JP2009501902A - Systems, circuits, and methods for extending the detection range by reducing thermal damage in inspection systems by avoiding detector and circuit saturation - Google Patents

Abstract

  Inspection systems, circuits, and methods are provided to enhance defect detection by addressing anode saturation as a limiting factor in the measurement detection range of a photomultiplier tube (PMT) detector. Also provided are inspection systems, circuits and methods for enhancing defect detection by addressing the saturation levels of amplifiers and analog and digital circuits as a limiting factor in the measurement detection range of inspection systems. In addition, an inspection system to enhance defect detection by reducing thermal damage to large particles by dynamically changing the incident laser beam power level delivered to the sample during the surface inspection scan A circuit and method are provided.

Description

  The present invention generally relates to circuits, systems, and methods for extending the measurement detection range of an inspection system used to inspect a sample. More specifically, the present invention relates to circuits, systems, and methods for extending the measurement detection range of an inspection system by addressing anode saturation of a photomultiplier tube (PMT) as a limiting factor of the measurement detection range.

  Manufacturing semiconductor devices such as logic elements, storage devices, and other integrated circuit devices generally uses several semiconductor manufacturing processes to form various features and multiple levels of semiconductor devices, Processing a sample such as a semiconductor wafer. For example, lithography is a semiconductor manufacturing process that typically requires that a pattern be transferred to a resist disposed on a semiconductor wafer. Additional examples of semiconductor manufacturing processes include, but are not limited to, chemical mechanical polishing, etch, deposition, ion implantation, and the like. A large number of semiconductor devices are manufactured as an arrangement on a semiconductor wafer and then separated into individual semiconductor devices.

  The inspection process is used in various steps during the semiconductor manufacturing process with the aim of detecting defects on the wafer in order to promote high yields in the manufacturing process and thus high profits. Inspection is always an important part of semiconductor manufacturing. However, as the dimensions of semiconductor devices become smaller, inspection becomes more important for the production of acceptable semiconductor devices. For example, even relatively small defects can cause undesired deviations in semiconductor devices and in some cases can cause the device to fail, making it increasingly necessary to detect defects of decreasing size It has become.

  Many different types of inspection tools have been developed for inspection of semiconductor wafers, including optical systems and E-beam systems. Optical inspection tools generally feature a dark field or bright field inspection system. Dark field systems are generally known to have a relatively high detection range. For example, a dark field system detects the amount of light that is incident on a sample from a standard or tilt angle and is scattered from the surface of the sample. The amount of scattered light detected by the system generally depends on the optical characteristics of the spot under inspection (eg, the refractive index of the spot) and any spatial deviations within that spot (eg, non-uniform surface topologies) . For dark field inspection, a smooth surface provides little collected signal, while a surface with protruding features (such as patterned features or defects) scatters more strongly (sometimes up to 6 times more). Tend. A bright field inspection system directs light to a sample at a specific angle and measures the amount of light reflected from the surface of the sample at a similar angle. In contrast to the dark field system, the deviation in the reflected signal collected by the bright field system is typically only about twice.

  In addition, most inspection tools are designed to inspect unpatterned semiconductor wafers or patterned semiconductor wafers, but are not designed to inspect both. Since tools are optimized to inspect a particular type of wafer, it is generally not possible to inspect different types of wafers for several reasons. For example, many inspection tools for unpatterned wafers produce a single output signal in which all of the light collected by the lens (or another collector) represents all of the light collected by the lens. Configured to be directed to the detector. Thus, light scattered from a pattern or feature on the patterned wafer will be combined with other diffuse light (eg, from a defect). In some cases, a single detector may saturate, and as a result may not produce a signal that is analyzed for defect detection. Even if a single detector does not saturate, scattered light from a pattern or other feature on the wafer cannot be separated from another scattered light, thereby providing defect detection based on the other scattered light. Even if not suppressed, this will be hindered.

  Tools used to inspect patterned wafers typically use at least two detectors to improve spatial resolution. However, especially when imaging with a dark field system, detectors used in patterned wafer inspection tools can also saturate. As noted above, the dark field scatter signal obtained from a patterned wafer is 6 times (or more) due to deviations in the surface topology from a smooth surface area (which appears dark) to a very textured area (which appears bright). And more) may be different. Especially for detection systems that operate at high data rates, it is often difficult to collect large signals from both very dark and very bright areas of the substrate being inspected without “on-the-fly” adjustment. .

  Most optical inspection tools are limited in detection range, detection sensitivity, or both. For example, inspection tools that use high gain detectors to obtain a higher detection range cannot detect smaller (ie, low light) signals. On the other hand, inspection tools that use low gain detectors achieve greater sensitivity at the cost of reduced detection range. That is, the low gain detector may be able to detect a smaller signal, but may be saturated if a larger signal is received. Other factors tend to limit the detection range in addition to the detector gain. For example, further limitations may be imposed by amplifier circuits or high speed analog to digital converters used to convert the scattered output signal into a form suitable for signal processing.

  One possible solution to this problem is to add non-linear amplification to the detector output signal to emphasize the low amplification signal range. This type of approach is described by Wolf in US Pat. No. 6,002,122, the disclosure of which is incorporated herein by reference. In the method described by Wolf, the output signal from the photomultiplier tube (PMT) is processed by a logarithmic amplifier or gain correction mechanism. Wolf has a low amplification signal range by changing the PMT gain "on the fly" (by changing the bias potential supplied to the dynode) to avoid anode saturation, which is a common detection range limitation of PMT detectors. To emphasize. This approach allows for improved visibility of small signal detection in dark field images, but does not extend the entire detection range of the inspection system. In addition, the “on-the-fly” gain modulation disclosed by Wolf allows PMTs to operate in a very non-linear manner, and thus complex (and expensive) drive electronics and advanced to compensate for non-linear and transient effects. Requires calibration.

  Another approach to extend the detection range of the inspection system is to utilize two or more detectors with separate detection channels. This type of approach is described by Almology et al. In US Patent Application 2003/0058433, the disclosure of which is incorporated herein by reference. Almology describes a defect detection system that utilizes at least two detectors. One of the detectors is optimized for high sensitivity, while the other detector is usually designed to have a high saturation level at the expense of sensitivity. Light scattered from the sample is split between the detectors by the addition of various optical components. Almogy can extend the detection range, but Almogy has additional optics and electronics, all of which consume additional space, increase complexity, and incur higher costs This is done by requesting multiple detectors.

  Accordingly, there remains a need for improved circuits and methods to extend the detection range of wafer inspection systems. Such improved circuits and methods are substantially free of the complexity and cost of real-time gain adjustment, as required by Wolf, or the additional detectors, optics, and electronics required by Almology. It is preferable to provide an extended measurement range. In addition, the improved inspection system will extend the detection range without sacrificing throughput or sensitivity. In some cases, the improved inspection system is used for patterned and unpatterned wafers.

  The following description of various embodiments of systems, circuits, and methods is in no way to be construed as limiting the subject matter of the appended claims.

  The inspection systems, circuits, and methods described herein enhance defect detection by addressing anode saturation as a limiting factor in the measurement detection range of a photomultiplier tube (PMT) detector. According to one embodiment of the present invention, a method for inspecting a sample includes directing light to the sample and detecting light scattered from the sample. In general, the steps of detecting include monitoring the anode current of the PMT detector and using the anode current to detect sample features, defects, or light scattering properties until the anode current reaches a predetermined threshold. including. The method may then use the dynode current of the PMT to detect sample features, defects, or light scattering properties.

  In some cases, the method may include selecting the dynode current used for detection before the light is directed to the sample. For example, a particular dynode current may be selected by providing an electrical connection to one of a plurality of dynodes included in the PMT. In one implementation, an intermediate dynode is selected from a plurality of dynodes to provide an intermediate amount of gain. In other embodiments of the invention, other dynodes may be selected, for example, depending on the range of light intensity scattered from the sample.

  In some cases, the method may include selecting an alternative dynode current that will be used to detect sample features, defects, or light scattering properties. For example, an alternative selection is made by providing another electrical connection to a different one of the dynodes during (or after) the detection step described above. The alternative dynode current may be selected to provide a smaller or greater amount of gain than the originally selected dynode current.

  According to another embodiment of the present invention, a circuit for detecting light scattered from a sample is described herein. In a typical embodiment, the circuit includes a photomultiplier tube (PMT) for receiving light scattered from the sample and converting it into an electrical signal. As is known in the art, a PMT includes a cathode and multiple stages of multiple dynodes and anodes. Thus, the electrical signal is amplified at each successive stage to produce an output signal at the anode stage.

  However, unlike conventional PMT detectors, the circuit described herein is capable of receiving an output signal along with an intermediate signal that represents an electrical signal that is amplified in one of the dynode stages of the PMT. Means may be included. Such means generally includes i) providing an output signal to a test system component as long as the output signal remains below a predetermined threshold, and ii) providing an intermediate signal to the test system configuration when the output signal reaches a predetermined threshold. Configured to feed the element. In most cases, the predetermined threshold is approximately the same as the current level at which the anode stage begins to saturate. Thus, the means claimed herein can be switched to provide an intermediate signal when the anode stage reaches saturation.

  According to another embodiment of the present invention, an inspection system for inspecting a sample is described herein. In general, an inspection system includes an illumination subsystem for directing light to a sample and a detection subsystem for detecting light scattered from the sample. The detection subsystem is configured to generate a first signal and a second signal in response to the detected light. However, unlike prior art inspection systems, the detection subsystem described herein can use only one detector to generate the first and second signals. Within the system claimed herein, the first signal is used until the first signal reaches a predetermined threshold level, and then the first signal is used to detect sample features, defects, or light scattering properties. A processor for the detection using two signals is included. For example, the processor is configured to i) compare the first signal to a predetermined threshold level and ii) switch to the second signal when the first signal reaches the predetermined threshold level.

  In a more specific embodiment, the detection subsystem includes a collector for collecting light scattered from the sample in the direction of the collector, and a detector for converting the collected light into first and second signals ( Or a photodetector). In most cases, the second signal is a small part of the first signal. For example, the photodetector includes a photomultiplier tube (PMT), a current / voltage converter, and a load. As is known in the art, a PMT includes a cathode, a plurality of dynodes, and an anode, and a current-to-voltage converter converts the anode current to a voltage for the purpose of producing a first signal. Combined for. However, unlike conventional methods, the load claimed herein is coupled to one of the dynodes to convert the intermediate dynode current to another voltage for the purpose of producing a second signal. Is done. The intermediate dynode current is much smaller than the anode current, so it is used to detect a much larger amount of light than that detected at the anode without causing dynode saturation.

  Further advantages of the present invention will become apparent to those skilled in the art from the following detailed description of the preferred embodiment and by reference to the accompanying drawings.

  While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will be described in detail herein. Drawings are not to scale. However, the drawings and detailed description thereof are not intended to limit the invention to the particular forms disclosed, but, on the contrary, fall within the spirit and scope of the invention as defined by the appended claims. It is intended to cover all modifications, equivalents, and alternatives.

  The methods and systems described herein include various limiting factors for measurement detection range including, but not limited to, detector saturation, amplifier saturation, and analog-to-digital converter (ADC) fixed bit range. Enhance defect detection by addressing Unlike some currently used inspection methods, the inspection system described herein maintains the signal linearity and stability, all of which consume the space, complexity, and cost of the inspection system. It is possible to extend the measurement detection range without the use of additional detectors, optics, or electronic components that undesirably increase.

  Various embodiments are described herein with respect to an optical inspection system or tool used to inspect a sample. The term “sample” refers to a wafer or reticle, or any other detected for defects, features, or other information known in the art (eg, amount of haze or film characteristics). Used to refer to the sample.

  As used herein, the term “wafer” generally refers to a substrate in the form of a semiconductor or non-semiconductor material. Examples of such semiconductor or non-semiconductor materials include, but are not limited to, single crystal silicon, gallium arsenide, and indium phosphide. Such substrates are typically found and / or processed at semiconductor manufacturing facilities.

  In some cases, the wafer may include only a substrate such as a virgin wafer. Alternatively, the wafer includes one or more layers formed on the substrate. Examples of such layers include, but are not limited to resists, dielectrics, conductor materials. The resist includes a resist patterned by optical lithography technology, electron beam lithography technology, or X-ray lithography technology. Examples of dielectrics include, but are not limited to, silicon dioxide, silicon nitride, silicon oxynitride, and nitrite. Additional examples of dielectrics can be found in Applied Materials, Inc. Black Diamond ™ commercially available from Santa Clara, California, and Novellus Systems, Inc. "Low dielectric constant" dielectrics such as CORAL ™, commercially available from San Jose, California, "Ultra low dielectric constant" dielectrics such as "Xerogel", "High dielectric constants" such as tantalum pentoxide Of dielectrics. In addition, examples of conductive materials include, but are not limited to, aluminum, polysilicon, and copper.

  One or more layers formed on the wafer may be “patterned” or “unpatterned”. For example, a wafer includes a plurality of dies having repeatable pattern features. The formation and processing of such a layer of material ultimately provides a finished semiconductor device. Thus, a wafer includes a substrate on which all layers of a complete semiconductor device are not formed, or a substrate on which all layers of a complete semiconductor device are formed. The term “semiconductor device” is used herein interchangeably with the term “integrated circuit”. In addition, other devices such as microelectromechanical (MEMS) devices are also formed on the wafer.

  A “reticle” is a reticle at any stage of the reticle manufacturing process, or a completed reticle that has been or has not been published for use in a semiconductor manufacturing facility. A reticle, or “mask”, is generally defined as a substantially transparent substrate having substantially opaque regions formed thereon and configured in a pattern. The substrate will include a glass material such as quartz, for example. The reticle is placed on the resist-coated wafer during the exposure step of the lithography process so that the reticle pattern is transferred to the resist. For example, a substantially opaque area of the reticle can protect the underlying area of the resist from exposure to an energy source.

  Referring now to the drawings, it should be noted that FIGS. 1, 2, 4, 5, 7-10 are not drawn to scale. In particular, the size of some elements in the figures is greatly exaggerated to emphasize the characteristics of the elements. Note also that FIGS. 1, 2, 4, 5, 7-10 are not drawn to scale. Similar elements shown in more than one figure are indicated using the same reference numerals.

  FIG. 1 illustrates a system used to perform the inspection methods described herein. The system shown in FIG. 1 illustrates a typical optical configuration used to inspect a sample in accordance with the methods described herein. The inspection system includes a dark field optical subsystem. It will be apparent to those skilled in the art that the illustrated system can be modified in various ways, while still providing the ability to perform the methods described herein. In addition, it will be appreciated by those skilled in the art that the illustrated system may include various additional components not shown in FIG. 1, such as stages, sample handlers, folding mirrors, polarizers, additional light sources, additional collectors, etc. Will be clear. All such modifications are within the scope of the invention described herein.

  The system illustrated in FIG. 1 includes an illumination subsystem. The illumination subsystem is configured to direct light to the sample. For example, the illumination subsystem includes a light source 10. The light source 10 includes, for example, a laser, a diode laser, a helium neon laser, an argon laser, a solid state laser, a diode pumped solid state (DPSS) laser, a xenon arc lamp, a discharge lamp, or an incandescent bulb. The light source is configured to emit near monochromatic light or broadband light. In general, the illumination subsystem samples light having a relatively narrow wavelength range (eg, near monochromatic light, ie, light having a wavelength range of less than about 20 nm, less than about 10 nm, less than about 5 nm, or even less than about 2 nm). It is configured to be oriented. Thus, if the light source is a broadband light source, the illumination subsystem includes one or more spectral filters that can limit the wavelength of light directed to the sample. The one or more spectral filters are bandpass filters and / or edge filters and / or notch filters.

  The illumination subsystem also includes various beam shaping and polarization control optics 12. For example, the illumination subsystem includes various optical systems for directing and providing an incident beam to a sample 14 having a particular spot size, for example. When the light source is configured to emit light of various polarizations, the illumination subsystem includes one or more polarization components that can change the polarization characteristics of the light emitted by the light source. In some cases, the light directed to the sample 14 may be coherent or non-interfering. The beam shaping / polarization control optics 12 includes several components not shown in FIG. 1, such as a beam expander, folding mirror, focus lens, cylindrical lens, beam splitter, and the like.

  In some cases, the illumination subsystem includes a deflector (not shown). In one embodiment, the deflector is an acousto-optic deflector (AOD). In other embodiments, the deflector includes a mechanical scanning assembly, electronic scanner, rotating mirror, polygon-based scanner, resonant scanner, piezoelectric scanner, galvo mirror, or galvanometer. The deflector scans the light beam across the sample. In some embodiments, the deflector can scan the light beam across the sample at an approximately constant scan rate.

  As shown in FIG. 1, the illumination subsystem may be configured to direct a beam of light to the sample at a standard angle of incidence. In this embodiment, the illumination subsystem may not include a deflector because the standard incident beam of light is scanned across the sample by relative movement of the optical system with respect to the sample and / or relative movement of the sample with respect to the optical system. Alternatively, the illumination subsystem is configured to direct the beam of light to the sample at an oblique incident angle. The system may be configured to direct multiple beams of light to the sample, such as a tilted incident beam of light or a standard incident beam of light. Multiple beams of light are directed to the sample substantially simultaneously or sequentially.

  The inspection system of FIG. 1 includes a single collection channel. For example, light scattered from the sample is collected by a collector 16, which can be a lens, a compound lens, or any suitable lens known in the art. Alternatively, the collector 16 is a reflective optical component or a partially reflective optical component such as a mirror. In addition, although one particular collection angle is illustrated in FIG. 1, it should be understood that the collection channels may be arranged at any suitable collection angle. The collection angle may vary depending on, for example, the incident angle and / or characteristics of the sample geometry.

  The inspection system also includes a detector 18 for detecting light scattered from the sample and collected by the collector 16. The detector 18 generally functions to convert scattered light into an electrical signal and thus includes virtually any photodetector known in the art. However, specific for use within one or more embodiments of the present invention based on the desired performance characteristics of the detector, the type of sample to be examined, and / or the configuration of the illumination subsystem Detectors are selected. For example, if the amount of light available for inspection is relatively small, a detector that enhances efficiency, such as a time delay integration (TDI) camera, can increase the signal-to-noise ratio of the system and increase throughput. . However, depending on the amount of light available for inspection and the type of inspection being performed, other devices such as electrified coupling device (CCD) cameras, photodiodes, phototubes, photomultiplier tubes (PMT), etc. A detector may be used. In at least one embodiment of the present invention, a photomultiplier tube is used to detect light scattered from the sample.

  The inspection system also includes various electronic components necessary to process the scatter signal detected by the detector 18. For example, the system shown in FIG. 1 includes an amplifier circuit 20, an analog-to-digital converter (ADC) 22, and a processor 24. The amplifier 20 is generally configured to receive output signals from the detector 18 and amplify the output signals by a predetermined amount. The ADC 22 converts the amplified signal into a digital format suitable for use within the processor 24. In one embodiment, the processor may be coupled directly to the ADC 22 by a transmissive medium, as shown in FIG. Alternatively, the processor may receive signals from other electronic components coupled to the ADC 22. In this way, the processor is indirectly coupled to the ADC 22 by the transmission medium and the intervening electronic components.

  Generally, the processor 24 is configured to detect sample features, defects, or light scattering properties using electrical signals obtained from the signal collection channel. The signal produced by a single collection channel represents the light detected by a single detector (detector 18). The term “single detector” is used herein to refer to only one detection range, or in some cases, several sensing ranges (eg, as found in a detector array or multiple anode PMT). Is used to describe a detector having Regardless of the number, the sensing range of a single detector is implemented within a single housing. In some cases, the inspection system described herein is used to inspect patterned and non-patterned samples. The processor includes any suitable processor known in the art. In addition, the processor may be configured to use any suitable defect detection algorithm or method known in the art. For example, the processor may use a die-database comparison or threshold algorithm to detect defects on the sample.

  The inspection systems described herein provide more feature, defect, or light scattering property information for a sample than other inspection systems, but trade off detection range for sensitivity (and vice versa). The same). That is, the inspection system described herein provides an extended detection range (eg, about 0 to about 3 times or more) without sacrificing sensitivity. The improved inspection system also maintains excellent signal linearity and stability and does not require complex calibrations or additional detectors or optics to extend the detection range. The improved inspection system accomplishes all this by addressing several factors that tend to limit the detection range of the inspection system. These factors include, but are not limited to, detector saturation, amplifier saturation, and a fixed bit range of the analog to digital converter. The limitations imposed by detector saturation will now be explained in the context of photomultiplier tubes. However, it should be recognized that the general concepts outlined below may be applicable to other types of detectors.

  Photomultiplier tubes (PMTs) are often used as detectors when the optical signal is hazy (ie, in low intensity applications such as fluorescence spectroscopy). A typical photomultiplier tube consists of a photoelectron emission cathode (photocathode) in a vacuum tube, followed by a focusing electrode, a plurality of dynodes (forming an electron multiplier), and an anode (forming an electron collector) Become. When light enters the PMT, the photocathode emits photoelectrons into the vacuum tube. The focusing electrode directs photoelectrons to the electron multiplier, where the electrons are multiplied by a secondary radiation process. For example, photoelectrons are accelerated from the photocathode toward the first dynode by an electric field. When the photoelectrons collide with the dynode, the photoelectrons liberate additional electrons and amplify the photoelectron signal. These secondary electrons move to the next dynode where they are amplified again. At the end of the dynode chain, electrons are collected by the anode and produce an electrical output signal proportional to the amount of light incident on the PMT. The output signal produced at the anode is generally large enough to be measured using conventional electronics, such as a transimpedance amplifier followed by an analog to digital converter.

The secondary emission process allows the photomultiplier tube to achieve high current amplification. That is, a very small photocurrent from the photocathode is observed as a large output current from the anode of the photomultiplier tube. Current amplification (also called gain) is simply the ratio of the anode output current to the photocurrent from the photocathode. The gain at each dynode is a function of the energy of the incident electrons proportional to the potential between that dynode and the previous stage. The total gain of the PMT is the product of the gains from all dynode stages. When a voltage (V) is applied between the cathode and anode of a photomultiplier tube with (n) dynode stages, the total gain is:
G (V) αV ^an Formula 1
Where α is a factor (usually in the range of 0.6 to 0.8) determined by the dynode material and geometric structure.

  In most cases, the photomultiplier will be operated with a single predetermined gain. For example, a bias voltage is generated for each of the dynodes by connecting a series of voltage divider resistors between the cathode, all dynodes, the anode, and ground. The resistor R is used as a scaling constant and is usually the same for all stages of the photomultiplier tube. A large negative voltage (usually -500V to -1500V) is then applied to the cathode, and the potential is divided evenly across the dynode by the voltage divider resistor chain. By doing this, each of the dynodes can be continuously maintained at a smaller negative potential, the difference between which defines the intermediate dynode gain. The total gain of the photomultiplier tube is altered by changing the voltage applied to the cathode, but this is generally not desired. For example, requiring a large voltage makes it difficult to quickly change the gain due to the parasitic capacitance and large resistance required to limit the power dissipation in the bias string. Thus, most users determine the tube gain in advance, set the appropriate cathode voltage, and then operate the tube at that voltage throughout the measurement operation.

  In this configuration, the detection range of the photomultiplier tube is limited at the low end by the noise and gain characteristics of the transimpedance amplifier and at the high end by the ability of the photomultiplier tube to deliver the anode current. In low intensity applications, the anode current is limited by the space charge effect in the tube, the power consumption of the bias string, and the consumption properties of the dynode coating. In high intensity applications, photomultiplier tubes are limited by anode saturation and sometimes cathode saturation. For example, a photomultiplier tube may give inaccurate results when a relatively large amount of light saturates the anode (or cathode). In the following embodiments, the present invention addresses anode saturation as a limiting factor for the detection range of an inspection system. As described in more detail below, the present invention, in one aspect, avoids measurement inaccuracies by providing a circuit and method designed to avoid anodic saturation, thereby providing a PMT detector. Extend the detection range.

  FIG. 2A illustrates one embodiment of a circuit 30 used to detect light scattered from the sample. Thus, the circuit 30 can be incorporated as a detector 18 in the inspection system of FIG. In the embodiment of FIG. 2A, circuit 30 includes a photomultiplier tube (PMT) 32 having a cathode 34, a plurality of dynodes 36, and an anode 38. Although shown with only eight dynodes, the PMT 32 may include virtually any number of dynodes, with a typical number ranging from about eight to about twenty. The PMT 32 is also shown in FIG. 2A as a head-on photomultiplier tube, particularly a linear-focused head-on PMT operated in transmission mode. However, those skilled in the art will clearly understand how the aspects of the invention described herein apply to other modes and / or types of PMTs. For example, the PMT 32 may alternatively be formed in a side-on configuration and / or operated in a reflective mode. In addition to the linear focusing PMT shown in FIG. 2A, embodiments of the present invention include a circular-cage type, a box-and-grid type, a Venetian brand type. It can also be applied to other types of PMTs, including but not limited to mesh types.

As with conventional PMT circuits, circuit 30 includes a voltage divider chain 40 having impedance elements (eg, Z1-Z9) coupled through a cathode, a majority of dynodes, and an anode. The impedance element may include an equivalent resistance, a tapered resistance, or a combination of resistors and capacitors as is known in the art. Thus, when a high negative voltage (V _S1 ) is applied to the cathode, the potential may be divided evenly or slightly unevenly across all dynodes, resulting in more dynodes in the middle of the chain Get less gain. When light (hv) is incident on the PMT, the cathode 34 emits photoelectrons that travel through the dynode chain 36 to produce an amplified photoelectron current (I _A ) at the anode 38. The current output from the anode 38 is converted into a voltage (V _A ) by a current / voltage converter 42. In most cases, the converter 42 has a feedback impedance (Z _F ) and a load impedance (Z _L ) such that the voltage (V _A ) generated at the output of the PMT is related to the anode current by the following equation: It may be an operational amplifier.
V _A = I _A ^* (Z _F / Z _L ) Formula 2

Unlike conventional PMT circuits, circuit 30 includes additional circuit elements for measuring intermediate dynode voltage (V _D ) in addition to anode voltage (V _A ). In one example, the intermediate dynode voltage is obtained by supplying the photocurrent (I _D ) generated in one of the dynode stages 36 to an additional impedance element (Z _D ) as follows: It is possible.
V _D = I _D ^* (Z _D ) Equation 3
In most cases, the intermediate dynode voltage is lower than the anode voltage because the dynode voltage represents the photocurrent before the current is fully amplified at the anode. As described in more detail below, the intermediate dynode voltage is used in place of the anode voltage to detect defects on the sample. For example, the intermediate dynode voltage is used to detect defects when photocurrent at the anode saturates the anode. Switching logic 44 is included in circuit 30 to monitor the anode voltage and switch to an intermediate dynode voltage when the anode current reaches saturation.

  The level at which anode 38 reaches saturation depends on several factors including, but not limited to, material configuration and anode geometry, the high voltage between the last dynode and anode, and the structure of the voltage divider chain. To do. As is known in the art, an element becomes “saturated” if further changes in the input no longer result in an apparent change in the power signal generated by that element. In some cases, anode 38 provides a high linear anode current up to about 10 mA. The anode can then saturate if the anode current is increased to an amount where the amount of light incident on the PMT exceeds approximately 10 mA. However, it should be noted that in other embodiments of the present invention, a lower saturation level or a higher saturation level is appropriate. For example, in some cases, a larger saturation level may be appropriate if the anode current is still substantially linear up to about 100 mA.

To avoid anodic saturation, switching logic 44 can switch to an intermediate dynode voltage (V _D ) when the anodic current (I _A ) approaches, reaches or exceeds the saturation level. For example, the switching logic 44 may use an anode saturation level or a slightly higher or slightly lower value as the predetermined threshold level. During circuit operation, switching logic 44 provides anode current (I _A ) while supplying anode current (V _A ) as a detector power signal (V _OUT ) to subsequent processing components (eg, ADC 22 or processor 24). ). When the anode current reaches a predetermined threshold level, the switching logic 44 uses the intermediate dynode voltage (V _D ) as the detector output signal (V _OUT ). Regardless of whether anode current or dynode current is used, the detector output signal transferred by switching logic 44 is amplified by ADC 22 by a substantially consistent amount.

The switching logic 44 may be implemented as hardware, software, or a combination of both (ie, firmware). Accordingly, the switching logic 44 may be located in or adjacent to the circuit 30 as a combination of logic elements that perform the functionality implemented in the switching logic 44. On the other hand, functionality may be implemented using program instructions that are stored and executed “on-chip” or “off-chip”. For example, the switching logic 44 may be implemented as a digital signal processing (DSP) chip coupled to the circuit 30 or as program instructions executed by the processor 24. Other configurations / embodiments of the switching logic 44 are possible and within the scope of the present invention. For example, the switching logic 44 may preferably monitor the anode voltage (V _A ) instead of the anode current (I _A ) and is therefore not coupled to receive the anode current, as shown in FIG. 2A. there is a possibility.

As described above, the intermediate dynode voltage is obtained from one of the plurality of dynodes 36. The particular dynode chosen may generally depend on the desired gain difference between the anode voltage and the dynode voltage. For example, each dynode is supplied by a specific bias voltage generated by the voltage divider chain 40 and a negative high voltage power supply (V _S1 ) coupled to the voltage divider chain 40. The high voltage power supply may typically be selected to provide a certain amount of overall (or “tube”) gain. In most cases, since the series resistance experienced along the voltage divider chain increases, each dynode 36 is supplied with a continuous smaller negative potential. For example, when a negative power supply of approximately −1000V is supplied, impedance elements (eg, Z1 to Z9) in the voltage divider chain 40 are supplied with −900V to the first dynode and −800V to the second dynode. , And so forth, and is configured to continuously supply a smaller negative potential to the dynode chain. The potential difference between the voltage supplied to the selected dynode and the voltage supplied to the anode determines the gain difference between them.

The desired gain difference, and thus the desired dynode, may be selected based on the deviation in light intensity incident on the PMT. For high intensity applications, a dynode that is close to the beginning of the chain is selected to create a larger gain difference between the dynode voltage and the anode voltage. Such a large gain difference allows a substantially larger amount of light to be detected after the anode is saturated. On the other hand, dynodes further down the chain may be selected to produce smaller gain differences that may be sufficient for smaller light intensity ranges to be measured. In some embodiments, a dynode near the middle of the dynode chain may be selected to produce an intermediate gain signal. Thus, the circuits and methods described herein can provide a user with a range of possible gain differences. This range is approximately equivalent to about G ^{((nm) / n)} , where G is the total PMT gain (cathode to anode) and m is the selected intermediate dynode (counted from the cathode). Where n is the total number of dynodes.

In the embodiment of FIG. 2A, the selected dynode is disconnected from the voltage divider chain and the intermediate dynode voltage is measured across an additional impedance element (Z _D ). The additional impedance element can be implemented by virtually any passive or active element, but a simple resistance is preferred. In general, the size of the additional impedance element should be selected such that the signal measured across it is substantially equal to the signal level at the anode. As shown in FIG. 2A, an additional power supply (V _S2 ) may be required to measure the intermediate dynode voltage. In some cases, the additional power supply level may be set to provide a potential that will be supplied to the selected dynode if the dynode is not disconnected from the voltage divider chain.

FIG. 2B illustrates an alternative embodiment of the circuit 30. Similar reference numerals are used to indicate similar features between the embodiments shown in FIGS. 2A and 2B, and therefore a detailed description of such features will not be repeated for the sake of brevity. Similar to FIG. 2A, the embodiment shown in FIG. 2B includes a PMT detector 32 having a cathode 34, a plurality of dynodes 36, and an anode 36. In some embodiments, the photocurrent (I _A ) generated at the anode is provided to a current to voltage converter 42 and switching logic 44. When the anode current (I _A ) reaches a predetermined threshold level (related to anode saturation), the switching logic can be switched to provide an intermediate dynode voltage (V _D ) to subsequent processing components. In other embodiments, anode voltage (V _A ) is monitored instead of anode current (I _A ). In such an embodiment, the switching logic 44 may not be coupled to receive the anode current, as shown in FIG. 2B.

The embodiment shown in FIGS. 2A and 2B differs in the way the intermediate dynode voltage is generated. Instead of a single voltage divider chain 40, the embodiment of FIG. 2B divides the chain into a first portion 40a and a second portion 40b. The first power supply voltage (V _S1 ) is supplied to the first portion 40a, while the second power supply voltage (V _S2 ) is supplied to the second portion 40b. In this configuration, the intermediate dynode voltage is measured across the last impedance element (Z5) in the first portion 40a. The advantage of this configuration is that the saturation of the anode and the portion of the voltage divider chain that supplies current to the higher dynode (eg, portion 40b) affects the portion of the chain that supplies the lower dynode (eg, portion 40a). It is a point that will not be given.

FIG. 2C illustrates yet another alternative embodiment of the circuit 30. As before, like reference numerals are used to indicate similar features between the embodiments shown in FIGS. 2A-2C, and thus a detailed description of such features is repeated for brevity. Absent. Similar to FIGS. 2A and 2B, the embodiment shown in FIG. 2C includes a PMT detector 32 having a cathode 34, a plurality of dynodes 36, and an anode 36. In some embodiments, the photocurrent (I _A ) generated at the anode is provided to the current to voltage converter 42 and the switching logic 44. When the anode current (I _A ) reaches a predetermined threshold level (related to anode saturation), the switching logic can be switched to provide an intermediate dynode voltage (V _D ) to subsequent processing components. In other embodiments, anode voltage (V _A ) is monitored instead of anode current (I _A ). In such an embodiment, the switching logic 44 is not coupled to receive the anode current, as shown in FIG. 2C.

FIG. 2C illustrates yet another way in which the intermediate dynode voltage (V _D ) is generated. For example, the circuit 30 includes an operational amplifier circuit 46 for modifying the gain of the intermediate dynode voltage (V _D ). The operational amplifier has a pair of inputs, each coupled to a different terminal of an additional impedance element (Z _D ). In some cases, the load impedance and the feedback impedance are coupled to at least one of the operational amplifier inputs. In this configuration, the intermediate dynode voltage measured across the additional impedance element (Z _D ) is modified to produce a signal with a somewhat higher or lower gain. For example, the configuration shown in FIG. 2C is used to bring the intermediate dynode voltage from a high negative potential to a ground level that is the same level as the anode output.

  FIG. 3 is a flow diagram illustrating an exemplary method for inspecting a sample using the inspection system of FIG. 1 and the circuitry shown in FIGS. 2A and 2B. The various method steps shown in FIG. 3 may be performed by components included within the inspection system, although some steps may be performed by a user of the inspection system.

  In some embodiments, the method begins by selecting (in step 50) the dynode that will be used to measure the intermediate dynode voltage. In some cases, a particular dynode is selected by the user of the inspection system to provide the desired gain difference between the selected dynode and anode of the PMT. The desired gain difference is an assumption, a reasonable guess, or a predetermined amount based on the expected or previous level of scattered light intensity. In other cases, a particular dynode may be automatically selected by an inspection system component (eg, processor 24) based on an expected or previous level of scattered light intensity.

  At step 52, the light is directed to the sample under consideration. As noted above, the term “sample” refers to a wafer, reticle, or any other information that is inspected for defects, features, or other information known in the art (eg, amount of haze or film properties). It is a sample. In the embodiments described herein, the system used to inspect the sample is a dark field optical inspection system that measures scattered light rather than reflected light. Thus, the method continues (at step 54) to detect light scattered from the sample using an anodic current from a photomultiplier tube (PMT). The PMT may be configured as shown in FIG. 2A or 2B. As shown in FIG. 3, the anode current is used to detect scattered light as long as the anode current remains below a predetermined threshold level. This threshold level can be set manually by the user of the system or automatically by system components based on the type / configuration of the PMT. In most cases, the predetermined threshold level may be close to or equal to the saturation level associated with the anode.

  If the anode current remains below a predetermined threshold (at step 56), the anode current is used (at step 58) to detect sample features, defects, and / or light scattering properties. Otherwise, (in step 60) the current from the selected dynode is used for such detection. Using a dynode current allows a much larger amount of scattered light to be measured at the dynode than is measured at the anode. This means that there is a smaller amplification at the dynode that allows a large signal of approximately 1 to 1000 times to be measured at the dynode without dynode saturation (eg, using an intermediate danode). caused by. However, in order to maintain accuracy, the dynode current must be multiplied by a calibration ratio approximately equal to the anode and intermediate dynode gain ratio. Referring to FIG. 1, the calibration ratio may be applied to the dynode current by the system processor 24 or other system component (eg, an analog board or a data processing board). Regardless of how the calibration ratio is applied, the use of the calibration ratio can be used to detect defects by the dynode current, as if it were generated at the anode.

  In some cases, the method may end at step 60. However, in other cases, if the originally selected dynode is insufficient for any reason for the current or subsequent measurement, the method includes additional steps (eg, steps 62, 64, 66). . For example, a relatively low gain dynode is insufficient to detect very low optical signals. On the other hand, relatively high gain dynodes can saturate when very high intensity light is supplied to the PMT. If it is determined that (in optional step 62) the initially selected dynodes are insufficient, an alternative dynode is selected (in optional step 64) and the specimen features, defects, (in optional step 66). Used to detect other light scattering properties. An alternative dynode may be selected by a user or component of the inspection system during the current ongoing measurement operation or in preparation for the next measurement operation to be performed.

  1-3 illustrate exemplary inspection systems, circuits, and methods for improving defect detection by increasing the measurement detection range of a PMT detector while maintaining linearity and stability of the detection signal. Indicates. For example, the anodic current can be used for detection of scattered light on the low intensity side, while the intermediate dynode current is used for detection on the high intensity side. Unlike conventional PMT detectors, where the detection range is limited by unavoidable anode saturation, the PMT detectors described herein switch to higher gain dynode currents when the anode is saturated (higher Extend the measurement detection range (on the intensity side). Thus, the PMT detector described herein has higher sensitivity, lower saturation detection signal (obtained from anode current) and lower sensitivity, higher saturation detection signal (obtained from dynode current). By switching dynamically between, it is used to detect substantially more features, defects, or light scattering properties of the sample. Anode current can be used to detect features such as small defects with low scattering intensity, while large defects that are highly scattered may be detectable by dynode current.

  In addition to extending the detection range, the PMT detector described herein is a simple but highly effective solution that avoids anodic saturation, thereby increasing the linearity and stability of the detection signal. To maintain. In addition, the PMT detector of the present invention is required in prior art designs to compensate for non-linear effects and transient effects that may result when attempting to change the PMT gain “on the fly”. Avoid elaborate calibration and complicated circuits, often. Furthermore, the present invention provides these advantages while using a single detector, and thus the additional optics required when using multiple detectors to extend the measurement detection range. And control circuit. Thus, the present invention provides significant savings in both space consumption and cost.

  FIGS. 4-6 illustrate exemplary circuits, systems, and methods for overcoming detection range limitations typically defined by amplifier saturation and a fixed bit range of an analog-to-digital converter (ADC). Although a relatively small number of embodiments are shown, those skilled in the art will readily understand how the various concepts described herein can be applied to create alternative embodiments with similar functionality. Like.

  4A and 4B show an exemplary embodiment of a detector circuit and an amplifier circuit used in the inspection system to detect light scattered from the sample. The detector circuit 70 shown in FIGS. 4A and 4B includes only one photodetector to detect light scattered from the sample and to convert the light into an electrical signal. As described above, a “single detector” may have only one detection area or several detection areas mounted in a single housing. Accordingly, the detector shown in FIGS. 4A and 4B includes a single detector having a single detection region or a single detector array having multiple detection regions. In general, detector circuit 70 may include virtually any technique suitable for detecting light scattered from a sample. Exemplary detectors include, but are not limited to, photodiodes, phototubes, photomultiplier tubes (PMT), time delay integration (TDI) cameras, charge coupled device (CCD) cameras.

  The amplifier circuit 72 is generally configured to produce a pair of output signals in response to the electrical signal generated by the detector circuit 70. For this reason, the amplifier circuit 72 may be referred to herein as a “dual output amplifier”. In the illustrated embodiment, the amplifier circuit 72 is configured to generate a high resolution (high gain) output signal and a low resolution (low gain) output signal from the electrical signal. As described in more detail below, the high resolution signal is used to detect sample features, defects, and / or light scattering properties when the scattered light is incident within a low intensity range. When the high resolution signal saturates, the low resolution signal tends to scatter more strongly (ie, when scattered light is incident within the high intensity range) to detect additional features or characteristics of the sample Used for. During a sample inspection system scan, the inspection system dynamically switches between high and low resolution signals, thereby overcoming the limitations typically provided by conventional amplifier and ADC circuits. Means for extending the detection range are provided below (eg, in FIGS. 5-6).

A first embodiment of the dual output amplifier 72 is shown in FIG. 4A as including a pair of operational amplifiers (74, 76) and a voltage divider network implemented, for example, by resistors R ₁ and R ₂ . It is. The operational amplifiers (or “op amps”) 74, 76 are configured as voltage followers with negative feedback. The positive terminals (+) of operational amplifiers 74, 76 are coupled to receive the photodetector current (I _S ) generated by detector 70 multiplied by some resistance value. In the embodiment of FIG. 4A, the photodetector current is multiplied by the value of resistor R ₂ for op amp 76 and by the values of resistors R ₁ and R ₂ for op amp 74. Because of the voltage follower configuration, the voltage present at the output terminals of the operational amplifier is substantially equal to the voltage supplied to their positive terminals. In this configuration, the dual output amplifier 72 can generate a high resolution output signal (V _H ) and a low resolution output signal (V _L ) substantially equivalent to the following.
V _H = V ₁ = I _S ^* (R ₁ + R ₂ ) and Equation 4
V _L = V ₂ = I _S ^* R ₂ Formula 5
The values of resistors R ₁ and R ₂ are selected to provide output signals having substantially different gains. In some cases, for example, to generate a high resolution output signal (V _H ) having a gain 16 times greater than a low resolution output signal (V _L ), the value of resistor R _{1 is} the value of resistor R ₂ . It may be 15 times larger than the value. In one embodiment, resistor R ₁ may have a value of about 7.5 kΩ and resistor R ₂ may have a value of about 500 Ω.

In the example provided above, the values of R ₁ and R ₂ are selected to increase the gain difference (and hence the circuit detection range) by about 2 to about 16 times. In other embodiments of the present invention, other values may be selected to produce the same amount of gain, or to provide more or less gain. For example, the resistor value may be selected to increase the gain difference (and hence the circuit's detection range) from about 2 to about 1024 times the original range provided by the photodetector alone. The amount by which the detection range is expanded is generally desired at the resolution of the subsequent analog-to-digital converter (eg, ADC 22 in FIG. 1) and the switching point between the high resolution signal and the low resolution signal. Depends on resolution. For example, the gain difference is extended by about 1024 times with a 14-bit converter (up to 16383 ADC) and the desired overlap resolution of 16 ADC.

Another embodiment of the dual output amplifier 72 is shown in FIG. 4B. In this embodiment, amplifier 72 includes at least three operational amplifiers (78, 80, 82). The first operational amplifier (78) is coupled to receive the photodetector current (I _S ) at the negative terminal (−) of the operational amplifier and the ground potential at the positive terminal (+). Resistor R ₁ is placed in the negative feedback of op amp 78 to produce an output voltage substantially equivalent to:
V _N = I _S ^* R ₁ Formula 6
The second and third operational amplifiers (80, 82) are coupled to generate a pair of output signals based on the contact voltage (V _N ) provided by the operational amplifier 78. For example, resistors R ₂ and R ₃ may be included in op amp 80 to generate a high resolution output signal (V _H ) that is substantially equivalent to:
V _H = −V _N [R ₃ / R ₂ ] Equation 7
Similarly, resistors R ₄ and R ₅ may be included in op amp 82 to generate a low resolution output signal (V _L ) that is substantially equivalent to:
V _L = −V _N [R ₅ / R ₄ ] Formula 8
As in the above embodiment, the values of resistors R ₁ , R ₂ , R ₃ , R ₄ , R ₅ may be selected to provide output signals having substantially different gains. In one embodiment, the values of resistors R ₃ and R ₅ may be set equal to each other such that Equation 8 is:
V _L = −V _N [R ₃ / R ₄ ] Formula 9
At this time, the values of resistors R ₂ and R ₄ may be selected to provide the desired gain difference. In one embodiment, the value of resistor R ₂ is 16 of the value of resistor R ₄ to produce a high resolution output signal (V _H ) having a gain 16 times greater than the low resolution output signal (V _L ). It may be twice as small. For example, resistor R ₂ has a value of about 62.5Ω, resistors R ₁ and R ₄ (which may be equivalent in some embodiments) have a value of about 1 kΩ, and resistor R ₃ And R ₅ may have a value of about 500Ω. In other embodiments of the invention, other values may be selected to produce the same amount of gain or to provide more or less gain. For example, the resistor value is selected to increase the gain difference (and hence the circuit's detection range) from about 2 to about 1024 times the original range provided by the photodetector alone. In some embodiments, one or more additional amplifier circuits are coupled in series with operational amplifiers 80 and 82 to increase / decrease the gain difference between the high resolution output signal and the low resolution output signal. .

The high and low resolution output signals generated by the dual output amplifier 72 (eg, as shown in FIG. 4A or 4B) are fed to separate processing channels (84, 90) where the filters Processed and converted into a pair of digital signals. For example, as shown in FIG. 5, the high resolution signal (V _H ) generated by the dual output amplifier 72 is provided to a processing channel 84 where it is filtered by an analog filter 86 for analog to digital conversion. Is digitized by a device (ADC1) 88. Similarly, the low resolution signal (V _L ) generated by the dual output amplifier 72 is fed to the processing channel 90 where it is filtered by the analog filter 92 and by the analog to digital converter (ADC2) 94. Digitized. The analog filters 86 and 92 may be anti-aliasing filters such as Butterworth filters and Bessel filters. Virtually any N-bit analog-to-digital converter (88, 94) may be used to convert the high and low resolution analog signals into corresponding digital signals. The gain difference between the analog signals makes the high resolution digital signal somewhat larger than the low resolution digital signal.

  At least one of the digital signals may be provided to subsequent inspection system components for further processing (such as processor 24 of FIG. 1). For example, as shown in FIG. 5, multiplexer 96 is coupled to select a high resolution digital signal from processing channel 84 or a low resolution digital signal from processing channel 90. Selection is most often controlled by switching logic 98 coupled to a processing channel 84 for monitoring high resolution digital signals.

As described above, each of the ADCs can process up to N digital bits. In some cases, as long as the digital value stays below a predetermined threshold associated with an N-bit ADC (eg, 2 ⁸ = 256 for an 8-bit ADC), the switching logic 98 outputs a high resolution signal for output. select. When the high resolution signal reaches a predetermined threshold, switching logic 98 selects the low resolution signal for output. In this configuration, the inspection system uses a larger (ie, higher resolution) digital signal to detect “low scatter” features (eg, small particles or surface defects), Alternatively, light scattering characteristics can be detected. When a predetermined threshold is reached, a small (ie, low resolution) digital signal is used to detect features that tend to scatter more strongly (eg, large particles or surface defects). The lowest resolution is maintained at the “switch point” between the high resolution signal and the low resolution signal by selecting a “predetermined threshold” smaller than the highest ADC output (2 ^N ). For example, a maximum threshold value of substantially 2 ^N-1 or less is selected to maintain sufficient resolution at the switching point.

  As shown in FIG. 5, switching logic 98, when combined, is implemented as logic and / or storage elements that perform the functions performed within switching logic 98. In other embodiments, the functionality may be implemented by program instructions that are stored and executed “on-chip” or “off-chip”. For example, switching logic 98 may be implemented as a digital signal processing (DSP) chip coupled to processing channels 84, 90, or as one or more program instructions executed by processor 24. In the latter case, each digital signal is instead fed to the processor 24 and the multiplexer 96 is excluded from the block diagram of FIG. Signal selection is then performed through program execution in the processor 24. Other configurations / implementations of the switching logic 98 are possible and within the scope of the present invention.

  FIG. 6 is a flow diagram illustrating an exemplary method for inspecting a sample using the inspection system of FIG. 1 and the circuits and systems shown in FIGS. The various method steps shown in FIG. 6 can be performed by components included within the inspection system, although some steps may be performed by a user of the inspection system.

  In some embodiments, the method begins by directing light (at step 100) to the sample under consideration. As noted above, the term “sample” is used to refer to wafers, reticles, or defects, features, or any other information known in the art (eg, amount of haze characteristics or film characteristics). This is a sample. In the embodiments described herein, the system used to inspect the sample is a dark field optical inspection system that measures scattered light rather than reflected light. Accordingly, the method continues (at step 102) by detecting light scattered from the sample. Unlike some prior art inspection systems that attempt to extend the detection range using multiple detectors, the systems and methods described herein use only one detector for such detection. use. A single detector as used herein includes virtually any photodetector that is commonly used to detect scattered light and to convert light into an electrical signal.

At step 104, the electrical signal generated by the detector is converted into a pair of unequal amplified signals. For example, a dual output amplifier similar to that shown in FIGS. 4A-4B is used to generate a first signal and a second signal in response to an electrical signal. In some cases, the first signal has a higher resolution (higher gain) than the second signal to detect substantially lower levels of scattered light. On the other hand, the second signal has a lower resolution (lower gain) than the first signal in order to detect substantially higher levels of scattered light. Thus, the lower level light detected by the first signal is in the first detection range, while the higher level light detected by the second signal is in the second detection range. . In order to maintain sufficient resolution at the “switching point”, the second detection range may at least partially overlap the first detection range. As described above, this is accomplished by selecting a predetermined threshold value less than the maximum value (eg, 2 ^N ) associated with the N-bit analog to digital converters 88,90. In some cases, ADCs 88, 90 each have a different number of bits, one N greater than the other, such as M. In such a case, sufficient resolution is still maintained at the switch point by selecting a threshold value less than the maximum value (eg, 2 ^N , 2 ^M ) associated with N-bit ADC 88 and M-bit ADC 90.

  In most cases, the first (high resolution) signal is used (at step 106) to detect sample features, defects, or light scattering properties until the first signal reaches a predetermined threshold. . However, when the first signal reaches the predetermined threshold (at step 108), the second (low resolution) signal (at step 110) is used to detect sample features, defects, or light scattering properties. used. For example, while the first signal is being used for detection, the switching logic 98 compares the digital value of the first signal to a predetermined threshold. The switching logic 98 then switches to the second signal when the digital value of the first signal reaches and / or exceeds a predetermined threshold.

  4-6 illustrate exemplary inspection systems, circuits, and methods for improving defect detection by increasing the measurement detection range of amplifiers and analog / digital circuits included in the inspection system. For example, dual output amplifiers are used to generate high and low resolution output signals from electrical signals provided by a single photodetector. The systems, circuits, and methods described herein saturate amplifiers and analog-to-digital circuits by dynamically switching between high and low resolution output signals during an inspection scan. To avoid. This essentially increases the measurement detection range and allows to detect more features, defects, or light scattering properties of the sample by extending the range of defect sizes detected by the two output signals To do. In one example, the high resolution signal may be used to detect defects in the range of about 40 nm to about 155 nm, while the low resolution signal is used to detect defects in the range of about 80 nm to about 500 nm. May be. Other embodiments are able to detect somewhat smaller or larger defects. Factors that determine the size of defects detected by the systems and methods of the present invention include, but are not limited to, laser power, wavelength, surface topology, spot size, polarization, collection angle, detector efficiency, and noise. . Overlapping the detection range ensures that there will be sufficient resolution at the “switch point” between the high resolution signal and the low resolution signal.

  The systems, circuits, and methods described herein also attempt to extend the measurement detection range by using expensive optics to split scattered light between multiple detectors. Avoid additional space consumption and costs associated with inspection systems. By using only one photodetector, the present invention increases the sensitivity of the detected signal by avoiding the signal-to-noise reduction that occurs when the scatter is split between multiple detectors. It is also possible to improve.

  In addition to the embodiments shown in FIGS. 4-6, the concepts described herein extend to more than one amplifier output. For example, the amplifier design shown in FIGS. 4A and 4B generates three or more output signals (eg, high resolution, medium resolution, and low resolution signals) from the electrical signal generated from the detector 70. Will be corrected. If a PMT detector similar to that shown in FIGS. 2A and 2B is used to implement the detector 70, the system shown in FIG. 5 may be an intermediate dynode if, for example, the low resolution signal is saturated. It can also be modified to switch to voltage. Other configurations and / or embodiments are possible and are therefore considered within the scope of the present invention.

  FIG. 7 shows another example of an inspection system used to perform a method for inspecting a sample, as described herein. In particular, FIG. 7 illustrates one example of an inspection system used to reduce and / or prevent thermal damage to a sample during a surface inspection scan. Thermal damage is generally seen in prior art systems when incident laser light directed to the sample is absorbed and poorly dispersed by features on the sample (such as large particles or defects). In addition to preventing thermal damage, the inspection system of FIG. 7 is used to provide yet another means for extending the measurement detection range of the inspection system.

  FIGS. 7-12 illustrate during a surface inspection scan using a high power laser-based inspection system by dynamically reducing the incident laser power before scanning over a large particle that is highly scattered. 2 illustrates an exemplary circuit, system, and method for reducing particle breakage. Although a relatively small number of embodiments are shown, those skilled in the art will readily understand how the various concepts described herein can be applied to produce alternative embodiments with similar functionality. Will understand.

In high power laser based inspection systems, the power density of the incident laser beam typically ranges from about 1 kW / cm ² to about 1000 kW / cm ² . Unfortunately, particle breakage often occurs during surface inspection scans with high power density laser beams due to the rapid power transfer from the laser beam to the particles (or part of the particles) on the sample. To do. Particles that cannot disperse large amounts of power tend to become hot quickly and often burst due to insufficient power distribution. For example, organic materials (such as photoresist particles) tend to disperse significantly less power than inorganic materials (such as metal particles) and therefore tend to suffer more damage. Unfortunately, the ruptured particles become debris that can spread large areas of contamination throughout the sample.

  The methods used to reduce thermal damage are reduced laser power (eg, power that is less than 10) or increased spot size (eg, spot size (eg, power that is known to provide damage) , 2.5x to 5.7x large spot size). Another method is to exclude the center of the wafer from inspection by suppressing a portion of the incident laser beam. This method typically reduces thermal damage by removing high power levels supplied to the central area of the wafer. However, these methods reduce sensitivity by reducing the signal-to-noise ratio (when reducing laser power, ie increasing spot size), or result in incomplete inspection of the wafer. (Ie if you exclude the center). Thus, the aforementioned methods often miss defects on the wafer due to insufficient sensitivity or complete exclusion.

In contrast, the inventive concepts described herein are based on the consideration that larger particles (generally> 5 μm) are more likely to be damaged by the incident laser beam than smaller particles. . For example, larger particles have a wider surface area and therefore tend to absorb significantly more power than smaller particles with less surface area. Larger particles also tend to scatter significantly more light than smaller particles because of the larger area and / or increased surface irregularities. For example, the amount of light scattered from a particle of radius R is relatively proportional to the particle radius increased to a power of ⁶ (ie, R ⁶ ).

  The inventive concept described herein takes advantage of the large scattering properties of large particles to reduce thermal damage during surface inspection scans. As described in more detail below, thermal failure is avoided by detecting the presence of large particles and reducing the incident laser beam power before the main part of the beam reaches the large particles. In one embodiment, the power reduction is obtained by a fast laser power attenuator that can be used to reduce the incident laser power to a “safe” level when scanning across large particles. The laser power attenuator can be removed to maintain (or increase) the incident laser power at “total” power when scanning the lower scattering portion of the wafer.

Referring to the drawings, FIG. 7 illustrates an inspection system similar to the system shown in FIG. 1 and described above. Elements common to FIGS. 1 and 7 are indicated using similar reference numerals and the description thereof will not be repeated here. Unlike the inspection system of FIG. 1, the inspection system shown in FIG. 7 is specific to a high-power optical inspection system. Thus, the light source 10 includes any number of light sources with output power densities ranging from about 1 kW / cm ² to about 1000 kW / cm ² . Examples of potential high power laser based sources used for the light source 10 are diode lasers, solid state lasers, diode pumped solid state (DPSS) lasers (helium neon lasers, argon lasers, etc.) This includes, but is not limited to, various gas lasers. In some cases, light source 10 is implemented using a high power non-laser based source, such as an arc lamp, a high or low pressure mercury lamp, an LED array, a bulb. The beam of light generated by the light source 10 (ie, “generated light”) is directed to the surface of the sample 14 via the beam shaping / polarization control optical system 12. To remove confusion, the light that reaches the surface of the sample will be referred to herein as “incident light” or “incident laser beam”. As described in more detail below, “incident light” differs from “generated light” in one or more ways, including polarization, intensity, spot size and shape, and the like.

  In addition to the elements shown in FIG. 1, the inspection system shown in FIG. 7 includes means for dynamically changing the power level of incident light supplied to the sample. For example, the laser power attenuator 26 is disposed between the light source 10 and the optical system 12 to dynamically change the power level of the incident laser beam during a surface inspection scan. In general, laser power attenuator 26 is implemented with a selectively transmissive optical component that can be adapted to transmit a portion of incident light based on the polarization of the incident light. For example, in some embodiments, the laser power attenuator 26 includes a wave plate (such as a quarter wave plate) and a polarizing beam splitter. In this configuration, the waveplate is used to change the polarization of the incident light, while the beam splitter transmits one or more selective polarizations (eg, linearly polarized light) and all other light (eg, , Randomly polarized light, circularly polarized light, or elliptically polarized light). By reflecting a portion of the light, the waveplate and beam splitter function to reduce the intensity or power level of light transmitted through the waveplate. However, waveplates and similar optical components (such as a neutral density filter) cannot be turned on or off like a switch, but instead have two distinct power levels. Must be moved in and out of the beam path to provide In some cases, such movement may not be fast enough to provide a dynamic output change during a surface inspection scan.

  FIG. 8 illustrates one embodiment of a preferred laser power attenuator 26. In the illustrated embodiment, very fast laser power attenuation is provided by using electro-optic material 200 in the switch between the “on” and “off” states. When “on”, the electro-optic material 200 can change the polarization of the incident light into a predetermined polarization direction. This so-called “re-polarized light” is then fed to a polarizing beam splitter 210 that can transmit only a portion of the re-polarization, depending on the specific polarization output from the electro-optic switch. The remaining portion of the repolarized light is reflected and absorbed by the beam dump 220. In some cases, the electro-optic material can be switched between an “on” state and an “off” state within a time span of a few nanoseconds to a few microseconds. Thus, fast laser power attenuation is provided by using electro-optic switches rather than selectively moving optical elements that are selectively transmissive in and out of the beam path.

  In certain embodiments, the laser power attenuator 26 is implemented by a high speed electrically controlled optical shutter known as a Pockels cell. Initially, Pockels cell 200 is set to an “on” state to allow light generated by light source 10 to freely pass through laser power attenuator 26. However, when the presence of large particles is detected, the Pockels cell 200 switches to the “off” state, changing the polarization of the generated light to a different polarization that is at least partially transmitted by the polarizing beam splitter 210. In order to switch between an “on” state and an “off” state, an electrical voltage provided by variable power supply 230 is supplied to Pockels cell 200 to polarize light passing through an electro-optic material (usually an electro-optic crystal). change. As shown in FIG. 8, the voltage supplied to the Pockels cell is determined by a control signal input to the variable power source 230.

  In one example, the voltage supplied to Pockels cell 200 (ie, switching the cell to the “on” state) changes linearly polarized light to circularly polarized light (a phenomenon often referred to as “quarter wavelength phase shift”). Furthermore, it is possible to change the characteristics of the electro-optic crystal. When circularly polarized light is supplied to a beam splitter configured to primarily reflect circularly polarized light, the intensity or power level of the light output from the laser power attenuator 26 will cause the Pockels cell to be in an “on” state. Reduced by setting. On the other hand, the intensity or power level of the light output from the laser power attenuator 26 can be maintained (or increased) by setting the Pockels cell to the “off” state.

  However, the intensity of the light output from the power attenuator 26 depends on the phase shift produced by the polarizing beam splitter 210 and the Pockels cell 200. For example, beam splitters generally distinguish between two orthogonal polarizations, such as so-called “S” and “P” polarizations. However, other polarizations of light (such as C-polarized light) are partially transmitted and are therefore partially redirected (within the beam dump) by the beam splitter. When a voltage is applied so that the Pockels cell produces a quarter wavelength phase shift, the incident linearly polarized light (usually the laser output) becomes circularly polarized and half of the light is redirected while the other half is Pass the beam splitter. In the case of a half-wave shift, no light passes through the beam splitter, except for some leakage due to imperfections in the optical components. That is, if the Pockels cell is configured to produce a ½ wavelength shift (assuming all light passes through the beam splitter in the unpowered state), virtually all of the incident light is redirected. Will be.

  In some cases, the constant power laser beam produced by the light source 10 can be identified two times by dynamically switching an electro-optic shutter (such as a Pockels cell) between an “on” state and an “off” state. Power levels (eg, “safe” power levels and “all” power levels). The safe power level is much less than the full power level when scanning over large particles to prevent thermal damage. For example, the safety power level is a few percent of the total power level (eg, ranging from about 1% to about 50%). In one embodiment, the safety power level is about 10% of the total power level. Other possibilities exist and may generally depend on the incident laser power, as well as the size and material composition of the particle being scanned.

  In other cases, an electro-optic shutter (such as a Pockels cell) may be configured to generate more than two distinguishable power levels. For example, a Pockels cell may be driven to produce substantially any phase shift, and thus may be combined with a polarizing beam splitter to produce virtually any output power level. That is, the embodiment shown in FIG. 8 is used to produce virtually any number of identifiable power levels. In some cases, circuitry and / or software may be included to enable continuous power level adjustment, for example, in the form of a closed feedback loop, as shown in FIG.

  In addition to the examples described above and shown in FIGS. 7 and 8, other means may be used to dynamically change the power level of incident light supplied to the sample. For example, such means may include, but are not limited to, direct power adjustment of the light source, high speed micromirror, acousto-optic deflector (AOD), high speed mechanical shutter, and the like. Thus, such means can provide a relatively fast response (eg, on the order of a few nanoseconds to a few microseconds) and at least two distinct power levels (eg, “safe” power level and “total” power level). The present invention can encompass any suitable means for dynamically changing the power level of the laser beam. In general, the response time should be faster than the typical time taken to break the particles. Other factors that can affect the choice of fast laser power attenuators include, but are not limited to, light transmission, cost, reliability, and lifetime.

  In addition to the various means described above for dynamically changing the power level of the laser beam, the inspection system of FIG. 7 provides a means for controlling such changes. For example, the laser power controller 28 is coupled between one or more elements of the detector subsystem (eg, collector 16, photodetector 18, amplifier 20, ADC 22, processor 24) and the laser power attenuator 26. May be. As described in more detail below, the laser power controller 28 continuously monitors the light scattered from the sample 14 and detected by the detector subsystem so that the detected scattered light is predetermined. Determine whether the threshold level is exceeded or less. Based on such determination, the laser power controller 28 can transmit the incident light at a first power level (eg, “all” power level) or a second power level (eg, “safe” power level). The laser power attenuator 26 can be instructed to provide the sample. If more than two power levels are available and the situation warrants (or benefits) such a level, the laser power controller may connect to the laser power attenuator, eg, third, fourth, or It is also possible to provide the sample with a fifth (and so on) power level.

In general, the predetermined threshold level is set to reduce or prevent thermal damage that may occur if incident light directed to the sample is absorbed by features on the sample and poorly dispersed. The The predetermined threshold level is generally based on the incident laser power density, more specifically the power density associated with the onset of thermal failure applied on a feature of a certain size or on a particle. For example, the predetermined threshold level, relatively large particles (e.g.,> 5 [mu] m) in order to avoid damage to, be selected from the group of the incident laser power density ranging from about 1 kW / cm ² to about 1000 kW / cm ² Good. When scanning the organic material, a predetermined threshold level, in order to avoid damage to the large particles due to the relatively poor thermal radiation, range from about 1 kW / cm ² to about 100 kW / cm ^2. As shown in FIG. 7, the predetermined threshold level is supplied from the processor (or computer system) 24 to the laser power controller 28. The predetermined threshold level may be selected manually by a user of the system or automatically by the processor 24.

  If the detected scattered light remains below a predetermined threshold level, the laser power controller 28 can instruct the laser power attenuator 26 to maintain the incident laser beam at a “full” power level. . However, if the detected scattered light exceeds a predetermined threshold level (e.g., indicating that a large particle is nearby), the laser power controller 28 sets the power of the incident laser beam to a "safe" power level. Can be given instructions to reduce. When the incident laser beam scans across a large particle (or other highly scattering feature), the detected scattered light is retracted below a predetermined threshold level, and the laser power controller 28 causes the incident laser beam to “total” Instruct the laser power attenuator to restore power.

  Thus, the inspection system described herein detects relatively small size features by directing incident light onto a sample at a first power level (eg, “total” power), while , May be uniquely configured to detect relatively large size features by directing incident light onto the sample at a second power level (eg, “safe” power). In current systems, larger features are detected without causing thermal damage on these features by setting the second power level much lower than the first power level. If more than two power levels are available, the laser power controller 28 compares the detected scattered light to two or more thresholds and, based on such comparison, determines the appropriate incident laser beam. The laser power attenuator can be commanded to maintain, reduce, or increase to a power level.

FIG. 9 illustrates one embodiment of a preferred laser power controller 28. In the illustrated embodiment, the laser power controller 28 includes a divider 240 for normalizing the detector output with respect to the incident laser power, the detector gain. Thus, the divider 240 can be used to calculate normalized scattered power and is therefore sometimes referred to as a scattered power calculator. In one example, the scattering power is calculated by dividing the detector output by the incident laser power and the detector gain, or by:
Scattering power = detector output / (laser power) (detector gain) Equation 10
By normalizing the detector output in this way, the laser power controller causes the laser power attenuator to switch consistently at the same scattered light level, rather than switching at a given signal level. That is, as the detector gain increases, all signals become larger. If the detector output is not normalized as the detector gain increases, switching may occur with a smaller particle size than is actually intended. Normalizing the detector output to the incident laser power and detector gain allows for consistent switching when a particle of a given size is detected (eg, to a lower power level) To.

  As shown in FIG. 7, the divider 240 is the detector output (ie, detected from the sample) as an analog signal from the photodetector 18 or as a filtered and digitized signal from the ADC 22 or processor 24. Scattered light) can be received. The detector gain (ie, the current amplification associated with the detector) is provided by processor 24 to divider 240. It can be variable or fixed depending on the particular photodetector used. However, as described in more detail below, incident laser power may be provided to divider 240 in one of two ways.

  In some embodiments, the normalized scattered power signal generated by divider 240 is fed back to processor 24 as indicated by the dashed lines in FIGS. 7 and 9-11. That is, divider 240 is used to present the scatter data (ie, detector output) normalized to the incident laser power and detector gain as the scan result. By normalizing the data before being sent to the processor, the actual detected scatter power is used to accurately detect the size of the defect. For example, if the incident laser power decreases when scanning over a large particle, the ADC count (ie, the detector output) will necessarily be lower than if it is not attenuated. This means that the scan results supplied by the processor will show smaller defects than those actually present. Normalizing the scatter data allows the processor to more accurately determine the size of the defect.

  In some embodiments, an additional normalizer / divider 23 is used in the data collection path between the ADC 22 and the processor 24 to normalize the scattered power signal to changes in incident laser power. Also good. Additional divider 23 may be used in conjunction with or as an alternative to normalizer / divider (240) included within laser power controller 28. For example, if two dividers are used, divider 23 is placed in the data acquisition path to normalize the scattered power signal sent to processor 24, while divider 240 is another laser power. Placed in the threshold path to normalize the scattered power signal sent to the control component (eg, comparator 250 discussed above). However, only one divider is used (in the data collection path or threshold path) or no divider is used (in this case, the system supports dynamic range expansion, as discussed below) Other options may not exist.

  As described above, incident laser power is provided to divider 240 in one of two ways. In the embodiment of FIG. 9, the actual laser power of the incident beam is measured by a laser power detector 27 that may be placed in the beam path below the laser power attenuator 26. In this embodiment, a laser power detector 27 is included to monitor the actual intensity or power level output from the laser power attenuator 26. The power measured by the laser power detector (ie, measured power) is supplied to the laser power controller 28 to calculate the normalized scattering power. The laser power detector may be implemented using virtually any power detection means including, but not limited to, a photodiode or a photomultiplier tube (PMT), among others. However, as described in more detail below, the laser power detector 27 may not be included in all embodiments of the present invention.

  As shown in FIG. 9, the laser power controller 28 includes a comparator 250 for comparing the normalized scatter power with one or more predetermined threshold levels provided by the processor 24. As described above, the threshold level (s) may be used by the system user or processing component to effectively reduce thermal damage when scanning across large (or other highly scattered) particles. May be selected. In one embodiment, the comparator 250 may receive only one threshold level (referred to as the “safe scatter threshold”) that indicates the laser power density associated with the maximum amount of “safe” scatter power. After comparing the normalized scatter power to the safe scatter threshold, the comparator 250 maintains or changes the power mode supplied to the attenuator to bring the incident laser beam to an appropriate power level (eg, , "Total" power level or "safe" power level) can be commanded to the laser power attenuator 26 to reduce, or increase. In general, the power mode may be any control signal that causes the attenuator to maintain or change the output power level. For example, in the embodiment of FIG. 8, the power mode supplied to the attenuator functionally corresponds to a control signal input to the variable power supply 230.

  FIG. 10 illustrates another embodiment of a preferred laser power controller 28. In particular, FIG. 10 illustrates one way in which normalized scattered power is calculated when a laser power detector is not used to make a measurement of the actual power level output from the laser power attenuator 26. . Instead of receiving measured power, divider 240 may be coupled to the output of comparator 250 to receive a power mode control signal supplied to the attenuator. Based on the control signal, divider 240 can determine an appropriate incident laser power that will be used in the scattered power calculation, for example, using a look-up table. In the embodiment shown in FIGS. 9 and 10, the distributor 240 and the comparator 250 may be implemented by hardware, software, or a combination of both. In one example, divider 240 may be implemented in software, while comparator 250 may be implemented in hardware.

  FIG. 11 illustrates yet another embodiment of a preferred laser power controller 28. In particular, FIG. 11 illustrates one method in which continuous power adjustment is performed by the laser power controller 28. As with the previous embodiment, divider 240 is coupled to receive the detector gain and output signal and to generate a normalized scattered power signal in response. . However, instead of supplying a normalized scattered power signal to the comparator (as shown in FIGS. 9-10), the scattered power signal dynamically adjusts the output power mode based on the supplied signal. To the control loop feedback filter 260. In FIG. 11, the scattered power signal is used in the feedback loop 260 to adjust the incident laser power (eg, to achieve a constant detector output signal). Thus, instead of a fixed power level, the embodiment shown in FIG. 11 provides a continuously adjustable power level.

  The circuits and systems shown in FIGS. 7-11 reduce thermal damage to large particles (eg,> 5 μm) by dynamically adjusting the intensity or power level of the incident laser beam during a surface inspection scan. Can be reduced. In one embodiment, thermal damage is reduced to 100% for a fixed incidence laser beam inspection system. The circuits and systems described herein provide one or more preset threshold levels that are used to dynamically switch between two or more incident beam power levels during a scanning operation. By doing so, it is adjusted for various scanning operations. In this way, thermal damage is reduced or avoided by reducing the incident laser power to a low power level (eg, a “safe” power level) when scanning across large particles that are highly scattered. It is even possible. However, detection sensitivity is maintained by scanning a low scattering region at a higher power level (eg, “total” power level) that allows the system to detect smaller defects.

  In an alternative embodiment of the present invention, an adaptive learning process is used to change the threshold level and / or power level based on previous or current inspection scan results. One advantage is that the adaptive learning process can be used for switching electronics because decisions are not made “on the fly” just before switching is needed, but much before the actual switching event. Longer delay times. While accuracy may increase, such embodiments will obviously increase the complexity (and possibly cost) of the circuits and systems described above. As an intermediate alternative, in some embodiments it is possible to establish a scaled relationship between scattering power and incident power level to continually change the amount of incident power delivered to the sample. . This alternative is used to provide a scattered light signal that is always near the optimum range for the PMT by continuous adjustment of the incident power delivered to the sample.

  Although the use of a single photodetector is preferred in most embodiments of the present invention, in some embodiments an additional detector is used to select the appropriate power level that will be delivered to the sample. May be included. If included, a separate detector may be used to monitor the light scattered from the sample. However, unlike the original detector used to detect scattered light so that the incident power level is adjusted accordingly, the additional detector will then have the appropriate power that will be directed to the sample. Used to select a specific threshold level or “switch point” that is used to select a level.

In addition to reducing thermal damage, the circuits and systems shown in FIGS. 7-11 may be used to increase the measurement detection range of an inspection system. Usually, the detection range of the fixed power inspection system is limited to the detection range of the photodetector. However, by providing a variable power inspection system, the present invention significantly increases the measurement detection range to nearly below.
(Detection range of photodetector) × (detection range of attenuator) Equation 11
In some cases, the additional detection range provided by the laser power attenuator increases the overall measured detection range of the system from about 2 to 16 times. When combined with other means described herein (such as improved PMT detectors and / or dual output amplifiers), the overall measurement detection range of the inspection system ranges from about 2 to 10, compared to the prior art. It is improved up to 000 times.

  FIG. 12 is a flow diagram illustrating an exemplary method for inspecting a sample using a variable power inspection system, such as that shown in FIGS. 7-11 and described above. While the various method steps described in FIG. 12 can be performed by components included within the inspection system, some steps are performed by a user of the inspection system.

  In some embodiments, the method can begin by directing light (at step 300) to the sample at a first power level. For example, the incident laser beam is delivered to the sample at a relatively high power level (such as a “total” power level) so that relatively small features or defects are detected. As described in more detail below, the incident laser beam is lower later (such as a “safe” power level) so that these features or defects are detected without damaging relatively large features. It may be reduced to a power level. The method shown in FIG. 12 is modified to dynamically switch between more than two power levels as required.

  In most cases, the method can detect light scattered from the sample (at step 304) while simultaneously scanning light (at step 302) across the surface of the sample. For example, if the incident laser beam is directed to a current location on the sample, scattered light is detected from the sample. The scattered light detected at the current location is used (at step 306) to detect sample features, defects, or light scattering properties at that location. The beam position can then be scanned against a nearby location, where the process is repeated to detect features that may be found at the nearby location.

  In some embodiments, the scanning is performed by placing an optical deflector in the beam path leading to the sample. For example, the deflector may be included in the beam shaping / polarization control optical system 12 of FIGS. In some embodiments, the deflector is an acousto-optic deflector (AOD), mechanical scanning assembly, electronic scanner, rotating mirror, polygon-based scanner, resonant scanner, piezoelectric scanner, galvo mirror, or galvanometer. including. In some embodiments, the deflector can scan the light beam across the sample at a substantially constant scan rate. For example, the light beam can be scanned across the sample at a constant scan speed selected from a speed range of about 0 m / s to about 24 m / s. In other embodiments, the deflector can scan the light beam across the sample at various scan speeds ranging from about 0 m / s to about 24 m / s. However, the deflector need not perform a scan in all embodiments of the invention. For example, a normal incident beam of light may be scanned across the sample by relative movement of the beam shaping optics with respect to the sample and / or by relative movement of the sample with respect to the optical system.

  During the scanning process, light scattered from the sample (at step 304) is detected at the current location, while the method monitors (at step 308) light scattered from the sample at a nearby location. For example, the incident laser beam may be delivered to the sample by a power density distribution that peaks near the middle of the distribution and tapers near the edge of the distribution. As used herein, the middle of the distribution will be referred to as the “main beam” while the end is referred to as the “beam skirt”. One example would be a power density distribution with a main lobe and at least one side lobe on each side of the main lobe. The side lobes have a lower power density and therefore have less amplification than the main lobe. Another example is a bell or Gaussian distribution with a beam skirt whose one main lobe gradually tapers.

  When scanning across the entire surface of the specimen, the beam skirt of the incident laser beam can reach the particle or defect before the main beam reaches the particle or defect (eg, about 1 to a few microseconds) It is. For example, most particles or defects on the surface of the sample will be much smaller than the spot size of the laser beam. This allows the beam skirt to reach particles or defects before the main beam reaches. When the incident laser beam is scanned across the surface of the sample, the amount of light scattered from the sample (ie, the scattered power) will vary depending on which part of the beam covers the particle or defect. become. For example, if a low power density beam skirt reaches large particles or defects, assume that the high power density main beam covers a relatively smooth surface of the sample. Because the amount of light scattered from a relatively smooth surface is generally much smaller than the scattering attributed to large particles or defects, the amount of scattering power attributed to the main beam portion should be considered negligible. Can do. Thus, a significant increase in scattering power can indicate that the beam skirt has reached large particles or defects. That is, the presence of large particles or defects is detected at close locations by monitoring the scattering level from a low power density beam skirt.

  If (at step 310) the scattering level from the beam skirt is substantially greater than or equal to the safe scattering threshold, (at step 314) the incident laser beam is at a second power level that is lower than the first power level. It may be reduced. As noted above, the second power level (referred to as the “safe” power level) is generally about 1% to about 50% of the first power level (referred to as the “total” power level). It may be within the range and in a preferred embodiment it may be about 10% of the total power level. If the scatter level from the beam skirt is below the safe scatter threshold (at step 310), the first power level is maintained to allow detection of smaller features and maintain detection sensitivity. Will be.

  In some cases, once the surface inspection scan is complete (at step 316), the method ends (at step 318). Otherwise, the scanning process may continue (at step 320) so that the close position becomes the current position. While the incident laser beam moves across the defect, the method continues to monitor the scattering level of the beam skirt. If (at step 310) the dense center of the beam (ie, the main beam) has passed through the defect and the scatter level falls back below the safe scatter threshold (at step 312), the first power The level will continue to inspect the sample for small defects.

  By using a fast laser power attenuator, such as that described above and shown in FIGS. 7 and 8, the system and method described herein allows the main beam to reach a close position. In addition, it is easy to switch between a high power level and a low power level. For example, the scattering level of the beam skirt is used to indicate the presence of a large defect at that location a few microseconds before the main beam reaches a nearby location. For one thing, due to the relatively fast response (eg, on the order of a few nanoseconds to a few microseconds) of the laser power attenuator 26, the present invention allows the main beam to be monitored by monitoring the scattering level of the beam skirt. It is possible to reduce the incident beam power level before the defect is reached. By dynamically reducing the power level while scanning a large particle and increasing the power level as the particle is scanned, the present invention allows large particles to be maintained while maintaining system throughput and sensitivity. Reduce heat damage against

  Further modifications and alternative embodiments of various aspects of the invention will be apparent to those skilled in the art in view of this specification. For example, methods and systems are provided for extending the detection range of an inspection system. Accordingly, this specification is to be construed as illustrative only and is for the purpose of teaching those skilled in the art the general manner of carrying out the invention. It should be understood that the forms of the invention shown and described herein are presently preferred embodiments. It will be apparent to those skilled in the art after obtaining the benefit of the present invention that elements and materials may be substituted for those illustrated and described herein, and the parts and processes may be reversed, Some features of the invention may be utilized independently. Changes may be made in the elements described herein without departing from the spirit and scope of the invention as described in the appended claims.

Use the illumination subsystem to direct light to the sample, the detection subsystem to detect light scattered from the sample, and the detected light to detect sample features, defects, or light scattering properties 1 is a block diagram of an exemplary inspection system including a processor for FIG. 2 is a block diagram of an exemplary circuit included in the detection subsystem of FIG. 1 for detecting light scattered from a sample according to a first embodiment of the present invention. FIG. 3 is a block diagram of an exemplary circuit included in the detection subsystem of FIG. 1 for detecting light scattered from a sample according to a second embodiment of the present invention. FIG. 4 is a block diagram of an exemplary circuit included in the detection subsystem of FIG. 1 to detect light scattered from a sample according to a third embodiment of the present invention. 4 is a flow diagram of an exemplary method for inspecting a sample according to first, second, and third embodiments of the present invention. FIG. 6 is a block diagram of an exemplary circuit included in the detection subsystem of FIG. 1 for detecting light scattered from a sample according to a fourth embodiment of the present invention. 4B is a block diagram of exemplary electronics used for filtering, digitizing, and switching between signals generated by the circuits of FIGS. 4A and 4B. FIG. 6 is a flowchart of an exemplary method for inspecting a sample according to fourth and fifth embodiments of the present invention. FIG. 6 is a block diagram of another exemplary inspection system that includes means for controlling the amount of incident laser power delivered to a sample under observation. FIG. 8 is a block diagram of an exemplary embodiment of a laser power attenuator that may be included in the inspection system of FIG. FIG. 8 is a block diagram of an exemplary embodiment of a laser power control apparatus that may be included in the inspection system of FIG. FIG. 8 is a block diagram of another exemplary embodiment of a laser power control apparatus that may be included in the inspection system of FIG. FIG. 8 is a block diagram of yet another exemplary embodiment of a laser power control apparatus that may be included in the inspection system of FIG. 7 is a flow diagram of a method for dynamically changing the amount of incident laser power delivered to a sample under observation in order to reduce thermal breakage of large particles and extend the measurement detection range of the inspection system of FIG. is there.

Claims (89)

A method for inspecting a sample, comprising:
Directing light to the sample;
Detecting light scattered from the sample, comprising:
Monitoring the photomultiplier tube (PMT) anode current;
Detecting features, defects, or light scattering properties of the sample using the anode current until the anode current reaches a predetermined threshold, and then detecting using the dynode current of the PMT for the detection. And detecting.   Prior to the directing step, further comprising selecting the dynode current to be used for the detection by electrically connecting to one of a plurality of dynodes included in the PMT. The method of claim 1 comprising.   The method of claim 2, wherein the selected one of the plurality of dynodes is disposed approximately in the center of the plurality of dynodes.   Selecting an alternative dynode current to be used for the detection by making another electrical connection to a different one of the plurality of dynodes during or after the detection; The method of claim 2.   The method of claim 4, wherein the alternative dynode current provides a gain that is less than or greater than the initially selected dynode current. A circuit for detecting light scattered from a sample,
A photomultiplier tube (PMT) for receiving the light scattered from the sample and converting it into an electrical signal, wherein the PMT includes a cathode and a plurality of stages including a number of dynodes and an anode. A PMT in which the electrical signal is amplified at each successive stage to produce an output signal at the anode stage;
A circuit coupled to the PMT for receiving the output signal together with an intermediate signal representative of the amplified electrical signal at one of the dynode stages of the PMT. Said means is
Provide the output signal to an inspection system component as long as the output signal remains below a predetermined threshold, and supply the intermediate signal to the inspection system component when the output signal reaches the predetermined threshold The circuit according to claim 6, which is configured as follows.   8. The circuit of claim 7, wherein the predetermined threshold includes a maximum current level at which the anode stage begins to saturate, or a current level that is slightly less than a current level at which the anode stage begins to saturate.   The circuit of claim 8, wherein the predetermined threshold is in a range of values including about 1 μA to about 10 mA.   9. The circuit of claim 8, wherein the intermediate signal comprises the amplified electrical signal from the one dynode stage multiplied by a calibration ratio.   The circuit of claim 10, wherein the calibration ratio includes the current level at the anode stage divided by the amplified electrical signal from the one dynode stage.   Control logic configured to monitor the output signal while the means is supplying the output signal to the inspection system component and switch to the intermediate signal when the output signal reaches the predetermined threshold. The circuit of claim 7 comprising: Said means is
A first set of program instructions for monitoring the output signal while providing the output signal to the inspection system component;
8. The circuit of claim 7, including a processor configured to execute a second set of program instructions for switching to the intermediate signal when the output signal reaches the predetermined threshold.   The circuit of claim 7 further comprising a plurality of resistors coupled between the cathode and the anode of the PMT.   The plurality of resistors are coupled in parallel to a first power supply and in series with each other to form a voltage divider chain coupled in parallel to the cathode and the plurality of stages in the PMT. The circuit according to claim 14.   Most of the resistors are each in parallel between i) the cathode and the first dynode, ii) two adjacent odd and even dynodes, or iii) the last dynote and either of the anodes. 16. The circuit of claim 15, wherein two or more of the resistors are coupled in parallel between adjacent odd dynodes or adjacent even dynodes.   17. The circuit of claim 16, further comprising an intermediate resistor disposed between the adjacent odd dynodes or the adjacent even dynodes and an additional resistor coupled between a second power supply.   The circuit of claim 16, wherein the intermediate signal comprises a voltage signal measured across the additional resistor.   15. The circuit of claim 14, wherein the plurality of resistors are separated into a series-coupled first resistor portion and a series-coupled second resistor portion, each having an associated separate power supply. .   20. The circuit of claim 19, wherein the first resistor portion is coupled to the cathode and to one or more dynodes that include and terminate at the intermediate dynode of the PMT.   21. The circuit of claim 20, wherein the second resistor portion is coupled to a dynode immediately following the intermediate dynode and the anode of the PMT.   The circuit of claim 21, wherein the intermediate signal comprises a voltage signal measured across the last resistor in the first resistor portion. An illumination subsystem configured to direct light to the sample;
A detection subsystem configured to detect light scattered from the sample and generate a first signal and a second signal in response thereto, wherein the first and second signals are A detection subsystem comprising only one photodetector for generating;
The first signal is used to detect features, defects, or light scattering properties of the sample until the first signal reaches a predetermined threshold level, and then the second signal is used for the detection. An inspection system including a processor configured for use. In detecting the feature, defect, or light scattering property of the sample, the processor
24. The test of claim 23, configured to compare the first signal to the predetermined threshold level and to switch to the second signal when the first signal reaches the predetermined threshold level. system. The detection subsystem comprises:
A collector for collecting the light scattered from the sample;
A photodetector for converting the collected light into the first and second signals, wherein the second signal comprises only a fraction of the first signal;
24. The inspection system of claim 23, including an amplifier for amplifying the first and second signals by a substantially equivalent amount. The photodetector is
A photomultiplier tube (PMT) having a cathode and a plurality of dynodes and anodes;
A current-to-voltage converter for converting the anode current into a voltage to produce the first signal;
26. The inspection system of claim 25, including a load coupled to one of the plurality of dynodes for converting an intermediate dynode current to another voltage to produce the second signal. A circuit for detecting light scattered from the sample,
Only one photodetector configured to detect the light scattered from the sample and convert the light into an electrical signal;
An amplifier stage configured to generate at least a first signal and a second signal in response to the electrical signal, the first signal detecting a substantially lower level of the scattered light; And an amplifier stage having a higher resolution than the second signal.   28. The circuit of claim 27, wherein the only one photodetector is selected from the group consisting of a photodiode, a phototube, and a photomultiplier tube.   28. The circuit of claim 27, wherein the only one photodetector includes only one detection region or a plurality of detection regions arranged in an array.   28. The circuit of claim 27, wherein the amplifier stage includes a first operational amplifier and a second operational amplifier, each having a positive and negative input terminal.   The first and second operational amplifiers have voltages coupled to different nodes in a voltage divider network whose positive input terminals include at least a first resistor (R1) and a second resistor (R2). 32. The circuit of claim 30, comprising a follower.   The positive input terminal of the first operational amplifier is coupled to receive the electrical signal multiplied by a series of resistors [R1 + R2], and the positive input terminal of the second operational amplifier is 32. The circuit of claim 31, coupled to receive the electrical signal multiplied by a resistor of R2.   33. The value of claim 32, wherein the values of the first and second resistors are selected to provide a first signal having a relatively high gain and a second signal having a relatively low gain. circuit.   34. The circuit of claim 33, wherein the first and second resistors are selected to increase the detection range of the circuit by about 2 times to about 1024 times.   31. The circuit of claim 30, wherein the amplifier stage includes a third operational amplifier having positive and negative input terminals.   36. The circuit of claim 35, wherein the first operational amplifier is coupled to generate a contact voltage by multiplying the electrical signal by a first resistance value (R1).   The second operational amplifier is coupled to generate the first signal by multiplying the contact voltage by a third resistance value (R3) and dividing by a second resistance value (R2). Item 37. The circuit according to Item 36.   The third operational amplifier is coupled to generate the second signal by multiplying the contact voltage by a fifth resistance value (R5) and dividing by a fourth resistance value (R4). Item 38. The circuit according to Item 37.   The first, second, third, fourth, and fifth resistance values provide the first signal having a relatively high gain and the second signal having a relatively low gain. 40. The circuit of claim 38, which is selected.   40. The circuit of claim 39, wherein the first, second, third, fourth, and fifth resistance values are selected to increase the detection range of the circuit by about 2 times to about 1024 times. An illumination subsystem configured to direct light to the sample;
A single photodetector configured to detect light scattered from the sample and convert the light into an electrical signal;
An amplifier stage coupled to convert the electronic signal into a pair of analog signals, one having a relatively higher gain than the other.   42. The inspection system of claim 41, wherein the only one photodetector is selected from the group consisting of a photodiode, a phototube, and a photomultiplier tube.   42. The inspection system of claim 41, wherein the only one photodetector includes only one detection area or a plurality of detection areas arranged in an array.   42. The method of claim 41, further comprising a pair of analog to digital converters (ADC) coupled to convert the pair of analog signals into a pair of digital signals, one of which is a value greater than the other. The inspection system described.   The digital signal having the larger value is supplied to a signal processing subsystem until the ADC responsible for generating the larger digital signal reaches a predetermined threshold, and then the digital signal having the smaller value. 45. The inspection system of claim 44, further comprising selection logic coupled to provide a signal processing subsystem to the signal processing subsystem. 46. The inspection system of claim 45, wherein the predetermined threshold is substantially less than or equal to ^2N-1 , where N is the number of bits associated with each of the ADCs.   Configured to detect features, defects, or light scattering characteristics of the sample using the larger digital signal until the predetermined threshold is reached, and then use the smaller digital signal for the detection. The inspection system of claim 45, further comprising the signal processing subsystem. A method for inspecting a sample, comprising:
Directing light to the sample;
Detecting light scattered from the sample, comprising:
Using only one photodetector to detect the light scattered from the sample and convert the light into an electrical signal;
In response to the electrical signal, generating a first signal and a second signal, wherein the first signal is the second signal to detect a rather low level of the scattered light. A step having a higher resolution than
Use the first signal to detect a feature, defect, or light scattering property of the sample until the first signal reaches a predetermined threshold, and then use the second signal for the detection. And a step of detecting.   49. The method of claim 48, further comprising converting the first and second signals from analog signals to corresponding digital signals prior to the using step. During the using step,
Comparing the digital value of the first signal with a predetermined threshold;
50. The method of claim 49, further comprising switching to the second signal when the digital value of the first signal reaches the predetermined threshold. The predetermined threshold is substantially less than or equal to 2 ^N-1 , where N is the number of bits associated with the means used to convert the first and second signals from analog signals to corresponding digital signals. 51. The method of claim 50.   51. The method of claim 50, wherein a second signal having a lower resolution than the first signal is generated to detect a substantially higher level of the scattered light.   The lower level of light detected by the first signal is within a first detection range, and the higher level of light detected by a second signal is at least partially in the first detection. 53. The method of claim 52, wherein the method is within a second detection range that overlaps the range. An illumination subsystem for directing light to the sample at a first power level;
A detection subsystem for detecting light scattered from the sample;
A power attenuator subsystem for dynamically changing a power level directed to the sample based on the scattered light detected from the sample.   55. The inspection system of claim 54, wherein the power attenuator subsystem includes a light source for providing the light directed to the sample.   56. The inspection system of claim 55, wherein the light source is selected from the group of laser based sources including a diode laser, a helium neon laser, an argon laser, a solid state laser, a diode pumped solid state (DPSS) laser.   56. The inspection system of claim 55, wherein the light source is selected from the group of non-laser based sources including arc lamps, high pressure mercury lamps, low pressure mercury lamps, LED arrays, bulbs. The power attenuator subsystem includes a laser power attenuator disposed between the light source and the sample, and the laser power attenuator includes the sample if the detected scattered light remains below a predetermined threshold level; Maintaining the light directed at the first power level;
If the detected scattered light exceeds the predetermined threshold level, the light directed to the sample is reduced to a second power level that is less than the first power level, and the detected 56. The inspection system of claim 55, configured to increase the light directed to the sample back to the first power level when the scattered light recedes below the predetermined threshold level. The power attenuator subsystem further includes a laser power controller coupled between the detection subsystem and the laser power attenuator, wherein the laser power controller is configured to detect the scattered light at the predetermined threshold level. Continuously monitoring the detected scattered light to determine whether it is greater than or less than the predetermined threshold, and the light directed to the sample based on a determination by the laser power controller 59. The inspection system of claim 58, wherein the inspection system is configured to instruct the laser power attenuator to provide at the first power level or the second power level.   The predetermined threshold level is selected to minimize thermal damage that occurs when the light directed to the sample is absorbed or poorly dispersed by features on the sample, 60. The inspection system of claim 59, wherein the threshold level is based on an input laser power density associated with the onset of thermal failure applied on a feature of constant size. 61. The inspection system of claim 60, wherein the predetermined threshold level is selected from the group of incident laser power densities ranging from about 1 kW / cm < ² > to about 1000 kW / cm < ² >.   61. The inspection system of claim 60, wherein the second power level comprises a percentage of the first power level, and the percentage is selected from the group comprising about 1% to about 50%.   64. The inspection system of claim 62, wherein the second power level is substantially equivalent to about 10% of the first power level.   64. The inspection system of claim 62, wherein the inspection system is configured to detect relatively small size features by directing the light to the sample at the first power level.   The inspection system directs the light to the sample at the second power level instead of the first power level without causing thermal damage on relatively large size features. 65. The inspection system of claim 64 configured to detect features.   The power attenuator subsystem utilizes more than one predetermined threshold level to dynamically select an appropriate power level to be directed to the sample, and the appropriate power level is in use 61. The inspection system of claim 60, selected from three or more different power levels available for use. A circuit for dynamically controlling the power level of light delivered to the sample during a scan of the sample surface inspection;
Means for dynamically changing the power level of the light delivered to the sample;
Means for controlling the step of dynamically changing the power level based on a calculated amount of scattered light detected from the sample.   68. The means for dynamically changing the power level comprises a selectively transmissive optical component adapted to transmit a portion of the light based on a polarization of the light. The circuit described.   A wave plate and a polarizing beam splitter adapted to transmit a portion of the light based on the polarization of the light when combined with the means for dynamically changing the power level; 68. The circuit of claim 67 comprising: The means for dynamically changing the power level comprises:
An electro-optic crystal adapted to change the polarization of the light based on a variable control voltage supplied to the electro-optic crystal;
68. The circuit of claim 67, comprising a beam splitter adapted to transmit a portion of the light based on the polarization of the light output from the crystal. The means for controlling the step of dynamically changing comprises:
A photodetector for detecting the light scattered from the sample and converting the detected scattered light into an electrical output signal;
68. The circuit of claim 67 including a comparator for comparing the electrical output signal to a predetermined threshold level and generating a control signal in response thereto.   The means for controlling the step of dynamically changing further comprises a divider coupled between the photodetector and the comparator, the divider being a normalized version of the electrical output signal Wherein the normalized version corresponds to the calculated amount of scattered light detected from the sample, and the comparator represents the normalized version of the electrical output signal. 72. The circuit of claim 71, configured to generate the control signal in response to a predetermined threshold level.   The control signal generated by the comparator is supplied to the means for dynamically changing to select a power level to be supplied to the sample from a plurality of available power levels. 73. The circuit of claim 72.   The normalized version of the electrical output signal is generated by dividing the electrical output signal by i) the selected power level, and ii) a gain applied to the electrical output signal by the photodetector. 74. The circuit of claim 73, wherein:   75. The circuit of claim 74, wherein the control signal causes the sample to provide a first power level if the normalized version of the electrical output signal is less than the predetermined threshold level.   If the normalized version of the electrical output signal is greater than or equal to the predetermined threshold level, the control signal causes the sample to provide a second power level, and the second power level is substantially equal to 76. The circuit of claim 75, wherein the circuit is less than the first power level.   77. The circuit of claim 76, wherein the second power level comprises a percentage of the first power level and the percentage is selected from the group comprising about 1% to about 50%.   77. The circuit of claim 76, wherein the second power level is substantially equivalent to about 10% of the first power level.   The circuit of claim 72 further comprising a processor configured to execute one or more program instructions for detecting defects, features, or light scattering properties of the sample.   The processor is configured to receive the normalized version of the electrical output signal from the divider and to detect the defect, feature, or light scattering characteristic of the sample in response thereto. 79. The circuit according to 79.   80. The circuit of claim 79, further comprising an additional divider coupled to the photodetector to generate a second normalized version of the electrical output signal generated by the photodetector.   The processor receives the second normalized version of the electrical output signal from the additional divider and responsively detects the defect, feature, or light scattering characteristic of the sample. 82. The circuit of claim 81, configured as follows. A method for inspecting a sample, comprising:
Directing light to the sample at a first power level;
Scanning the light across the surface of the sample, wherein the method detects light scattered from the sample during the scanning;
If the detected light exceeds a predetermined threshold level, reducing the light directed to the sample to a second power level;
Using the detected light scattered from the sample to detect features, defects, or optical properties of the sample.   If the predetermined threshold level is exceeded, the method causes the light directed to the sample to be directed to the first if the detected scattered light recedes below the predetermined threshold level during the scan. The method of claim 83, further comprising increasing back to a power level.   Directing the light to the sample supplies the light to the surface of the sample by a power density distribution that reaches a peak near the middle of the distribution and tapers near an edge of the distribution, wherein the middle of the distribution is 85. The method of claim 84, referred to as a main beam and the end of the distribution referred to as a beam skirt.   86. The power density distribution includes a main lobe and at least one side lobe on each side of the main lobe, wherein the side lobe is less in power density than the main lobe. The method described in 1.   The method of claim 85, wherein the power density distribution comprises a Gaussian distribution.   Detecting the light allows the light directed to the sample before the main beam reaches a relatively large feature on the sample to be reduced to the second power level; 86. The method of claim 85, wherein the relatively large feature is susceptible to thermal failure when the main beam is delivered thereto at the first power level. During the scan, the method comprises:
Monitoring light scattered from the sample;
90. The method of claim 88, comprising using the monitored light to select the predetermined threshold level that is used to select an appropriate power level to be directed to the sample.

Images