ULTRAVIOLET LIGHT SOURCE

CROSS-REFERENCE TO RELATED APPLICATION

This application claims the benefit of U.S. Utility Application No. 15/137,933, filed on

April 25, 2016, and entitled REDUCING THE EFFECT GF PLASMA ON AN OBJECT IN AN EXTREME ULTRAVIOLET LIGHT SOURCE, which, is incorporated herein by reference in its entirety. TECHNICAL FIELD

This disclosure relates to reducing the effect of plasma on an object in an extreme ultraviolet (EUV) light source.

BACKGROUND

Extreme ultraviolet ("EUV") light, for example, electromagnetic radiation having wa velengths of around 50 mn or less (also sometimes referred to as soft x-rays), and including light at a wavelength of about 13 am, may be used in photolithography processes to produce extremely small features in substrates, for example, silicon wafers.

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 plasm state. In one such method, often termed laser produced plasma ("EPF"), the required plasma may be produced by irradiating a target material, for example, in the form of droplet, plate, tape, stream, or cluster of material, with a light beam that may 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

In one general aspect, a first target is provided to an interior of a vacuum chamber, the first target, including target, material that emits extreme ultraviolet (EUV) light in a plasma state; a first light beam is directed toward the first target to form a first plasma from the target material of the first target, the first plasma being associated with a directional flux of particles and radiation emitted from the first target along a first emission direction, the first emission direction being detemuned by a position of fee first target; a second target is provided to the ^■ interior of the vacuum chamber* tlie second target mchKung target material that emits extreme ultiavioletJight in a plasma state and a second light beam is directed toward the second target to form a second plasma from the target material of the second target, the second plasma being associated with a directional Sux of particles and radiation emitted from the second target along a second emission direction, the second emission direction being determined by a position of th second target tlie second emission direction being different from the first emission direction.

Implementations may include one or more of the following features. The target material of the first target may be arranged in a first geometric distribution, the first geometric distribution may have an extent along an axis oriented at a first angle relative to a separate and distinct object in the vacuum chamber, the target material of the second target may be arranged in a second geometric distribution, the second geometric distribution may have an extent along an axis oriented at a second angle relative to the separate and distinct object in the vacuum chamber, the second angle may be different from the first angle, the first emission direction may be determined by the first angle, and the second emission may be determined by the second angle.

In some implementations, providing a first target to an interior of a vacuum chamber includes; providing a first initial target to the interior of the vacuum chamber, the first initial target including target material in an initial geometric distribution; and directing an optical pulse toward the first initial target to form the first target, the geometric distribution of the first target being different from the geometric distribution of the first initial target, and providing a second target to an interior of a vacuum chamber includes; providing a second initial target to the interior of the vacuum chamber, the second initial target including target material in a second initial geometric distribution; and directing an optical pulse toward the second initial target to form the second target, the geometric distribution of the second target being different from the geometric distribution of the second initial target.

Tlie first initial target and the second initial target may be substantially spherical, and the first target and the second target may be disk shaped. The first initial target and the second initial target may be two initial targets of a pliirality of initial targets that travel along a trajectory, and the separate and distinct object in the vacuum chamber may be one of the plurality of initial targets other than the first initial target and the second initial target. A Said ipay be provided to t e interior of the vacuum chamber; the fluid occupying a volume is the vacuum chamber, arid the separate and distinct object the vacuum chamber may include a portion of the fluid. The fluid ^' may ^" be a flowing gas. In a target region that receives the target, the first light beam may propagate toward the first target and the second light ^' beam may propagate toward the second target in a propagation direction, and the flowing gas may flow in a direction that is parallel to the propagation direction.

The separate and distinct object in the vacuum chamber may include an optical element. The optical element may be a reflective element.

The separate and distinct object in the vacuum chamber may be a portion of a reflective surface of an optical element, and the portion being less than all of the reflective surface.

A fluid may be provided to the interior of the vacuum chamber based on a flow configuration, and, in these implementations, the fluid flows in the vacuum chamber based on the flow configuration. Tlie first light beam and the second light beam may be optical pulses in a pulsed light beam configured to provide an EUV burst duration, and the EUV burst dmation may be determined. A property of the fluid associated with the EUV burst duration may be determined, the property including one or more of a minimum flow rate, density , and pressure of the fluid, and the flow configuration of the fluid may be adjusted based on the determined property. The flow configuration may include one or more of a flow rate and a flow direction of the fluid, and adjusting the flow configuration of the fluid may include adjusting one or more of the flow rate and the flow direction.

In some implementations, the first target forms a plasma at a first time, the second plasma forms a target at a second time, the time between the first time and the second time being an elapsed time, and the light beam includes a pulsed light beam configured to provide an EUV burst duration. The EUV burst duration may be determined, a minimum flow rate associated with the EUV ^" burst dmation may be determined, and one or more of the elapsed time and the flow rate of the fluid may be adjusted based on the determined minimum flow rate of the fluid.

The first light beam may have an axis, and the intensity of the first light beam may be greatest at the axis. The second light beam may have an axis, and the intensity of the second light beam may greatest at the axis of the second light beam. The first emission direction may be determined by location of the first target relative to the axis of the first light beam, and the second emission direction may be dete &ned by a location of the second target illa tive to fee axis of the second beam.

The axis of the first light beam, and the axis of fee second light beam may be along the same direction, the first: target is at a location on a first side of the axis of the first light beam, and the second target is at a location on a second side of the axis of the first light beam.

The axis of the frrst. light bean and the axis of the second .light beam .may be alon different directions, and the first target and the second target may be at substantially the same location in the vacuum chamber at different times.

The first and second targets may be substantially spherical.

In another general aspect, the effect of plasm on an object in a vacuum chamber of an extreme ultraviolet (EUV) light source may be reduced. An initial target is modified, in the vacuum chamber, to form a modified target, the initial tar get including target materia! in an mitiai geometric distribution and fee modified target including target material in a different, modified geometric distribution. A light beam is directed toward fee modified target, the light beam ha ving an energy sufficient to convert at lea st some of fee target material in the modified target to plasma that emits EUV light, the plasma being associated with a directionally dependent flux of particles and radiation, the directionally dependent flux having an angular distribution relative to the modified target, the angular distr ibution being dependent on a position of fee modified target such that positioning the modified target in fee vacuum chamber reduces the effect of the plasma on the object.

Implementations may include one or more of the following features. The modified geometric distribution may have a first extent in a first direction and a second extent in a second direction, the second extent may be larger than the first extent, and the modified target may be positioned by orienting the second extent at an angle relative to the object. A second initial target also may be provided to an interior of the vacuum chamber , the initial target and the second initial target traveling along a trajectory. The separate and distinct object may be fee second initial target. The second initial target may be one target in a stream of targets that tr avel on the trajectory. The second initial target may be the target in the stream that is closest in distance to the initial target. In some implementations, the second mitiai target is modified to form a second modified target, the second modified target having the modified geometric distribution of target material, and the second extent of the second modified tar get being positioned with the second extent, oriented at a second, different angle relative to the separate and distinct obj ect. The separate and distinct object may be one of more of a portion of a volume of fluid that flow in the vacuum chamber and an optical element in the vacuum chamber.

The modified target may be positioned by directing a pulse of light a the initial target way from a center of the _. initial target suc that the target material of the initial target expands along the second extent and reduces along the first exten and the second extent tilts relative to the separate and distinct object.

A fluid may be provided to the interior of the vacuum chamber, the fluid occu ing ¾ volume in the vacuum chamber, and the separate and distinct object in the vacuum chamber may include a portion of the volume of the fluid.

In another general aspect, a control system for an extreme ultraviolet (EUY) light source includes one or more electronic processors; an electronic storage storing mstructions that, when executed, cause the one or more electronic processors to: declare a presence of a ^' first initial, target at a first time, the first, initial target having a distribution of target material that emits EUV light in a plasma state: direct a first light beam toward ate first initial targe at a second time based on the declared presence of the first initial target, a. difference between the first time and the second time being a first elapsed time; declare a presence of a second initial target at a t ird time, the third time occurring after the first time, the second initial target including target material that emits EUV light in a plasma state; direct the first light beam toward the second initial target at a fomth time based on the declared presence of the second initial target, the fomth time occurring afte the second time, a difference between the third time and the fourth time being a second elapsed time, where the first, elapsed time is different from the second elapsed time such that the first and second initial targets expand along different directions and have different orientations in a target region, the target region bein a region that receives a second light beam having energy sufficient to convert target material to plasma that emits EUV light.

Implementations of any of the techniques described above may include an apparatus, a method or process, an EUV light source, an optical lithography system, a control system for an optical source, or instructions stored on a computer-readable medium.

The details of one or more implementations are set forth in the accompanying drawings and the description below. Other features will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS

FIG. I is a. block diagram. of an e emplary optica! lithography system that i cludes an EUV light source.

FIG. 2A is a side eross-sectioimi view of an exemplary target.

FIG. 2B 2A.

FIGS. 2C and 2D axe iiiiistraiions of different exemplary positions of the target of FIG

2A.

FIG. 3 A is an illustration of energy emitted ^■ from plasma foraied from an exemplaiy target.

FIGS. 3B and 3C are block diagrams of an exemplary target in two different positions .

FIG. 3D is an example of an intensity profile of a light beam.

FIG. 3E and 3F are block diagrams of a light beam interacting with as exemplary target in two different positions.

FIG. 4 is a block diagram of an exemplary - .system that includes a control system for controlling a position of a target.

FIG. 5 i a flow chart of an exemplary process for generating EOV light.

FIG. 6A shows an exemplary initial target that is converted to a target.

FIG. 6B is a plot of an exemplary waveform, shown as energy versus time, for generating the target of FIG. 6A.

FIG. 6C shows side views of the initial target and the tar get of FIG. 6 A.

FIGS. 7 A and 7B are block diagrams of an exemplaiy vacuum chamber.

FIG. 7C is a block diagram of an exemplary optical element in the vacuum chamber of FIGS. 7 A and 7B.

FIG. 8 is a flow chart of an exemplaiy process for varying the positions of targets.

FIGS. 9A-9C ar e block diagr ams of an exemplaiy vacuum chamber thai includes a target that has a position that varies with time.

FIGS. I OA and I OB are block diagrams of an exemplaiy vacmnn ehamber that includes a target that has a position that varies with time.

FIG. 10C is a block diagram of an optical element and a path swept out by ^' a peak of a directionally dependent energy profile.

FIG. I I is a plot of exemplaiy data relating rnininiuin fluid flow and EUV burst duration. FIG. 12 is a flow chart of an ^■■ exemplary process for protectin an object in. a ..vacuum .chamber.

FIGS. I3A-13C are block diagrams of an exeiBplary vacuum chamber that includes a target that has a position and/or a target patli that varies with time.

FIG. 14 is a block diagram of an exemplary optical lithography system that includes an

EUV ligM source.

FIG. ISA is a block diagram of an exemplary optical lithography system that includes an EUV light source.

FIG. 15B is a block diagram of a optical amplifier system that can be used in the EUV light source of FIG. 15 A.

FIG. 16 is a block diagram of another implementation of the EUV light source of FIG. 1. FIG. 17 is a block diagram of an exemplary target material supply apparatus that can be used in an EUV ⁷ light source. DETAILED DESCRIPTION

Techniques for reducing the effect of plasma on objects in a vacuum chamber of an extreme ultraviolet (EUV) light source are disclosed. To produce EUV light, the EUV ^' light source converts target material in targets to plasma that emits EUV light. By varying a spatial orientation or position of the various targets such that the targets do not all have the same position or orientation, the effect of the plasma may be reduced. The described techniques may be used to. for example, protect objects inside of a vacuum vessel of an E ⁷ light source.

Referring to FIG. 1, a block diagram of an exemplary optical lithography system 100 is shown. The system 100 includes an extreme ultraviolet (EUV) light source 101 that provides EUV light 162 to a lithography tool 103. The EUV light source 101 includes an optical source 102 and a fluid delivery system 104. The optical source 102 emits a light beam 110, which enters a vacuum vessel 140 through an optically transparent opening 114 and propagates in a direction z (112) at a target region 130, which receives a target 120. The light beam 110 can be an amplified light beam.

The fluid delivery system 104 delivers a buffer fluid 108 into the vessel 140. The buffer fluid 108 may flow between an optical element 155 and the target region 130. The buffer fluid 108 may flow in the direction z or in any other direction, and the buffer fluid 10S may flow in multiple directions. The target region 130 receives the target 120 from a target supply system 116. The target 120 includes target material that emits EUV light 162 when in a plasma state, and an interaction between the target material asd the light beam 110 at the target region 130 converts at least some of the target material to plasma. The, optical element 155 directs EUV light 162 toward the lithography tool 103. A control system 170 can receive and provide electronic signals to the iloid delivery system 104, the optical source 102, and/or the lithography tool 103 to allow for control of any or all of these components. An example of the control system 170 is discussed below with respect to FIG. 4.

The target material of the target 120 is arranged in a geometric or spatial distribution, with a side or region 129 that receives (and interacts with) the light beam 110. As discussed above, the target material emits EUV light 162 when in a plasma state. Additionally, the plasma also emits particles (such as ions , neutral atoms , and/or clusters of the tar get ma terial) and/or radiation other than EUV light. The energy emitted by the plasma (including the particles and/or radiation that is other than EUV light) is non-isotropic relative to the geometric distribution of the target material. The energy emitted by the plasma may be considered to be a directionally dependent flux of energy with an angularly dependent distributio relative to the target 120. Thus, the plasma may direct a greater amount of energy toward some regions in the vessel 140 than others. The energy emitted from the plasma causes, for example, localized heating in the regions toward which it is directed.

FIG. 1 shows the vacuum vessel 140 at an instance of time. In the example shown, the target 120 is in the target location 130. At times before and or after the time of FIG. 1, other instances of the target 120 are in the target region 130. As discussed below, the other instances of the target 120 are similar to the target 120 except, as compared to the target 120, prior and/or subsequent instances of the target 120 have a different geometric distribution of target material, a different position in the vacuum vessel 140, and/or a different orientation of the geometric distribution of target material relative to an object or objects in the vacuum vessel 140. In other words, the geometric distribution, position, and/or orientation of a target that is present in the target region 130 varies among the instances and can be considered to vary over time. In this way, the direction along which the peak (maximum) of the directionally dependent flux extends may be changed over time. Thus, the peak of the directionally dependent flux may be directed

S away train a partictilar object, a particular portion of an object, and/or a region of the vessel 140, thereby reducing the effects of the plasma on thai obj ect, portion, or region.

Varying tlie position, geometric dishibiuion, aad/or orientation of the target material among the instances or over time increases the total, amount of area toward which energy is directed by the plasma. Thus, varying the position of the target and/or the target orientation over tisae allows Hie energy from the plasma to more closely approximate an isotropic energy profile relative to the target 120 such thai a particular region in the vessel 140 is not exposed (for example, heated) excessively compared to other regions. This allows an object or objects in the vicinity of the target region 130, such as optical elements in the vessel 140 (for example, the optical element 155), and oilier objects in the vessel 140, such as targets other than the target 120 (for example, subsequent or previous targets, such as targets 121a, 121b), and or the buffer fluid 108, to be protected from the plasma. Protecting objects from tlie plasma may increase the useful life of the object, and or make the light, source 101 perform more efficiently and/or reliably.

FIGS. 2A-2D discuss an example target that may be used as the target 120 to produce the plasma that emits EUV light 162. FIGS. 3A-3C, 3E, and 3F discuss examples of a directional flux that may be associated with the plasma.

Referring to FIG. 2 A, a side cross-sectional view (viewed along the direction ^x ) of an exemplary target 220 is shown. The target 220 ma be used in the system 100 as the target 120. Tlie target 220 is inside of a target region 230 that receives a light beam 210. The target 220 includes a target material (such as, for example, fin, litMurn, and/or xenon) thai emits EUV light when converted to plasma. The light beam 210 has energy sufficient to convert at least, a portion of the target ma terial in the target 220 to plasma.

The exemplary target 220 is an ellipsoid (a three-dimensional ellipse). In other words, the target 220 occupie a volume that is approximately defined a the interior of a surface that is a three-dimensional analo of an ellipse. However, the target 220 may have other forms. For example, the target 220 may occupy a volume that has the shape of all or part of a sphere, or the target 220 may occupy an arbitrarily shaped volume, such as a cloud-like form that does not have well-defined edges. For a target.220 that lacks well-defined edges, a volume that contains, for example, 90%, 95%, or more of the target material may be treated as the target 220. The target 220 may be asymmetric or symmetric. Additionally, the target 220 may have any spatial distribution of target material and may include non-target material (material that does not emit EUV light is a .plasma state). The target 220 may he a system of particles and/or pieces, an extended object thai is essentially a continuous and ^■' homogenous material, collection of particles, (including ions and/or electrons), a spatial disu fcutioii of material that includes continuous segments of molten metal, pre-plasma, and particles, ^■ and/or a segment of ^■ molten metal. The contents of the ^■ target ^" 220 - may have an spatial distribution. For example, the target 220 may be homogeneous in one or more directions. In some implementations, the contents of the target 220 are concentrated in a particular portion of the target 220 and the target 220 lias a non-uniform distribution of mass.

The target material can be a target mixture that includes a target substance and impurities such as non-target particles. The target substance is the substance that, when in a plasma state, has an emission line in the EUV range. The target substance 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. Tlie target substance can be, 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 substance can be the element tin. which can be used as pure tin (Sn); as tin compound, for example, SnBr ₄ , SnBr ₂ , SnH ₄ ; as a tin alloy, for example, tin-gallium alloys, tin-indium alloys, tin-indium-galliimi alloys, or any combination of these alloys. Moreover, in the situation in which there are no impurities, t e target material includes only the target substance.

The side cross-section of the target 220 shown in FIG. 2A is an ellipse with a major axis, which has a length equal to the largest distance that spans the entire ellipse, and a minor axis, which is perpendicular to the major axis. The target 220 has a first extent 222 that extends alon a direction 221, and a second extent 224 that extends along a direction 223 that is perpendicular to the direction 221. For the exemplary target 220, the extent 222 and the direction 221 are tlie length and direction, respectively, of the minor axis, and the extent 224 and the direction 223 are the length and direction, respectively, of tlie major axis.

Referring also to FIG. 2B, a front cross-sectional view of the target 220, viewed along the direction 221. is shown. Tlie target 220 has an elliptical y shaped front cross-section with the major axis extending in the direction 223 and having the extent 224. The front cross-section of the target 220 has m extent 226 in a third dimension in a direction 225. The direction 225 is perpendicular to the directions 221 and 223.

Referriri to FIG. 2A, the extent 224 of the target 220 is tilted relative to the direction 212 Of propagation of the light beam 210. Referring also to FIG, 2C, the d reetioa 223 of the extent 224 forms an angle 227 with the direction 212 of propagation of the ligh beam 2 JO. The angle 227 is measured relative to the light beam .210 as it travels in the direction 212 and impinges on the target 220. The angle 227 may be 0-180 degrees, in FIGS. 2A and 2C, the target 220 is tilted with the direction 223 being less than 90 degrees relative to the direction 212. FIG. 2D shows an example in which the angle 227 is between 90 and 180 degrees.

As discussed above, the target 220 may have other forms besides an ellipsoid. For targets that occupy a volume, the shape of the target may be considered to be a three-dimensional form. The form may be described with the three extents 222, 224. 226, which extend along the three mutually orthogonal directions 221, 223, 225, respectively. The lengths of the extents 222, 224, 226 may be the longest length across the form, fi om one edge of the form to an edge on another side of the form, in a particular direction that corresponds to one of the directions 221, 223, 225. The extents 222, 224, 226 and their respective directions 221, 223, 225 may be determined or estimated from visual inspection of the target 220. For example, the target 220 may be used as the target 120 in the system 100. in these implementations, visual inspection of the target 220 may occu by, for example, imaging the target 220 as it leaves the target material supply apparatus 116 and travels to the target regio 130 (FIG. 1).

In some implementations, the directions 221, 223. 225 may be considered to be mutually orthogonal axes that pass through the center of mass of the target 220 and correspond to the principal axes of inerti for the target 220. The center of mass of the target 220 is the point in space where the relative position of the mass of the target 220 is zero. In other words, the center of mass is the average position of the material that makes up the target 220. The center of mass does not necessary coincide with the geometric center of the target 220, but may when the target is a homogenous and symmetric volume.

The center of mass of the target 220 may be expressed as a function of products of inertia, which are a mea sure of imbalance of the spatial distributio of mass in the target 220. The products of inertia may be expressed as a matrix or a tensor. For a three-dimensional object, three mutually orthogonal axes mat pass through the center of mass exist for which the products of inertia are zero. That is, -iheproduct of inertia lies along a direction iii wliicii the mass is equally balanced on either side of a vector that extends along ^' that direction. The directions of flie products of in rtia imy be referred to as the prmcipal axes of inertia of the feee-dimensional object. The directions 221, 223, 225 may be tne principal axes of inertia for the target 220. In this implementation, the directions 221 , 223, 225 are the eigenvectors of the inertia! tensor or matrix of die products of inertia for tiie target 220. Tiie extents 222, 224, 226 may be deieneined from the eigenvalues of the inertia! tensor or matrix of the products of inertia.

In some implementations, the target 220 may be regarded as an approximately two- dimensional object. When the target 220 is two-dimensional, the target 220 may be modeled with two orthogonal principal axes and two extents along the directions of the principal axes. Alternatively or additionally, as for a three-dimensional target the extents and directions for a two-dimensional target may be determined through visual inspection.

The spatial distribution of energy emitted from plasma formed from the target material of a target such as the target 220 depends on the positioning or orientation of the target and/or the spatial distribution of the target material in the target. The position of the tar get is the location, arrangement, and or orientation of the target relative to an irradiating light beam and/or an object in the vicinity of the target. The orientation of the target ma be considered to be the

arrangement and/or angle of the target relative to an irradiating light beam and/or to an object in the vicinity of die target. The spatial distribution of the target is the geometric arrangement of the target material of the target.

Referring to FIG. 3A, an exemplary energy distribution 364A is shown. In the example of FIG. 3 A, the solid line depicts the energy distribution 364A. The energy distribution 364A is the angular distribution of energy emitted from a plasma formed from the target material in a target 320A. The energy is emitted from the plasma has a peak or a maximum in a direction along an axis 363. The direction along which the axis 363 extends (and thus the direction in which the energy is primarily emitted) depends on the positioning of the target 320A and/or the spatial distribution of target material in the target 320 A. The target 320A may be positioned such that an extent of the target in one direction forms an angle relative to a direction of propagation of a light beam. In another example, the target 320 A may be positioned relative to the most intense portion of the light beam, or the target 320A positioned with an extent of the target at an angle relative to an object in a vacuum chamber. The energy distribution 36 A is provided as as esaiaple, and other energy distributions may have different spatial characteristics. FIGS . 3B, 3 ,.3E, and 3P stow additional examples of spatial energy distributions.

Referring to FIGS. 3B and 3C, respectively, exemplary energy distiibtitioiis 364B and 364C with respective peaks (or maximums) 365B, 365C are shown. The .energy distributions 36 B, 364C represen a spatial distribution of energy emitted from plasma formed by an interaction between- a light beam 310, wliich propagates in the z direction at the target region 330, and target material in a target 32GB, 320C, respectively. The interaction converts at least some of the target material in the target 320 to plasma. The spatial distributions of energy 364B and 3 4C may represent the angular spatial distribution of the average energy or the total energy emitted from the plasm .

The target material of the targets 320B, 320C is arranged in a disk-like shape, such as an ellipsoid (similar to the target 220 of FIGS. 2 A and 2B) wife an elliptical cross-section in fee x-y plane. The target 320B has an extent 324 in the y direction, and an extent 322 in the z direction. The extent 324 is greater than the extent 322. In the example of FIG. 3B, the extent 322 is parallel to the direction of propagation of the light beam 310, and the target 320 is not tilted relative to the light beam 310. In the example of FIG. 3C, the target 320C is tilted relative to the direction of propagation of the light beam 310. For the target 320C, the extent 324 is along a direction 321, which is tilted at an angle 327 from the direction of propagation of the light beam 310. The extent 322 is along a direction 323. Tims, the example of FIGS. 3B and 3C shows targets that are positioned in two different, ways, and the energy distributions 364B and 364C show how the peaks 365B, 365C can be moved by changing the target position.

The plasma formed by Hie interaction between the target material and the light beam 310 emits energy, including EUV light, particles, and radiation other than EUV Iight. The particles and radiation may include, for example, ions (charged particles) formed from the interaction between the light beam 310 and the target material. The ions may be ions of the target material . For example, when the target material is tin, the ions emitted from the plasma may be tin ions. The ions may include high-energy ions that travel a relatively long distance from the target. 120, and relatively low-energy ions that have! a shorter distance from the target 120. The high- energy ions transfer their kinetic energy as heat into material that, receives them and create localized regions of heat in the material. A high-energy ion may be an ion that has an energy equal to or greater than, for example, 500 electron volts (eV). A low-energy ion may be an ion that lias an energy less than 500 eV,

As discussed above, tlie example distributions 364B and 36 C of FIGS. 3B and 3C, respectively, may be considered to show the spatial distribution of the total or average energy of the ions that are emitted from the plasma, hi the example of FIG: 3B, the energy caused by emission of the ions has &e distribution 364B in the y-z plane. The dis bution 364B represents the relative amount of energy emitted from the plasma as a function of angle relative to the center of the target 32QB. In the example of FIG. 3B, the extent 324 is perpendicular to the direction of propagation of the light beam 310 at the target region 330, and the greatest amount of energy are delivered in the direction of the peak 365B. In the example of FIG. 3B, the peak 365B is in the -z direction, which is parallel to the extent 322 and perpendicular to the extent. 324. The lowest amount of energy is emitted in the z direction, and it is possible that the low- energy ions are preferentially emitted in the z direction.

Relative to FIG. 3B, the position of the target 320C (FIG . 3C) is different, hi the example of FIG. 3C, the extent 324 is tilted at the angle 327 relative to the direction of

propagation of the beam 310. The profile 364B of the total or average ion energy is also different in the example of FIG. 3C, with the greatest amount of energy being emitted toward the peak 365C. As with the example of FIG. 3B, in the example of FIG. 3C, ions may be preferentially emitted along direction that extends away from a side 329 of the target 320 that receives the light beam 310 and is normal to the extent 324. The side 329 is the portion or side of the target 320 that receives the light beam 310 before any other portion of the target 320 or the portion or side of the target 320C that receives the most mdiation from the light bea 310. The side 329 is also referred to as the "heating side/'

Other particles and radiation emitted from the plasma may have a different profile in the y-z plane. For example, a profile may represent the profile of high-energy ions or low-energy ions. The low-energy ions may be preferentially emitted in a direction that is opposite to the direction in which the high-energy ions are preferentially emitted.

The plasma created by the interaction of the targets 320B, 320C and the light beam 310 thus emits a directionally dependent flux of mdiation and/or particles. The direction in which the highest portion of the radiation and/or particles is emitted depends on the position of the target 320B, 320C. By adjusting or changing the position or orientation of the target 320, the direction in which fee greatest amount of radiation and/or particles is emitted is also changed, allowing tlie heating effects of the drrectionally dependent flux on other objects to he minimized or eliminated.

The spatial distribution of energy emitted from the plasma also ma be chang b changing the relative position of the target a.nd the light beam.310.

FIG; ^' 3D shows an example intensity profile for the light beam 310, The intensity profile 350 represents the intensity of the light beam 310 as a fimetion of position in the x-y plane, which is perpendicular to the direction of propagation at the target region 330 (the direction z). The intensify profile has a maximum 351 in the x-y plane along an axis 352. The intensify decreases on either side of the maximum 351.

FIG. 3E and FIG. 3F show a target 320E and a target 320F, respectively, interacting with the light beam 310. The targets 320E and 320F are substantially spherical and contain target material that emits EUV light when hi a plasma state. The target 320E (FIG. 3E) is at a location 328E, which i displaced from the axis 352 in the x direction. The target 320F (FIG. 3F) is at a location 328F, which is displaced from the axis 352 in the -x direction. Tims, the targets 320E and 320F are on different sides of the axis 352. The portion of the target 320E, 320F closest to the axis 352 (which is the most intense portion of the light beam 310) evaporates and converts to plasma before the remaining portions of the target 320E, 320F. The energy of the plasma generated from the target 320E is primarily emitted from the portion of the target 320E that is closest to the axis 352 and in a direction that is toward the axis 352. In the example shown, the energy emitted from the plasma generated from the target 320E is primarily emitted along a direction 363E, and tlie energy emitted from the plasma generated from the target 320F is primarily emitted along a direction 363F. The directions 363E, 363F are different from each other. As such, the relative placement of the target and the light beam also may be used to direct the energy emitted from the plasma in a particular direction. Additionally, although the targets 320E. 320F are shown as being spherical, targets of other shapes emit plasma direetionally based on the " locatio relative to the light beam 310.

FIGS. 3 A-3C show the profiles 364A-364C, respectively, in the y-z plane and in two dimensions. However, it is contemplated mat the profiles 364A-364C may occupy three dimensions and may sweep out a volume in three dimensions. Similarly, the energy emitted from the targets 320E and 320F may occupy a three-dimensional volume. FIG; 4 is a -block diagram of a system 400 that can control Hie position -of targets during use of an EUV light -source; FIG. 5 is a flow chart of an exemplary process 500 for controlling the positioning of a target dining use of an EUV light -source. FIGS. 6A-6C illustrate an ..example of the process 500 for a target.

The control system 470 is used to reduce or eliminate the effects of a plasma 442, which is generated in a vacuum- chamber 440, on as object 444 in the vacuum chamber 440. Tlie plasma 442 is produced from an interaction between a light beam arid target material at a target region i the vacuum chamber. Tlie target material is released into the vacuum chamber 440 from a target source, and the target material travels from the target source (such as tlie target material supply apparatus 116 of FIG. 1) to the target region along a trajectory. The object 444 can be any object hi the vacuum chamber 440 that is exposed to the plasma 442. For example, the object 444 can be another target for producing additional plasma, an optical element in the vacuum chamber 440, and/or a fluid 408 that flows in the vacuum chamber 440.

The system 400 also includes a sensor 448, which observes the interior of the vacuum chamber 440. The sensor 448 may be located in the vacuum chamber 440 or outside of tlie vacuum chamber 440. For example, t e sensor 448 may be placed outside of the vacuum chamber at a viewport window that allows visual observation of the interior of the vacuum chamber 440. The sensor 448 is capable of sensing the presence of target material hi the vacuum chamber. In some implementations, the system 400 includes an additional light source that produces a light beam or a sheet of light that intersects the trajectory of the target material. The light of the light beam or the sheet of light is scattered by the target material, and the sensor 448 detect the scattered light. The detection of the scattered light may be used to determine or estimate the location of the target material in the vacuum chamber 440. For example, fee detection of the scattered light indicates that the target material is in a location that where the light beam or light sheet intersects the expected target material trajectory. Additionally or alternatively, the sensor 448 may be positioned to detect the light sheet or light beam, and the temporary blocking of the light sheet or light beam by the target material may be used as an indication that the target material is in a location that where the light beam or light sheet intersects fee expected target materiai trajectory.

Tlie sensor 448 may be a camera, photo detector, or another type of optical sensor that is sensitive to wavelengtiis in the light beam or light sheet that intersects the trajectory of the target material The sensor 448 produces a representation of the iateriof of the vacuum chamber -440 (for example, a ^■' representation that indicates the detection of scattered light or an indication of Eglit being blocked), and provides the representation to the control system 470. From the representation, the control system 70 may determine or estimate the location of the target material within the vacuum chamber 440 and declare that the target, material is in a certain portion of the vacuum chamber 440. The / ^' location where the light beam or light sheet, intersects the expected target material trajectory may be at any part of the trajectory. Further, in some implementations, other techniques for detemihmig that the target material is in a particular portion of the vacuum chamber 440 may be used.

The system 400 includes a control system 470 that communicates with light-generation module 480 to provide one or more light beams to a vacuum chamber 440. In the example shown, the light generation module 4S0 provides a first light beam 410a and a second light beam 410b to the vacuum chamber 440. hi other examples, the light generation module 480 can provide more or fewer light beams.

The control system 470 controls the timing and or direction of propagation of pulses of light emitted from the light-generation module 480 such that the positioning of a target in the vacuum chamber 440 can be changed from target-to-target. The control system 470 receives the representation of the interior of the vacuum chamber 440 from the sensor 448. From the representation, the control system 470 may determine whether target material is present in the vacuum chamber 440 and'or the position of the target material in the vacuum chamber 440. For example, the control system 470 may determine that target material is in a particular location of the vacuum chamber 440 or in a particular location in tlie vacuum chamber 440. When target material is determined to be in the vacuum chamber 440 or in a particular location in the vacuum chamber 440, the target material may be considered to be detected. Tlie control system 470 may cause pulses to be emitted from the light- generation module 480 based on a detection of target, material. The detection of target material may be used to time the emission of pulses from the light-generation module 480. For example, the emission of a pulse may be delayed or advanced based on detecting target material in a particular portion of the vacuum chamber 470. In another example, the direction of propagation of a pulse may be determined based on the detectio of tar set material. The control system 470 includes a light beam control module 471, a flow control module 472, an electronic storage 473* an electronic processor 474, and an input/output interface 475. The electronic processor 474 includes one or more processors suitable for the execution, of a computer program such as a general or special prtrpase microprocessor, and any one or more processors of any kind of digital computer. Generally, an electronic processor receives

instructions and data from a read-only memory or a random access memor or bom. The electronic processor 474 can be any type of electronic processor.

T!ie electronic storage 473 can be volatile memory, such as RAM, or non-volatile memory. In some implementations, and the electronic storage 473 can include non-volatile and volatile portions or components. The electronic storage 473 can store data and forrnation that is used in the operation of the control system 470 and or components of the control system 470. For example, the electronic storage 473 can store timing information that specifies when the first and second beams 410a, 410b are expected to propagate to specific locations in the vacuum chamber 440, a pulse repetition rate for the first and/or second beams 410a, 410b (in

implementations hi which the first and/or second beams 410a, 410b are pulsed light beams). and/or information that specifies a direction of propagation for the first and second beams 410a, 410b in the vkiirity of die target (for example in a target region such as the target region 330).

The electronic storage 473 also can store instructions, perhaps as a computer program, that, when executed, cause the processor 474 to communicate with components in the control system 470, the light-generation module 480, and/or the vacuum chamber 440. For example, the instructions can be instructions that cause the electronic processor 474 to provide a trigger signal to the light-generation module 480 at certain times that are specified by the timing information stored on the electronic storage 473. The trigger signal can cause the light generation module 480 to emit a beam of light. The timing information stored on the electronic storage 473 may be based on information received from the sensor 448, or the timing information may be predetermined tinting infonnation that is stored on the electronic storage 473 when the control system 470 is initially placed into service or through the actions of a human operator.

The I/O interface 475 is any kind of electronic interface that allows the control system 470 to receive and/or provide data and signals with an operator, the light-generation module 480, the vacuum chamber 440, and or an automated process running on another electronic device. For example, the I/O interface 475 can include one or more of a visual display, a keyboard, or a coiBiBii icatioHS interface.

The light beam control module 471 coffimiinicates with the light-generation module 480, the electronic storage 473, and/or the electronic processor 474 to direct pulses of ^' light into the vacuum chamber 440.

The light generation module 480 is any device or optical source that, is capable of producing- pulsed light beams, at least some of which have energy sufficient to convert target material to plasma that emits EUV light. Additionally, the light-generation module 480 can produce other light beams that do not necessarily transform target material to plasma, such as light beams that are used to shape, position, orient, expand, or otherwise condition an initial target into a target that is converted into plasma that emits EUV light.

In the example of FIG. 4, the light-generation module 480 includes two optical subsystems 481a. 481b, which produce first and second light beams 410a, 410b, respectively. In the example of FIG. 4, the first, light, beam 410a is represented by a solid line and the second light beam 410b is represented by a dashed line. The optical subsystems 481a, 481b can be, for example, two lasers. For example, the optical subsystems 481a. 481b can be two carbon dioxide (C0 ₂ ) lasers. In other implementations, the optical subsystems 481a, 481b can be different types of lasers. For example, the optical subsystem 481a can be a solid state laser, and the optical subsystem 481b can be a C0 ₂ laser. Either or both of the first and second light, beams 410a, 410b can be pulsed.

The first and second light beams 481a, 481b can have different wavelengths. For example, in implementations in which the optical subsystems 481a, 481b include two C0 ₂ lasers, the wavelength of the first light beam 410a can be about 10.26 micrometers (μηι) and the wavelength of the second light beam 410b can be between 10.18 μηι and 10.26 pm. Tlie wavelength of the second light beam 410b can be about 10.59 «m. In these implementations, tlie light beams 410a, 410b are generated from different lines of the C0 ₂ laser, resulting in the light beams 410a, 410b having different wavelengths even though both beams are generated from the same type of source. Hie light beams 410a, 410b also can have different, energies.

The light-generation module 480 also includes a beam combiner 482, which directs the first and second beams 410a, 410b onto a beam path 484. The beam combiner 482 can be any optical element or a collection of optical elements capable of directing the first and second beams 410a, 410b onto the beam path 484. For example, the beam combiner 482 can be a collection of.miirors, some of which are positioned to direct tlie first beam.410a onto the ^' beam path.484 and others of which, are po sitioned to direct the second beam.410b onto the beam path 4 _. 84. The. light-generation module 480 also can include a pre-anrplifier 483, which amplifie the first and second beams 410a, 1 Ob within the _: ^' light-generation . module 480.

The first and second beams 410a, 410b can propagate on the path 484 at different times, ίίί the example shown in FI6. 4, the first and second beams 410a, 410b follow the pat 484 in the light-generation module 480, and both beams 410a, 410b traverse substantially the same spatial region through the optical amplifier 483. In other examples, the beams 410a and 410b can travel along different paths, including through two different optical amplifiers.

The first and second light beams 10a, 410b are directed to the vacuum chamber 440.

T!ie first and second beams 410a, 410b are angularly disbursed by a beam delivery system 485 such that the first beam 410a is directed toward an initial target region, and the second beam 410b is directed toward a target region (such as the target region 130 of FIG. 1). The initial target region is a volume of space in the vacuum chamber 440 that receives the first light beam 410a and initial target material, which is conditioned by the first light beam 410a. The target region is a volume of space hi the vacuum chamber 440 that receives the second light beam 410b and a target that is converted into plasma. The initial target region and the target region are at different locations within the vacuum chamber 440. For example, and referring to FIG. 1, the initial target region can be displaced in the -y direction relative to the target region 130 such that the initial target region is between the target region 130 and the target material supply 1 16. The initial target region and the target region can partially spatially overlap, or the initial target region and the target region can be spatially distinct without any overlap. FIG. 14 includes an example of first and second light beams bein displaced from each other within a vacuum chamber, in some implementations, the beam delivery system 485 also focuses the first and second beams 410a, 410b to locations within or near the initial and modified target regions, respectively.

In other implementations, the light-generation module 480 includes a single optical subsystem that generates both the first and second light beams 410a, 410b. In these

implementations, the first and second light beams 10a. 410b are generated by the same optical source or device. However, the first and second light beams 410a, 410b can have the same wavelength or different wavelengths. For example, the single optical subsystem can be a carbon dioxide (CO ?) laser,, and the first and second light beams 410a, 410b can be generated by different lines of the€Q ₂ laser and can be different wavelengths.

In some implementations, the light-generation module 480 does not emit the first ligh beam 410a and there is no initial target region, in these implementations, the target is received in the target region without being pre-conditioned by the first light beam 410a. An example of soc an implementation is shown in FIG, 17.

A fluid 408 can flow in the vacuum chamber 440. The control system 470 also may control the flow of the fluid 408 in the vacuum chamber 440. The fluid 40S may be, for

example, hydrogen and or other gasses. The fluid 408 can be the object 444 (or one of the objects 444 in the case where multiple objects in the vacuum chamber 440 are to be protected from the effects of the plasma 442). In these implementations, the control system 470 also can include a flow control module 472, which controls a flow configuration of the fluid 408. The flow control module 472 can set, for example, the flow rate and or flow direction of the fluid 408.

The light beam control module 471 controls the light generation module 480 and determines whe the first light beam 410a is emitted fiom the light-generation module 480 (and, thus, when the first light beam 410a reaches the initial target region and the target region). The light beam control module 471 also can determine a direc tion of propaga tion of the first light beam 410a. By controlling the timing and or direction of the first light beam 410a, the light beam control module 471 also can control a position of a target and the direction in which particles and/ or radiation are primarily emitted.

FIGS. 5 and 6A-6C discuss a technique for positioning the target using a pre-pulse, or a pulse of light that reaches the target prior to a pulse of radiation that conveits the target material to plasma that emits EUV light.

Referring to FIG. 5, a flow chart of an exemplary process 500 for generating EUV light is shown. The process 500 can also be used to tilt a target (such as the target 120 of FIG. 1, the target 220 of FIG. 2 A, or the target 320 of FIGS. 3 A and 3B). The target is provided at a target region (510). The target has a first extent along first direction and a second extent along a second direction. The target includes target material that emits EUV light when converted to plasma. An amplified light beam is directed toward the target region (520). FIGS. 6A-6C show an example of the process 500. As discussed ^' etow, a target 620 is provided to a target region 630 {FIG. 6€), and an amplified tiglii beam 610 is directed toward the target region 630.

Referring to FIGS. 6 A and 6B, an exemplary waveform 602 transforms an initial targe 618 itito the target 620. The initial, target 6 I -8 and the target 620 include target material that emits EUV light 660 when converted to plasm . through irradiation with an amplified light beam 610 (FIG. 6C). The discussion below provides an example in which the initial target 618 is a droplet made of molten nietal. For example, the initial target 618 can be substantially spherical and have a diameter of 30-35 μηι. However, the initial target 618 can take other forms.

FIGS. 6 A and 6C show a time period 601 during which the initial target 618 physically transforms into the target 620 and then emits EUV light 660. The initial target 618 is

transformed through interaction with the radiation delivered in time according to the waveform 602. FIG. 6B is a plot of the energy in the waveform 602 as a function of time over the time period 601 of FIG. 6A . As compared to the initial target 618, the target 620 has side cross section with an extent that is less in the z direction. Additionally, the target 620 is tilted relative to the z direction (the direction 612 of propagation of the amplified beam 610 that converts a t least part of the target 620 to plasma).

The waveform 602 includes a representation of a pulse of radiation 606 (a pre-pulse 606). The pre-pulse 606 can be, for example, a pulse of the first light beam 41.0a (FIG. 4). The pre-- pulse 606 c an be any type of pulsed radiation that .has sufficient energy to act on the initial target 618, but the pre-pulse 606 does not convert a significant amount of the target material to plasma that emits EUV light. The interaction of the first ^■ pre-pulse 606 and the iniiial target 618 can deform the initial target 618 into a shape that is closer t a disk. After about.1-3 microseconds (its), this deformed shape expands into a disk shaped piece or form of molten metal. Hie amplified light beam 610 can be referred to as the main beam o the main pulse. The ^' amplified light beam 610 has sufficient energy to convert target material in the target 620 to plasma that emits EOT light.

The pre-pulse 606 and the amplified light beam 610 are separated in time by a delay time 611, with the amplified light beam 610 occurring at time t ₂ , which is after the pre-pulse 606. The pre-pulse 606 occurs at a time t=ti and has a pulse duration 615. The pulse duration 615 can be represented by the full width at half maximum, the amount of time that the pulse has an intensity that s at least half of the ¾ maximum intensity of the pulse. However, other metrics can ^" be used to determine the pulse duration 61 .

Before discussing .the teclniique of providing the target 620 to the target region 630, a discussion of the interactions of the pulses of radiation, including the pre-pulse 606. with the initial target 618 is provided.

When a laser ulse impipges (strikes) a .metallic target material droplet, "the leading edge of the pulse sees (interacts with) a surface of the droplet that is a reflective metal. The leading edge of the pulse is the part of the pulse that interacts with the target material first, before any other parts of the pulse. The initial target 618 reflects most of the energy in the leading edge of the pulse and absorbs little. Tlie small amount of light that is absorbed heats the surface of the droplet, evaporating and ablating the surface. Tlie target material that is evaporated from the surface of the droplet forms a cloud of electrons and ions close to the surface. As the pulse of radiation continues to impinge on the target material droplet, the electric field of the laser pulse can cause the electrons in the cloud to move. The moving elections collide with nearby ions, heating the ions through t e transfer of kinetic energy at a rate that is roughly proportional to the product of the densities of the electrons and the ions in the cloud. Through tlie combination of the moving electrons striking the ions and the heating of the ions, the cloud absorbs the pulse.

As the cloud is exposed to the later parts of the laser pulse, the electrons in the cloud continue to move and collide with ions, and the ions in the cloud continue to heat. The electrons spread out. and transfer heat to the surface of the target material droplet (or bulk material that, underlies the cloud), further evaporating the surface of the target material droplet. The electron density in the cloud increases in the portion of the cloud that is closest to the surface of the target material droplet. The cloud can reach a point where the density of electrons increases such that the portions of the cloud reflect the lase pulse instead of absorbing it.

Referring also to FIG. 6C, the initial target 618 is provided at an initial target region 631.

Tlie initial target 618 can be provided at the initial target region 631 by, for example, releasing target material from the target material supply apparatus 116 (FIG. 1). In the example shown, the pre-pulse 606 strikes the initial target 618, transforms the initial target 618, and the transformed initial target drifts or moves into the target region 630 over time.

Tlie force of the pre-pulse 606 on the initial tar get 618 causes the initial target 618 to physically transform into a geometric distribution 652 of target material. The geometric distribution 652 can ^' include, a materia! that is not ionized (a -material mat is nor a lasnia). The geometric -distribution 652 can be, for .example, a disk of liquid or molten metal a continuous segment of target material mat does no have voids or substantial gaps, a mist of micro- or nano- partieles, or a cloud of ^' atomic apor. The geometric distributio 652 .further expands during the dela time 611 and becomes the target 620. Spreading the initial target.618 can have three effects.

First, as compared to the initial target 618, the target 620 generated by the interaction with the pre-pulse 606 has a form that presents a larger area to an oncoming pulse of radiation (such as the amplified light beam 610). The target 620 has a cross-sectional diameter in the y direction that is larger than the cross-sectional diameter in the y direction of the initial target 61S. Additionally, the target 620 c an have a thickness that is thinner in a direction of propaga tion (612 or z) of the amplified light beam 610 at the target 620 than the initial target 618. The relative thinness of the target 620 in the direction z allows the amplified light beam 610 to irradiate more of the target material that is in the target 618.

Second, spreading the initial target 618 out in space can minimize or reduce the

occurrence of regions of excessively high material density during heating of the plasma by the amplified light beam 610. Such regions of excessively high material density can block generated EUV light, if the plasma density is high throughout a region that is irradiated with a laser pulse, absorption of the laser pulse is limited to the portions of the region that receives the laser pulse first. Heat generated by this absorption may be too distant from the bulk target material to maintain the process of evapora ting and heating of the target material surface long enough to utilize (for example, evaporate and/or ionize) a meaningful amount of the bulk target material din ing the finite duration of the amplified light beam 610.

In instances where the region has a high election density, the light pulse only penetrates a fraction of the way into the region before reaching a "critical surface" where the electron density is so high that the light pulse is reflected. The light pulse cannot travel into those portions of the region and little EUV light is generated from target material in those regions. The region of high plasma density can also block EUV light that is emitted from the portions of the region that do emit EUV hght. Consequently, the total amount of EUV light that is emitted from the region is less than it would be if the region lacked the portions of high plasma density. As such, spreading the initial target 618 into the larger volume of the target 620 means that an incident light beam reaches more of the material in the target 620 before being: reflected. This can increase the amount of EUV light produced.

Tiiird, the interaction of ilie pre-pulse 606 and the initial target 618 causes the target 620 to arrive at the target region 630 tilled at an angle 627 with, respect to the direction of

propagation 612 of the amplified light beam 610. The fflitiai target 618 has a center of mass 619, and the pre-pulse 606 strikes the initial target 618 sueh that the majority of the energy in the pre- pulse 606 fells on one side of the center of mass 619. The pre-pulse 606 applies a force to the initial target 618. and, because the force is on one side of the center of mass 61 . the initial target 618 expands along a different set of axes than the target would if the pre-pulse 606 struck the initial target 618 at the center of mass 619. The initial target 618 flattens along the direction from which it is hit by the pre-pulse 606. Thus, striking the initial target 618 off-center or away from the center of mass 619 produces a tilt. For example, when the pre-pulse 606 interacts with the initial target 618 away from the center of mass 619, the initial target 618 does not expand along the y axis and instead expand along an axis y', which is tilted at an angle 641 relative to the y axis while moving toward the target region 630. Thus, after the time period has elapsed, the initial target 618 has transformed into the target 620, which occupies an expanded volume and is tilted at the angle 627 with respect to the direction 612 of propagation of the amplified light beam 610.

FIG. 6C shows side cross-section of the target 620. The target 620 has an extent 622 along a direction 621 and an extent 624 along a direction 623, which is orthogonal to the direction 621. The extent 624 is gr eater than tlie extent 622, and the extent 624 forms the angle 627 with the direction 12 of propagation of the amplified light beam 610. The target 620 can be placed so that part of the target 620 is hi a focal plane of the amplified light beam 610, or the target 620 can be placed away from the focal plane. In some implementations, the amplified light beam 610 can be approximated as a Gaussian beam, and the target 620 can be placed outside of the depth of focus of the amplified light beam 610.

In the example hown in FIG. 6C, the majority of the intensity of the pre-pulse 606 strikes the initial target 618 above (offset in the -y direction) the center of mass 619. causing tlie target material in the initial target 618 to expand along the y ^* axis. However, in other examples, the pre-pulse 606 can be applied below (offset in the y direction) the center of mass 619, causing the target 620 to expand along an axis (not shown) that is counterclockwise compared to the y' axis. Is -file example shown is FIG. 6C, the initial target 618 drifts through the initial targe region 631 white traveling along the direction. Thus, the portion of the initial target 618 upon. ^■ which the pre-pulse 606 is incident can be controlled wit the timing of the pre-pxilse 606. For example, releasing the pre-puise 606 at a earlier time than the example shown in FIG. 6C (that is, increasing the delay .time 61 1 of FIG. 6B), causes the pre-pulse 606 to strike the lower portion of the initial target 618.

The pre-pnlse 606 can be any type of radiation that can act on the initial target 618 to form the target 620. For example, the pre-pnlse 606 can be a pulsed optical beam generated by a laser. The pre-pulse 606 can have a wavelength of 1-10 μιη. The duration 612 of the pre-puise 606 can be, for example, 20-70 nanoseconds (ns), less than 1 ns, 300 picoseconds (ps), between 100-300 ps, between lQ-50ps, or between 10-100 ps. The energy of the pre-pulse 606 can be, for example. 15-60 mi li Joules (mJ), 90-110 mJ, or 20-125 ml. When the pre-pnlse 606 has a duration of 1 ns or less, the energy of the pre-pulse 606 can be 2 mJ. The delay time 611 can be, for example, 1-3 microseconds (.us).

The target 620 may have a diameter of, for example, 200-600 urn, 250-500 μηι, or 300-

350 pin. The initial tar get 18 may travel toward the mitial target region 631 with a velocity of, for example, 70-120 meters per second (mis). The initial target 618 may travel at a velocity of 70 ni/s or 80 m/s. The target 620 may travel at a higher or lower velocity than the initial target 610. For example, the tar get 620 may travel toward the target region 630 at a velocity that 20 m s faster or slower than the initial target 610. In some implementations, the target 620 travels at the same velocity as the initial target 610. Factors that influence the velocity of the target 620 include the size, shape, and/or angle of the target 620. The width of the light beam 610 at. the target region 630 in the y direction may be 200-600 pin. In some implementations, the width of the light beam 610 in the y direction is approximately the same as the width of the target 620 in the y direction at the target region 630.

Although the waveform 602 is shown as a single waveform as a function of time, various portions of the waveform 602 can be produced by different sources. Furthermore, although the pre-pulse 606 is shown as propagating in the direction 612, this is not necessarily the case. The pre-pulse 606 can propagate in another direction and still cause the initial target 618 to tilt. For example, the pre-pulse 606 can propagate in a direction that is at the angle 627 relative to the z direction. When the pre-pulse 606 travels in this direction and impacts the mitial tar get 618 at the center of mas 619, th initial target 618 expands along tlie y' axis and. is ^' tilted Thus, in some .implementations,, the. initial . target ^' 6 Ϊ 8 ca be tilted relative to the direction of propagation of the amplified light beam 610 by striking the initial target 618 on-center or at the center of mass 619. Striking the initial target 618 in this manner causes the initial target 618 to flatten or expand along a direction that is perpendicular to the direction in which die pre-pulse 606 propagates, thus angling or tilting the initial target 618 relative to the z axis. Additionally, in other examples, the pre-pnlse 606 can propagate in oilier directions (for example, out of the page of FIG. 6C and along the x axis) and cause the initial target 618 to flatten and tilt relative to the z axis.

As discussed above, the impact of the pre-palse 606 on tlie initial target 61 S deforms tlie initial target 618. in implementations in which the initial target 618 is a droplet of molten metal, the impact transforms the initial target 618 into a shape that is similar to a disk, the disk expands into the target 620 over the time of the delay 611. The target.620 arrives in the target region 630.

Although FIG. 6C illustrates an implementation in which the initial target 618 expands into tlie target 620 over the delay 611, in other implementations, tlie target 620 is tilted and expanded along a directio that is orthogonal to the direction of propagation of the pre-pnlse 606 by adjusting the spatial position of tiie pre-pulse 606 and the initial target 618 relative to each other, and without necessarily using the delay 611. In this implementation, tlie spatial position of the pre-pulse 606 and the initial target 618 are adjusted relative to each other. Due to this spatial offset, an interactio between the pre-pulse 606 and the initial tar get 618 causes the initial target 618 to tilt in a direction that is orthogonal to the direction of propagation of the pre-pulse 606. For example, the pre-pulse 606 can propagate into the page of FIG. 6C to expand and tilt the initial target.618 relative to the directio of propagation of the amplified ligh t beam 610.

FIG. 8 discusses an example of causing a position of at least two targets in a stream of droplets to be different. Before turning to FIG. 8, FIGS. 7 A and 7B provide an example of a system in which the position of a target remains the same over time (that is, each target that arrives in the target, region has substantially the same orientation and or position in the vacuum chamber).

Referring to FIGS. 7 A and 7B, an interior of an exemplary vacuum chamber 740 is shown at two times. The example of FIGS. 7A and 7B illustr ates the effect of a directionally dependent flux of particles and or radiation associated with a plasma on objects in the vacuum chamber 740 when- the ositions of the- targets that eater the target region is not varied or .changed over time by the control system 470, In the example of FIGS. 7 A and 7B, the objects are a fluid 708 and targets 720 in a stream 722,

The fluid 70S is between a target region 730 and an optical element 755 and is intended to act as a buffer that protects the optical element 755 from the plasma. The fluid 708 .may be a gas, such as, for example, hydrogen. The. fluid 708 may be introduced into the : vacuum chamber 740 by a fluid delivery system 704. The fluid 708 has a flow configuration, which describes the intended characteristics of the fluid 70S. The flow configuration is intentionally selected such that, the fluid 708 protects the optical element 755. The flow configuration may be defined by, for example, a flow rate, a flow direction, flow location, and/or a pressure or density of the fluid 708. in the example of FIG. 7 A, the flow configuration results in the fluid 708 flowing through t e region between the tar get region 730 and the optical element 755 and feinting a uniform volume of gas between the target region 730 and the optical element 755. The fluid 708 may flow in any direction. In the example of FIG. 7 A, the fluid 708 flows in the y direction based on the flow configuration.

Referring also to FIG. 7B, the interaction between the target 720 and the light beam 710 produces the directionaily dependent flux of particles and/or radiation. The distribution of the particles andor radiation is represented by the profile 764 (FIG. 7B). The distribution profile 764 is substantially the same shape and position for each target 720 that is converted to plasma in the target region 730. The particles and or radiation emitted from the plasma enter the fluid 708 and may change the flow configuration. These changes can result in damage to the optical element 755 and/or changes to the trajectory 723.

For example , as discussed above, the directionaily dependent flux of particles and/or radiation may include high-energy ions that are primarily emitted in a direction that is determined by the position of the target 720, which remains constant for all targets entering the target region 730 for the example of FIGS. 7A and 7B. High-energy ions released from the plasma travel in the fluid 708, and may be stopped by the fluid 708 before reaching the optical element 755. Ions stopped in the fluid transfer kinetic energy into the fluid 708 as heat. Because the majority of the high-energy ions are emitted in the same direction and travel approximately the same distance into the fluid 708, the high-energy ions can form a heated localized volume 757 within the fluid 708 that is wanner than the rest of the fluid 70S. The viscosity of the fluid 708 increases with temperature. Thus, the viscosity of the fluid hi the heated loeahzed volume 757 is greater than the viscosity of the surrounding fluid 708. Due to the higher viscosity, fluid flowing toward the volume 757 experiences a greater resistance HI the ^■ volume 757 than the surroimdmg region. As a result, the fluid tends to flow around the volume 757, deviating from the intended So w coaiiguratio of the fluid 70S.

Additionally, hi instances hi which the heated localized volume 757 arises from metallic ion deposits, the volume 757 may include a gas that contains a high amount of the metallic material that produced tlie ions. In these instances, if the direction of the profile 764 remains constant over time, the amount of metallic material in the volume 757 can become so high that the flowing fluid 708 is no longer able to carry the metal material away from the volume 757. When the fluid 708 is no longer able to carry the metallic material a way from the volume 757, the metallic material can escape from die volume 757 and impact a region 756 of the optical element 755, resulting in contamination of the region 756 of the optical element 755. The region 756 can be referred to as the "contamination region/'

Referring also to FIG. 7C. the optical element 755 is shown. The optical element 755 includes a reflective surface 759 and an aperture 758 through which the light beam 710 propagates. The contamination region 756 is formed on a portion of the reflective surface 759. The contamination region 756 can be any shape and can cover any portio of the reflective surface 759, but the location of the contamination region 756 on the reflective surface 759 depends on the distribution of tlie directional flux of particles and/or radia tion .

Referring to FIG. 7B, the prese ce of the heated localized volume 757 also may change the location and/or shape of the trajectory 723 by changing the amount of drag o the targets that travel o the trajectory 723. As shown in FIG. 7B, in the presence of the heated localized volume 757, the targets 720 may travel on a trajectory 723B, which is different from the expected trajectory 723. By traveling on the changed trajectory 723B, the targets 720 may arrive in the target region 730 at the wrong time (for example, when the light beam 710 or a pulse of the light beam 710 is not in the target region 730) and or not arrive at the target region 730 at all, leading to reduced or no EUV light production.

Thus, it is desirable to spatially distribute the heating caused by the directional flux of particles and/or radiation. Referring to FIG. 8, an exemplary process 800 for varying the position of a target that arrives in the target region as compared to the position of other targets that arrive in the target region is shown. In this way, fee target position is considered to be varied over time, and any of die positions of the targets can be different from the positions of the other targets. By varying the positions of the various targets, the heat produced by the plasma is spread out. in space, thereby protecting an object in a vacuum chamber from the effects of the plasma. The process can be performed b the cojitrol system 470 (TIG. 4). The process 800 can he used to reduce the effect of plasma on one or more objects in a vacuum, chamber in which plasma is formed, such as a vacuum chamber of an EUV light source. For example, the process 800 can be used to protect objects in the vacuum vessel 140 (FIG. 1), 440 (FIG. 4), or 740 (FIG.

7>- FIGS. 9A-9C are an example of using the process 800 to protect the fluid 708 (by ensuring that the fluid 708 remains in its intended flow configuration) and the optical element 755 by varying the position of the target 720. Although the process S00 can be used to protect any object in a vacuum chamber from the effects of plasma, the process 800 is discussed with respect to FIGS. 9A-9C for purposes of illustration.

A first target is provided to an interior of a vacuum chamber (810). Referring also to

FIG. A, at the time tl , the target 720A is provided to the target region 730. The target 720A is an instance of the target 720 {FIG. 7A). The target 720A is an example of a first target. The target 720A includes target material arranged in a geometric distribution. The target materia! emits EUV light when in a plasma state, and also emits particles and/or radiation other than EUV light. The geometric distribution of target material in the target 720A has a first extent in a first direction, and a second extent hi a second direction, which is perpendicular to the first direction. The first and second extents can be different. Referring to FIG. A. the target 720A has an elliptical cross -section in the y-z plane, and the larger of the first and second extent is along a direction 923A. As discussed below, the instances 720B and 720C of the target 720 at the later times of t2 and t3 (FIGS . 9B and 9C, respectively) have a different position than the instance 720A at the time tl (FIG. 9 A). The targets 720B and 720C have substantially the same

geometric distribution of target material as the target 720 A. However, the position of the targets 720A, 720B, 720C is different. As shown in FIG. 9B, at the time t2, the target 720B has the larger extent along a direction 923B, which is different from the direction 923A. At the time t3 (FIG. 9C), the target 720C has the larger extent along a direction 923C, which is different from 923A and 923B. Providipg any of the. targets 720A, 72GB, 720C to the target region 730 eaa include shaping, positioning, and/or orienting the target prior to the target reaching the target region 730. For example, and referring also to FIGS. 10A, and 1QB, the target material snpply apparatus 716 can provide ^' an initial target Ϊ 018 to an initial target region 1031. i the example of FIGS. 1ΘΑ and IOB, the initial target region 1031 is between the target region 730 and the target material supply apparatus 716. hi Hie example of FIG. l O A, a target 920A is formed. In the example of FIG. IOB, a target 920B is formed. The target 920A and 920B are similar but are positioned differently in the vacuum chamber, as discussed below.

Referring to FIG. 10A, the control system 470 causes a pulse of the first optical beam 410a to propagate toward the initial target region 1031. The control system 470 causes the pulse of the first optical beam 410a to be emitted at a time such that the first optical beam 410a arrives in the initial target region 1031 when the initial target 1018 is in the initial target region 1031 but positioned such that the first light bea 410a strikes the initial target above (displaced in the— y direction) the center of mass 1019. For example, the control system 470 may receive a representation of the interior of the vacuum chamber 740 from the sensor 448 (FIG. 4} and detect that a the initial target 1018 is near or in the initial target region 1031 and then cause the emission of the pulse of the first light beam 410a based on the detection such that the first light beam 410a is displaced in the -y direction relative to the center of mass 101 . The initial target. 1018 expands to form first and second extents along perpendicular directions, and tlie larger of these two extents extends in the direction 1023 A.

Referring to FIG. IOB. to change the position of tlie next target (a target that arrives in the initial target region 1031 at a later time), the control system 400 causes another pulse of the first optical beam 410a to be emitted froni the light-generation module 480 at a time such that the first light beam 410a reaches the initial target region 1031 when the next initial target 1018 is in the region 1031 and positioned within the region 1031 such that the first light beam 410a strikes the initial tar get 1018 below (displaced in the y direction) the center of mass 1019. For example, the control system 470 may receive a representation of the interior of the vacuum chamber 740 from the sensor 448 (FIG. 4) and detect that the next initial target 101 S is near or in the initial target region 1031 and then cause the emission of the pulse of the first light beam 410a based on the detection such that the first light beam 410a is displaced in the y direction relative to the center of mass 1019. Tlie next initial target 1018 expands to form first and second extents along perpendicular directions, and the larger of these two ext ents extends in the direction 1023B, which is different from the direction 1023A.

As compared to a light beam that strikes the initial target 1018 a t the center of mass 1019, the control system 470 causes the light beam 10a or a pulse of the light beam 410a t arrive earlier to orient the larger extent of the target 92ΘΑ along the direction 1023A (FIG. lOA) and to arrive later to orient Hie larger extent of tlie target 920B along the direction 1023B (FIG, _. Γ0Β).

Thus, a target can be positioned by irradiating an initial target with a light beam with a timing that is controlled with the control system 470 prio to the target arriving in the target region 730. in other implementations, the target can be positioned by changing the direction of propagation of the first light beam 410a. Additionally, in some implementations, the target can be provided to the target region 730 at a particular orientation (and the orientation can be varied from target-to-target) without the use of an initial target. For example, the target can be oriented through manipulation of the target material supply apparatus 716 and/or formed prior to being released from the target material supply apparatus 716.

Retaining to FIGS. 8 and 9 A, the light beam 710 is directed to the target region 730

(820). The light beam 710 has an energy sufficient to convert at least some of the target material in the target 720A to plasma. The plasma emits EUV light and also emits particles and/or radiation. The particles and/or radiation are emitted non-isotropicaily and are primarily emitted in a particular direction, toward a first peak 965A (FIG. 9 A).

The first and second extents of the first target are positioned relative to a separ ate and distinct object in the vacuum chamber. Fo example, the target 720A of FIG. 9A has an ellipticaify shaped cross-section in the y-z plane and a maximum extent in the y-z plane hi a direction 923A. The direction 923 A (and a direction perpendicular to the direction 923A) forms an angle with respect to a surface normal of the window 714. In this way, the target 720A can be considered to be positioned or angled relative to the window 714. hi another example, the direction 923 A forms an angle relative to a space in the fluid 408 that is marked with the label 909. In yet another example, the direction 923 A forms a angle with a surface normal at a region (marked with the label 956) on the optical element 755

As discussed above, the location of the peak 965 A depends on the position of the target 920. Thus, the location of the peak 965B can be changed by changing the position of the target 920. A second target is provided to the interior of tile ^' vacuum chamber 740 (830). Tlie second target lias a different position than the first target. Referring to FIG. 9B, at the time t2, the target 720B as an elh tieaieross-seetioo is the y-z plane, with the ellipse having a major axis. The largest extent of the second target in the y-z plane is along the major axis in a direction 92.3B. The direction 923B is different frorn the direction 923 A. Tims, a compared to the ^' first target, the second target is positioned differently lela live to the window 714 and oilier objects in the vacuum chamber 740. In this example, the direction 923B is perpendicular to the z direction. Tlie target 720B can be positioned to have the larger extent in tlie direction 923B by. for example, controlling the light beam control module 471 to emit the first light beam 410a at a time such that tlie first light beam 410a stiikes an initial target (such as the initial target 1018 of FIGS. 10A and lOB) at its center of mass.

Tlie light beam 710 is directed toward the target region 730 to form a second plasma from the second target (840). Because the position of the second target is different from the position of the first target, the second plasma primarily emits particles and/or radiation toward a peak 965B, which is in a different location than the peak 965A.

Thus, by controlling the position of the target over time with the control system 470, the direction in which particles and radiation are emitted from the plasma can also be controlled.

The process 800 can be applied to more than two targets, and the proces 800 can be applied to determine the position of any or all of the targets that enter the target region 730 during operation of the vacuum chamber 740. For example, as shown in FIG. 9C, the target 720C in the target region 730 at the time t3 has a different position than the targets 720A and 720B. Plasma formed from the target 720C at the time t3 emits particles and or radiation primarily toward a peak 965C. Tlie peak 965C is at a different location in the vacuum chamber 740 than the peaks 965 A and 965B. Thus, continuing to vary the target orientation or position over time can further spread out the heating effects of the plasma. For example, the peak 965 A is pointed toward the region of the fluid 708 labeled 909, but the peaks 965B and 965C are not. In other example, the peak 965C i pointed at the region 956 on the optical element 755, but the peaks 965A and 965B are not. In this way, the region 956 may avoid becoming contaminated.

The process 800 can be used to continuously change the position of targets that enter the target region 730. For example, the position of any target in the target region 730 can be different than the position of the immed ately precedin and/or hnmediately subsequent targets. ίιι other examples, the position of each target that reaches the target region 730 is not necessarily different. I these examples, the position of any target in the target region 730 can he different ftom the position of at least one other target in the target region 730. Further, the change in. position can he incremental with the angle relative to a particular object increasing or decreasin with each change until a maximuin and/or mrnhmrni angle is reached, other implementations, the change in position among -the "various targets reaching the target region 730 ma be a random or pseudo-random amount of angle variation .

Furthermore, and referring to FIG. IOC. the position of the targets .may be changed such that, the direction along which the peak directional flux is emitted sweeps out a .three-dimensional region in the vacuum vessel 740. FIG. IOC shows a view of fee optical element 755 looking from the target region 730 (looking in the -z direction), with the direction along which the peak directional flux is emitted over time represented by a path 1065. Although the directional fhijs did not necessarily reach the optical element 755, the path 1065 illustrates that the targets that come into the target region 730 over time can have different positions from each other and the different positions can result in the peak emission directio sweepin out a tlnee-diineiisionaJ region in the vacuum vessel 740.

Additionally, the process S00 can change the position of targets that enter the target region 730 at a rate that does not necessarily result in the positioning of any tar get being different than the positioning of the immediately preceding and/or immediately subsequent targets, but that changes the position of the targets that enter the target region 730 at a rate that, prevents damage to objects in the vacuum chamber based on the operating conditions or desired operating parameters.

For example, the amount, of fluid 708 and the flow rate of the fluid 708 needed to protect the optical element 755 from high-energy ion deposits depends on the duration of the plasma generation in the vacuum chamber. FIG. 11 is an example plot 1100 of the relationship between oi iimum acceptable fluid flow and EUV emission duration. The EUV emission duration also can be referred to as an EUV burst dur ation, and the EUV burst can be formed from converting a plurality of successive targets into plasma. The y-axis of the plot 1100 is the fluid flow rate, and the x-axis of the plot 1100 is the duration of an EUV light burst generated m the vacuum chamber 740. The x-axis of the plot 1100 is in log scale. Data relating the imiEnrum flow rate to the EUV emission duration (such .as data thai forms a plot such, as the plot Ϊ 100) may be stored on. the electronic storage 473 of the control system. 70 and used by the control system 470 to determine how often the position of the target 720 should be changed to mimmize conju ction of the fluid 708 while still protecting the objects in the vacuum chamber 740. For example, the data used for the plot 1100 indicate a inisiniimi Sow rate to prevent eoB airiinatioJi n a system that uses an EXJV ^' bursi having various durations. Hie minimum flow rate needed may be reduced by changing the position of one or more of the targets that are used to produce the EUV burst relative to the position of the other targets thai are used to produce the EUV bur st. The plot 1100 may be used to determine how often the target in the target region should be repositioned to achieve the desired minimum flow rate. For example, if the desired minimum flow rate corresponds with a lower EUV bur st duration than the source is operating at the targets arriving in the target region may be repositioned such that the directional flux of particles and/or radiation produced by any individual tar get or collection of targets is directed into a particular region of the vacuum chamber for an amount of time that is the same as that lower EUV burst duration. In this way, the EUV hurst duration experienced by any particular region of the vacuum chamber may be reduced and the minimum flo w ra te of the fluid 708 also msy be reduced.

FIG. 11 shows an example relationship between the flow rate of the fluid 708 and the EUV burst duration. Other properties of the fluid 70S, such as, for example, pressure and/or density, may vary with the EUV burst duration. In this way, the process 800 also can be used to reduce the amount of fluid 708 that is needed to protect the optical element 755.

Referring to FIG. 12, a flow chart of an example proces 1200 is shown. The process 1200 positions a target in a vacuum chamber such that the effect of plasma on an object in the vacuum chamber are reduced or eliminated. The process 1200 can be performed by die control system 470.

An initial target is modified to form a modified tar get (1210). The modified target and the initial target include target material, but the geometric distribution of the target material is different than that of the modified target . The initial tar get can be, for example, an initial tar get such as the initial targets 18 (FIG. 6C) or 1018 (FIGS. 10A and 10B). The modified target can be a disk-shaped target formed by irradiating the initial target with a pre-pulse (such as the pre- pulse 606 of FIGS . 6A-6B) or with a light beam, such as the first light beam 410a of FIG. 4, that does aotneeessarily convert tlie target material hi the mitial target to a plasma that emits EUV but does condition the mitial target.

Tiie modified target may be positioned relative to a separate and distinct object. The interaction between the initial target and a light beam can determine the position of the modified target. For example, as discussed above with respect to FIGS. 6A-6C, HG. 8, and FIGS. 10A and 10B, a disk-shaped target with a particular position can be t½ined by dhecliiig a light beam to a paiticuiar part of the initial target. The separate and distinct object is any object in a vacuum chamber. For example, fee separate and distinct object may be a buffer fluid, a target in a stream of targets, and/or an optical element.

A light beam is directed toward the modified target (1220). The light beam may be an amplified light beam, such as the second light beam 410b (FIG. 4). The light beam has an energy sufficient to convert at least some of the target material in the modified target to plasma that eniits EUV light. Tlie plasma is also associated with a directiooally dependent flux of particles and/or radiation, and the direcfionally dependent flux has a maximum (a location, region, or direction into which the highest portion of the particles and or radiation flow). Tlie maximum is referred to as the peak direction, and the peak direction depends on the position of the modified target. The particles and radiation may be preferentially emitted from the heated side of the modified target, which is the side that receives the light beam first. Thus, for a disk- shaped target that receives the light beam at one of the flat faces of the disk, the peak direction is in a direction that is normal to the face of the disk that receives the light beam. The modified target may be positioned such that the effect of the plasma on the object is reduced. For example, orienting the modified target such that the heating side of the target points away from the object to be protected will result in the fewest possible high-energy ions being directed toward the object.

The process 1200 can be performed for a single target or repeatedly. For

implementations in which the process 1200 is perfomied repeatedly, the position of the modified target for any particular instance of the process 1200 can be different from positions of previous or subsequent modified targets.

Referring to FIGS. 13 A- DC, the process 1200 may be used to protect targets in stream of targets from the effects of the plasma. FIGS. I3A-I3B, which are block diagrams of an interior of a vacuum chamber 1340, illustrate ho w a target in the vac uum cliamber may be protected from the effects of plasma. F16. 13 A shows a stream 1322 of targets, which tra els in the■ ^■ vacuum chamber in a direction y toward a target region 1330. The direction along which the stream 1322 travels may be referred to as the target trajectory or the target path. A light beam 1310 propagates in a direction z toward the target region 1330. The target 1320 is the target in the stream .1322 in the target region 1330. The interaction between the light beam 1310 and the target 1320 converts the target material in the target 1320 to plasma that emits EUV light

Additionally, the plasma emits a directionally dependent flux of particles and/or

mdiation, represented by the profile 1364. In the example of FIG. 13 A, profile 1364 shows that the particles and/or radiation are primarily emitted in a direction opposite to the z direction, and the greatest effect of the plasma is in tiiis direction. However, the plasma also has an impact on objects that are displaced in the y direction, including a target 1322a, which is the target in the stream 1322 that is closest to (but outside of) the target region 1330 when the plasma is formed. In other words, in the example of FIG. 13 A, the target 1322a is the next incoming target or the target that will he in the target region 1330 after the target 1320 is consumed to produce plasma.

The effect of the plasma on the target 1322a may be direct, such as the tar get 1322a experiencing ablation from the radiation in the directionally dependent flux. Such ablation may slow the target and or change the shape of the target. Tlie radiation from the plasma can apply a force to the target 1322a, resulting in the target 1322a reaching the target region 1330 later than expected. The light beam 1310 may be a pulsed light beam. Thus, if the target 1322a reaches the target region 1330 later than expected, the light beam 1310 and the target may miss each other and no plasma is produced. Additionally, the force of the plasma radiation may change the shape of the target 1322a unexpectedly and may interfere with intentional shape changes that condition the targets in the stream 1322 prior to reaching the tar get region 1 30 to increase plasma production.

The effect of the plasma on the target 1322a also may be indirect. For example, a buffer fluid may flow in the vacuum chamber 1340, and the directionally dependent flux may heat the fluid, and the heating of the fluid may change the trajectory of the targets (such as discussed with respect to FIGS. 7 A and 7B). Indirect effects also may interfere with the proper operation of the light source.

Tlie effects of the plasma on the target 1322a can be reduced by orienting a heating side

1329 of the target 1320 away from the target 1322a. The heating side 1329 of the target 1320 is the side of fee target 1320 that initially receives the light beam 1310, and the : ^■ particles and/or radiation are eirsitted primarily from the beating side 1329 and in a direction that is normal to the target material disnibution at the heating side 1329. The portion P of the radiation emitted by the plasma: at a ^■ particular ^■ angle relative to the target 1320 may approximate the relationship of Equation 1 :

Ρ(θ) = 1 - cos ^a (6) (1), where n is an integer number, and Θ is the angle between the normal to the target on the heating side 1329 and the direction of the target trajectory between the centers of mass of the target

1320 and the target 1322a. Other angular distributions of the radiation are possible.

Referring to FIG. 13B. the position of the target 1320 is changed as compared to the position in FIG. 13 A such that the heating side 1325) points away from the tar get 1322a. As a result of this positioning, the particles and/or radiation are emitted in a dir ection 1351, away from the target 1322a. Referring to FIG. 13C, the effect on the target 1322a is further reduced by positioning the heating side 1329 of the target 1320 away from the target 1322a and positioning the path of the target stream 1322 such that the target 1322a is located in a region that has the fewest particles and/or the least radiation from the plasma. In the example of FIG.

13C, this region is a region that is in a direction opposite the direction 1351 (behind the target 1320), and the targets in the target stream 1322 travels along the direction 1351.

Tims, the effects of the plasma on other targets in the vacuum chamber may be reduced by orienting the target and/or positioning of the target path.

FIGS. 14, 15 A, and 15B are additional examples of systems in which the processes 800 and 1200 can be performed.

Referring to FIG. 14, a block diagram of an exemplary optical imaging system 1400 is shown. The optical imaging system 1400 includes an LPP EUV light source 1402 that provides

EUV light to a lithography tool 1470. The light source 1402 can be similar to, and or include some or all of the components of, the light source 101 of FIG. 1.

The system 1400 includes an optical source such as a drive laser system 1405, an optical element 1422, a pre-pulse source 1443, a focusing assembly 1442, and a vacuum chamber 1440.

The drive laser system 1405 produces an amplified light beam 1410. The amplified light beam 1410 lias energy sufficient t ccnyerl target material :ia a target 1420 into- plasma that emits ETJV ^" Light. Any of the targets discussed above can be used as tlie target- 1420.

Tlie pre-puise source 1443 emits poises of radia tion 1417, The poises of radiation, can be. used as the pre-puise 606 (FIG. 6A-6C). The pre-puise source 1443 can-be, for example, a Q- switched d: Y AG laser that operates at a 50 kHz repetition rate, and the pul es of radiation 1417 can be pulses from Hi Nd:YAG laser that have a ..wavelength of 1.06 μηι. The repetition rate of tlie pre-puise source 1443 indicates how often the pre-puise source 1443 produces a pulse of radiation. For the example where the pre-puise source 1443 has a 50 kHz repetition rate, a pulse of radiation 1417 is emitted every 20 microseconds (jis).

Oilier sources can be used as tlie pre-puise source 1443. Fo example, the pre-puise source 1443 can be any rare-earth-doped solid state laser other that an Nd.YAG, such as an erbium-doped fiber (Englass) laser. In another example, the pre-puise source can be a carbon dioxide laser that produces pulses having a wavelength of 10.6 um. The pre-puise source 1443 can be any oilier radiation or light source thai produces light pulses that have an energy and wavelength used for the pre-pulses discussed above.

The optical element 1422 directs the amplified light beam 1410 and the pulses of

radiation 1417 front the pre-puise source 1443 to the chambe 1440. The optical element 1422 is any element that can direct, the amplified light beam 1410 and tlie pulses of radiation 1417 along similar or the same paths. In the example shown in FIG. 14. the optical element 1422 is a dichroic beamsplitter that, receives the amplified light beam 1410 and reflects it toward the chamber 1440. The optical element 1422 receives the pulses of radiation 1417 and transmits the pulses toward the cliamber 1440. Tlie dichroic beamsplitter has a coating that reflects tlie wavelength(s) of the amplified light beam 1410 and transmits the wavelengih(s) of the pulses of radiation 1417. The dichroic beamsplitter can be made of. fo example, diamond.

In other implementations, the optical element 1422 is a mirror that defines an aperture

(not shown). In this implementation, the amplified light beam 1410 is reflected from the mirror surface and directed toward the chamber 1440, and the pulses of radiation pass through the aperture and propagate toward the chamber 1440.

In still other implementations, a wedge-shaped optic (for example, a prism) can be used to separate the main pulse 1410 and the pre-puise 1417 into different angles, according to their wavelengths. Tlie wedge-shaped optic can be used in addition to the optical element 1422, or it can fee used as the optical element 1 22. Th wedge-shaped optic can he positioned j st iipstream (in the -z direction) of fee focusing assembly 1442.

Additionally, fee pulses 1417 can he delivered to the .chamber 1440 in other ways. For example, the piilses.1417 can travel troug ^' optical fibers that deliver the pulses 1417 to the chamber 1 40 and/or the focusing assembly 1442 without the use of the optical element 1422 or other directing elements. In these implementations, fee fibers bring fee pulses of radiation 1417 directly to an interior of the chamber 1440 through an opening formed in a wail of the chamber 1440.

The amplified light beam. 1410 is reflected from the optical element 1422 and propagates through fee focusing assembly 1442. The focusing assembly 1442 focuses the amplified light beam 1410 at a focal plane 1446, which may or may not coincide with a target region 1430. The pulses of radiation 1417 pass through the optical element 1422 and are directed through the focusing assembly 1442 to the chamber 1440. The amplified light beam 1410 and the pulses of radiation 1417, are directed to different locations along the y direction hi the chamber 1440 and arrive in the chamber 1440 at different nines.

In the example shown in FIG. 14, a single block represents the pre-pulse source 1443. However, the pre-pulse source 1443 can be a single light source or a plurality of light sources. For example, two separate sources can be used to generate a plurality of pre-pulses. The two separate sources can be different types of sources that produce pulses of radiation having different wa velengths and energies. For example, one of fee pre-pulses can have a wavelength of 10.6 μπι and be generated by a C<¾ laser, and the other pre-pulse can have a wavelength of 1.06 urn and be generated by a rare-eart h -doped solid state laser.

In some implementations, the pre-pulses 1417 and the amplified light beam 1410 can be generated by the same source. For example, the pre-pulse of radiation 1417 can be generated by the drive laser system 1405. In this example, the drive laser system can include two C0 ₂ seed laser subsystems and one amplifier. One of the seed laser subsystems can produce an amplified light beam having a wavelength of 10.26 μηι, and the other seed laser subsystem can produce an amplified light beam having a wavelength of 10.59 ηι. These two wavelengths can come from different lines of the C0 ₂ laser. In other examples, other lines of the CO? laser can be used to generate the two amplified light beams. Both amplified light beams from the two seed laser subsystems are amplified in the same power amplifier chain and then angularly dispersed to reach different locations within the chamber 1440. Tlie amplified light beam .with the wavelength of 10.26 μι« can be used as the pre-pulse 1 17, and the amplified ^' light beam with the ^■ wavelengt of 10.59 μ¾_ can ^' be used as the amplified light beam 1 10. In ksplen eniations feat employ a plurality of pre-pnises, three seed lasers can be used, one of which is used to generate each of the .amplified light beam 1.410, a _. first pre-pulse, and a second, separate pre-pulse.

The amplified lightheani 1410-.and.iiie.pre-p.alse of radiation 1417 can all be amplified in the same optica! amplifier. For example, the three or more power amplifiers can be used to amplify the amplified light beam 1410 and the pre-pulse 1417.

Referring to FIG. 15 A, an LPP EUV light source 1500 is shown. The EOT light source 1 00 can be used with the light sources, processes, and vacuum chambers discussed above. The LPP EUV light source 1500 is formed by irradiating a target mixture 1514 at a target region 1505 with an amplified light beam 1510 that travels along a beam path toward the target mixture 1514. The target region 1505, which is also referred to as the irradiation site, is within an interior 1507 of a vacuum chamber 1530. When the amplified light, beam 1510 strikes the target mixture 15 i 4, a target material within the target mixture 1514 is converted into a plasma state that has an element with an emission line in the EUV range. Tlie created plasma has certain characteristics that depend on the composition of the target material within the target mixture 1514. These characteristics can include the wavelength of the EUV light produced by the plasma and the type and amount of debris released from the plasma.

The light source 1500 also includes a target material delivery system 1525 that delivers, controls, and directs the target mixture 1514 hi the form of liquid droplets, a liquid stream, solid particles or clusters, solid particles contained within liquid droplets or solid particles contained within a liquid stream. The target mixture 1514 includes the target, material such as, for example, water, tin, lithium, xenon, or any material that, when converted to a plasma state, has an emission line hi the EUV range. For example, the element tin can be used as pure tin (Sn); as a tin compound, for example, SnBr ₄ , SnBr ₂ . S11H ₄ ; as a tin alloy, for example, tin-gallium alloys, tm-indiuni alloys, tm-hidiuni-galhuni alloys, or any combination of these alloys. The target mixture 1514 can also include impurities such as non-target particles. Thus, in fee situation in which there are no impurities, the target mixture 1514 is made up of only the target material. The target mixture 1514 is delivered by the target material delivery system 1525 into the interior 1507 of tlie chamber 1530 and to the target region 1 05. The light source 1560 includes a drive laser system 1515 that produces the amplified light bean 1510 due to a population .inversion within the gain .mediiim or mediunis of the ..laser system 1515. The light ^' .source 1500 includes a beam delivery■system ..between the laser system 1515 aad the target region 1505. the beam delivery system inehiding a beam , transport system 1 20 and a focus assembly 1522. The beam transport system 1.5.20 receives the amplified light beam 1510 ^• from the laser system 1515. aad steers and modifies fee amplified light beam 10 as needed and outputs the amplified light beam 1510 to the focus assembly 1522. The focus assembly 1522 receives the amplified light beam 1510 aad focuses the beam 1510 to the target region 1505.

In some implementations, the laser system 1515 can include one or more optical amplifiers, lasers, and/or lamps for providing one or more main pulses and, ia some cases, oae or more pre-pulses. Each optical amplifier includes a gain medium capable of optically amplifying the desired wavelength at a high gain, aa excitation source, and internal optics. The optical amplifier may or may not have laser mirrors or other feedback devices that form a laser cavity. Thus, the laser system 1515 produces an amplified light beam 1510 due to the population inversion in fee gain media of the laser amplifiers even if there is ao laser cavity. Moreover, fee laser system 1515 can produce an amplified light beam 1510 that is a coherent laser beam if there is a laser cavity to provide enough feedback to the laser system 151 . The term "amplified light beam" encompasses one or more of: light from the laser system 1515 that is merely amplified but not necessarily a coherent laser oscillation and light from the laser system 1515 that is amplified and is also a coherent laser oscillation.

The optical amplifiers in the laser system 151 caa include as a gain medium a filling gas that, includes C0 ₂ and can amplify light at a wa velength of between about 100 aad about. 11000 mil, and in particular, at about 10600 am, at a gain greater than or equal to 1500. Suitable amplifiers and lasers for use in the laser system 151 can include a pulsed laser device, for example, a pulsed, gas-discharge C<¼ laser device producing radiation at about 9300 nm or about 10600 nm, for example, with DC or RF excitation, operating at relatively high power, for example, lOfcW or higher and high pulse repetition rate, for example, 40kHz or aiore. The optical amplifiers in the laser system 1515 can also include a cooling system such as water that can be used when operating the laser system 1515 at higher powers.

FIG. 15B shows a block diagram of an example drive laser system 1580. The drive laser system 1580 can be used as part of the drive laser system 151 in the source 1500. The drive laser system: 1580 includes three power amplifiers 1581, 1582, and 1583. Any or all of the power amplifiers Ϊ 581. 1582, and 1583 ^' can include internal optical elements (not shows).

Light 1584 exits from ^' the power amplifier 1581 through an output window 1585 and is reflected off a. curved mirror 1586. Afte reflection, the light 1584 passes through a spatial filter 1587, is reflected off of a curved mirror 1588, and enters the power amplifier 1 82 through an input window 1 89. The light 1584 is ainplifie in the power amplifier 1 82 and redirected out of the power amplifier 1582 through a output window- 1590 as light 1591. The light 1591 is diiected toward the ampliiier 1583 with a fold minor 1 92 and enters the amplifier 1583 through an input window 1593. The amplifier 1583 amplifies the light 1591 and directs the light 1591 out of the amplifier 1583 through an output window 1594 as an output beam 1595. A fold mirror

1596 directs ihe output beam 1595 upward (out of the page) and toward the beam transport system 1520 (FIG. 15A).

Referring again to FIG. 15B, the spatial filter 1587 defines an aperture 1597, which can be, for example, a circle having a diameter between about 2.2 mm and 3 mm. The curved mirrors 1586 and 1588 can be. for example, off-axis parabola mirrors with focal lengths of about 1.7 m and 2.3 m, respectively. The spatial filter 1587 can be positioned such that the aperture

1597 coincides with a focal point of the drive laser system 1580.

Referring again to FIG. 15 A, the light source 1500 includes a collector mirror 1535 having an aperture 1 40 to allow the amplified light beam 1510 to pass through and reach the target region 1505. The collector mirror 1535 can be, for example, an ellipsoidal mirror that has a primary focus at the target region 1505 and a secondary focus at an intermediate location 1545 (also called an intermediate focus) where the EUV light can be output from the light source 1500 and can be input to, for example, an integrated circuit lithography fool (not shown). The light source 1500 can also include an open-ended, hollow conical shroud 1550 (for example, a gas cone) that tapers toward the target region 1505 from the collector minor 1535 to reduce the amount of plasma-generated debris tha t enters the focus assembly 1522 and/or the beam

transport system 1520 while allowing the amplified light beam 1510 to reach the target region 1505. For this purpose, a gas flow can be provided in the shroud that is directed toward the target region 1505.

The light source 1500 can also include a master controller 1555 that is connected to a droplet position detection feedback system 1556, a laser control system 1557, and a beam control system 1558. The /light source 1500 can include one or more target or droplet imager 1560 that provide an output indicative of the position of a droplet, for example, relative to the target region 1505 and provide this oelpoi to the droplet position detection feedback system 1556, which can, for example, compute a droplet position a d trajectory from which a droplet position ; error can be computed either on a droplet .by droplet basis or on a verage. The droplet position detection feedback system 1556 thus provides -the droplet position .error as an mpiit to the master controlle 1555. The master controller 1555 can therefore provide a laser position, direction, and timing correction signal, for example, to the laser control system 1557 that can be used, for example, to control the laser timing circuit and/or to the beam control system 1558 to control an amplified light beam position and shaping of the beam transport system 1520 to change the location and or focal power of the beam focal spot within the chamber 1530.

The target material delivery system 1525 include a target material delivery control system 1526 that is operable, in response to a signal from the master controller 1 55, for example, to modify the release point of the droplets as released by a target material supply apparatus 1527 to correct for errors in the droplets arriving at the desired target region 1505.

Additionally, the light source 1500 can include light source detectors 1 65 and 1570 that measures one or more EUV light parameters, including but not limited to, pulse energy, energy distribution as a function of wavelength, energy within a particular band of wavelengths, energy outside of a particular band of wavelengths, and angular distribution of EUV intensity and/or average power. The light source detector 1565 generate a feedback signal for use by the master controller 1555. The feedback signal can be, for example, indicative of the errors in parameters such as the timing and focus of the laser pulses to properly intercept the droplets hi the right place and time for effective and efficient EUV light production.

The light source 1500 can also include guide laser 1575 that can be used to align various sections of the light source 1500 or to assist in steering the amplified light beam 1510 to the target region 1505. In connection with the guide lase 1575, the light source 1500 includes metrology system 1524 that is placed within the focus assembly 1522 to sample a portion of light from the guide lase 1575 and the amplified light beam 1510. In other implementations, the metrology system 1524 is placed within the beam tiansport. system 1520. The metrology system 1524 can include an optical element tha t samples or re-directs a subset of the light such optical element being made out of any material mat can withstand the powers of the guide laser beam and the amplified light beam.1510. A beam analysis system is. formed .from the metrology system ί 524 and the master ^' controller 1555 since the master controller 1555 analyzes the sampled light from the guide laser 1575 and uses this 'information to adjust components within the focus assembly 1522 through the beam control system 1558.

Thus, in suniniary, the .light scarce 1500 produces an amplified light beam 1510 that is directed along the beam, path- to irradiate the target .-mixture 1514 at the target regio 1505 to convert the target material within the mixture 1514 into plasma thai emits light in the EUV range. The amplified light beam 1510 operates at a particular wavelength (that is also referred to as a chive laser wavelength) that is determined based on the design and properties of the laser system 151 . Additionally, the amplified light beam 1510 can be a laser beam when the target material provides enough feedback back into the laser system 1515 to produce coherent laser light or if the drive laser system 1515 includes suitable optical feedback to form laser cavity.

Other miplementations are within the scope of the claims. For example, the fluid 108 and 708 is shown as flowing in the y direction and perpendicular to the direction of propagation of a light beam that converts target material to plasma. However, the fluid 108 and 708 may flow in any direction as determined by the flow configuration associated with a set of operating

conditions. For example, referring to FIG. 16, an alternate implementation of the light source 101 is shown in which the fluid 108 of the vacuum chamber flows in the z direction.

Additionally, any of the characteristics of the flow that are part of the flow configuration

(including the direction of flow) can be intentionally changed during operation of the light source 101.

Additionally, although the examples of FIGS. 6 A-6C and 10A and I0B show using a pre- pulse to initiate tilting of an initial target, as discussed above, a tilted target can be delivered to the target regions 130, 730. and or 1330 with oilier techniques that do not employ a pre-pulse. For example, as shown in FIG. 17, a disk-shaped target 1720 that includes target material that emits EUV light when converted to plasma is pre-formed and provided to a target region 1730 by releasma the disk target 1720 with a force that results in the disk target 1720 moving through the target region 1730 tilted relative to an amplified light beam 1710 that is received in the target region 1730.

FIGS. 7A and 7B show the vacuum chamber in the y-z plane and in two dimensions.

However, it is contemplated that the profile 764 (FIG. 7B) may occupy three dimensions and may sweep out a volume in three disieasions. _. Similarly, FIGS. ¾\, 9C, 10A, 1 OB, aad 13A-13C show a vacuum chamber in the y-z plane and in two dimensions. However, it is contemplated that the targets ia the vacuum chambers may lilt in my direction in three dimeosioas and the directional flux of .particles- and/or radiation ma sweep out a space in three dimensions.