CROSS REFERENCE TO RELATED APPLICATION

[0001] This application claims priority to U.S. Application No. 62/839,922, filed April 29, 2019 and titled METROLOGY APPARATUS AND METHOD USING MECHANICAL FILTER, and which is incorporated herein in its entirety by reference.

TECHNICAL FIELD

[0002] The disclosed subject matter relates to a metrology apparatus and method that uses a mechanical filter to discriminate between two types of light in an extreme ultraviolet light source.

BACKGROUND

[0003] In semiconductor lithography (or photolithography), the fabrication of an integrated circuit (IC) includes performing a variety of physical and chemical processes on a semiconductor (for example, silicon) substrate (which is also referred to as a wafer). A photolithography exposure apparatus or scanner is a machine that applies a desired pattern onto a target portion of the substrate. The wafer is irradiated by a light beam that extends along an axial direction, and the wafer is fixed to a stage so that the wafer generally extends along a lateral plane that is substantially orthogonal to the axial direction. The light beam can have a wavelength in the ultraviolet (UV) range, for example, from about 10 nanometers (nm) to about 400 nm, and specifically in either the deep UV (DUV) range, from about 100 nm about 400 nm or in the extreme ultraviolet (EUV) range, less than about 100 nm (or around 50 nm or less, and including 13 nm). The light beam travels along the axial direction (which that is orthogonal to the lateral plane along which the wafer extends).

[0004] Methods to produce EUV light include, but are not necessarily limited to, converting a material that has an element, for example, xenon, lithium, or tin, with an emission line in the EUV range in a plasma state. In one such method, often termed laser produced plasma (“LPP”), the required plasma can be produced by irradiating a target material, for example, in the form of a droplet, plate, tape, stream, or cluster of material, with an amplified light beam that can be referred to as a drive laser. For this process, the plasma is typically produced in a sealed vessel, for example, a vacuum chamber, and monitored using various types of metrology equipment. SUMMARY

[0005] In some general aspects, a metrology apparatus includes: a diagnostic apparatus configured to interact a diagnostic probe with a current target at a diagnostic region before the current target enters a target space; a detection apparatus; and a control system in communication with the detection apparatus. The detection apparatus includes: a light sensor having a field of view overlapping with the diagnostic region and configured to sense light produced from the interaction between the diagnostic probe and the current target at the diagnostic region; and a mechanical filter between the diagnostic region and the light sensor. The mechanical filter includes an optical beam reducer and an optical mask defining an aperture positioned between the optical beam reducer and the light sensor. The control system is configured to estimate a property of the current target based on the output from the light sensor.

[0006] Implementations can include one or more of the following features. For example, the mechanical filter can be configured to angularly separate diagnostic light emitted from the diagnostic region from non-diagnostic light emitted from the target space. The diagnostic light can be produced from an interaction between the current target and the diagnostic probe at the diagnostic region. The non-diagnostic light can include light emitted from a plasma produced by a previous target in the target space. A transverse extent of the aperture can be about the same as or greater than a transverse extent of the diagnostic light in the plane of the optical mask, and a transverse extent of the optical mask can be greater than or about the same as a transverse extent of the non-diagnostic light in the plane of the optical mask. The optical mask can be positioned so that the non-diagnostic light emitted from the target space is substantially blocked by the optical mask while the diagnostic light substantially passes through the aperture.

[0007] The mechanical filter can include an optical collimator between the diagnostic region and the optical beam reducer. The optical beam reducer can be an afocal beam reducer and can be configured, in conjunction with the optical collimator, to project a finite object to infinity. The optical collimator and components of the optical beam reducer that have positive focal length nearest the optical collimator can be integrated into a single refractive element. The optical beam reducer can be configured to maintain a collimated status of light. [0008] The light sensor can include one or more: photodiodes, photo-transistors, light-dependent resistors, photomultiplier tubes, multi-cell photoreceivers, quad-cell photoreceivers, and cameras.

[0009] The diagnostic probe can include at least one diagnostic light beam, and the light sensor is configured to sense diagnostic light produced from the interaction between the current target and the at least one diagnostic light beam. The diagnostic light can include the diagnostic light beam that is reflected from, scattered from, or blocked by the current target.

[0010] The detection apparatus can include one or more of a spectral filter and a polarization filter.

[0011] The diagnostic probe can include first and second diagnostic light beams, each configured to interact with the current target before it enters the target space, each interaction occurring at a distinct region and a distinct time.

[0012] The optical beam reducer can include a refracting telescope, a reflecting telescope, or a catadioptric telescope. The refracting telescope can include: a positive focal length lens arrangement and a negative focal length lens arrangement that are separated by the sum of their focal lengths; or a pair of positive focal length lens arrangements that are separated by the sum of their focal lengths.

[0013] The optical beam reducer can be configured to reduce a transverse size of impinging light by a factor of at least five, at least ten, at least twenty, or about ten.

[0014] The aperture can include a circular opening, an elliptical opening, a polygonal opening, or an elongated slit opening.

[0015] The detection apparatus can be positioned external to a chamber of an extreme ultraviolet (EUV) light source, the diagnostic region can be inside the chamber, and the detection apparatus can receive light from the chamber through an optical window in a wall of the chamber. A distance between the diagnostic region and the optical window can be about 200-500 times the size of a distance between the diagnostic region and the target space.

[0016] The detection apparatus can include a focusing lens at an output of the aperture, the focusing lens configured to focus the sensed light onto the light sensor.

[0017] The aperture can have an extent that is at least 2 millimeters (mm). The aperture can be positioned to receive diagnostic light at a location at which the diagnostic light is collimated or is non-converging and non-diverging. [0018] In another general aspect, a metrology method includes: interacting a diagnostic probe with a current target at a diagnostic region before the current target enters a target space;

collecting diagnostic light produced from the interaction between the diagnostic probe and the current target at the diagnostic region, the collecting including also collecting non-diagnostic light produced from the target space; collimating the diagnostic light and the non-diagnostic light; angularly separating the diagnostic light and the non-diagnostic light from each other including reducing a transverse extent of the diagnostic light and the non-diagnostic light;

sensing the diagnostic light at a sensing region transversely displaced from a non-sensing region traversed by the non-diagnostic light after the diagnostic light and the non-diagnostic light have been angularly separated; and estimating a property of the current target based on the sensed diagnostic light.

[0019] Implementations can include one or more of the following features. For example, the diagnostic probe can be interacted with the current target at the diagnostic region by interacting one or more diagnostic light beams with the current target at the diagnostic region; and diagnostic light can be collected by collecting one or more diagnostic light beams that have been reflected from, scattered from, or blocked by the current target at the diagnostic region.

[0020] The metrology method can further include filtering the diagnostic light based on one or more of its spectral properties and its polarization state.

[0021] The diagnostic region can be inside a hermetically-sealed chamber of an extreme ultraviolet (EUV) light source and collecting the diagnostic light including also collecting non diagnostic light can include receiving the diagnostic light including the non-diagnostic light transmitted through an optical window in a wall of the chamber.

[0022] The transverse extent of the diagnostic light and the non-diagnostic light can be reduced by reducing the transverse extent of the diagnostic light and the non-diagnostic light by at least a factor of five, at least ten, at least twenty, or by a factor of about ten.

[0023] The metrology method can also include blocking or redirecting the non-diagnostic light at the non-sensing region. The transverse extent of the diagnostic light and the non-diagnostic light can be reduced by one or more of: refracting the light and reflecting the light.

[0024] The metrology method can also include focusing the diagnostic light at the sensing region. [0025] The transverse extent of the diagnostic light and the non-diagnostic light can be reduced by maintaining a collimation state of the diagnostic light and the non-diagnostic light.

[0026] The metrology method can also include passing the diagnostic light through an aperture of an optical mask after the diagnostic light and the non-diagnostic light are angularly separated from each other, the aperture having an extent that is greater than an extent of the diagnostic light, and before the diagnostic light is sensed.

DESCRIPTION OF DRAWINGS

[0027] Fig. 1 is a schematic illustration of a metrology apparatus including a detection apparatus that has a mechanical filter for collecting diagnostic and non-diagnostic light produced within an environment, the diagnostic light produced from a target in a diagnostic region and the non diagnostic light produced from a target in a target space distinct from the diagnostic region;

[0028] Fig. 2A is a schematic illustration of an implementation of a mask that can be used in the mechanical filter of Fig. 1 ;

[0029] Fig. 2B is a schematic illustration of an implementation of a mask that can be used in the mechanical filter of Fig. 1 ;

[0030] Fig. 3 is a schematic illustration of an implementation of the detection apparatus of Fig. 1 that includes an optical beam reducer that is designed as a refracting Galilean telescope;

[0031] Fig. 4 is a schematic illustration of an implementation of the detection apparatus of Fig. 1 that includes an optical beam reducer that is designed as a refracting Keplerian telescope;

[0032] Fig. 5 is a schematic illustration of an implementation of the metrology apparatus of Fig. 1, in which the detection apparatus includes one or more spectral filters and polarization filters;

[0033] Fig. 6A is a schematic illustration and block diagram of an implementation of a diagnostic apparatus producing a single diagnostic light beam;

[0034] Fig. 6B is a schematic illustration and block diagram of an implementation of a diagnostic apparatus producing two diagnostic light beams from a single light source;

[0035] Fig. 6C is a schematic illustration and block diagram of an implementation of a diagnostic apparatus producing two diagnostic light beams from respective light sources;

[0036] Fig. 7 is a schematic illustration of an implementation of the metrology apparatus of Fig. 1 implemented in an extreme ultraviolet (EUV) light source; [0037] Fig. 8 is a block diagram of an implementation of a control apparatus of the metrology apparatus of Fig. 1 ;

[0038] Fig. 9 is a schematic illustration of an implementation of a lithography apparatus that receives EUV light output from the EUV light source of Fig. 7 ;

[0039] Fig. 10 is a flow chart of a procedure performed by the metrology apparatus of Fig. 1 to estimate one or more properties of a current target;

[0040] Fig. 11 is a schematic illustration of an implementation of the detection apparatus of Fig.

1 that includes an optical beam reducer that is designed as a refracting Galilean telescope and in which an optical collimator and the positive focal length lens of the optical beam reducer are integrated into a single refractive element;

[0041] Fig. 12 is a schematic illustration of an implementation of the metrology apparatus of Fig. 1, in which the diagnostic probe provides a back illumination for the target so that a shadow of the target is formed from the target obscuring at least a portion of the diagnostic probe;

[0042] Fig. 13 is a schematic illustration of an implementation of the detection apparatus of Fig.

1 that includes an optical beam reducer that is designed as a reflecting off-axis telescope;

[0043] Fig. 14 is a schematic illustration of an implementation of the detection apparatus of Fig.

1 that includes an optical beam reducer that is designed as a catadioptric off-axis telescope; and [0044] Fig. 15 is a schematic illustration of an implementation of the metrology apparatus of Fig. 1, in which the mechanical filter includes a mask that blocks the diagnostic light while permitting the non-diagnostic light to pass through to a sensor.

[0045] Applicant notes that the drawings may not be to scale. For example, distances between optical elements in the schematic illustrations may be greater or less than what is shown.

DESCRIPTION

[0046] Referring to Fig. 1, a metrology apparatus 100 is configured for estimating one or more properties of a current target 105c that is traveling along a trajectory TR toward a target space 110 within an environment 115. The metrology apparatus 100 is configured to estimate at least one property (such as a speed, a location, a velocity, a direction) of the current target 105c by analyzing light (referred to as diagnostic light 120) that is produced due to an interaction between one or more diagnostic probes 125 and the current target 105c before the current target 105c enters the target space 110. However, non-diagnostic light 122 is produced at the same time as and in the same vicinity as the diagnostic light 120. This non-diagnostic light 122 can saturate a light sensor 130 within the metrology apparatus 100 or can interfere with operation of the light sensor 130 in the metrology apparatus 100. Because of this, the non-diagnostic light 122 can reduce an accuracy of the analysis performed by the metrology apparatus 100 and therefore cause errors in the estimated properties of the current target 105c.

[0047] The metrology apparatus 100 is able to more effectively and accurately estimate a property of the current target 105c because it is able to more effectively discriminate between the diagnostic light 120 and the non-diagnostic light 122. To this end, the metrology apparatus 100 includes a detection apparatus 135, which includes the light sensor 130, and a mechanical filter 140 between a diagnostic region 145 (at which the current target 105c interacts with the diagnostic probe 125) and the light sensor 130. The mechanical filter 140 includes an optical beam reducer 150 and a light-opaque optical mask 155 defining a light-transparent aperture 160. The aperture 160 is positioned between the optical beam reducer 150 and the light sensor 130. In order to properly sense the diagnostic light 120, the light sensor 130 is positioned so that its field of view overlaps with the diagnostic region 145. The metrology apparatus 100 also includes a diagnostic apparatus 165 that produces the diagnostic probe 125 and a control system 170 in communication with the detection apparatus 135. The control system 170 receives the output from the sensor 130 and performs an analysis on this output to estimate the one or more properties of the current target 105c.

[0048] The mechanical filter 140 includes an optical collimator 142 that forms respective collimated light beams from the diagnostic light 120 and the non-diagnostic light 122, and then the optical beam reducer 150 optically reduces a size of these collimated light beams to form reduced-sized collimated light beams 121, 123, respectively. The optical beam reducer 150 increases an angular separation between images produced by the collimated diagnostic light beam 121 and the collimated non-diagnostic light beam 123. The non-diagnostic light 122 originates from a location outside the location of the current target 105c (that produces the diagnostic light 120). For example, the diagnostic light 120 originates from the diagnostic region 145 while the non-diagnostic light 122 originates outside the diagnostic region 145 such as from the target space 110. Because of this, the non-diagnostic light 122 enters the mechanical filter 140 at a slightly different angle from the angle at which the diagnostic light 120 enters the mechanical filter 140. This fact can be exploited in the design of the mechanical filter 140, which can further separate the images arising from the diagnostic light 120 and the non-diagnostic light 122 by increasing the angular separation between the respective collimated light beams 121, 123. The increase in angular separation permits greater discrimination between the two images at the aperture 160 after the light beams have progressed a length of an optical path 152 between the optical beam reducer 150 and the mask 155. Thus, in this implementation, the mask 155 is placed so that the aperture 160 permits the passage of the diagnostic light beam 121 (formed from the diagnostic light 120) to the sensor 130 while the mask 155 blocks the passage of the non-diagnostic light beam 123 (formed from the non-diagnostic light 122) to the sensor 130.

[0049] The optical collimator 142 forms the respective collimated light beams for input to the optical beam reducer 150. A collimated light beam is a beam that has a beam divergence that is low enough so that the beam radius does not undergo significant changes within moderate propagation distances. In this case, in the absence of any additional beam shaping (thus, in the absence of the optical beam reducer 150), the beam radius of each collimated light beam output from the optical collimator 142 would not undergo a significant change within a distance that extends to the sensor 130. For example, the beam radius of a collimated light beam output from the optical collimator 142 changes less than 1%, less than 5%, or less than 10% over a distance along the Z direction to the sensor 130 (in the absence of any intermediate optical elements). In some examples, a distance between the optical collimator 142 and the sensor 130 along the Z direction is on the order of one meter, but it can be shorter or longer, depending on the design of the optical beam reducer 150.

[0050] The optical beam reducer 150 is an afocal system (that is, a system without a focus), which means that the optical beam reducer 150 produces no net convergence or divergence of the collimated light beams that are input to the optical beam reducer 150. That is, the optical beam reducer 150 can be considered to have an infinite effective focal length that preserves or maintains a collimated status of the light beams output from the optical collimator 142. This type of system can be created with a pair of optical elements where the distance d between the elements is equal to the sum of each element’s focal length fl, f2 (that is, d = fl+f2). Although the afocal system does not alter the divergence of the collimated light beams, it does alter the width of the beam, increasing or decreasing its magnification. Overall, the optical beam reducer 150 can reduce the transverse size (that is, the size in the XsYs plane) of the collimated light beams output from the optical collimator 142 by at least a factor of five, at least a factor of ten, or at least a factor of twenty. In some implementations, the optical beam reducer 150 reduces the transverse size of the collimated light beams by a factor of about 10 or about 20.

[0051] The optical beam reducer 150 can be, for example, a refracting telescope, a reflecting telescope, or a catadioptric telescope.

[0052] A refracting telescope uses refracting optics such as lenses or prisms to form an image of the respective collimated light beams at the plane of the mask 155. In some implementations, the refracting telescope is a Galilean telescope that includes a positive focal length lens arrangement and a negative focal length lens arrangement that are separated by the sum of their focal lengths. In other implementations, the refracting telescope is a Keplerian telescope that includes a pair of positive focal length lens arrangements that are separated by the sum of their focal lengths.

Examples of refracting telescopes for use as an optical beam reducer 150 are discussed below with reference to Figs. 3 and 4.

[0053] A reflecting telescope includes a single or a combination of curved mirrors that reflect the collimated light beams output from the optical collimator 142 and form respective images at the plane of the mask 155. For example, in some implementations, the reflecting telescope is a Gregorian, a Newtonian, or a Cassegrain (and its variants) telescope. In other implementations, the reflecting telescope is an off-axis design such as a Herschelian or Schiefspiegler (which is a variant of the Cassegrain). An example of a reflecting telescope is shown and described with respect to Fig. 13.

[0054] A catadioptric telescope is one in which refraction and reflection are combined in an optical system, usually implementing lenses (that is, dioptrics) and curved mirrors (that is, catoptrics). For example, in some implementations, a catadioptric telescope includes Schmidt- Cassegrain telescopes and Maksutov-Cassegrain telescopes. In other implementations, a catadioptric telescope is a catadioptric variant of a Herschelian telescope, which uses both lenses and a mirror or an off-axis variant of a Stevick-Paul telescope. An example of a catadioptric telescope is shown and described with respect to Fig. 14.

[0055] The environment 115 can be a vacuum environment within a chamber of an extreme ultraviolet (EUV) light source, such as the EUV light source discussed below with reference to Fig. 7. In some implementations, the detection apparatus 135 is placed external to the chamber of the EUV light source while the diagnostic region 145 is within the chamber, and the detection apparatus 135 receives the diagnostic light 120 and the non-diagnostic light 122 through an optical window in a wall of the chamber. The optical window is transparent to the wavelength of the diagnostic light 120. This is discussed with reference to Fig. 7, below.

[0056] As discussed with reference to Fig. 7, the EUV light source supplies EUV light to an output apparatus, which can be a lithography apparatus. EUV light is formed in the environment 115 by converting targets 105 as they reach the target space 110 into plasma that emits the EUV light, and this EUV light is collected and transmitted to the lithography apparatus. The targets 105 that reach the target space 110 are converted by interacting the targets 105 with radiation pulses in the target space 110, the radiation pulses providing enough energy to the targets 105 to convert them to plasma. A continuous stream 106 of targets (each generally designated as 105) is directed from a target supply apparatus 175 along the trajectory TR toward the target space 110.

[0057] The trajectory TR extends along a direction that can be considered as a target (or axial) direction, which lies in a three-dimensional X, Y, Z coordinate system that is defined by a physical aspect of the environment 115. Thus, the C,U,Z coordinate system can be defined by a wall or spot of the chamber that defines the environment 115. The axial direction of each target 105 generally has a component that is parallel with the -X direction of the coordinate system of the environment 115. The axial direction of each target 105 also can have components along one or more of the directions Y and Z that are perpendicular to the -X direction. Additionally, each target 105 released by the target supply apparatus 175 can have a slightly different actual trajectory and the trajectory depends at least in part on the physical properties of the target supply apparatus 175 at the time of release of the target 105 as well as the environment 115.

[0058] On the other hand, the detection apparatus 135 defines a local three-dimensional Xs, Ys, Zs coordinate system and this local coordinate system can be defined by an image plane of the sensor 130.

[0059] Each target 105 includes a component that emits EUV light when converted to plasma. These targets 105 travel (for example, ballistically) from the region of generation (such as from the target supply apparatus 175) toward the target space 110. The properties (such as the speed, location, velocity, direction, arrival, or motion) of the current target 105c are estimated by the metrology apparatus 100 by probing the current target 105c as it travels along the trajectory TR with the diagnostic probe 125 produced by the diagnostic system 165, detecting or sensing the diagnostic light 120 produced from the interaction between the diagnostic probe 125 and the current target 105c, and analyzing the detected diagnostic light 120. [0060] As mentioned above, non-diagnostic light 122 can be present as well, and this non diagnostic light 122 can interfere with an accurate estimation of the properties of the current target 105c. This non-diagnostic light 122 can include broadband optical radiation that is emitted by plasma that is produced by one or more previous targets 105p that entered the target space 110 before or while the current target 105c is interacting with the diagnostic probe 125.

Moreover, the intensity of the non-diagnostic light 122 can be much greater than the intensity of the diagnostic light 120.

[0061] The non-diagnostic light 122 can include, for example, EUV light emitted from the plasma of previous targets 105p, light having a wavelength range that overlaps with the wavelength of the diagnostic light 120, and/or any light present and having a wavelength range that includes the range of the wavelength that can be detected by the sensor 130.

[0062] On the other hand, the diagnostic light 120 is produced from the interaction between the diagnostic probe 125 and the current target 105c and the diagnostic light 120 has a spectral bandwidth that is substantially narrower than the spectral bandwidth of the non-diagnostic light 122. For example, the spectral bandwidth of the diagnostic light 120 can be several hundred times lower than the overall spectral bandwidth of the non-diagnostic light 122. In some implementations, such as shown in Figs. 6A-6C, the diagnostic light 120 is produced from a portion of the diagnostic probe 125 reflecting off or scattered from the current target 105c.

[0063] In general, the sensor 130 can include one or more of a photodiode, a photo-transistor, a light-dependent resistor, and a photomultiplier tube. In other implementations, the sensor 130 includes one or more thermal detectors such as a pyroelectric detector, a bolometer, or a calibrated charged coupled device (CCD) or CMOS. In other implementations, the sensor 130 includes multi-cell photoreceivers, quad-cell photoreceivers, or cameras.

[0064] As shown in the implementation of Fig. 1, the optical mask 155 is positioned so that the collimated non-diagnostic light beam 123 (produced from the non-diagnostic light 122) is substantially or mostly blocked while the collimated diagnostic light beam 121 substantially passes through the aperture 160. Another implementation in which the collimated non-diagnostic light beam 123 substantially passes through the aperture while the collimated diagnostic light beam 121 is substantially blocked by the aperture is shown and discussed with respect to Fig. 15.

[0065] Referring to Fig. 2A, in some implementations, the mask 155 is a mask 255A that defines an aperture 260A having a circular shape in the XsYs plane. In other implementations, such as shown in Fig. 2B, the mask 155 is a mask 255B that defines an aperture 260B having a shape of a slit, which is not rotationally symmetric and has an extent along the Ys direction that is greater than an extent along the Xs direction. The design of Fig. 2B can be useful in circumstances in which the collimated diagnostic light beam 121 is moving, oscillating, or being perturbed such that its image plane moves along the Ys direction. The extent along the Ys direction

accommodates this fluctuation in the image plane of the collimated diagnostic light beam 121. The mask 155 can be configured to define other shapes of the aperture 160 in the image plane (the XsYs plane) such as elliptical apertures and polygonal openings.

[0066] In order to properly permit the collimated diagnostic light beam 121 to pass to the sensor 130, the aperture 160 along the XsYs plane is at least as large as a transverse extent of the collimated diagnostic light beam 121 in the XsYs plane. Additionally, in order to properly block the collimated non-diagnostic light beam 123, the optical mask 155 should have an extent along the XsYs plane that is at least as large as a transverse extent of the collimated non-diagnostic light beam 123 in the XsYs plane. This means that, with reference to Fig. 2 A, an extent 261 A of the aperture 260A in the XsYs plane is as large as the transverse extent of the collimated diagnostic light beam 121 in the XsYs plane and an extent 256A of the mask 255 A is large enough to block the full extent of the collimated non-diagnostic light beam 123 in the plane XsYs. As another example, with reference to Fig. 2B, a shortest extent 261B of the aperture 260B in the XsYs plane is as large as the transverse extent of the collimated diagnostic light beam 121 in the XsYs plane and an extent 256B of the mask 255B is large enough to block the full extent of the collimated non-diagnostic light beam 123 in the plane XsYs.

[0067] In some implementations, an extent 261 A, 26 IB of the respective aperture 260A, 260B in the XsYs plane is at least 2 millimeters (mm) or about 4 mm. An extent of the collimated diagnostic light beam 121 in the XsYs plane at the aperture 260A or 260B is about 3 mm. In some implementations, an extent 256A, 256B of the respective mask 255A, 255B is greater than 3 mm.

[0068] Unlike a spatial filter, in which light is focused at the aperture of a mask, in the mechanical filter 140, the light passing through the aperture 160 (such as aperture 260A or 260B) (the diagnostic light beam 121) is collimated and therefore has a larger transverse extent than light that is focused at an aperture of a spatial filter. Accordingly, the size of the aperture 160 (such as aperture 260A or 260B) in the XsYs plane can be much larger than an aperture that is used in a spatial filter to block focused non-diagnostic light and to pass focused diagnostic light. Because of this, unwanted particles (such as dirt in the environment 115) have much less impact on the performance of the aperture 160 (such as aperture 260A or 260B) than such particles would have for an aperture of a spatial filter. For example, the size of the aperture 160 in the XsYs plane is on the order of millimeters, and is much larger than a size of unwanted particles. On the other hand, a typical size of an aperture of a spatial filter can be a fraction of a millimeter (for example, 100 pm in extent) and an unwanted particle can have a comparable size. Because of this, interference between unwanted particles and the collimated light beam 121 in the mechanical filter 140 is reduced when compared with spatial filters.

[0069] Additionally, it is easier to direct the collimated diagnostic light beam 121 to pass through the aperture 160 (260A, 260B) because the tolerance in a relative positioning between the collimated diagnostic light beam 121 and a 4 mm aperture 160 (or 260A, 260B) is on the order of 0.2-0.4 mm. On the other hand, the tolerance in a relative positioning between the collimated diagnostic light beam 121 and a 100 pm aperture of a spatial filter is on the order of 5-10 pm.

[0070] Referring to Fig. 3, an implementation 335 of the detection apparatus 135 is shown. In this illustration, the coordinate system XsYsZs of the detection apparatus 335 is such that Ys is out of the page and Zs extends perpendicularly to the imaging region of the sensor 330. The detection apparatus 335 includes an optical collimator 342 that produces, for each of the diagnostic light 120 and the non-diagnostic light 122, a respective collimated light beam for input to the optical beam reducer 350. As discussed above, in the absence of the optical beam reducer 350, the beam radius of each collimated light beam output from the optical collimator 342 would not undergo a significant change within a distance that extends from the optical collimator 342 to the sensor 330. The optical collimator 342 in this example is a doublet lens (two lenses 342a, 342b). The focal length or curvature radius of the doublet lens 342 is chosen such that the originally curved wavefronts of the diagnostic light 120 and the non-diagnostic light 122 become flat or substantially flat for at least a distance that extends the length to the sensor 330.

[0071] In this implementation, the optical beam reducer 350 is designed as a Galilean-type refracting telescope. The optical beam reducer 350 optically reduces a size of the collimated light beams output from the optical collimator 342 to form the reduced-sized collimated light beams 121, 123, respectively. The optical beam reducer 350 includes a positive focal length lens arrangement 351 (which can be a convergent lens arrangement that includes one or more of a positive bi-convex lens, plano-convex lens, or meniscus lens) at the input side. The optical beam reducer 350 includes a negative focal length lens arrangement 353 (which can be a divergent lens arrangement that includes a concave lens) at the output side. The positive focal length lens arrangement 351 and the negative focal length lens arrangement 353 are separated by the sum of their focal lengths. The convergent lens arrangement 351 in this example is a compound lens (having convex lens 351a and concave lens 351b), which can help to correct distortions in the light beams. The optical beam reducer 350 lacks an intermediary focus (there is no focus between the convergent lens arrangement 351 and the divergent lens arrangement 353). Though not required, in this implementation, the divergent lens arrangement 353 includes a secondary lens 354 in between the convergent lens 351 and the divergent lens 353. The secondary lens 354 can be used in combination with the divergent lens arrangement 353 to collimate the beam from the positive focal length lens arrangement 351. The secondary lens 354 can be a positive focal length lens such as a meniscus lens (as shown), a bi-convex lens, or a plano-convex lens.

Overall, the optical beam reducer 350 can reduce the transverse size (that is, the size in the XsYs plane) of the collimated light beams output from the optical collimator 342 by at least a factor of five, at least a factor of ten, or at least a factor of twenty.

[0072] The optical beam reducer 350 outputs the reduced collimated diagnostic light beam 121 and the reduced collimated non-diagnostic light beam 123, which then progress a length of an optical path 352 to the mask 355. The longer the optical path 352, the greater is the separation between the respective images from these light beams 121, 123 at the mask 355. In this implementation, the mask 355 is placed so that the aperture 360 permits the passage of the diagnostic light beam 121 (formed from the diagnostic light 120) to the sensor 330 while the mask 355 blocks the passage of the non-diagnostic light beam 123 (formed from the non diagnostic light 122) to the sensor 330. The collimated diagnostic light beam 121 that has passed through the aperture 360 can be focused by way of a convergent lens 357 onto the imaging region of the sensor 330.

[0073] Referring to Fig. 4, another implementation 435 of the detection apparatus 135 is shown. In this illustration, the coordinate system XsYsZs of the detection apparatus 435 aligns with the page. The detection apparatus 435 includes an optical collimator 442 that produces, for each of the diagnostic light 120 and the non-diagnostic light 122, a respective collimated light beam for input to the optical beam reducer 450 of the detection apparatus 435. As discussed above, the beam radius of each collimated light beam output from the optical collimator 442 does not undergo a significant change within a distance that extends to the sensor 430. The optical collimator 442 in this example is similar to the optical collimator 342 and includes a doublet lens (two lenses 442a, 442b). The focal length or curvature radius of the doublet lens 442 is chosen such that the originally curved wavefronts of the diagnostic light 120 and the non-diagnostic light 122 become flat or substantially flat for at least a distance that extends the length to the sensor 430.

[0074] In this implementation, the optical beam reducer 450 is designed as a Keplerian-type refracting telescope. The optical beam reducer 450 optically reduces a size of the collimated light beams output from the optical collimator 442 to form the reduced-sized collimated light beams 121, 123, respectively. The optical beam reducer 450 includes an input positive focal length lens arrangement 451 (which can be a convergent lens arrangement that includes one or more of a bi convex lens, a plano-convex lens, or a meniscus lens) at the input side. The optical beam reducer 450 includes an output positive focal length lens arrangement 453 (which can be a convergent lens arrangement that includes one or more of a bi-convex lens, a plano-convex lens, or a meniscus lens) at the output side. For example, the positive focal length lens arrangement 453 is shown as an aspheric lens element in Fig. 4. It is possible for the positive focal length lens arrangement 453 to be a compound lens group. The positive focal length lens arrangements 451, 453 are separated by the sum of their focal lengths. The convergent lens arrangement 451 in this example is a compound lens (having convex lens 451a and concave lens 451b), which can help to correct distortions in the light beams. An intermediary focus IF (or intermediary focal plane IF) is between the input convergent lens arrangement 451 and the output convergent lens arrangement 453. Overall, the optical beam reducer 450 can reduce the transverse size (that is, the size in the XsYs plane) of the collimated light beams output from the optical collimator 442 by at least a factor of five, at least a factor of ten, or at least a factor of twenty. The extent of the optical beam reducer 450 along the Zs direction tends to be longer than the extent of the optical beam reducer 350 along the Zs direction, and thus, space requirements can determine which design of optical beam reducer 350 or 450 is more suitable. Moreover, the optical beam reducer 350 may be a more suitable design if the power of the diagnostic light 120 or the non-diagnostic light 122 is too high to impose an intermediary focus IF (such as present in the optical beam reducer 450).

[0075] The optical beam reducer 450 outputs the reduced collimated diagnostic light beam 121 and the reduced collimated non-diagnostic light beam 123, which then progress a length of an optical path 452 to the mask 455. The longer the optical path 452, the greater the separation between the respective images from these light beams 121, 123 at the mask 455. In this implementation, the mask 455 is placed so that the aperture 460 permits the passage of the diagnostic light beam 121 (formed from the diagnostic light 120) to the sensor 430 while the mask 455 blocks the passage of the non-diagnostic light beam 123 (formed from the non diagnostic light 122) to the sensor 430. The collimated diagnostic light beam 121 that has passed through the aperture 460 can be focused by way of a convergent lens 457 onto the imaging region of the sensor 430.

[0076] Referring to Fig. 5, in other implementations, the detection apparatus 135 is a detection apparatus 535 that also includes one or more spectral filters 543 and polarization filters 544, which can be arranged serially or in parallel with the mechanical filter 140. A spectral filter 543 is an optical filter that passes light in a particular range of wavelengths, such as a bandpass filter. A polarization filter 544 is an optical filter that passes light having a particular polarization. For example, the diagnostic light 120 may have a distinct polarization that is dependent on the polarization of the diagnostic probe 125 while the non-diagnostic light 122 can be unpolarized. Thus, a polarization filter can select the polarization of the diagnostic light 120 to pass.

[0077] Referring to Fig. 6A, in some implementations, the diagnostic apparatus 165 is designed as a diagnostic apparatus 665A. The diagnostic apparatus 665A produces, as the one or more diagnostic probes 125, a single probe light beam 625 A from a light source 626A. The probe light beam 625A is directed as a light curtain to cross the trajectory TR at a position x so that each of the targets 105 passes through the light curtain on their way to the target space 110. The light source 626A produces a single light beam 611 A, which is directed through one or more optical elements 627A (such as mirrors, lenses, apertures, and/or filters) that modify the light beam 611 A to form the single probe light beam 625 A.

[0078] The light source 626A can be a solid-state laser such as a YAG laser, which can be a neodymium-doped YAG (Nd:YAG) laser operating at 1070 nm and at 50 W power. In this example, when the current target 105c passes through the probe light beam 625 A at time t, at least some of the probe light beam 625 A is reflected or scattered from the current target 105c to form the diagnostic light 620A, which is detected by the detection apparatus 135. The control system 170 uses the information from the sensor 130 to estimate a moving property of the current target 105, which can be used to estimate the arrival time of a present target (which can be the current target 105c or a subsequent target) at the target space 110. This estimation can be used to adjust characteristics of a radiation pulse that is directed to the target space 110 to ensure that the radiation pulse interacts with the present target in the target space 110. The control system 170 can also rely on some assumptions about the path of the present target to perform the calculations to estimate the arrival time of the present target at the target space 110.

[0079] The probe light beam 625A can be a Gaussian beam so that its transverse profile of the optical intensity can be described with a Gaussian function. The focus or beam waist of the probe light beam 625A can be configured to overlap at the trajectory TR or the -X direction. Moreover, optical elements 627A can include refractive optics that ensure that the focus (or beam waist) of the probe light beam 625A overlaps the trajectory TR.

[0080] Referring to Fig. 6B, in some implementations, the diagnostic apparatus 165 is designed as a diagnostic apparatus 665B. The diagnostic apparatus 665B produces, as the one or more diagnostic probes 125, two probe light beams 625B_1 and 625B_2. The probe light beam

625B_1 is directed as a first light curtain to cross the trajectory TR at a first location (for example, the location xl along the X axis) so that each of the targets 105 pass through the first light curtain on their way to the target space 110. The probe light beam 625B_2 is directed as a second light curtain to cross the trajectory TR at a second location (for example, the location x2 along the X axis) so that each of the targets 105 pass through the second light curtain on their way to the target space 110 and after having already passed through the first light curtain. The probe light beams 625B_1, 625B_2 are separated by a distance Ad that is equal to x2-xl at the trajectory TR. This dual-curtain diagnostic apparatus 665B can be used to determine not only location and arrival information of the target 105 but also can be used to determine the speed or velocity of the target 105.

[0081] In some implementations, the diagnostic apparatus 665B includes a single light source 626B that produces a single light beam 61 IB and one or more optical elements 627B that receive the single light beam and split the light beam 61 IB into the two probe light beams 625B_1, 625B_2. Additionally, the optical elements 627B can include components for directing the probe light beams 625B_1, 625B_2 toward the respective locations xl, x2 along the trajectory TR.

[0082] In some implementations, the optical components 627B include a beam splitter that splits the single light beam from the single light source 626B into the two probe light beams 625B_1, 625B_2. For example, the beam splitter can be a dielectric mirror, a beam splitter cube, or a polarizing beam splitter. One or more of the optical components 627B can be reflective optics placed to redirect either or both of the probe light beams 625B_1, 625B_2 so that both probe light beams 625B_1, 625B_2 are directed toward the trajectory TR.

[0083] In other implementations, the optical components 627B include a splitting optic (such as a diffractive optic or a binary phase diffraction grating, a birefringent crystal, an intensity beam splitter, a polarization beam splitter, or a dichroic beam splitter) and a refractive optic such as a focusing lens. The light beam 61 IB is directed through the splitting optic, which splits the light beam 61 IB into two light beams, which travel along distinct directions and are directed through the refractive optic to produce the probe light beams 625B_1, 625B_2. The splitting optic can split the light beam 61 IB so that the probe light beams 625B_1, 625B_2 are separated by a set distance (for example, 0.65 mm along the X direction) at the trajectory TR. In this example, x2-xl=0.65 mm. Moreover, the refractive optic can ensure that the foci (or beam waist) of each of the probe light beams 625B_1, 625B_2 overlaps the trajectory TR.

[0084] As shown in this example, the probe light beams 625B_1, 625B_2 are directed so that they intersect the trajectory TR at different locations xl, x2, but generally intersect at substantially similar angles relative to the X axis. For example, the probe light beams 625B_1, 625B_2 are directed at about 90° relative to the X axis. In other implementations, it is possible to use the splitting optic and the refractive optic to adjust the angle at which the probe light beams 625B_1, 625B_2 are directed relative to the X axis so that they fan out toward the trajectory TR and intersect the trajectory TR at different and distinct angles. For example, the probe light beam 625B_1 can intersect the trajectory TR at approximately 90° relative to the -X direction while the probe light beam 625B_2 can intersect the trajectory TR at an angle that is less than 90° relative to the -X direction.

[0085] Each of the probe light beams 625B_1, 625B_2 can be a Gaussian beam so that the transverse profile of the optical intensity of each probe light beam 625B_1, 625B_2 can be described with a Gaussian function. The focus or beam waist of each probe light beam 625B_1, 625B_2 can be configured to overlap at the trajectory TR or the -X direction.

[0086] The light source 626B can be a solid-state laser such as a YAG laser, which can be a neodymium-doped YAG (Nd:YAG) laser operating at 1070 nm and at 50 W power. In this example, the current target 105c passes through the first probe light beam 625B_1 at time tl (and location xl), and at least some of the probe light beam 625B_1 is reflected or scattered from the current target 105c to form diagnostic light 620B_1, which is detected by the detection apparatus 135 (by way of the mechanical filter 140). Additionally, the current target 105c passes through the second probe light beam 625B_2 at time t2 (and location x2), at least some of the probe light beam 625B_2 is reflected or scattered from the current target 105c to form the light 620B_2, which is detected by the detection apparatus 135 (by way of the mechanical filter 140).

[0087] The separation Ad between the probe light beams 625B_1, 625B_2 at the trajectory TR can be adjusted or customized depending on the rate at which the targets 105 are released from the target supply apparatus 175 as well as the size and material of the targets 105. For example, separation Ad can be less than the spacing between adjacent targets 105. As another example, the separation Ad can be determined or set based on the spacing between adjacent targets 105 to provide for greater precision in the measurements that are performed based on the interactions between the probe light beams 625B_1, 625B_2 and the current target 105c. Up to a point and in general, the larger the separation Ad the higher the precision in the measurements that are performed. For example, the separation Ad can be between about 250 pm and 800 pm.

[0088] The interactions between the probe light beams 625B_1, 625B_2 and the current target 105c enable the control system 170 to determine a moving property such as a velocity V of the current target 105c along the -X direction. It is also possible to determine trends in the velocity V or the changing velocity V over many targets 105. It is also possible to determine a change in a moving property of the current target 105c along the -X direction using only the probe light beams 625B_1, 625B_2 if some assumptions about the motion of the current target 105c are made.

[0089] The wavelength of the diagnostic probe 125 (such as the probe light beam 625 A and the probe light beams 625B_1, 625B_2) produced by the diagnostic apparatus 165 should be distinct enough from the wavelength of the radiation pulses directed to the target space 110 (for interaction with the target 105) to facilitate discrimination between the diagnostic light 120 and non-diagnostic light 122. In some implementations, the wavelength of the probe light beam 125, 625A, 625B_1, and 625B_2 is 532 nm or 1550 nm.

[0090] In other implementations such as shown in Fig. 6C, instead of having a single light source such as light source 626B in the diagnostic apparatus 665B, a diagnostic apparatus 665C includes a pair of light sources 626C_1, 626C_2 (such as two lasers), each producing a light beam 611C_1, 611C_2, respectively. Each of the light beams 611 C_ 1, 611C_2 pass through a respective one or more optical elements 627C_1, 627C_2, which can alter or adjust

characteristics of the light beams 611C_1, 611C_2. The output of each of the one or more optical elements is a respective probe light beam 625C_1, 625C_2. The optical components 627 C_l,

627 C_2 can include components for directing the respective probe light beams 625C_1, 625C_2 toward the respective locations xl, x2 along the trajectory TR. Examples of the optical components 627 C_l, 627C_1 are discussed above with reference to optical components 608B.

[0091] As discussed above, and referring to Fig. 7, in some implementations, the metrology apparatus 100 is implemented as a metrology apparatus 700 in an EUV light source 776 to measure one or more properties of targets 105. The EUV light source 776 includes the target supply apparatus 775 that produces the continuous stream 706 of targets (each generally designated as 105) along the trajectory TR toward the target space 710 inside vacuum

environment 715 defined by a chamber 716. The EUV light source 776 supplies EUV light 777 that has been produced by an interaction between a target 105 and a radiation pulse 778 to an output apparatus 779. As discussed above, the metrology apparatus 700 measures and analyzes one or more moving properties (such as speed, velocity, and acceleration) of the current target 105c as the current target 105c travels along the trajectory TR toward the target space 710. The trajectory TR extends along a direction that can be considered as a target (or axial) direction, which lies in the three-dimensional X, Y, Z coordinate system that is defined by the chamber 716. As discussed above, the axial direction of a target 105 generally has a component that is parallel with the -X direction of the coordinate system of the chamber 716. However, the axial direction of the target 105 also can have components along one or more of the directions Y and Z that are perpendicular to the -X direction. Additionally, each target 105 released by the target supply apparatus 775 can have a slightly different actual trajectory and the trajectory depends on the physical properties of the target supply apparatus 775 at the time of release of the target 105 as well as the environment 715 within the chamber 716. [0092] The EUV light source 776 generally includes an EUV light collector 780, an optical source 781, an actuation system 782 in communication with the optical source 781, and a control apparatus 783 in communication with the control system 770 of the metrology apparatus 700 as well as the target supply apparatus 775, the optical source 781, and the actuation system 782.

[0093] The EUV light collector 780 collects as much EUV light 784 emitted from the plasma 785 as possible and redirects that EUV light 784 as collected EUV light 777 toward the output apparatus 779. The light collector 780 can be a reflective optical device such as a curved mirror that is able to reflect light having EUV wavelength (that is, the EUV light 784) to form the produced EUV light 777.

[0094] The optical source 781 produces one or more beams of radiation pulses 778 and directs the one or more beams of radiation pulses 778 to the target space 710 generally along the Z direction (although the beam of radiation pulses 778 can be at an angle relative to the Z direction). In Fig. 7, which is a schematic representation, the beam of radiation pulses 778 is shown as being directed along the -Y direction. The optical source 781 includes one or more light sources that produces radiation pulses 778, a beam delivery system that includes optical steering components that change a direction or angle of the beam of radiation pulses 778, and a focus assembly that focuses the beam of radiation pulses 778 to the target space 710. Exemplary optical steering components include optical elements such as lenses and mirrors that steer or direct the beam of radiation pulses 778 by refraction or reflection, as needed. The actuation system 782 can be used to control or move the various features of the optical components of the beam delivery system and the focus assembly as well as adjust aspects of the optical source 781 that produces the radiation pulses 778.

[0095] The optical source 781 includes at least one gain medium and an energy source that excites the gain medium to produce the radiation pulses 778. The radiation pulses 778 constitute a plurality of optical pulses that are separated from each other in time. In other implementations, the beam output from the optical source 781 can be a continuous wave (CW) beam. The optical source 781 can be or include, for example, a solid-state laser (for example, Nd:YAG laser, an erbium-doped fiber (Englass) laser, or a neodymium-doped YAG (Nd:YAG) laser operating at 1070 nm and at 50 W power).

[0096] The actuation system 782 is coupled to components of the optical source 781 and also in communication with and under control of the control apparatus 783. The actuation system 782 is able to modify or control a relative position between a radiation pulse 778 and the target 105 in the target space 710. For example, the actuation system 782 is configured to adjust one or more of a timing of a release of the radiation pulse 778 and a direction at which the radiation pulse 778 travels.

[0097] The target supply apparatus 775 is configured to release a stream (or plurality 706) of targets 105 at a particular rate. The metrology apparatus 700 takes this rate into account when determining the total amount of time needed to perform the measurement and analysis on the moving property (or properties) of the current target 105c as well as affecting a change to other aspects or components of the EUV light source 776 based on the measurement and analysis. For example, the control system 170 can communicate the results of the measurement and analysis to the control apparatus 783, which determines how to adjust one or more signals to the actuation system 782 to thereby adjust one or more characteristics of the radiation pulse 778 directed to the target space 710.

[0098] The adjustment to the one or more characteristics of the radiation pulse 778 can improve a relative alignment between a present target 105’ and the radiation pulse 778 in the target space 710. The present target 105’ is the target that has entered the target space 710 at the time that the radiation pulse 778 (which has just been adjusted) arrives in the target space 710. Such adjustment to the one or more characteristics of the radiation pulse 778 improves the interaction between the present target 105’ and the radiation pulse 778 and increases the amount of EUV light 784 produced by such interaction. As shown in Fig. 7, a previous target 105p has already interacted with a prior radiation pulse (not shown) to produce plasma 785 that emits the non diagnostic light 122 (in addition to the EUV light 784).

[0099] In some implementations, the present target 105’ is the current target 105c. In these implementations, the adjustment to the one or more characteristics of the radiation pulse 778 happens in a relatively shorter time frame. A relatively shorter time frame means that the one or more characteristics of the radiation pulse 778 are adjusted during the time after the analysis of the moving properties of the current target 105c is completed to the time that the current target 105c enters the target space 710. Because the one or more characteristics of the radiation pulse 778 are able to be adjusted in the relatively shorter time frame, there is enough time to affect the interaction between the current target 105c (the moving properties of which have just been analyzed) and the radiation pulse 778. [0100] In other implementations, the present target 105’ is another target, that is, a target other than the current target 105c, and following the current target 105c in time. In these

implementations, the adjustment to the one or more characteristics of the radiation pulse 778 happens in a relatively longer time frame such that it is not feasible to affect the interaction between the current target 105c (the moving properties of which have just been analyzed) and the radiation pulse 778. On the other hand, it is feasible to affect the interaction between the other (or later) target and the radiation pulse 778. A relatively longer time frame is a time frame that is greater than the time after the analysis of the moving properties of the current target 105c is completed to the time that the current target 105c enters the target space 710. Depending on the relatively longer time frame, the other target could be adjacent to the current target 105c. Or, the other target could be adjacent to an intermediate target that is adjacent to the current target 105c. In these other implementations, an assumption is made that the other target (which is not the current target 105c) is traveling with a moving property that is similar enough to the detected or estimated moving property of the current target 105c.

[0101] Each of the targets 105 (including the previous target 105p and the current target 105c, and all other targets produced by the target supply apparatus 775 (or 175)) includes a material that emits EUV light when converted to plasma. Each target 105 is converted at least partially or mostly to plasma through interaction with the radiation pulse 778 produced by the optical source 781 within the target space 710. Each target 105 produced by the target supply apparatus 775 (or 1750 is a target mixture that includes the target material and optionally impurities such as non target particles. The target material is the substance that is capable of being converted to a plasma state that has an emission line in the EUV range. The target 105 can be, for example, a droplet of liquid or molten metal, a portion of a liquid stream, solid particles or clusters, solid particles contained within liquid droplets, a foam of target material, or solid particles contained within a portion of a liquid stream. The target material can include, for example, water, tin, lithium, xenon, or any material that, when converted to a plasma state, has an emission line in the EUV range. For example, the target material can be the element tin, which can be used as pure tin (Sn); as a tin compound such as SnBr4, SnBr2, SnH4; as a tin alloy such as tin-gallium alloys, tin-indium alloys, tin-indium-gallium alloys, or any combination of these alloys. In the situation in which there are no impurities, then each target 105 includes only the target material. The discussion provided herein is an example in which each target 105 is a droplet made of molten metal such as tin. However, each target 105 produced by the target supply apparatus 775 (or 175) can take other forms.

[0102] The target 105 can be provided to the target space 710 by passing molten target material through a nozzle of the target supply apparatus 775 (or 175), and allowing the target 105 to drift along the trajectory TR into the target space 710. In some implementations, the target 105 can be directed to the target space 710 by force (either in addition to or despite gravity). As discussed below, the present target 105’ (which can be the current target 105c) that interacts with the radiation pulse 778 can also have already interacted with one or more prior radiation pulses. Or, the present target 105’ that interacts with the radiation pulse 778 can reach the target space 710 without having interacted with any other radiation pulses.

[0103] In this implementation, the detection apparatus 735 is positioned external to the chamber 716 while the diagnostic region 745 is inside the environment 715 of the chamber 716. A wall of the chamber 716 is fitted with an optical window 736 that is transparent to the wavelength of the diagnostic light 120 and is able to withstand any pressure differentials at the wall. The optical window 736 can be held in a mount and hermetically sealed in the wall to maintain the pressure within the environment 715. For example, the optical window 736 can be made of a crown glass having a relatively low refractive index and low dispersion, such as a borosilicate glass (BK7 or N-BK7), or fused silica. The optical window 736 has an aperture that is large enough to accommodate the extent of the diagnostic light 120. A distance between the diagnostic region 745 and the optical window can be on the order of several hundred millimeters (or about 600-700 mm) while a distance between the diagnostic region 745 and the target space 710 can be on the order of a few millimeters (or about 1-5 mm). Thus, the distance between the diagnostic region 745 and the optical window can be 200-500 times the distance between the diagnostic region 745 and the target space 710.

[0104] The control apparatus 783 communicates with the control system 170 and also communicates with other components (such as the actuation system 782, the target supply apparatus 775, and the optical source 781) of the EUV light source 776. Referring to Fig. 8, an implementation 883 of the control apparatus 783 is shown and an implementation 870 of the control system 170 is shown. The control apparatus 883 includes the control system 870, but it is possible for the control system 870 to be physically separate from the control apparatus 883 and still remain in communication. Moreover, features or components of the control apparatus 883 can be shared with the control system 870, including features not shown in Fig. 8.

[0105] The control system 870 includes a signal processing module 871 configured to receive the output from the detection apparatus 735 (or 135, 335, 435). The control system 870 includes a diagnostic control module 872 in communication with the diagnostic apparatus 765 (or 165). For example, the signal processing module 871 receives a signal from the sensor 130 within the detection apparatus 735 (135, 335, 435), where the signal is a voltage signal related to current produced from the detected light at the sensor 130. Generally, the signal processing module 871 analyzes the output from the sensor 730, and determines one or more moving properties of the current target 105c based on this analysis. The diagnostic control module 872 controls operation of the diagnostic apparatus 765. For example, the diagnostic control module 872 can provide a signal to the diagnostic apparatus 765 for adjusting one or more characteristics of the diagnostic apparatus 765 and also for adjusting one or more characteristics of the diagnostic probe(s) 725.

[0106] The signal processing module 871 also determines whether an adjustment needs to be made to a subsequent radiation pulse 778 output from the optical source 781 based on the determination of the one or more moving properties of the current target 105c. And, if an adjustment is needed, the signal processing module 871 sends an appropriate signal to an optical source actuation module 884, which interfaces with the optical source 781 or the actuation system 782. The optical source actuation module 884 can be within the control apparatus 883 (as shown in Fig. 8) or it can be integrated within the control system 870.

[0107] The signal processing module 871 can include one or more field-programmable hardware circuits, such as field-programmable gate arrays (FPGAs). A FPGA is an integrated circuit designed to be configured by a customer or a designer after manufacturing. The field- programmable hardware circuit can be dedicated hardware that receives the value or values of the time stamps, performs a calculation on the received values, and uses one or more lookup tables to estimate a time of arrival of the present target 105’ at the target space 710. In particular, the field-programmable hardware circuit can be used to quickly perform a calculation to enable the adjustment to the one or more characteristics of the radiation pulse 778 in the relatively shorter time frame to enable the adjustment of the one or more characteristics of the radiation pulse 778 that interacts with the current target 105c, the moving properties of which have just been analyzed by the signal processing module 871. [0108] The control apparatus 883 includes a target delivery module 885 configured to interface with the target supply apparatus 775. Moreover, the control apparatus 883 and the control system 870 can include other modules specifically configured to interface with other components of the EUV light source 776 not shown.

[0109] The control system 870 generally includes or has access to one or more of digital electronic circuitry, computer hardware, firmware, and software. For example, the control system 870 can have access to memory 873, which can be read-only memory and/or random access memory. Storage devices suitable for tangibly embodying computer program instructions and data include all forms of non-volatile memory, including, by way of example, semiconductor memory devices, such as EPROM, EEPROM, and flash memory devices; magnetic disks such as internal hard disks and removable disks; magneto-optical disks; and CD-ROM disks. The control system 870 can also include or interface with one or more input devices 874i (such as a keyboard, touch screen, microphone, mouse, hand-held input device, etc.) and one or more output devices 874o (such as speakers and monitors).

[0110] The control system 870 can also include or have access to one or more programmable processors, and one or more computer program products tangibly embodied in a machine- readable storage device for execution by a programmable processor. The one or more

programmable processors can each execute a program of instructions to perform desired functions by operating on input data and generating appropriate output. Generally, the processor receives instructions and data from memory 873. Any of the foregoing may be supplemented by, or incorporated in, specially designed ASICs (application-specific integrated circuits).

[0111] Moreover, any one or more of the modules can include their own digital electronic circuitry, computer hardware, firmware, and software as well as dedicated memory, input and output devices, programmable processors, and computer program products. Likewise, any one or more of the modules can access and use the memory 873, the input devices 874i, the output devices 874o, the programmable processors, and the computer program products.

[0112] Although the control system 870 is shown as a separate and complete unit, it is possible for each of its components and modules to be separate units. The control apparatus 883 can include other components, such as dedicated memory, input/output devices, processors, and computer program products, not shown in Fig. 8. [0113] Referring to Fig. 9, an implementation 979 of the lithography apparatus 779 is shown. The lithography apparatus 979 exposes a substrate (which can be referred to as a wafer) W with an exposure beam B. The lithography apparatus 979 includes a plurality of reflective optical elements Rl, R2, R3, a mask M, and a slit S, all of which are in an enclosure 10. The enclosure 10 is a housing, tank, or other structure that is capable of supporting the reflective optical elements Rl, Rl, R2, the mask M, and the slit S, and is also capable of maintaining an evacuated space within the enclosure 10.

[0114] The EUV light 777 enters the enclosure 10 and is reflected by the optical element Rl through the slit S toward the mask M. The slit S partly defines the shape of the distributed light used to scan the substrate W in a lithography process. The dose delivered to the substrate W or the number of photons delivered to the substrate W depends on the size of the slit S and the speed at which the slit S is scanned.

[0115] The mask M also may be referred to as a reticle or patterning device. The mask M includes a spatial pattern that represents the features that are to be formed in a photoresist on a substrate W. The EUV light 777 interacts with the mask M. The interaction between the EUV light 777 and the mask M results in the pattern of the mask M being imparted onto the EUV light 777 to form the exposure beam B. The exposure beam B passes through the slit S and is directed to the substrate W by the optical elements R2 and R3. An interaction between the substrate W and the exposure beam B exposes the pattern of the mask M onto the substrate W, and the photoresist features are thereby formed at the substrate W. The substrate W includes a plurality of portions 20 (for example, dies). The area of each portion 20 in the Y-Z plane is less than the area of the entire substrate W in the Y-Z plane. Each portion 20 may be exposed by the exposure beam B to include a copy of the mask M such that each portion 20 includes the electronic features indicated by the pattern on the mask M.

[0116] The lithography apparatus 979 can include a lithography control system 30 that is in communication with the control apparatus 783 of the EUV light source 776.

[0117] Referring to Fig. 10, a procedure 1090 is performed by the metrology apparatus 100 (or the metrology apparatus 700). The diagnostic probe 125 is interacted with the current target 105c in the diagnostic region 145 before the current target 105c enters the target space 110 (1091). Diagnostic light 120 that is produced from the interaction (1091) is collected, and at the same time, some unwanted non-diagnostic light 122 produced from the target space 110 is also collected (1092). For example, a pupil at the entrance of the mechanical filter 140 the diagnostic light 120 and non-diagnostic light 122. The diagnostic light 120 and the non-diagnostic light 122 are collimated (1093). The optical collimator 142 collimates the diagnostic light 120 and the non-diagnostic light 122 that have entered the mechanical filter 140. This collimated diagnostic light and non-diagnostic light are angularly separated from each other (1094). The optical beam reducer 150 reduces the transverse extent (along the XsYs plane) of the collimated diagnostic light beam and the collimated non-diagnostic light beam and these reduced light beams 121, 123, respectively, exit the optical beam reducer 150 and their angular separation increases as they travel along the optical path 152 toward the mask 155. The collimated diagnostic light beam 121 is sensed at a sensing region (for example, the sensor 130) that is in optical communication with an open image plane, the open image plane being transversely displaced (in the XsYs plane) from a closed image plane that blocks the collimated non-diagnostic light beam 123 (1095). For example, the sensor 130 senses the collimated diagnostic light beam 121, the sensor 130 is in optical communication with the aperture 160 (the open image plane), and the aperture 160 is transversely displaced in the XsYs plane from the mask 155, onto which the collimated non diagnostic light beam 123 impinges. The property of the current target 105c is estimated based on the sensed diagnostic light (1096). In particular, the control system 170 analyzes the output from the sensor 130 to estimate one or more moving properties of the current target 105c.

[0118] As discussed above with reference to Figs. 6A-6C, the diagnostic probe 125 can be one or more probe light beams that is directed to cross the trajectory of the targets 105. Thus, the interaction between the diagnostic probe 125 and the current target 105c (1091) can be between these probe light beams and the current target 105c. In some implementations, the diagnostic light 120 that is collected (1092) can be a portion of the probe light beam 125 that is reflected or scattered from the current target 105c. In other implementations, the diagnostic light 120 that is collected (1092) is light that is produced by the current target 105c. In other implementations, the diagnostic light 120 that is collected (1092) is light that is blocked by the current target 105c (as shown in Fig. 12).

[0119] The impact of the non-diagnostic light 122 on the analysis of the output from the sensor 130 is reduced by blocking or redirecting the non-diagnostic light 122 with the mechanical filter 140, and thus the non-diagnostic light 122 is prevented from impinging upon the sensor 130 and instead is blocked by the mask 155. Moreover, the collimation state of the collimated diagnostic light beam and the collimated non-diagnostic light beam output from the collimator 142 is maintained ah the way to the plane of the mask 155, while the collimated diagnostic light beam 121 is further focused onto the sensor 130 after passing through the aperture 160.

[0120] Referring to Fig. 11, in other implementations, the optical collimator 142 and

components of the optical beam reducer 150 that have positive focal length and are nearest the optical collimator 142 are integrated into a single refractive element 1142/1151. This integrated single refractive element 1142/1151 can be applied to the Galilean-type refracting telescope (such as that of Fig. 3 and shown in Fig. 11) or to the Keplerian-type refracting telescope such as that of Fig. 4.

[0121] Referring to Fig. 12, in other implementations, the diagnostic light 120 is produced from a portion of a diagnostic probe light beam 1225 that is blocked by the current target 105c. In this implementation of the metrology apparatus 1200, the diagnostic probe 1225 provides a back illumination of the current target 105c. The sensor 130 is a two-dimensional (for example, imaging) sensor 1230 such as a camera. Thus, when the current target 105c crosses the diagnostic probe 1225, a shadow of the target 105c is formed from the target 105c obscuring at least a portion of the diagnostic probe 1225, as shown in the inset of Fig. 12. This sort of arrangement can be considered a shadowgraph arrangement. In such an implementation, the sensor 1230 is arranged on a side of the target trajectory TR that is opposite to the side on which the diagnostic apparatus 1265 is arranged. The sensor 1230 is a camera that captures the two- dimensional representation (which can be considered an image) of the diagnostic light 1220. Thus, for example, the sensor 1230 includes a two-dimensional array of thousands or millions of photo-sites (or pixels). The diagnostic light 1220 is directed onto the photo- sensitive area of each pixel where it is converted into electrons that are collected into a voltage signal and the array of these signals forms the two-dimensional image. The non-diagnostic light 1222 is substantially blocked from reaching the sensor 1230, as discussed above.

[0122] Referring to Fig. 13, in other implementations, the optical beam reducer 150 is designed as a reflective beam reducer 1350 that receives the collimated light beams from the optical collimator such as collimator 342. The reflective beam reducer 1350 includes a concave reflective element (a curved mirror) 1351 that converges the light beams toward the convex reflective element (a curved mirror) 1353, which collimates the light as collimated diagnostic light beam 121 and collimated non-diagnostic light beam 123, which travel in distinct directions (or angles) along the optical path 1352 toward the mask 155.

[0123] Referring to Fig. 14, in other implementations, the optical beam reducer 150 is designed as a catadioptric (hybrid) beam reducer 1450 that receives the collimated light beams from the optical collimator such as collimator 342. The hybrid beam reducer 1450 includes a flat reflective element (a flat mirror) 1451a that directed the beams onto a curved reflective element (a curved mirror) 1451b that converges the light beams toward an intermediary focus IF between the curved mirror 1451b and a convex or convergent refractive element (a lens) 1453. The lens 1453 collimates the light as collimated diagnostic light beam 121 and collimated non-diagnostic light beam 123, which travel in distinct directions (or angles) along an optical path toward the mask 155.

[0124] Referring to Fig. 15, in other implementations, the mask 155 is designed as a mask 1555, which defines an aperture 1560 that aligns with the pathway of the collimated non-diagnostic light beam 123 while the mask 1555 acts to stop or prevent the passage of the collimated diagnostic light beam 121 from reaching the sensor 130. Such a design can be useful if it is necessary to analyze aspects relating to the non-diagnostic light 122.

[0125] Other aspects of the invention are set out in the following numbered clauses.

1. A metrology apparatus comprising:

a diagnostic apparatus configured to interact a diagnostic probe with a current target at a diagnostic region before the current target enters a target space;

a detection apparatus comprising:

a light sensor having a field of view overlapping with the diagnostic region and configured to sense light produced from the interaction between the diagnostic probe and the current target at the diagnostic region; and

a mechanical filter between the diagnostic region and the light sensor, the mechanical filter comprising an optical beam reducer and an optical mask defining an aperture positioned between the optical beam reducer and the light sensor; and

a control system in communication with the detection apparatus and configured to estimate a property of the current target based on the output from the light sensor.

2. The metrology apparatus of clause 1, wherein the mechanical filter is configured to angularly separate diagnostic light emitted from the diagnostic region from non-diagnostic light emitted from the target space, wherein the diagnostic light is produced from an interaction between the current target and the diagnostic probe at the diagnostic region.

3. The metrology apparatus of clause 2, wherein non-diagnostic light includes light emitted from a plasma produced by a previous target in the target space.

4. The metrology apparatus of clause 2, wherein a transverse extent of the aperture is about the same as or greater than a transverse extent of the diagnostic light in the plane of the optical mask, and a transverse extent of the optical mask is greater than or about the same as a transverse extent of the non-diagnostic light in the plane of the optical mask.

5. The metrology apparatus of clause 2, wherein the optical mask is positioned so that the non diagnostic light emitted from the target space is substantially blocked by the optical mask while the diagnostic light substantially passes through the aperture.

6. The metrology apparatus of clause 1, wherein the mechanical filter comprises an optical collimator between the diagnostic region and the optical beam reducer.

7. The metrology apparatus of clause 6, wherein the optical beam reducer is an afocal beam reducer and is configured, in conjunction with the optical collimator, to project a finite object to infinity.

8. The metrology apparatus of clause 6, wherein the optical collimator and components of the optical beam reducer that have positive focal length nearest the optical collimator are integrated into a single refractive element.

9. The metrology apparatus of clause 6, wherein the optical beam reducer is configured to maintain a collimated status of light.

10. The metrology apparatus of clause 1, wherein the light sensor comprises one or more:

photodiodes, photo-transistors, light-dependent resistors, photomultiplier tubes, multi-cell photoreceivers, quad-cell photoreceivers, and cameras.

11. The metrology apparatus of clause 1, wherein the diagnostic probe comprises at least one diagnostic light beam, and the light sensor is configured to sense diagnostic light produced from the interaction between the current target and the at least one diagnostic light beam.

12. The metrology apparatus of clause 11, wherein the diagnostic light comprises the diagnostic light beam that is reflected from, scattered from, or blocked by the current target.

13. The metrology apparatus of clause 1, wherein the detection apparatus further comprises one or more of a spectral filter and a polarization filter. 14. The metrology apparatus of clause 1, wherein the diagnostic probe comprises first and second diagnostic light beams, each configured to interact with the current target before it enters the target space, each interaction occurring at a distinct region and a distinct time.

15. The metrology apparatus of clause 1, wherein the optical beam reducer comprises a refracting telescope, a reflecting telescope, or a catadioptric telescope.

16. The metrology apparatus of clause 15, wherein the refracting telescope comprises:

a positive focal length lens arrangement and a negative focal length lens arrangement that are separated by the sum of their focal lengths; or

a pair of positive focal length lens arrangements that are separated by the sum of their focal lengths.

17. The metrology apparatus of clause 1, wherein the optical beam reducer is configured to reduce a transverse size of impinging light by a factor of at least five, at least ten, at least twenty, or about ten.

18. The metrology apparatus of clause 1, wherein the aperture includes a circular opening, an elliptical opening, a polygonal opening, or an elongated slit opening.

19. The metrology apparatus of clause 1, wherein the detection apparatus is positioned external to a chamber of an extreme ultraviolet (EUV) light source, the diagnostic region is inside the chamber, and the detection apparatus receives light from the chamber through an optical window in a wall of the chamber.

20. The metrology apparatus of clause 19, wherein a distance between the diagnostic region and the optical window is about 200-500 times the size of a distance between the diagnostic region and the target space.

21. The metrology apparatus of clause 1, wherein the detection apparatus comprises a focusing lens at an output of the aperture, the focusing lens configured to focus the sensed light onto the light sensor.

22. The metrology apparatus of clause 1, wherein the aperture has an extent that is at least 2 millimeters (mm).

23. The metrology apparatus of clause 1, wherein the aperture is positioned to receive diagnostic light at a location at which the diagnostic light is collimated or is non-converging and non diverging.

24. A metrology method comprising: interacting a diagnostic probe with a current target at a diagnostic region before the current target enters a target space;

collecting diagnostic light produced from the interaction between the diagnostic probe and the current target at the diagnostic region, the collecting including also collecting non-diagnostic light produced from the target space;

collimating the diagnostic light and the non-diagnostic light;

angularly separating the diagnostic light and the non-diagnostic light from each other including reducing a transverse extent of the diagnostic light and the non-diagnostic light;

sensing the diagnostic light at a sensing region transversely displaced from a non-sensing region traversed by the non-diagnostic light after the diagnostic light and the non-diagnostic light have been angularly separated; and

estimating a property of the current target based on the sensed diagnostic light.

25. The metrology method of clause 24, wherein:

interacting the diagnostic probe with the current target at the diagnostic region comprises interacting one or more diagnostic light beams with the current target at the diagnostic region; and

collecting diagnostic light comprises collecting one or more diagnostic light beams that have been reflected from, scattered from, or blocked by the current target at the diagnostic region.

26. The metrology method of clause 24, further comprising filtering the diagnostic light based on one or more of its spectral properties and its polarization state.

27. The metrology method of clause 24, wherein the diagnostic region is inside a hermetically- sealed chamber of an extreme ultraviolet (EUV) light source and collecting the diagnostic light including also collecting non-diagnostic light comprises receiving the diagnostic light including the non-diagnostic light transmitted through an optical window in a wall of the chamber.

28. The metrology method of clause 24, wherein reducing the transverse extent of the diagnostic light and the non-diagnostic light comprises reducing the transverse extent of the diagnostic light and the non-diagnostic light by at least a factor of five, at least ten, at least twenty, or by a factor of about ten.

29. The metrology method of clause 24, further comprising blocking or redirecting the non diagnostic light at the non-sensing region. 30. The metrology method of clause 24, wherein reducing the transverse extent of the diagnostic light and the non-diagnostic light comprises one or more of: refracting the light and reflecting the light.

31. The metrology method of clause 24, further comprising focusing the diagnostic light at the sensing region.

32. The metrology method of clause 24, wherein reducing the transverse extent of the diagnostic light and the non-diagnostic light comprises maintaining a collimation state of the diagnostic light and the non-diagnostic light.

33. The metrology method of clause 24, further comprising passing the diagnostic light through an aperture of an optical mask after the diagnostic light and the non-diagnostic light are angularly separated from each other, the aperture having an extent that is greater than an extent of the diagnostic light, and before the diagnostic light is sensed.

[0126] Other implementations are within the scope of the following claims.