavelength is considered to be less than the wavelength of the deep ultraviolet (DUV) wavelength, i.e., less than about 190 nm. Although there is no universal cutoff for the bottom end of the VUV range, some in the field can consider VUV to be the end of the termination and extreme ultraviolet (EUV) range (for example, EUV can be defined as a wavelength less than 100 nm) . Although the principles described herein can be applied to wavelengths greater than 100 nm, these principles are generally applicable to wavelengths less than 100 nm. Thus, as used herein, it will be understood that the term VUV means to indicate a wavelength that is substantially less than about 190 nm, whereas VUV is not meant to exclude lower wavelengths. Thus, as used herein, VUV is generally meant to include wavelengths generally less than about 190 nm without excluding low end wavelengths. In addition, low end VUV can generally be understood as a wavelength below about 140 nm.

Substantially all forms of matter (solids, liquids, and gases) exhibit increasing optical absorption characteristics at the VUV wavelength are generally true. In part, it is this rather basic material property that is itself responsible for the increased sensitivity that can be used for VUV optical metrology techniques. This results in small changes in process conditions that produce undetectable changes in the optical behavior of the material at longer wavelengths that can induce substantial and easily detectable changes in the measurable characteristics of such materials at the VUV wavelength. This highly nonlinear dependence of the photon absorption cross-section facing the wavelength presents a great opportunity for VUV optical metrology test equipment; unfortunately, it also introduces correlation complication.

One such complication is related to the fact that VUV radiation cannot be transmitted via standard atmospheric conditions. O ₂ and H ₂ O strongly absorb VUV photons, and thus such materials must be maintained at a sufficiently low level (e.g., typically sub-PPM) to allow transmission along the optical path of the metrology test apparatus. For this purpose, a vacuum or purification method using a non-absorbed purge gas such as nitrogen, argon or helium is generally used. The purification process can be quite effective in reducing the concentration of the gases O ₂ and H ₂ O, but it is difficult to remove the absorbed water in a timely manner without using a vacuum technique. Although the method of vacuum is more effective to remove the absorbed H ₂ O, but due to the increase of the mean free path of the substance subjected to reduced pressure at the surface contaminants inadvertently promote migration.

In considering these different considerations, it follows that neither the individual vacuum method nor the individual purification methods constitute the most efficient method for generating and subsequently maintaining the controlled environment required to perform optical measurements under VUV. Instead, a program that combines the components of both technologies is needed to ensure optimal tool performance. Briefly, a vacuum is initially used to rapidly reduce the concentration of moisture and oxygen species to an acceptable level, followed by backfilling the pressure within the apparatus with a non-absorbent gas.

In one embodiment, in order to reduce the absorbed moisture concentration to an acceptable level, it is generally desirable to reduce the pressure in the tool to somewhere near 1 x 10 ^-5 to 1 x 10 ^-6 Torr. Care must be taken during this time to ensure that the optical surface of the device is not exposed to VUV radiation as this can result in the formation of a layer of contaminants for photodeposition. This situation can be easily avoided by placing a baffle to block or close the VUV source before evacuating.

The time required to achieve the necessary vacuum conditions depends on many aspects of the system (ie, temperature, surface area of the equipment capacity, pumping speed of the vacuum system, etc.), but it will be greatly affected by the sample upload process The surface is pushed by the exposure of the atmosphere. Thus, it follows that the system evacuation time will be minimized via the use of the intelligence of the upload lock mechanism; thereby increasing system throughput and reducing the migration of contamination.

The pumping cycle time can also be shortened by applying energy to the absorbed water bed via mechanical, thermal or radiation methods. Mechanical energy can be applied during the initial portion of the pumping cycle by venting in a controlled purge gas. Thermal energy can be applied by heating the walls of the device, however this approach promotes migration migration and introduces mechanical instability. UV lamps can also be used to transfer energy directly to the absorbed water molecules, but at the same time can cause photodeposition of contaminants.

Once the concentration of the absorbing material within the device capacity is substantially reduced, the controlled environment required to support the operation (ie, allowing sufficient transmission of VUV photons) is achieved by backfilling the device with a high purity non-absorbed gas. . While pollution considerations can encourage the pressure on the equipment to be maintained at a higher level, mechanical considerations generally limit the practice operating pressure to somewhere near atmospheric conditions. Therefore, pressures in the range of 300 to 700 Torr are usually used. Therefore, optical performance can be improved by increasing optical transmission. Optical transmission can be increased by reducing the oxygen and/or moisture content by using vacuum techniques. In addition, optical performance can also be increased by inhibiting migration of the absorbing material (i.e., contaminants) from the surface via the use of backfilling techniques. Absorbing contaminants that migrate from the surface of the device and adhere to the optical surface can significantly degrade the performance of such components, resulting in a mirror having reduced reflectivity and a window having reduced transmission characteristics.

Over time, it is expected that the quality of the controlled environment within the equipment will be degraded for a variety of reasons; including (but not limited to) degassing from internal surfaces, penetration through materials, penetration through vulnerabilities (real and virtual) Leakage and degassing from the sample itself. Any of these mechanisms may result in an increase in the concentration of the absorbing material within the device and a corresponding reduction in the optical flux of the system. It follows that the state of the environment is monitored so that appropriate steps can be taken to make it possible to restore it when needed; thus ensuring that there is no compromise measurement accuracy.

While there are many independent methods, systems, and sensors for monitoring the quality of a closed environment, the most straightforward and certifiable most useful approach is to utilize a reference sample to utilize the optical components of the metrology device to track optical flux. Flowchart 100 in Figure 1 illustrates how environmental monitoring can be achieved in this manner. First, at step 110, the capacity of the apparatus is evacuated to a predetermined pressure P _L using a suitable vacuum system. The device is then backfilled to a predetermined measured pressure P _H using a non-absorbed gas at step 120. As shown at step 130, once the pressure has been measured at time t ₁ immediately recording the intensity of the spectrum from the reference sample. It is assumed that the concentration of the absorbing material is at its lowest achievable level at this point in time. Subsequently thereafter as shown at step 150 at time t ₂ is re-collected from a reference sample intensity spectrum of step 140 for a predetermined period of performing measurements of the test sample. As shown at steps 160 and 170, the ratio of the two intensity spectra (time t ₁ and time t ₂ ) from the reference sample is then calculated and analyzed to determine the concentration of the absorbing material. At step 180, the determined concentration is then compared to a user defined threshold to determine if the environment in the device is suitable for supporting further measurements. As indicated at step 190, if the environment is suitable, the measurement of the test sample is again performed for another predetermined period by returning control to step 140. Conversely, if the environment is no longer suitable (as indicated at step 195), the environment is regenerated by resuming the evacuation/backfill procedure at step 110.

Figure 2 presents reference measurements collected in a non-absorbent atmosphere containing oxygen and water at a minor concentration (1, 5, 10, and 20 PPM as indicated by curves 200, 205, 210, and 215, respectively). The ratio of reference measurements collected under non-absorbed gas. As is apparent from the graph, the transmission through the controlled environment decreases significantly at wavelengths below 190 nm as the concentration of oxygen and water increases. In the case of prior knowledge of optical path lengths, equipment pressures and absorption cross sections for oxygen and water, the concentration of such substances can be readily determined by analysis.

Alternatively, the ratio of reference measurements at discrete wavelengths can also be used to provide simple monitoring of the quality of the controlled environment. In Figure 3, for VUV wavelengths of 124.6 nm, 144.98 nm, and 177.12 nm, reference measurements performed under non-absorbed gases containing traces of oxygen and water are compared to reference measurements collected under pure non-absorbed gases. The ratio is depicted as a function of oxygen and water concentration (shown as curves 300, 305, and 310, respectively). In practice, the measured transmission value can be compared to a user defined threshold to determine the state of the controlled environment. The actual wavelength used in the comparison can be selected based on the absorption cross section of the absorbing material.

The environmental monitoring program described herein assumes that the spectral intensity of the VUV source has no appreciable change between the initial reference measurement time and the final reference measurement time. While this may be a reasonable assumption in many instances, it is noted that the spectral intensity of the source can be independently monitored to account for intensity fluctuations in the event of a significant change expected.

The period in which the device can be operated (and sample measurements can be reliably performed) before the re-production of the controlled environment can vary significantly depending on the way the device is designed and operated. For a preferred design (ie, sealed) system that is operated to minimize exposure of the inner surface (ie, where the sample is introduced via the upload locking mechanism), the change in the controlled environment will typically occur at a significant longer than the amount The time scale of the time scale required to measure a given sample. It is therefore generally possible to measure many of these samples before re-generating the controlled environment. In any case, the reference measurement interval can be adjusted as needed so that short intervals can be used in less stable environments and longer intervals can be used in more stable environments.

The controlled "measurement" environment established in the process outlined in the flow chart of Figure 1 can be generated by backfilling the device capacity to a predetermined pressure. Once this state is reached, the flow of purge gas to the device can be interrupted. An alternative to operating this device may be to equip the device to a particular "release" pressure using a purge valve device such that the purge gas can flow continuously through the device. In principle, the need to "regenerate" the measurement environment can be reduced or eliminated altogether by limiting fouling of the absorbing material like oxygen and water. In practice, the implementation of this approach is difficult due to the high flow rate required to maintain a sufficiently low concentration of absorbing material (due to backflow of contaminants via purged exhaust). In addition, continuous purification can initiate pressure fluctuations that can adversely affect measurement stability.

Furthermore, a second difficulty associated with the use of VUV photons in optical metrology test equipment is related to problems with surface contamination. Thinly contaminated layers that only marginally affect the performance of the optical surface at longer wavelengths can significantly degrade the response of these components at the VUV wavelength. In addition to the absorbed layers that are expected to form on the optical surface under normal atmospheric conditions, when the VUV photon is used in a contaminated atmosphere, the organic film and polyfluorene may be undesirably photodeposited on such surfaces. Base film.

An example of the effect of a VUV response of a contaminated layer on an optical surface is provided in Figure 4, which shows the reflectivity of a "clean" surface with "slightly contaminated" and "more polluting" as shown by curves 400, 405 and 410, respectively. The reflectance of the surface is compared. It is apparent from the figure that the reflectance of the Si surface in the VUV region is significantly degraded with the accumulation of contamination. Since the photons in the optical metrology tool typically encounter many of these surfaces as they travel from the source to the sample and ultimately to the detector, it follows that even a small reduction in the optical performance of each surface can seriously The overall optical flux of the device.

Fortunately, in many cases, such fouled deposits on the optical surface may be reduced, removed or completely eliminated via VUV illumination in an atmosphere containing traces of oxygen. Upon exposure to the VUV wavelength, the diatomic oxygen is dissociated into atomic oxygen, which in turn reacts with the diatomic oxygen to form ozone. Both atomic oxygen and ozone are highly reactive and can oxidize surface contaminants to form a gaseous product which can then be released.

This photolithography cleaning process is schematically illustrated in FIG. In FIG. 5A, a contaminated optical surface 500 having contamination 505 is shown. In FIG. 5B, the contaminating optical surface 500 is exposed to VUV radiation 510 under an oxygen containing atmosphere, thereby causing contamination 505 to be removed from the surface via the reactions outlined above. In Figure 5C, the resulting "clean" optical surface 520 is presented.

In the case of certain contaminants (eg, halogenated organic compounds and organopolyoxins), the photodeposition reaction is irreversible and thus it is not possible to completely remove it via a photolithographic process. In such instances, the use of VUV photons to illuminate the surface will only result in continuous growth of the contaminated layer, under an atmosphere consisting primarily of oxygen, even with trace levels of contaminating compounds.

In a more general case, it can be contaminated by oxygen, which contributes to the pollution of photodeposition and photoetching (that is, contamination formed by reversible reaction) and photodeposition but not by photoetching (ie, via irreversible reaction). A surface is exposed to VUV radiation in an atmosphere of contamination. There are at least three different processes occurring in such environments; reversible deposition processes, irreversible deposition processes, and reverse reaction processes via etching. The three processes are graphically depicted in FIG. 6 by a hollow circle 610 and a solid circle 620 (reversible deposition and irreversible deposition, respectively) moving toward the surface and by a hollow circle 610 (etching) moving away from the surface. The relative rates associated with such processes will depend on various factors including the surface concentration of the absorbed contaminants, the oxygen concentration, the absorption cross-section of the contaminants, and the associated VUV photon flux.

Since oxygen plays a key role in the photolithography process, it follows that it can be beneficial to monitor and control the concentration of oxygen contained within the device's capacity. In this manner, trace levels of oxygen (or clean dry air) can be intentionally added to the backfilled non-absorbent gas to facilitate the cleaning process without significantly compromising the VUV photon flux for measurement purposes. Depending on the relative rate of contamination and etching, it may be possible to first operate the VUV optical test apparatus in a manner that does not promote fouling of the undesirable material (i.e., where the etch rate exceeds the contamination rate). In the event that a significant amount of contaminant is already present on the optical surface, the cleaning time required to remove such films can be greatly reduced by temporarily increasing the oxygen concentration above the level normally used for data retrieval.

The concentration of trace amounts of oxygen intentionally added to the controlled environment of the device can be monitored using the environmental monitoring method of Figure 1 to track the unintentional accumulation of oxygen and moisture (due to loopholes, etc.). A mass flow controller can be used to accurately add trace amounts of gas, such as oxygen, to the capacity of the device. A practical implementation can be achieved by modifying the method of Figure 1 to add a fixed amount of oxygen to the device volume directly prior to the sample concentration test. To verify that an appropriate amount of oxygen is added to the system, the intensity spectrum from the reference sample can be recorded and compared to the intensity spectrum obtained immediately after backfilling with non-absorbed gas. Thus, for example, trace amounts of oxygen (eg, in the range of 1 ppm or less and preferably in the range of 0.1 ppm or less) can be added to a controlled environment to accelerate multiple cleaning mechanisms. In an embodiment, the controlled environment may be in sub-atmospheric pressure.

In principle, all optical surfaces in VUV metrology equipment are susceptible to contamination effects. This includes not only the optical components (i.e., windows, beam splitters, mirrors, etc.) but also the surface of the sample itself. Since the concentration of contaminants in the metrology equipment can be expected to vary significantly with changes in the atmosphere of the tool and through the introduction of the sample, it is concluded that in order to achieve optimal system performance, the accumulation of such contaminants over time (or removal) is monitored. It can be beneficial to take appropriate cleaning methods.

If such surfaces are to be effectively cleaned via a photolithography process, it follows that the VUV relative flux distribution received by the optical surface during cleaning will closely match the VUV relative flux distribution received during the initial photolithography process. Therefore, in order to achieve optimal cleaning results, it may be desirable to accurately configure and align the VUV cleaning system with the VUV optical metrology device in a manner that ensures a close match to the VUV flux distribution received by the optical surface.

Therefore, it can be advantageous to integrate the VUV cleaning capability into an optical metrology test apparatus without the need for precise configuration and alignment. Moreover, system capabilities and cost requirements can be greatly reduced if such capabilities can be incorporated into a component that requires few additional components in addition to components already present in the optical metrology tool. One of the innovative methods used to accomplish this is to combine a reference sample to utilize the optical components of the metrology device itself to track the pollution status of the system.

Flowchart 700 in Figure 7 illustrates how contamination monitoring can be achieved in this manner. First, a series of n measurement measurements of reference sample strengths are performed on the reference samples as shown in step 710 and recorded. Each measurement exposes the optical path of the device (and each of the optical components encountered along the way) to a flux of VUV radiation for a known time interval. Thus, each measurement can be considered to impart a specific dose of VUV radiation to the optical surface of the device. Next, the reference sample intensity can be analyzed as a function of the number n of measurements. For example, the reference sample intensity can be plotted as measured in step 720 as a function of the number n of measurements. Thus, by recording the intensity at the detector as a function of the number of measurements, it is possible to effectively track the optical flux of the system as a function of VUV exposure. In an embodiment, the number of measurements may be 10 or less. If the system environment is fully controlled, the number of measurements can be only 2.

After such reference measurements, the recorded results can be analyzed to determine if the cleaning process is complete and thus the optical flux of the tool is stable. For example, as shown in step 730, a curve of the reference sample intensity as a function of the number n of measurements can be analyzed to determine the state of contamination, and if unstable, the time t _{clean is} required to achieve stability. If the system is found to be stable, sample measurements can be performed. For example, as shown in step 740, if the difference between the reference sample intensities of two successive measurements (n and n-1) is in a range (the range is in a desired amount associated with the "clean" state Within the repeatability test, the state is considered to be stable. However, as shown in step 745, if the difference between the reference sample intensities of two successive measurements (n and n-1) is greater than the range (the range is in a desired repeat associated with the "clean" state Within the sex), the state can be considered unstable. If the system is found to be unstable (indicating that the cleaning process is not completed), the exposure dose required to achieve system stability can be estimated. As indicated in step 750, the exposure dose can be converted to an effective measurement time of the exposeable system (and reference sample). This exposure can be performed in a single event in one embodiment as compared to exposure through a series of individual reference measurements. After the system is exposed to the necessary cleaning dose, the series of monitoring measurements can be performed again on the reference sample as indicated by the re-evaluation step 760. This process can be repeated until it is confirmed that the device is in a stable "clean" state. As shown in the exemplary technique of FIG. 7, the change between the two intensity measurements is determined by subtracting one intensity measurement from the other. However, it should be understood that the difference or variation of the two intensity measurements can be identified by comparing the wide range of two measurements, and thus the variation between the two measurements can be quantified in a wide range of ways. For example, a ratio can also be used to quantify the change. Therefore, it should be understood that a wide range of mechanical methods can be used to analyze, compare, and quantify the measurement data while still utilizing the concepts described herein. In addition, although the description is made regarding the evaluation of two successive measurements (n and n-1), it should be understood that the two measurements are not necessarily sequential but only two measurements can be evaluated to determine one measurement to another measurement. Changes.

It follows that by tracking the stability of the system (varied with time, usage, etc.), the details associated with the cleaning process can be adjusted (ie, the reference measurement frequency, the effective dose per exposure, and the presence in the device capacity). Trace concentrations of oxygen, etc.) to ensure effective equipment cleaning and thus enhanced system stability. In this mode, the tool can be operated in a way that optimizes device performance.

An example of a potential result of this process is presented as curve 800 in Figure 8, where the normalized intensity from a reference sample at a single VUV wavelength is plotted as a function of the effective dose. It is apparent from the graph that for a period prior to stabilization, the intensity from the reference sample increases substantially linearly after exposure. It follows that if the cleaning is not performed, the ability of the optical device to perform an accurate measurement will be significantly compromised before this cleaning is completed.

It is also considered that the non-optical surfaces within such devices can act as a source of contaminants that can be absorbed on the surfaces and can be released at a later time. Therefore, it is desirable to fabricate such surfaces in a manner that minimizes the probability of adhesion to potential contaminants. This may involve coating of a particular processing process and/or a suitable coating to ensure vacuum compatibility is achieved.

In order to reduce the migration of contaminants from unrelated surfaces within the device to the optical elements, it is advantageous to operate the tool in a manner that minimizes the evaporation rate of the material. At a given temperature, the rate of evaporation of the molecules will increase as the pressure of the atmosphere decreases. In addition, the average free path of these molecules will also increase under these conditions. As a result, these molecules will exhibit a greater tendency to distribute themselves throughout the available capacity, thereby increasing the likelihood that the optical surface it encounters will adhere to. Therefore, from a pollution point of view, it is desirable to minimize the time required to maintain the capacity of the VUV optical test equipment at a reduced pressure.

Just as optical components in VUV metrology tools are susceptible to contamination effects, the samples to be measured are themselves susceptible to pollution effects. The degree of contamination of the sample will primarily depend on the environment in which the sample is exposed after it is produced and the duration of the exposure. Therefore, in order to assist in achieving accurate measurement of sample performance, it is necessary to properly address the surface contaminant layer that may be present. This is especially important in the case of ultra-thin films where the sample under investigation contains a thickness comparable to the thickness of the contaminant layer itself. This importance is further emphasized in the case where the contaminant layer exhibits a high absorbance relative to the ultra-thin film being studied.

A known approach to solving this problem prior to initial measurement is to attempt to completely remove the contaminated layer from the entire surface of the sample. Typically, the entire wafer cleaning is performed using thermal energy heating, microwave radiation, UV radiation, or some combination of these or other techniques. There are many difficulties associated with such overall wafer cleaning methods. Such systems tend to be relatively large and thus may be configured as separate systems located outside of the controlled environment of the VUV optical metrology device. Thus, the sample needs to be transferred between the cleaning system and the metrology equipment, thereby causing the possibility of recontamination.

In addition, although moisture can be easily removed using the entire wafer cleaning method, complete and uniform removal of other contaminants can prove problematic. This results from the difficulty of generating a strong flux correlation with sufficient energy, amplitude, and spatial uniformity to ensure complete removal of contaminants from all regions of the sample. Residual contaminants after cleaning can lead to inaccurate measurements of spatial uniformity of sample performance and misleading conclusions.

Individual point cleaning techniques are subject to shortcomings even when they are integrated into a controlled environment of VUV metrology equipment. To ensure accurate measurement results, a precision alignment point cleaning system and an optical metrology test equipment may be required to align the cleaning point location with the measurement point location.

Therefore, it would be advantageous if the point cleaning capability could be integrated into the optical metrology tool by using a common optics module to avoid alignment. Moreover, system capabilities and cost requirements can be greatly reduced if such capabilities can be incorporated into a component that requires few additional components in addition to components already present in the optical metrology tool.

The novel components that accomplish this will use the measurement radiation itself to clean and characterize the sample. Thus, discrete locations on the sample can be cleaned by exposure to the measurement radiation directly prior to measurement. In addition to completely eliminating cleaning/measurement alignments, this approach also avoids unnecessarily "handling" most of the sample surface area. However, such techniques as provided herein can still be utilized via the use of separate light sources for use as a clean light source and a metrology source. One or both of these sources may or may not be VUV sources.

Another benefit of this technology is its ability to easily measure and characterize the contaminant layer itself. The manner in which this can be accomplished is illustrated in the flow chart of FIG. By obtaining an optical response from the contaminated sample before and after removal of the contaminated layer and then analyzing the results, it is possible to determine the properties (thickness, optical properties, composition, roughness, etc.) of the contaminant layer. In the case of this knowledge of gripping, data can be collected from other contaminated locations on the sample and analyzed to characterize the performance of the sample. In other words, once the performance of the contaminant layer on a given sample is determined, it is in principle possible to accurately characterize other locations on the sample without first cleaning it. In addition to the obvious throughput advantages, this technique can also provide important information about the contaminant layer itself that can be used to improve the sample processing method due to the reduction in combined cleaning/measurement cycle times.

As shown in step 910 of Figure 9, the reflectance data can first be collected from a first location of a "dirty" sample having a contaminant film present. Next, at step 920, the first location on the sample is cleaned by exposing the location to the measurement radiation. At step 930, "clean" material may be collected from the first location after the contaminated film is removed in step 920. At step 940, clean data from the first location of the sample can be analyzed to determine the performance of the sample. At step 950, the measured data from the first location can be utilized to analyze the dirty data from the first location (from the data of step 910) to determine the performance of the contaminant membrane. Next at step 960, "dirty" data can be collected from another location of the sample in which the contaminant film is present. Next, in step 970, the performance of the sample can be determined for other locations in step 970 by using the "dirty" data obtained from other locations in step 960 and the measured performance of the contaminant film obtained in step 950. No cleaning steps are required in other locations. Multiple locations can be analyzed in this manner by returning control from step 970 to step 960 and repeating the process.

It follows that once the effectiveness of the pathway has been demonstrated, the method of Figure 9 can be used in a manner to determine the performance of the contaminant layer at a location or a number of locations on a given sample or series of samples, And then used during analysis of subsequent measurement locations on the same or different samples. Therefore, it is not necessary to analyze the contaminant layer at all measurement locations. Alternatively, where it is expected that the nature of the contaminant layer will vary significantly from one location to the next, each measurement location may be cleaned prior to data collection. In addition, it is also possible to measure each position on the sample before and immediately after cleaning to determine the performance of the contaminant layer and the underlying sample.

Of course, it is possible to combine this integrated point cleaning method with the use of other separate, entire wafer or spot cleaning techniques. This approach may be advantageous in situations where the sample is heavily contaminated and/or will measure many locations on a given sample. In such environments, a combination of cleaning methods can help speed up the cleaning process.

The same sample cleaning method as outlined herein can also be used to prepare calibration and/or reference samples utilized by the optical metrology system to ensure a high level of measurement accuracy is achieved. In addition, the performance of such "cleaned" samples can be monitored over time to track the "jiankang status" of such samples. In this way, repairs and/or substitutions of such samples can be scheduled to be consistent with other preventive maintenance activities.

In some instances, a further understanding of the characteristics of the contaminated layer itself can be obtained by observing the response of the layers to the cleaning process. That is, by recording the removal rate of the contaminated film as being varied with the accumulated dose, it is possible to partially determine the chemical nature of the contaminated film. Once the additional analysis data is used for correction, a library of functions for cleaning the response distribution can be generated. This library of functions can then be referenced during subsequent measurements that characterize unknown contaminants.

For example, Figure 10 presents exemplary cleaning response profiles 1000, 1010, and 1020 for three different contaminated layers. An exemplary cleaning response profile illustrates the distribution changes visible from different contaminated layers. Using the distributions stored in the clean response function library, the distribution from unknown contaminant membranes can be classified by comparison. These distributions can be directly compared or can be equipped using a parametric model in which the resulting parameter values can be used to distinguish pollutants.

Thus, as described above, the performance of the sample layer can be altered by exposing the sample to optical radiation, and the changes can be characterized via the use of measurements performed before and after exposure to optical radiation. In the examples described above, the change can include removing the contaminant layer from the sample. However, it should be understood that other variations in the sample can be characterized. Thus, for example, the following changes can be characterized: the contaminant layer does not need to be present but there is some other layer or part of the sample. In one embodiment, the performance of the layer or portion of the sample to be analyzed via the use of the optical metrology tool can be characterized. In this embodiment, the layer or portion may remain after exposure to optical radiation, although some state of the layer or portion may vary. Information about the original performance of the layer or portion can be obtained by characterization of the changes.

For example, exposure to optical radiation can alter the bond structure, material concentration, or other physical properties of the sample. In such instances, the original bond structure, material concentration/migration or other physical properties may be quantifiable based on the amount of change detected using a dose of optical radiation. Thus, how the layer reacts with optical radiation can reproduce useful information about the original properties of the layer. Layer changes can be analyzed via quantitative measurements of detected changes, or via comparisons with known optical response profiles, such as stored in a library of response distribution functions, or other techniques.

In one example, a nitrogen-containing cerium oxide film can be characterized via such techniques. For example, VUV optical exposure can preferentially affect the nitrogen contained in such films by migrating nitrogen and/or changing the bond structure of nitrogen held in such films. Thus, the detected optical response changes before and after exposure to optical radiation can provide useful information about the original state of the nitrogen bond (tight bonds, loose bonds, etc.), nitrogen concentration, or the like. Accordingly, it should be understood that the techniques provided herein in the general form provide detection of the same characteristics via analysis of the film before and after its performance is altered due to exposure to optical radiation.

To facilitate accurate and repeatable results from optical metrology devices operating under VUV, the environmental monitoring method of FIG. 1, the contaminant monitoring method of FIG. 7, and the sample cleaning method of FIG. 9 can be combined to form the device under operation. basis.

An example of a VUV optical metrology apparatus that is preferably adapted to benefit from the use of the methods described herein is disclosed in U.S. Patent Application Serial No. 10/668,642, filed on Sep. 23, 2003, which is hereby incorporated by reference. In U.S. Patent Application Serial No. 10/909,126, filed on Jul. 30, 2004, the entire disclosure of which is hereby incorporated by reference. The metrology device can be a broadband reflectometer that is specifically designed to operate over a wide range of wavelengths including VUV.

An example of such a device 1100 is presented in FIG. It is apparent that source 1110, beam conditioning module 1120, optics (not shown), spectrometer 1130, and detector 1140 are contained in an environmentally controlled device (or optics) chamber 1102. Sample 1150, additional optics 1160, motorized/sample chuck 1170 (with optional integrated desorption tower capability), and reference sample 1155 are masked in a separate environmentally controlled sample chamber 1104 for sample uploading and Unloading without contaminating the quality of the equipment chamber environment. The device and sample chamber are connected via a controllable coupling mechanism 1106 that allows for the transfer of photons, and gas exchange occurs if photons are transferred. The equipment chamber 1102 and sample chamber 1104 are coupled to a vacuum and purification subsystem 1175 that includes a suitable vacuum connection 1176, valve, purge connection 1177, and pressure gauge 1178 so that environmental control can be independently performed in each chamber. In this manner, the environmental vacuum and backfilling techniques described above can be accomplished independently for each chamber or all of them. Thus, vacuum and backfill techniques (one or both) can be performed for the test device/optical chamber, sample chamber, and/or two chambers.

Additionally, a processor (not shown) located outside of the controlled environment can be used to coordinate and facilitate the automated monitoring method and analyze the measured data. It will be appreciated that the processor can be any of a variety of computing components that provide for proper data processing and/or storage of collected data.

Although not explicitly shown in Figure 11, it should be noted that the system can be equipped with robots and other associated mechanized components to facilitate uploading and unloading samples in an automated manner to further increase the throughput. Moreover, as is known in the art, the sample chamber can be utilized in conjunction with the sample lock chamber to improve environmental control and increase system throughput for interchangeable samples.

In operation, light from source 1110 is modified by beam conditioning module 1120 and guided through transfer optics through coupling mechanism 1106 and into sample chamber 1104, where light is focused by focusing optics 1160 to Sample 1150. Light reflected from the sample 1150 is collected by the focusing optics 1160 and redirected out via the coupling mechanism 1106, wherein the light is dispersed by the spectrometer 1130 and recorded by the detector 1140. The entire optical path of the device is maintained within a controlled environment that acts to remove the absorbing material and allow transmission of VUV photons.

A more detailed schematic of the optical aspect of the device is presented in FIG. The device is configured to collect the reference broadband reflectance data in the VUV and two additional spectral regions. In operation, light from these three spectral regions can be obtained in parallel or in series. When operating in a serial manner, the reflectance data from the VUV is first obtained and referenced, after which the reflectance data from the second region and then the third region is collected and referenced. Once all three data sets are recorded, they are joined together to form a single broadband spectrum. In parallel operation, reflectance data from all three zones are simultaneously collected, referenced, and recorded prior to data bonding.

The device is divided into two environmentally controlled chambers, an equipment chamber 1102 and a sample chamber 1104. The equipment chamber 1102 covers most of the system optics and is not regularly exposed to the atmosphere. The sample chamber 1104 masks the sample and the sample and the reference optics, and it is periodically open to help change the sample. For example, device chamber 1102 can include mirrors M-1, M-2, M-3, and M-4. The Flip-in mirrors FM-1 and FM-3 can be utilized to selectively select which of the light sources 1201, 1202, and 1203 (each having a different spectral region). One of the spectrometers 1204, 1216, and 1214 can be selectively selected using the swivel mirrors FM-2 and FM-4 (depending on the selected spectral region). As shown, mirrors M-6, M-7, M-8, and M-9 can be used to assist in guiding the beam. The windows W-1 and W-2 couple light between the device chamber 1102 and the sample chamber 1104. Windows W-3, W-4, W-5, and W-6 couple light into device chamber 1102. As shown, the beam splitter BS and the baffles S-1 and S-2 are used to selectively direct light to the sample 1206 or reference member 1207 with the aid of mirrors M-2 and M-4 (in one embodiment The reference piece can be mirrored). The sample beam passes through the compensation plate CP. A compensation plate CP is included to eliminate a phase difference that will occur between the sample path and the reference path, the phase difference being passed through the beam splitter substrate only once by the light traveling in the sample channel, and the light traveling in the reference channel is worn Produced by the fact that the beam splitter substrate is three times (due to the operational properties of the beam splitter). Therefore, the compensating plate can be constructed from the same material and the same thickness as the beam splitter. This ensures that light traveling through the sample channel also passes through the beam splitter substrate material of the same total thickness.

When operating in a serial mode, first switch to the "out" position by transferring the second spectral region into the source mirror FM-1 and the third spectral region into the source mirror FM-2 to allow for the source from the VUV source. Light is collected, collimated, and redirected to the beam splitter element BS to obtain VUV data. The light striking the beam splitter is split into two components using a near-balanced Michelson interferometer configuration: sample beam 1255 and reference beam 1265. The sample beam is reflected from the beam splitter BS and travels to the sample chamber 1104 via the compensator plate CP, the sample baffle S-1 and the sample window W-1, wherein the light is redirected via the focusing mirror M-2 and focused to the sample 1206 . The reference flapper S-2 is closed during this time. The sample window W-1 is constructed from a material that is sufficiently transparent to the VUV wavelength to maintain high optical flux.

Light reflected from the sample is collected, collimated, and redirected back through the sample window by sample mirror M-2, with light passing through the sample baffle and compensator plate. Then the light continues to be blocked from the first spectral region turning into the detector mirror FM-2 and the second spectral region into the detector mirror FM-4 (switching to the "out" position), wherein the light is focused by the mirror M-3 Redirect and focus to the entrance slit of VUV spectrometer 1214. At this point, the light from the sample beam is scattered by the VUV spectrometer and recorded by its associated detector.

After collecting the sample beam, the reference beam is measured. This is done by closing the sample baffle S-1 and opening the reference baffle S-2. This causes the reference beam to travel into the sample chamber 1104 via the beam splitter BS, the reference baffle S-2 and the reference window W-2, wherein the light is redirected by the mirror M-4 and focused to the plane serving as the reference Reference mirror 1207. The reference window is also constructed from a material that allows the VUV wavelength to be sufficiently transmitted to maintain high optical flux.

Light reflected from the surface of the planar reference mirror 1207 returns to travel to the focus reference mirror M-4, where the light is collected, collimated and redirected through the reference window W-2 and the reference baffle S-2 to the beam splitter BS . Light is then reflected by the beam splitter to the focusing mirror M-3, where the light is redirected and focused to the entrance slit of the VUV spectrometer 1214.

The path length of the reference beam 1265 is specifically designed to match the path length of the sample beam 1255 in each of the environmentally controlled chambers. It follows that the quality of the controlled environment of the device can be evaluated by monitoring the intensity from the reference arm as previously described in FIG. As described above with respect to Figure 7, the amount of optical radiation of the various optical components of the exposure system can affect the amount of surface contamination that can be present on the various optical components. Therefore, to assist in further balancing the reference path traveled by the reference beam with the sample path traveled by the sample beam, it may be desirable to balance the exposure of the optical radiation dose for each path. In this way, in addition to the optical path length and the balance of the components, the contamination associated with the optical components can be relatively balanced between each path. Additionally, techniques for determining the pollution status of the optical paths described herein can be performed independently for each path to monitor the status of each of the paths.

After measuring the VUV data set, the second spectral region data set is obtained in a similar manner. During the collection of the second region spectral data, the second spectral region source is transferred to the mirror FM-1 and the second spectral region detector is turned into the mirror FM-2 to switch to the "in" position. As a result, light from VUV source 1201 is blocked and light from second spectral region source 1203 is allowed to pass through window W-3 after being collected, collimated, and redirected by its focusing mirror M-6. Similarly, the second spectral region detector is turned into the mirror FM-2 to switch to the "in" position to direct the light from the sample beam (the sample baffle is open and the reference baffle is closed) and the reference beam (refer to this reference) The baffle is open and the sample baffle is closed) through the associated window W-6 and onto the mirror M-9 that focuses the light onto the entrance slit of the second spectral region spectrometer 1216, wherein the light is passed through its detector Disperse and collect.

The data from the third spectral region is collected in a similar manner, by turning "in" the third spectral region source into the mirror FM-3 and the third spectral region detector into the mirror FM-4, while rotating The "out" second spectral region source is turned into the mirror FM-1 and the second spectral region detector is turned into the mirror FM-2.

Once sample measurements and reference measurements have been performed for each of the spectral regions, a processor (not shown) can be used to calculate the reference reflectance spectra in each of the three regions. Finally, the individual reflectance spectra are combined to produce a single reflectance spectrum comprising the three spectral regions.

When operating in parallel mode, an appropriate beam splitter is used instead of the source and the detector is turned into the mirror so that data from all three spectral regions is simultaneously recorded.

Just as the auxiliary light source can be easily added to the reflectometer of Figure 12, a particular VUV source can be integrated with the shared optical module for system and/or sample cleaning if deemed necessary. For example, in an alternative case, an auxiliary VUV source can be utilized for system and/or sample cleaning. The auxiliary VUV source can be a source having a higher intensity than the primary VUV source. The auxiliary VUV source can also be a single wavelength line source or have other wavelength characteristics than the primary VUV source. Use a higher intensity light source to improve cleaning flux. In one embodiment relating to the use of an auxiliary VUV source, it may be desirable to configure this source to direct light from the source to the mirror M-1 via a turn-in mirror, baffle or the like (not shown). This will allow the mirror M-1 to be cleaned. In this configuration, the auxiliary source will encounter most, if not all, of the optical paths of the primary VUV source utilized for sample measurement. It will be appreciated that the benefits of the techniques disclosed herein may be achieved without encountering all of the elements of the optical path of the primary VUV source, although a greater number of such elements may be encountered. In addition, it will be appreciated that if an auxiliary VUV source is utilized, it will be desirable to couple the source to the system in such a manner that the path of the light is contained within the environmentally controlled optical path.

The systems of Figures 11 and 12 can be utilized as separate tools or can be integrated with another processing tool. In one embodiment, only the mechanism that allows the sample to be transferred between a processing tool and the metrology tool sample chamber can be used to attach only the systems of Figures 11 and 12 to the processing tool. In another alternative, the sample chamber can be constructed in a manner common to processing tools, such that the metrology tool can be more closely integrated with the processing tool. For example, the test device/optical chamber can be in communication with the same chamber via a window, gate valve, or other coupling mechanism that is formed with a processing tool. In this manner, the sample need not leave the environment of the processing tool, and the sample can be contained within the processing tool (such as one of the processing chambers, the transfer zone, or other zones within the processing tool).

It will be apparent that the systems presented in Figures 11 and 12 contain exemplary components to facilitate the environmental monitoring method outlined in Figure 1, the system contamination monitoring method outlined in Figure 7, and the contamination presented in Figure 9. Sample measurement method. In general, all three effects (ie, the absorption of material within the device's capacity, the contamination of the optical surface within the device, and the accumulation of sample contamination) can significantly affect the optical data in the VUV, and therefore all three can be used simultaneously. The method is to achieve optimal system performance by means of equipment operating in this spectral region. The operational flow diagram 1300 presented in Figure 13 provides an example of how this can be done.

The capacity of the device is first evacuated to a predetermined pressure at step 1305. The device is then backfilled to a predetermined measurement pressure 1310 using a non-absorbent gas. Once the measured pressure has been obtained, the system is prepared as indicated at step 1315 to perform the measurement using the system contamination monitoring method outlined in FIG. Accordingly, at step 1315, the various system contaminant steps 710-760 of FIG. 7 can be performed. Once the step 1315 and the system state is considered stable, "clean", then at step 1320 to times t ₁ immediately obtained from a reference sample of the intensity of the spectrum. At this time, it is assumed that the concentration of the absorbing substance is at its lowest achievable level. At this time, the obtained intensity spectrum can be used as part of the environmental monitoring process corresponding to step 130 of FIG. The measurement of the test sample can be performed for a predetermined period of time using the contaminated sample measurement method outlined in steps 910-970 of FIG. 9 at step 1325.

Once the predetermined measurement time has elapsed, the state of the controlled environment is again evaluated at step 1330 by the preparation system to perform measurements using the system contamination monitoring method outlined in FIG. Performing the system contaminant monitoring and cleaning steps again can place a variety of optical surfaces in a state similar to that at step 1315. This allows the environmental monitoring step to be completed under conditions that more accurately match the system conditions of the first intensity measurement recorded at step 1320. Thus, after step 1330, as indicated at step 1335, at time t ₂ again recorded intensity of the spectrum from the reference sample. The ratios of the two intensity spectra (time t ₁ and time t ₂ ) from the reference sample are then calculated and analyzed at steps 1340 and 1345 to determine the concentrations of the absorbing materials N ₁ , N _{2 ,} and the like. At step 1350, the determined concentration is then compared to the threshold to determine if the environment in the device is suitable for supporting further measurements. As indicated by step 1355, if the environment is deemed suitable (i.e., the determined concentration is less than the threshold), then further measurements may be performed once the system is again prepared for measurement via the method of FIG. Conversely, if the environment is found to be insufficient to support further measurements as indicated by step 1360, the environment may be regenerated by reinitiating the process of FIG. 13 by returning control to step 1305. Thus, steps 1335-1360 correspond to the environmental monitoring steps 150-195 of FIG.

As shown in FIG. 13, various concepts of environmental monitoring and regeneration of FIG. 1, system contamination monitoring and cleaning of FIG. 7, and sample cleaning of FIG. 9 can be integrated for use in controlling optical metrology tools. It will be appreciated that the order of the various techniques may be varied and that FIG. 13 only illustrates one way of combining the various concepts. For example, the concepts can be implemented serially rather than being integrated together as shown in the technique of FIG. Moreover, these concepts need not be fully utilized. For example, in an alternate embodiment, only one or both of these concepts may be utilized.

The degree of pollution effects (environment and surface/sample) that will degrade the performance of VUV optical metrology test equipment will generally depend on a wide range of factors including, but not limited to, tool design, method of operation, measurement frequency, sample upload method, and sample. characteristic. Therefore, it is contemplated that the operational procedures outlined in Figure 13 can be modified on a case-by-case basis to ensure that optimal device performance is maintained.

Other modifications and alternative embodiments of the present invention will be readily apparent to those skilled in the art. Therefore, the description is to be construed as illustrative only and illustrative of the embodiments of the invention. It should be understood that the form of the invention shown and described herein is considered to be the presently preferred embodiment. Equivalent elements may be substituted for the elements described and illustrated herein, and certain features of the invention may be utilized independently of the use of other features, which will be readily apparent to those skilled in the <RTIgt; And so on.

500. . . Contaminated optical surface

505. . . Pollution

510. . . VUV radiation

520. . . "clean" optical surface

610. . . Hollow circle

620. . . Solid circle

1100. . . device

1102. . . Equipment chamber

1104. . . Sample chamber

1106. . . Controllable coupling mechanism

1110. . . source

1120. . . Beam adjustment module

1130. . . spectrometer

1140. . . Detector

1150. . . sample

1155. . . Reference sample

1160. . . Optics

1170. . . Electric table / sample chuck

1175. . . Purification subsystem

1176. . . Vacuum connector

1177. . . Purification connector

1178. . . pressure gauge

1201. . . light source

1202. . . light source

1203. . . light source

1204. . . spectrometer

1206. . . sample

1207. . . Reference piece / plane reference mirror

1214. . . spectrometer

1216. . . spectrometer

1255. . . Sample beam

1265. . . Reference beam

BS. . . Beam splitter

CP. . . Compensation board

FM-1, FM-2, FM-3, FM-4. . . Turn into the mirror

M-1, M-2, M-3, M-4, M-6, M-7, M-8, M-9. . . Mirror

S-1, S-2. . . Baffle

W-1, W-2, W-3, W-4, W-5, W-6. . . window

FIG. 1 is an exemplary flow chart of an environmental monitoring.

Figure 2 illustrates an exemplary environmental monitoring result using a wideband VUV data set.

Figure 3 illustrates an exemplary environmental monitoring result using a selected wavelength VUV data set.

Figure 4 illustrates exemplary reflectance data obtained from a clean crucible surface and a contaminated crucible surface.

Figure 5 (a) illustrates an optical surface having a contaminated layer.

Figure 5(b) illustrates VUV exposure in an oxygen-containing atmosphere resulting in cleaning.

Figure 5(c) illustrates a clean optical surface after processing.

Figure 6 illustrates the reversible deposition of one of the contaminants and the irreversible deposition of the second contaminant by VUV radiation in an atmosphere containing contaminants and oxygen.

Figure 7 is a flow chart showing an exemplary system contamination monitoring.

Figure 8 illustrates an exemplary pollution monitoring data using a selected VUV wavelength.

Figure 9 is a flow chart showing an exemplary contamination sample measurement.

Figure 10 illustrates an exemplary cleaning response distribution for three different contaminated membranes.

Figure 11 illustrates an exemplary schematic of a VUV reflectometer.

Figure 12 illustrates a wideband reference reflectometer covering three spectral regions including VUV.

Figure 13 is a flow chart showing an exemplary operation of a VUV optical metrology device.

Claims (33)

A method for controlling an atmosphere in an optical metrology tool, comprising: providing at least a first environmentally controlled chamber and a second environmentally controlled chamber, the first environmentally controlled chamber and The second environmentally controlled chamber is configured to pass a beam of light having a wavelength below the DUV wavelength; reducing the first environmentally controlled chamber and the second environmentally controlled chamber by utilizing a vacuum evacuation technique a concentration of one of the at least one of the optical absorption species, the at least one of the first environmentally controlled chamber and the second environmentally controlled chamber being a controlled atmosphere chamber; (trace levels) gas is added to the controlled atmosphere chamber; the controlled atmosphere chamber is backfilled using a non-absorbent gas to increase the pressure in one of the controlled atmosphere chambers to a vacuum evacuation pressure level The above improves the optical performance; and transmits the light beam having a wavelength lower than the DUV wavelength when the controlled atmosphere chamber is in the backfill state. The method of claim 1, wherein the optical performance is improved by increasing optical transmission. The method of claim 2, wherein the optical transmission is increased by inhibiting the absorbed material from degassing from the surface of the optical tool. The method of claim 1, wherein the optical performance is improved by reducing the migration of contaminants. The method of claim 4, wherein reducing the migration limit of the contaminant is adhered to Degradation of optical surface reflection properties caused by contaminants of such optical surfaces. The method of claim 4, wherein the optical performance is also improved by increasing optical transmission. The method of claim 1, wherein the first chamber is the same chamber and the second chamber is an optics chamber, the decrease in the concentration of an optically absorbing material and the backfilling in the sample chamber get on. The method of claim 1, wherein the optical metrology tool is a separate optical metrology tool. The method of claim 1, wherein the first chamber is the same chamber and the second chamber is an optics chamber, the decrease in the concentration of an optically absorbing material and the backfilling in the optics chamber In progress. The method of claim 9, wherein the sample chamber is integrated into a processing tool. The method of claim 1, wherein the first chamber is the same chamber and the second chamber is an optics chamber, the decrease in the concentration of an optically absorbing material and the backfilling in the optical chamber The chamber is carried out in the sample chamber. The method of claim 1, wherein the optically absorbing material is moisture or oxygen. The method of claim 1, wherein the vacuum evacuation pressure level is less than 1 x 10 ^-5 Torr. The method of claim 1, further comprising applying energy to the optically absorptive material during the reducing step. The method of claim 14, wherein the method is by mechanical, thermal or radiation Apply this energy. The method of claim 1, wherein the backfilling increases the pressure in the controlled atmosphere chamber to a range of 300-700 Torr. The method of claim 1, further comprising continuing to provide a purge gas to the controlled atmosphere chamber after the backfilling step. The method of claim 1, wherein the trace level of gas promotes a surface cleaning process. A method for controlling an atmosphere in an optical metrology tool, comprising: providing at least one environmentally controlled sample chamber and an environmentally controlled optics chamber, the sample chamber and the optics chamber Passing a light beam having a wavelength lower than the DUV wavelength; the reduction occurs by reducing the concentration of moisture or oxygen in at least one of the sample chamber and the optical member chamber from an ambience state by a vacuum evacuation technique The at least one of the sample chamber and the optics chamber is a controlled atmosphere chamber; the controlled atmosphere chamber is backfilled with a VUV non-absorbent gas to enable one of the controlled atmosphere chambers The pressure is increased above a vacuum evacuation pressure level to reduce contaminant migration; a purge gas is continuously supplied to the controlled atmosphere chamber after the backfilling step; and the transmission has a wavelength when the controlled atmosphere chamber is in the backfill state The beam is below the DUV wavelength. The method of claim 19, wherein the reducing the migration limit of the contaminant is adhered to Degradation of optical surface reflection properties caused by contaminants of such optical surfaces. The method of claim 19, wherein the decrease in the concentration of moisture or oxygen and the backfilling are performed in the sample chamber. The method of claim 19, wherein the optical metrology tool is a separate optical metrology tool. The method of claim 19, wherein the decrease in the concentration of moisture or oxygen and the backfilling are performed in the optics chamber. The method of claim 19, wherein the sample chamber is integrated into a processing tool. The method of claim 19, wherein the decrease in the concentration of moisture or oxygen and the backfilling are performed in the optics chamber and the sample chamber. The method of claim 19, further comprising adding a trace level of gas to the controlled atmosphere chamber. The method of claim 26, wherein the trace level of gas promotes a surface cleaning process. A method for controlling an atmosphere in an optical metrology tool, comprising: providing at least one environmentally controlled sample chamber and an environmentally controlled optics chamber, the sample chamber and the optics chamber Transmitting a beam having a wavelength lower than a DUV wavelength; providing an optical path of the present beam and a reference optical path, the optical path of the sample beam and the optical path length of the optical path of the reference beam are matched; Reducing the concentration of moisture or oxygen in at least one of the sample chamber and the optics chamber from an ambience state by vacuum evacuation techniques occurs in the sample chamber and the optics chamber The at least one is a controlled atmosphere chamber; a trace level of gas is added to the controlled atmosphere chamber; the controlled atmosphere chamber is backfilled with a VUV non-absorbed gas to enable the controlled atmosphere chamber One of the pressures is increased above a vacuum evacuation pressure level to improve optical performance; and the light beam having a wavelength below the DUV wavelength is transmitted while the controlled atmosphere chamber is in the backfill state. The method of claim 28, wherein the optical performance is improved by increasing optical transmission. The method of claim 29, wherein the optical transmission is increased by inhibiting the absorbed material from degassing from the surface of the optical tool. The method of claim 28, wherein the optical performance is improved by reducing the migration of contaminants. The method of claim 31, wherein reducing contaminant migration limits degradation of optical surface reflection properties caused by contaminants adhered to the optical surfaces. The method of claim 31, wherein the optical performance is also improved by increasing optical transmission.