Sensor, imaging system, and method for forming a sensor

Abstract

Wire-bonded sensor dies limit the number of interconnections and readout speeds. [Solution] A sensor comprises a substrate 104 and one or more components 110 mounted on the substrate 104. The sensor also includes a sensor die 108 having a thin back surface 222 and an energy-sensitive element 116 configured to detect energy irradiating the thin back surface 222 of the sensor die 108. The sensor further comprises a separate heat conduction structure 212 formed between the front surface 224 of the sensor die 108 and the substrate 104 by a flip-chip process, thereby bonding the sensor die 108 to the substrate 104 and giving the thin back surface 222 of the sensor die 108 a pre-selected shape. At least a portion of the separate heat conduction structure 212 electrically connects the sensor die 108 to one or more components 110.

Description

Technical Field

[0001] The present invention relates generally to sensors, imaging systems, and methods of forming sensors. Particular embodiments relate to sensor shape control through sensor assembly.

Background Art

[0002] The following description and examples are not to be considered prior art by virtue of their inclusion in this paragraph.

[0003] Backside-illuminated image sensors can achieve high quantum efficiency (QE) and good modulation transfer function (MTF), and are widely used for inspection of various semiconductors and other substrates. To enable high-speed operation, this sensor is closely connected to an application-specific integrated circuit (ASIC), which can perform one or more of the following functions: analog-to-digital (A / D) conversion, signal conditioning, digital signal processing, and communication with an external computer.

[0004] Such a sensor configuration presents challenges with respect to proper control of the shape of the photoactive area of the sensor. For example, in a backside-illuminated sensor, the photoactive area is a thin film that may become mechanically unstable. Flip-chip assembly of a sensor die enables high-speed operation compatible with backside illumination. Flip-chip assembly of the sensor onto a ceramic substrate may cause the sensor to become convex, concave, or wrinkled. The effect of flip-chip assembly on sensor shape is highly uncontrolled and can present challenges for optical systems that require a simple shape. In optical applications, a curved image surface with a specific curvature may be preferred. Therefore, the ability to control sensor shape during assembly and to design an assembly targeting a specific sensor shape can be important to the imaging performance of such systems.

[0005] Optical system design typically produces negative curvature, positive curvature, or a flat image plane. However, image sensors can be assembled on ceramic substrates, and the shape of these substrates is difficult to control due to the manufacturing process. Failure to control the shape of the sensor assembly can reduce the effective field of view, decrease the system-level optical tolerance, and increase the amount of optical aberration.

[0006] Therefore, the drawbacks of currently used sensor assembly methods include the inability to easily control the shape of the ceramic substrate, while high-performance optical design may require high sensor planarity or a specific sensor shape. A further drawback of currently used assembly methods is the relatively low coplanarity of the die, which can make it difficult to properly mount the sensor die, affecting the thermal performance of the assembly. Another drawback of currently used sensor assembly methods is heat dissipation, which is important for high-speed, low-noise operation. A further drawback of currently used sensor assembly methods is that image field curvature can make it difficult to achieve a telecentric image space, which can be important in measurement applications. In addition, current assembly methods do not provide good control or repeatability of the sensor shape.

[0007] Currently, several methods have been proposed to control the shape of back-illuminated thin sensor dies for mobile phone and astronomical applications. However, all of these applications utilize wire-bonded sensor dies. Wire-bonded sensor dies limit the number of interconnections and readout speeds, making them less suitable for the optical inspection applications described above. The methods currently used for controlling the shape of such sensors may also not be suitable for achieving good thermal contact, high-density interconnected flip-chips, or use in a vacuum. [Prior art documents] [Patent Documents]

[0008] [Patent Document 1] U.S. Patent Application Publication No. 2018 / 0076257 [Patent Document 2] U.S. Patent Application Publication No. 2018 / 0102312 [Overview of the project] [Problems that the invention aims to solve]

[0009] Therefore, it would be beneficial to develop sensor systems and methods, imaging systems, and methods for forming sensors that do not have any of the drawbacks described above. [Means for solving the problem]

[0010] The following descriptions of various embodiments should not be construed as limiting the subject matter of the attached claims.

[0011] One embodiment relates to a sensor comprising a substrate and one or more components mounted on the substrate. The sensor also comprises a sensor die comprising a thin back surface and an energy-responding element configured to detect energy irradiating the thin back surface of the sensor die. The sensor further comprises a separate thermal conduction structure formed between the front surface of the sensor die and the substrate by a flip-chip process, thereby bonding the sensor die to the substrate and giving the thin back surface of the sensor die a pre-selected shape. At least a portion of the separate thermal conduction structure electrically connects the sensor die to one or more components. The sensor may further be configured as described herein.

[0012] Another embodiment relates to an imaging system comprising an energy source configured to generate energy directed onto a sample by an illumination subsystem. The imaging system also comprises a sensor configured to detect energy from the sample and generate an output corresponding to the detected energy. The sensor is further configured as described above. The imaging system may further be configured as described herein.

[0013] Another embodiment relates to a method for forming a sensor. This method includes the steps of forming a separate heat conduction structure on a substrate and modifying the shape of the separate heat conduction structure based on a pre-selected shape of the thin back surface of a sensor die. This method also includes the step of bonding the front surface of the sensor die to the substrate via the separate heat conduction structure, thereby giving the thin back surface of the sensor die a pre-selected shape. At least a portion of the separate heat conduction structure electrically connects the sensor die to one or more components mounted on the substrate. The sensor die includes an energy-sensitive element configured to detect energy irradiating the thin back surface of the sensor die.

[0014] Each step of this method may be further carried out as described herein. This method may include any other steps of any other method described herein. This method may be carried out by any of the systems described herein.

[0015] Further advantages of the present invention will become apparent to those skilled in the art thanks to the detailed description of the following preferred embodiments and by referring to the following accompanying drawings. [Brief explanation of the drawing]

[0016] [Figure 1] This is a schematic diagram showing a cross-sectional side view and a plan view of a sensor assembly of one embodiment. [Figure 2-1] This is a flowchart illustrating one embodiment of a method for forming a sensor assembly. [Figure 2-2]1 is a flow diagram illustrating one embodiment of a method of forming a sensor assembly. [Figure 3] 3 is a flow diagram illustrating an embodiment of interconnect formation using solder balls and gold studs. [Figure 4] 6 is a flow diagram illustrating one embodiment of a sensor assembly method adapted to the profile of concave or convex solder balls. [Figure 5] 9 is a schematic diagram illustrating a cross-sectional view of an embodiment of the geometry of a free-standing sensor die and the same sensor die in contact with a solder bump. [Figure 6] 12 is a schematic diagram showing a cross-sectional view of one embodiment of a portion of a complete sensor assembly, with overlaid arrows indicating the magnitude and direction of heat flux within the sensor assembly. [Figure 7a] 15 is a schematic diagram showing a side view of an embodiment of a camera lens subsystem coupled to an embodiment of a sensor described herein. [Figure 7b] 18 is a schematic diagram showing a side view of an embodiment of a camera lens subsystem coupled to an embodiment of a sensor described herein. [Figure 7c] 21 is a contour plot of surface sag for an example of curvature of an embodiment of a sensor. [Figure 7d] 24 is a plot of an example of the root mean square (RMS) spot size geometry across the entire field of view (FOV) for the embodiment shown in FIGS. 7a and 7b. [Figure 8a] 27 is a schematic diagram showing a side view of an embodiment of a tube lens subsystem coupled to an embodiment of a sensor described herein. [Figure 8b] 30 is a schematic diagram showing a side view of an embodiment of a tube lens subsystem coupled to an embodiment of a sensor described herein. [Figure 8c] 33 is a contour plot of surface sag for an example of curvature of an embodiment of a sensor. [Figure 8d]Figures 8a and 8b show an example plot of the geometric RMS of the spot size across the entire FOV of the embodiment shown. [Figure 9a] This is a schematic perspective view of one embodiment of a part of an imaging system comprising two or more sensors configured as described herein. [Figure 9b] Figure 9a shows a plot of contour lines representing surface sag, an example of the curvature of a portion of the image plane in an embodiment of the imaging system shown. [Figure 9c] Figure 9a shows a contour plot representing the surface sag, an example of curvature in one of the multiple sensor embodiments shown. [Figure 9d] Figure 9a shows a contour plot of surface sag, an example of optimized sensor curvature in multiple sensor embodiments. [Figure 10] This is a schematic diagram showing plan views of different grid embodiments in an image space that can be projected onto a sample space. [Figure 11] This is a schematic diagram showing a side view of an embodiment of an imaging system configured as described herein. [Figure 11a] This is a schematic diagram showing a side view of an embodiment of an imaging system configured as described herein. [Figure 12] This diagram shows one embodiment of a non-temporary computer-readable medium for storing program instructions for causing a computer system to execute the methods described herein on a computer. [Modes for carrying out the invention]

[0017] While various modifications and alternative forms are possible for the present invention, specific embodiments of the invention are shown in the drawings as examples and described in detail herein. The drawings may not be to scale. However, the drawings and the detailed description thereof are not intended to limit the invention to any particular form disclosed, but rather to cover all modifications, equivalents, and alternatives that fall within the spirit and scope of the invention as defined by the appended claims.

[0018] Now, looking at the drawings, please note that they are not drawn to scale. In particular, the scale of some elements in the drawings has been significantly exaggerated to emphasize the characteristics of those elements. Also note that the drawings are not drawn to the same scale. Elements shown in multiple drawings that can be similarly constructed are indicated using the same reference number. Unless otherwise stated herein, any suitable commercially available elements may be included in the descriptions and illustrations.

[0019] The embodiments described herein generally relate to sensors, imaging systems, and methods for forming sensors. More specifically, the embodiments described herein relate to methods for sensor assembly and shape control for applications such as inspection and measurement. The embodiments provide image sensors, including, but are not limited to, time delay integration (TDI) sensors using deep ultraviolet (DUV) and extreme ultraviolet (EUV) light, which can operate in partial vacuum or other controlled environments and are assembled on various ceramic substrates. The embodiments described herein are useful because they demonstrate methods for realizing such sensors having a controlled sensor shape while maintaining relatively fast and relatively low-noise operation. In addition, the embodiments described herein provide methods for controlling the sensor shape while enabling substantially good thermal contact of the sensor die, flip-chip assembly with substantially high-density interconnections, and adaptation for use in vacuum.

[0020] As used herein, the term “energy-sensitive element” is defined as a sensor element that is sensitive to or responds to one of the types of energy described herein, including light, electrons, and other charged particles. Such energy-sensitive elements may be formed in the form of various components, depending on the type of energy used for detection. Many embodiments and examples are described herein using the term “photosensitive element,” but no use of this term implies that such embodiments and examples exclude any other types of energy-sensitive elements described herein. In other words, the terms “energy-sensitive element” and “photosensitive element” are used interchangeably herein for convenience, and any example of the term “photosensitive element” should be interpreted more broadly as “energy-sensitive element” as described herein.

[0021] The embodiments also include imaging systems, such as inspection systems based on such sensors, which achieve superior imaging performance, and therefore higher defect sensitivity and throughput, compared to currently available inspection systems. By enabling sensor assemblies with controlled curvature as described herein, it is beneficial to expand the optical field of view (FOV) of the imaging system, relax the optical tolerance at the imaging system level, and reduce the amount of optical aberration in the system. The embodiments described herein are also beneficial because they enable relatively large sensors and tiled sensor arrays designed for curved or other non-flat image spaces, providing higher sensitivity for applications such as inspection with sufficient throughput (and therefore lower cost of ownership).

[0022] As will become apparent from the following description of various embodiments, the sensors described herein have several further advantages over sensors currently in use. These further advantages include the ability to bond sensor dies having substantially high sensor die planarity or a specific pre-selected sensor die shape to a substrate such as those described herein, even though the shape of the ceramic substrate cannot be easily controlled, thereby making the sensors particularly suitable for substantially high-performance optical designs. Another advantage of the embodiments described herein is that, regardless of the pre-selected sensor die shape, the embodiments described herein allow for sufficient contact with the sensor die, optimizing the thermal performance of the assembly, thereby enabling relatively fast and relatively low-noise operation of the sensor die. A further advantage of the embodiments described herein is that the sensor die can realize a telecentric image space regardless of any field curvature of the imaging system, which is important for measurement applications. These advantages and other advantages described herein are brought about by the sensor assembly method described herein, which allows for substantially precise control of the shape of the sensor die.

[0023] As further described herein, embodiments of a sensor die include a thin back surface and a photosensitive element configured to detect energy illuminating the thin back surface of the sensor die. The energy detected by the sensor die is directed to the thin back surface, then passes through the body of the sensor die, and as a result, a charge can be collected by an element formed on the front surface. The shape of the back surface defines the location of the image plane of the sensor. Thus, the shape of the sensor die is almost certainly one of the most important characteristics of the sensor die, for reasons including those further described herein.

[0024] As used herein, the term “pre-selected shape of the thin back surface of the sensor die” is used interchangeably with the terms “sensor shape” and “sensor die shape.” “Pre-selected shape of the thin back surface of the sensor die” is defined herein as the position of the thin back surface of the sensor die relative to some reference or coordinate system, and as a function of the position across the entire sensor die. “Pre-selected shape of the thin back surface of the sensor die” can be defined, for example, by the depth or vertical height of the thin back surface of the sensor die, which is a function of the position across the entire sensor die. This function of depth or height can be defined in two dimensions (2D) across the entire sensor die. “Pre-selected shape of the thin back surface of the sensor die” therefore also defines the 2D outline of the height or depth of the back surface of the sensor. In other words, “pre-selected shape of the thin back surface of the sensor die” efficiently defines the position of the back surface of the sensor die in the z direction as a function of x and y.

[0025] As further described herein, the “pre-selected shape of the thin back surface of the sensor die” can be realized by the novel and advantageous sensor assembly method described herein, and it is preferable that the thickness or vertical height of the sensor film (photosensitive element) is never changed. The shape of the film is constant, and the shape of other dimensions is fixed. In other words, the sensor assembly method described herein does not change the shape of the sensor die by changing any properties of the photosensitive element (although negligible changes may occur). On the contrary, it is preferable that the photosensitive element has the same properties before and after sensor assembly. Thus, a change in the height (or depth) of the back surface of the sensor die after sensor assembly will be similar to a change in the height (or depth) of the front surface of the sensor die. Conversely, a change in the external shape of the front surface due to mounting to another component will result in a similar change in the external shape of the back surface.

[0026] One embodiment of the sensor comprises a substrate, one or more components mounted on the substrate, and a sensor die bonded to the substrate. The image sensor die is therefore packaged together with other dies, assembled on a common substrate as shown in Figure 1, in one embodiment. Figure 1 includes a side view 100 and a bottom view 102 of the sensor assembly. This embodiment of the sensor assembly comprises a substrate 104 with an electrical interconnect 106 formed inside, as shown in the side view. At least a portion of the electrical interconnect 106 electrically connects a sensor die 108 mounted on one side of the substrate to one or more components 110 mounted on the other side of the substrate. The substrate may also be mounted on a heatsink 112 on the side of the substrate opposite to the sensor die. The bottom view 102 shows the sensor assembly without the heatsink, thereby further showing one or more components 110 mounted on the substrate 104. Although this figure shows four components, the sensor may comprise any number and any arrangement of one or more components, as can be further described herein.

[0027] As further shown in side view 100, the sensor die 108 comprises a thinned back surface 114 (not shown in Figure 1) and an energy-sensitive element 116 configured to detect energy 120 (such as light or electrons) illuminating the thinned back surface of the sensor die. The energy-sensitive element 116 is shown very generally herein, as embodiments can be applied to many different sensor configurations. The energy-sensitive element 116 generally comprises a combination of different elements (not shown), some of which may have different functions. The energy-sensitive element may include, for example, elements that actually detect energy as well as elements that store a signal or charge in response to the detected energy. In such an example, the energy-sensitive element may be configured such that energy is converted into a charge near the back surface 114 of the sensor die (or, in some cases, deeper in the bulk of the sensor die), and the signal charge is collected and stored in elements on or near the front surface 118 of the sensor die.

[0028] As further described herein, embodiments are particularly suitable for situations in which the sensor die operates in a vacuum, such as embodiments in which the energy-sensitive element is configured to detect DUV light, vacuum ultraviolet or extreme ultraviolet (VUV) light, electron beams, and / or X-rays. These embodiments are also suitable for non-vacuum applications, such as when the energy-sensitive element is configured to detect visible light or infrared (IR) light.

[0029] The sensor die can be configured as a charge-coupled device (CCD), a TDI sensor, or a complementary metal-oxide-semiconductor (CMOS) image sensor die. The sensor die can also be made from silicon (Si), indium gallium arsenide (InGaAs), indium antimony (InSb), cadmium telluride (CdTe), or any other compound suitable for energy detection across the entire spectrum, including but not limited to X-ray, VUV, DUV, visible, and IR light. While some embodiments described herein may relate to silicon-based sensor dies, the embodiments described herein can be applied to sensors made from any other suitable material.

[0030] In one embodiment, one or more components 110 are configured to perform one or more functions on the output generated by the energy-sensitive element 116, corresponding to the energy detected by the energy-sensitive element. One or more other components (or other dies) may be analog-to-digital (A / D) chips, digital-to-analog (DAC) components, image signal processing dies, application-specific integrated circuits (ASICs), or a combination thereof. One or more functions performed by one or more components may include, for example, amplification, A / D conversion, signal conditioning, digital image processing, and communication to an external computer. One or more functions may therefore be as simple as transferring the output of the sensor die to an external component of the sensor assembly, or they may include converting the output from one type to another, modifying the sensor output in some way, and so on. The assembly may utilize a variety of interfaces, including but not limited to pin grid arrays (PGA), ball grid arrays (BGA), flexible circuits, and land grid arrays (LGA), as described herein. In one embodiment, the substrate is formed of a ceramic material. The substrate is preferably a ceramic based on, for example, glass, alumina, aluminum nitride, or other materials selected as further described herein.

[0031] Figure 2 shows one embodiment of a method for forming a sensor. This figure includes the steps of fabricating a ceramic substrate and creating a sensor assembly from the substrate, but embodiments described herein may include fewer steps than all the steps shown in Figure 2. For example, this method may begin with step g) described below, rather than starting with step a) described below, and the steps prior to step g) may be performed by another method or system.

[0032] As shown in step a), the ceramic substrate 200 is fabricated. In this step, the substrate is generally fabricated thicker than the nominal design to include sacrificial material which will be removed in the next step. In step b), the top surface of the substrate is polished to the desired shape, thereby forming a ceramic substrate 202 with a polished top surface. This polishing process may expose internal vias (not shown in Figure 2) formed within the substrate. In step c), the bottom surface of the substrate is polished until it becomes a flat surface, thereby forming a ceramic substrate 204 with two polished sides. In Figure 2, the ceramic substrate is shown with both sides of the substrate polished to a specific shape, but the shapes of the two sides of the substrate may differ from those shown in Figure 2 and may be selected as further described herein.

[0033] In step d), metal 206 is deposited on the upper surface of the substrate to form a pattern. Pattern formation on a flat surface can be achieved using standard lithography methods known in the art. Pattern formation on a concave surface can be achieved using direct drawing methods for pattern formation, such as direct imaging, also known in the art. In step e), the same process as in step d) may be carried out on the bottom surface of the substrate, thereby forming metal 208 on the bottom surface of the substrate. Metals 206 and 208 may be formed from any suitable material known in the art and may have any suitable configuration known in the art. In step f), one or more components 210, such as an ASIC chip, are assembled on the bottom surface of the substrate using any suitable flip-chip process known in the art.

[0034] This method includes the step of forming isolated heat conduction structures on a substrate. For example, in step g), solder bumping is performed on the upper surface, thereby forming isolated heat conduction structures 212 on the upper surface of the substrate. This method also includes the step of modifying the shape of the isolated heat conduction structures based on a pre-selected shape of the thin back surface of the sensor die. For example, in step h), the isolated heat conduction structures, such as solder balls, may be stamped (coined) using a tool 214 having a polished surface 216 of a desired shape (a curved shape shown in Figure 2) that is lowered in the direction indicated by arrow 218, until the surface 216 contacts and exerts force on the isolated heat conduction structure.

[0035] The method further includes the step of bonding the front surface of the sensor die to the substrate via a separate thermal conductive structure, thereby giving the thin back surface of the sensor die a pre-selected shape. The sensor thus comprises a separate thermal conductive structure formed between the front surface of the sensor die and the substrate by a flip-chip process, thereby bonding the sensor die to the substrate and giving the thin back surface of the sensor die a pre-selected shape. At least a portion of the separate thermal conductive structure electrically connects the sensor die to one or more components 210 mounted on the substrate 204. For example, not all of the separate thermal conductive structure can be electrically connected to a component or device. In areas where a photosensitive element is formed (e.g., a film area), the conductive structure may provide mechanical and thermal advantages but may not have an electrical purpose. The sensor die may further be configured as described herein. For example, as shown in step i), the sensor die 220 may comprise a thin back surface 222, a front surface 224, and an energy-sensitive element (not shown in Figure 2) configured to detect energy irradiating the thin back surface of the sensor die.

[0036] In step i), the periphery of the sensor die 220 is soldered to the substrate using a contact method such as a hotbar or similar. The hotbar 226 may, for example, press down on the back surface 222 of the sensor die in the direction indicated by arrow 228 so that the periphery of the sensor die contacts and is soldered only to isolated heat conduction structures located near the periphery of the sensor die. In this way, the sensor die may be bonded to only a portion of the isolated heat conduction structures after this step, and the sensor die may be bonded to the remaining isolated heat conduction structures in a later step.

[0037] In one embodiment, the sensor comprises an underfill material formed around a separate heat conduction structure between the front surface of the sensor die and the substrate. In such an embodiment, the underfill material is configured to stabilize the sensor die when the sensor die is exposed to a vacuum. One novel and advantageous feature of the embodiments described herein is that, for example, a curved image sensor can be vacuum-based by being packaged on a ceramic substrate with underfill. The underfill resin 230 is added between the sensor die 220 and the substrate as shown in step j) to reinforce the solder joint. The underfill resin thus stabilizes the solder joint and can help maintain the shape of the sensor die even when a vacuum or other pressure is present on or to which the sensor is otherwise exposed. The sensor may need to be exposed to such a vacuum if, for example, the energy being detected is VUV light, EUV light, electrons, etc. In step k), the flow cell 232 is used to pressurize the sensor die 220 to establish contact between the thinner portion of the sensor die and the substrate. Furthermore, the underfill can also be cured in this step.

[0038] As shown in steps j) and k), the back surface of the sensor die may have a curved shape after the sensor die is bonded to all of the separate heat conduction structures, which are stamped or coined so that they combine to have a pre-selected shape. One novel and advantageous feature of the embodiments described herein is that a thin, curved image sensor with a back surface can thus be assembled by a flip-chip process, enabling both back-illumination such as DUV and VUV, and substantially high-speed operation.

[0039] The steps shown in Figure 2 may be modified in one or more methods further described herein. For example, in step b), if the application may require a substantially flat sensor die, the front surface of the ceramic may be polished to a flat surface rather than a curved surface. In that case, the method shown in Figure 2 may be applied just as similarly, but the manufacturing and assembly processes of the ceramic may be greatly simplified.

[0040] Each step of this method may be further performed as described herein. This method may also include any other steps that can be performed using the sensors, imaging systems, computer subsystems, components, etc., described herein. The sensors formed by the method described above, and the imaging systems comprising such sensors, may be configured according to any of the embodiments described herein. This method may be performed by any of the embodiments of the systems described herein.

[0041] The general idea of ​​assembling an integrated circuit (IC) on a substrate using coined solder bumps is described in U.S. Patent Application Publication No. 2012 / 0309187 by Sri-Jayantha et al., published on December 6, 2012, and is incorporated herein by reference as if it were fully described. However, a challenge in implementing such a method for image sensors arises from the undesirable fact that mechanical contact on the sensor die surface can damage the pixel array and result in a relatively low assembly yield. Figure 3 shows one embodiment of a method by which a substrate may be formed, and Figure 4 shows one embodiment of the process flow for sensor assembly.

[0042] In step 300 of Figure 3, the ceramic substrate 302 is fabricated using a co-firing process, which may be any preferred process known in the art. The substrate will exhibit warping and non-flatness due to the difference in shrinkage of the ceramic layer and the conductive ink (not shown) within the ceramic. In the next step, an interface material is added. The interface material is preferably soft enough to allow formation by stamping or coining. Examples of such interface materials include, but are not limited to, solder bumps and gold studs. For example, in step 304, solder bumps 306 may be formed on the ceramic substrate 302. In the case of gold studs, in step 308, gold studs 310 may be formed on the ceramic substrate 302.

[0043] The interface material is then coined using a forming tool. For example, in step 312, the solder bump 306 can be stamped or coined by moving the tool 314 in the direction indicated by arrow 316, bringing the tool into contact with the solder bump, and applying force to the solder bump. Similarly, in step 318, the gold stud 310 can be stamped or coined by moving the tool 320 in the direction indicated by arrow 322, bringing the tool into contact with the gold stud, and applying force to the gold stud.

[0044] One novel and advantageous feature of the embodiments described herein is that the shape of a thin-backface image sensor can be substantially precisely controlled by stamping / coining bumps beneath the film, thereby allowing the sensor shape to conform to a specific image field curvature. For example, in one embodiment, isolated thermal conduction structures are formed on the substrate before the sensor die is bonded to the substrate in a flip-chip process, and one or more of the shapes of the isolated thermal conduction structures are deformed so that the combination of isolated thermal conduction structures has substantially the same shape as a pre-selected shape. In another embodiment, the pre-selected shape is determined before the flip-chip process, and one or more of the shapes of the isolated thermal conduction structures formed on the substrate are modified based on the pre-selected shape before the sensor die is bonded to the substrate in a flip-chip process. As shown in Figure 3, the surface shape of the tool used to stamp or coin the isolated thermal conduction structures may differ in various embodiments and may vary depending on the pre-selected shape. In particular, the shape of the surface of the tool that contacts the solder bump or gold stud may be substantially the same as a pre-selected shape, such that the pre-selected shape is transferred to the solder bump or gold stud, and then transferred to the sensor die when the sensor die is joined to the solder bump or gold stud.

[0045] In one embodiment, the surface of the substrate on which the separated heat conduction structures are formed has a shape different from the pre-selected shape. For example, in the embodiment shown in Figure 3, the shape of the solder bumps is modified by stamping or coining, performed using a tool having a substantially flat surface, so that the combination of solder bumps has a substantially flat surface after stamping or coining. Such stamping or coining of solder bumps would be appropriate if the pre-selected shape is substantially flat. As can be seen from Figure 3, stamped or coined solder bumps can also have a substantially flat surface throughout the combination of solder bumps, even if the ceramic substrate is not flat or is warped. In this way, when the front surface of the sensor die is joined to the stamped or coined solder bumps, the sensor assembly can have a substantially flat outline, even if the ceramic substrate is not flat or is warped.

[0046] In contrast, as shown in Figure 3, the shape of the gold studs is modified by stamping or coining performed with a tool having a curved surface, resulting in a combination of gold studs having a curved surface after stamping or coining. Such stamping or coining of gold studs would be appropriate if the pre-selected shape is a curved shape. As can be seen from Figure 3, stamped or coined gold studs can also have a curved surface across the entire combination of gold studs, which differs from the shape of the ceramic substrate. In this way, when the front surface of the sensor die is bonded to the stamped or coined gold studs, the sensor assembly can have a curved shape despite the ceramic substrate being non-flat or warped, and despite the difference between the pre-selected curved shape of the back surface of the sensor die and the shape of the surface of the ceramic substrate. Thus, by modifying the shape of the isolated heat conduction structure, as described herein, constraints on the shape of the ceramic substrate are reduced.

[0047] Figure 4 further illustrates how solder can be coined onto any surface of any shape, whether concave (top), convex (bottom), or otherwise arbitrary. In other words, Embodiment 400 shows a method for assembling a sensor that conforms to the concave shape of the solder ball, and Embodiment 402 shows a method for assembling a sensor that conforms to the convex shape of the solder ball. In both cases, the solder ball 404 may be formed on a substrate 406. As shown in both embodiments, the substrate does not need to have the same shape as the shape selected for the sensor die, and therefore has a shape that is neither concave nor convex. The solder ball and substrate may be formed as described herein and may be configured as further described herein. Although Figure 4 illustrates and illustrates a solder ball, other isolated heat conduction structures described herein may be modified in shape as shown in this figure.

[0048] The shape of the molding tool and the separated heat conduction structure will determine the precise shape of the sensor die assembled on the substrate. Once the sensor die is placed on the molded, separated heat conduction structure, the separated heat conduction structure is heated and reflowed, establishing a permanent connection between the sensor die and the substrate. During heating, the sensor die needs to be pressed against the substrate, which can be done by various means, such as pressing the periphery of the sensor die with a hot bar or applying gas pressure through a dedicated jig. In the embodiment of Figure 4, the partial housing 408, in combination with a sealing ring 410 that contacts the sensor die 412, forms a flow chamber 414. For example, by applying pressure to the sensor die through the flow chamber by a gas flow 416 that controls the pressure within the flow chamber, the sensor die is forced to conform to the shape of the coined, separated heat conduction structure.

[0049] Embodiments further described herein illustrate practical examples in which the sensor die is designed to have a specific shape. As further described herein, in one embodiment, the pre-selected shape is a curved shape. In another embodiment, the pre-selected shape is defined by a higher-order polynomial. It is worth highlighting, for example, the ability to describe the sensor shape with a higher-order polynomial, which can increase the flexibility of optical system design.

[0050] In some embodiments, the surface of the substrate on which the isolated heat conduction structure is formed has a shape determined based on a pre-selected shape. For example, if the ceramic substrate is substantially out of the desired shape and solder balls cannot fill the gap between the shape of the desired sensor die and the shape of the ceramic, the ceramic substrate can be polished to the desired shape and an upper metal pattern defining the solder pads can be patterned on the surface. This can be done as described and illustrated in steps b) and d) above with reference to Figure 2.

[0051] In some embodiments, the substrate is formed of a material selected based on the coefficient of thermal expansion (CTE) of the material, which is determined by the size and pre-selected shape of the sensor die. In another embodiment, the isolated heat conduction structure is formed of a material selected based on the reflow temperature of the material, which is determined by the size and pre-selected shape of the sensor die. Herein, we describe how to select substrate and solder materials that can help ensure that the assembled sensor can reliably meet performance requirements.

[0052] Since solder reflow is performed at high temperatures, both the sensor die and the ceramic substrate shrink when cooled to room temperature or operating temperature. This temperature change presents several challenges to both achieving the desired die shape and ensuring the reliability of the solder joint due to stress. The conditions for a substantially stress-free solder joint can be determined based on geometric considerations using a simple 1D model.

[0053] An embodiment showing an initially flat sensor die and a fully fitted final shape is shown in Figure 5. The sensor die likely has an initial length Ls, indicated by dimension 500 in Figure 5, during solder reflow in step i) of Figure 2 at a relatively high temperature. The sensor die will then conform to the shape of the coined bumps, assuming a final length Ls', indicated by dimension 502 in Figure 5, after cooling to the target temperature and performing the subsequent steps j) and k) of Figure 2. The substrate will also shrink due to the temperature difference, so the length between the outer bump pads will shrink from Lc, indicated by dimension 504, to Lc', indicated by dimension 506. The mismatch between the shrinkage differences will result in residual stress at the solder joints, which will be worst in the outer bumps near the periphery of the die.

[0054] In this example, minimizing the difference (Lc'-Ls') will minimize the stress on the outer solder joint. When Lc'-Ls'=0, a stress-free solder joint is achieved. The requirements to satisfy this condition can be modified by simple geometric considerations. In the example shown in 1D, assuming no curvature of the ceramic substrate, the following is obtained:

number

[0055] This example, though simplified and involving several approximations, illustrates how to select key process conditions to minimize solder stress in the proposed assembly. These conditions can be obtained using numerical modeling for actual 2D geometry. In the simple case of a flat sensor die (R=infinite), optimal conditions are obtained when the CTEs between the sensor die and the ceramic substrate are nearly identical. Ceramic IC substrates containing silicon with perfectly matched CTEs are also commercially available from several suppliers. These materials include both oxide and non-oxide ceramics across a range of CTEs. Non-oxide ceramics include aluminum nitride (CTE approximately 4.4–4.7 ppm / °C), silicon carbide (CTE approximately 3.7–3.9 ppm / °C), and silicon nitride (CTE approximately 2.8–3.5 ppm / °C). This range represents various compositions with different CTEs for each type. Oxide-based ceramics, such as those commercially available from Kyocera Corporation in Kyoto, Japan, include materials with CTEs ranging from 3.4 to 12.3 ppm / °C. For silicon dies, the CTE is 2.6 ppm / °C. Given the die size and curvature, the material with the optimal CTE can be selected using any suitable numerical modeling known in the art.

[0056] To further minimize discrepancies, solder at the desired reflow temperature is used. A variety of solder materials are commercially available across the entire temperature range from approximately 60°C to over 220°C. Solder materials include indium-bismuth-tin (In-Bi-Sn), indium-bismuth (In-Bi), indium-tin (In-Sn), tin-silver-copper (SACx), and other alloys, and are commercially available from multiple suppliers. By selecting the appropriate solder and ceramic material, how parameter α c The selection of ΔT will reveal whether the solder joint can be brought close to a substantially stress-free state. A separate heat-conducting structure made of solder may contain multiple materials with distinctly different melting points. It is advantageous that one of the materials may be a higher-melting-point solder that can more easily maintain the desired shape, and the other may be a lower-melting-point solder that can electrically connect to a component or device at a much lower temperature. Such a combination of materials can be selected in any preferred way from any suitable commercially available solder material.

[0057] To enable substantially high-speed operation, a relatively large number of interconnect signals must be extracted from the sensor die to one or more components, such as an ASIC. Transistors are required to drive many circuits. An example of such an embodiment is presented in U.S. Patent No. 10,764,527, granted to Chuang et al. on September 1, 2020, which is incorporated herein by reference as if it were described in its entirety. The sensors described herein may further be configured as described herein. To enable substantially high-speed operation and a relatively large number of signals, the packaging technology needs to provide relatively high-density routing of the ceramic substrate and relatively low channel parasitic capacitance. Both of these are simultaneously possible by flip-chip assembly of the sensor die on low-temperature cofired ceramic (LTCC). Such technology makes interconnects with a pitch of 150 μm or less available. Such interconnects can transfer data exceeding 10 gigasamples per second (GS / sec) across the entire ceramic substrate to the ASIC.

[0058] Figure 6 shows a cross-section of a complete sensor assembly comprising a sensor die 600, a ceramic substrate 602, a heat sink 604, an interface material comprising a resin 606 and an interface material 608, and a bump 610 for bonding the sensor die to the substrate. The embodiment shown in Figure 6 may include other elements not shown in this figure but described further herein. For example, a portion of the complete sensor assembly shown in Figure 6 corresponds to the central portion of the sensor assembly shown in Figure 1, which lies between the electrical interconnects and one or more components, such as an ASIC. Only a portion of the entire sensor assembly is shown in Figure 6 so that further details regarding heat transfer within the sensor assembly can be more clearly illustrated.

[0059] The superimposed arrows in this figure indicate the magnitude and direction of the heat flux within the sensor assembly. Embodiments of the assembly method described herein have thermal advantages for the sensor assembly. For example, the bump 610 supporting the sensor die not only defines the shape of the sensor die but also functions as an efficient heat conduit for the heat generated by the image sensor die. The gap between the sensor die and the ceramic substrate may be filled with a resin to solidify and reinforce the solder joints for increased reliability. All resins developed in the electronics industry for this purpose exhibit relatively low thermal conductivity, typically less than 1 W / mK (watts per meter Kelvin). The thermal conductivity of the solder bump or gold pillar is substantially higher, meaning that the majority of the heat flux is conducted through the isolated heat conduction structure of the solder bump or gold pillar. Figure 6 shows the result of thermal modeling such an interface with resin and solder bumps. The thickness shown in the cross-sectional view of the assembly is not proportional to the actual size for clarity. The superimposed arrows indicate the direction of the heat flow, and the length of the arrows is proportional to the magnitude of the heat flux. As shown in this diagram, the thermal conductivity from the sensor die to the ceramic substrate will be determined primarily by the bumps, not the resin.

[0060] In one embodiment, the sensor comprises thermally conductive and conductive vias formed within a substrate, at least a subset of which are configured to connect at least a portion of a separated thermal conduction structure to one or more components, thereby connecting the sensor die to one or more components. The ceramic substrate 602 may, for example, comprise an array of thermally conductive and conductive vias (not shown in Figure 6) to further improve the thermal performance of the sensor assembly. In another embodiment, the underfill material comprises a resin containing dispersed particles formed from a dielectric material having high thermal conductivity. The underfill resin 606 may preferably contain, for example, dispersed particles (not shown). Such particles are preferably made from a dielectric material having high thermal conductivity, such as aluminum nitride, sapphire, or diamond. As used herein, the term “high thermal conductivity” is generally defined as a thermal conductivity of 30 W / mK to 2000 W / mK.

[0061] Another embodiment relates to an imaging system. The imaging system generally comprises an energy source (e.g., a light source, an electron beam source, etc.) configured to generate energy directed onto a sample by an illumination subsystem. Such energy sources and illumination subsystems are further described herein and may be configured as shown in Figures 11 and 11a. In some embodiments, the sample is a wafer. The wafer may include any wafer known in the semiconductor technology field. While some embodiments described herein may be for one or more wafers, the embodiments are not limited to the samples that can be used. The embodiments described herein may be used for samples such as reticles, planar panels, personal computer (PC) boards, and other semiconductor samples.

[0062] The system also includes a sensor configured to detect energy from a sample and generate an output corresponding to the detected energy. The sensor is configured as described further herein. The energy detected by the sensor may include any energy described herein, such as electrons, charged particles, X-rays, VUV light, EUV light, DUV light, visible light, and IR light. The types of energy detected by the sensor may also include specularly reflected light, scattered light, or both, depending on the system configuration, as described further herein. The output generated by the sensor may include any preferred output, such as image data, image signals, non-image data, non-image signals, or any combination thereof. One or more elements of the sensor and the imaging system coupled to the sensor may be configured as described further herein.

[0063] In one embodiment, the imaging system includes a camera lens subsystem configured to direct energy from a sample to a sensor. For example, an image sensor having a pre-selected curved shape may be used in the camera lens system, as shown in Figure 7a, or an image sensor having a pre-selected flat shape may be used in the camera lens system, as shown in Figure 7b. In Figure 7a, the camera lens subsystem includes an image sensor 710 having a curved shape, four refractive lenses 700, 702, 704, and 706 that work together to focus light 712, and an aperture diaphragm 708. Similarly, in Figure 7b, the camera lens subsystem includes an image sensor 724 having a substantially flat shape, four refractive lenses 714, 716, 718, and 720 that work together to focus light 726, and an aperture diaphragm 722. Figures 7a and 7b show two designs optimized for curved image sensors versus flat image sensors. In the case of a curved image sensor, the curvature of the sensor is also optimized. For such a system, the following system specifications can be selected as an example: FOV of ±14 degrees, aperture diaphragm diameter of 20 mm, sensor format of 50 mm diagonal, image space F-number (f / #) = 5.5, and effective focal length (EFL) 728 of 100 mm. These parameters represent an example of typical application requirements for narrowband designs that can be selected for wavelengths targeted in the DUV, visible, or IR spectrum.

[0064] Although the camera lens subsystem is shown in Figures 7a and 7b as comprising four refractive lenses, the camera lens subsystem may comprise a different number of refractive lenses. The camera lens subsystem shown in Figures 7a and 7b may also be modified to include one or more reflective lens elements (not shown) in place of, or in combination with, one or more refractive lens elements. Furthermore, as can be seen from Figures 7a and 7b, the shapes of the refractive lens elements, which are only schematically shown in these figures, may be modified depending on the shape of the image sensor. In particular, as can be seen from Figures 7a and 7b, refractive lenses 704 and 718 have substantially different shapes, and even refractive lenses 706 and 720 have at least slightly different shapes. The shapes of all refractive elements shown in Figures 7a and 7b are not intended to limit or indicate the properties of any actual refractive lens or reflective mirror that may be used with an image sensor. Rather, as will be apparent to those skilled in the art, the refractive lenses and any other elements provided in the camera lens subsystem can be optimized not only according to the sensor configuration but also according to the overall configuration of the imaging system, in addition to the application in which the sensor is used (e.g., scattered light vs. specular light, inspection vs. measurement, DUV light vs. VUV light, etc.).

[0065] Figure 7c shows the curvature of the image sensor die, indicated by a surface sag of 730, which is an optimized result. Such a surface shape can be achieved using the method described above. The exact values ​​shown in this plot, and the exact curvature to which these values ​​correspond, are not important to understanding this embodiment. These values ​​and surface sag can be determined and optimized in any preferred way known in the art based on considerations further described herein, and this plot is included here to illustrate, with illustration, how the shape characteristics of the sensor die can be determined and optimized. Figure 7d shows the geometric mean square (RMS) of the spot size over the entire FOV. The RMS of the spot size is a measure of the amount of aberration introduced by the system. The two plots correspond to two cases: a flat sensor die (solid line) versus a curved sensor die (consisting of both dashed lines and dots). The other lines shown in this figure (consisting only of dashed lines) correspond to the diffraction limit of imaging in the imaging system. As is clearly evident from this plot, a curved sensor die can significantly reduce image blur compared to a flat sensor die. In inspection systems, reduced blur leads to a higher level of defect detection, i.e., higher inspection sensitivity. Again, the exact values ​​shown in the plot of Figure 7d, and the precise aberrations to which those values ​​correspond, are not important for understanding this embodiment. This plot is included here simply to illustrate, using diagrams, how various sensor die shapes can affect the aberrations of an imaging system.

[0066] Aperture diaphragms based on curved image sensor dies can also be larger than those based on flat sensor dies. In the example shown in Figure 7a, the aperture diaphragm can be increased to a maximum diameter of 26 mm while accommodating the aberrations of the initial design. This allows the light-gathering power of the camera lens subsystem to be (26 / 20). 2This means a doubling, or approximately 70%, increase. This enhanced light-gathering capability improves the detected optical signal, resulting in a higher level of defect detection. Such lenses can be optimized for imaging applications of wafers, panels, or IC substrates, intended for film measurement, for visual inspection and review of relatively large defects.

[0067] In another embodiment, the system comprises a tube lens subsystem configured to direct energy from a sample to a sensor. Such a configuration may be used for the tube lens of a DUV camera used with a microscope. In this embodiment, an image sensor having a pre-selected shape is used with the tube lens subsystem, as shown in Figures 8a and 8b. For example, as shown in Figure 8a, an image sensor die having a pre-selected curved shape may be used in the tube lens system, and as shown in Figure 8b, an image sensor die having a pre-selected flat shape may be used in the tube lens system. In Figure 8a, light may be incident on the tube lens subsystem through an aperture 800, and the tube lens subsystem comprises three refractive lenses 802, 804, and 806 that, in combination, focus the light 808 on a curved image sensor die 810. Similarly, in Figure 8b, light may be incident on this embodiment of the tube lens subsystem through aperture 814, and the tube lens subsystem comprises three refractive lenses 816, 818, and 820 that combine to focus the light 822 onto an image sensor die 824 having a substantially flat shape. As an example, the following system specifications were selected: FOV = ±10 degrees, aperture diameter = 8 mm, sensor format diagonal 12 mm, image space F-number (f / #) is 4.2, EFL 828 is 34 mm, and total lens track is <40 mm.

[0068] Although the tube lens subsystem is shown in Figures 8a and 8b as comprising three refractive lenses, the tube lens subsystem may comprise a different number of refractive lenses. The tube lens subsystem shown in Figures 8a and 8b may also be modified to include one or more reflective and / or diffractive lens elements (not shown) in place of, or in combination with, one or more refractive elements. Furthermore, as can be seen from Figures 8a and 8b, the characteristics of the elements, which are only schematically shown in these figures, may be modified depending on the shape of the image sensor die. In particular, as can be seen from Figures 8a and 8b, the refractive lenses 806 and 820 have distinct shapes. The design examples shown in Figures 8a and 8b are not intended to limit or indicate any design or lens material required for an image sensor. Rather, as will be apparent to those skilled in the art, the elements provided in the tube lens subsystem can be optimized not only according to the application in which the sensor is used (e.g., scattered light versus specular light, inspection versus measurement, DUV light versus VUV light, etc.), but also according to the overall configuration of the imaging system as well as the configuration of the sensor.

[0069] A tube lens subsystem comprising three elements was optimized for two cases: one in which the curvature of the sensor die is permitted, as shown in Figure 8a, and another in which the sensor die is constrained to be flat, as shown in Figure 8b. The results of these optimizations are shown in Figures 8c and 8d, respectively. Figure 8c shows the optimized contour plot of the sensor surface sag in millimeters. This corresponds to a radius of curvature of approximately 50 mm. A sensor die having such a shape can also be obtained using embodiments of the methods described herein. By using only three optical elements with two conical surfaces (L1-L and L3-L), a diffraction-limited design can be obtained over the entire FOV.

[0070] The aberrations of this design are approximately 10 times smaller compared to a comparable design based on a planar image sensor die, as shown in the RMS spot size plot 832 in Figure 8d. Most importantly, this design reduces distortion by more than twofold, substantially reduces the angle of the principal ray, and results in more uniform sensor response across the entire FOV. Geometric aberrations are in the submicron range, substantially below the pixel size of any actual image sensor. Smaller aberrations are particularly beneficial for optical inspection applications of patterned wafers, panels, or IC substrates. This example demonstrates how the use of a curved image sensor die makes it possible to design simple tube lenses for these applications. In particular, using the curvature of the sensor reduces the number of reflective elements required and expands the actual image plane size possible in systems using EUV illumination. Such systems are highly constrained and require expensive elements and manufacturing methods.

[0071] The exact values ​​shown in the plot of Figure 8c, and the precise curvatures to which those values ​​correspond, are not important to understanding this embodiment. These values ​​and surface sag can be determined and optimized in any preferred manner known in the art, based on considerations further described herein, and this plot is included to illustrate, graphically, how the characteristics of the sensor die shape can be determined and optimized. Figure 8d shows the geometric RMS of the spot size over the entire FOV. The RMS of the spot size is a measure of the amount of aberration introduced by the system. The two plots correspond to two cases: a flat sensor die (solid line) versus a curved sensor die (dashed line). As is clearly visible from this plot, a curved sensor die can significantly reduce blur compared to a flat sensor die. Again, the specific values ​​shown in this plot of Figure 8d are not important to understanding this embodiment. This plot is included to illustrate, graphically, how various sensor die shapes can affect the aberrations of the imaging system.

[0072] Embodiments described herein may include an array of curved sensors. For example, embodiments described herein may include two or more of the sensors described herein, the sensor dies of which may have the same characteristics, such as the same shape and size, or may have one or more different characteristics, such as different shapes and / or different sizes, that are selected in advance. In such an embodiment, the imaging system includes an additional sensor configured to detect additional energy from a sample and produce an output corresponding to the detected additional energy. The additional sensor may be configured to detect additional energy and produce an output, as further described herein. The energy detected by the two sensors may have one or more different characteristics, such as the type of energy (scattering versus specular reflection), wavelength, and polarization. For example, different detection channels may include different sensors, each of which may be configured as described herein, as further described below. However, one particularly advantageous embodiment of the multiple sensor embodiments comprises multiple sensors coupled to the same collector or collection subsystem such that the multiple sensors detect the energy of the same image plane, even if the image plane is curved or otherwise has some non-flat shape.

[0073] An additional sensor comprises an additional substrate and one or more additional components mounted on the additional substrate. The additional sensor also comprises an additional sensor die having a thin back surface, and an additional energy-sensitive element configured to detect additional energy from a sample illuminating the thin back surface of the additional sensor die. The additional sensor further comprises an additional isolated thermal conduction structure formed between the front surface of the additional sensor die and the additional substrate by a flip-chip process, thereby bonding the additional sensor die to the additional substrate and giving the thin back surface of the additional sensor die an additional pre-selected shape. At least a portion of the additional isolated thermal conduction structure electrically connects the additional sensor die to one or more additional components. Each such element of the additional sensor may further be configured as described herein.

[0074] Using two or more of the sensor embodiments described herein in a single detection channel may be particularly beneficial in several situations, such as when light from a sample is directed over a relatively large area of ​​the image plane, and / or when the image plane has curvature that cannot be easily achieved with a single sensor. In any case, the pre-selected shape and the additional pre-selected shape (the pre-selected shape of the back surface of the sensor die of another sensor) may be different or the same. In addition, as further described herein, in some embodiments the imaging system is configured to control the position of the sensors of the imaging system and the additional sensors separately. The imaging system may be configured to control the position of each sensor in any preferred manner using any suitable software and / or hardware known in the art of image system control.

[0075] Figure 9a shows one embodiment in which multiple image sensors with varying curvatures are arranged in a tiled configuration. In this embodiment, the rearmost optical element 900 of the imaging system (not shown in Figure 9a) in front of the image plane directs light 902 to four image sensors 904, 906, 908, and 910. In this embodiment, the array of curved sensors is positioned at the focal plane. Examples of possible sensor arrangements and their applications to inspection can be found in U.S. Patent Application Publication No. 2004 / 0175028 by Cavan, published September 9, 2004, and U.S. Patent No. 9,077,862 granted to Brown et al., July 7, 2015, which are incorporated herein by reference as if they were described in their entirety. The imaging system may further be configured as described in these references. In systems with a relatively wide field of view and relatively high magnification, the image space may be substantially large, and multiple sensors may be used to cover the entire image. Such an image space will typically have a relatively large curvature, represented by a superposition of Zernike functions known in the art, or a superposition of even and odd polynomials. To obtain a substantially high-quality image, the focal plane may be curved in accordance with the field curvature.

[0076] Figure 9a shows an example of an array of four image sensors. These image sensors would preferably have curvature that precisely mimics the curvature of the image plane shown in Figure 9b. This figure shows a contour plot 912 of the surface sag. The contour plot represents the surface shape optimized for best optical performance. The exact optimal shape will vary depending on the specific imaging system design, but most imaging systems will produce curvature essentially similar to that shown in the figure. The focal plane (or focal surface) is imaged by the array of sensors for the reasons described in the references above. Figure 9c shows examples 914 of curvatures 916, 918, 920, and 922 for the image sensors 904, 906, 908, and 910 shown in Figure 9a, respectively, where they may need to approach the best focal surface.

[0077] The amount of surface sag for each individual sensor, as shown in Figure 9c, may be too large to achieve using the assembly process described herein. Similarly, the deviation of surface sag across each sensor, represented by the density of contour lines, may be too large to achieve using the proposed method. To address this problem, the following enhancements can be implemented. It is not necessary to assemble different image sensors on a common substrate; each sensor may have its own substrate, and the position of the sensor can be changed in all six degrees of freedom relative to all other sensors. For example, each of the sensors 904, 906, 908, and 910 shown in Figure 9 may be formed on its own substrate. Alternatively, two or more sensors 904, 906, 908, and 910 may be formed on a single substrate, and any sensor formed on the same substrate may be separated, for example, by cutting the substrate, thereby generating four distinct sensors on four distinct substrates. In either case, the position of each sensor along the optical axis (Z axis), as well as the tilt angle of each sensor in the XZ and YZ planes, can be manipulated to achieve the minimum surface sag of the image sensor relative to the substrate. Figure 9d shows the result of such optimization. In this case, plot 924 shows the curvatures 926, 928, 930, and 932 of the image sensors 904, 906, 908, and 910 shown in Figure 9a, and the direction 934 in which the swath is scanned on the sample. The sparse contour lines of the surface sag in this figure (compared to Figure 9c) indicate a substantially gentle curvature and surface sag of the image sensor that can be more easily achieved using the proposed method.

[0078] Figures 9a to 9d show specific configurations, but the method for optimizing the image sensor shape is essentially general. The shape of the focal surface can always be optimized to obtain the best imaging performance. The position and tilt angle of each image sensor along the Z-axis can always be optimized to minimize curvature and surface sag. Such optimization typically reduces the magnitude of surface sag by an order of magnitude, making the assembly methods outlined herein easier to use. Therefore, in most common cases, the precise surface shape of each sensor can be expressed using Zernike coefficients. Since each sensor is off-axis with respect to the optical system, the surface will generally not be axially symmetric.

[0079] It should be noted that the various numerical values ​​in Figures 9b to 9d are not important for understanding and fully disclosing the embodiments described herein. These values ​​are not intentionally illegible, but rather due to the original versions of these plots and their reproducibility. The inclusion of these values ​​in this application is merely to illustrate how embodiments of multiple sensors can be configured and optimized based on the image plane shape within the image plane.

[0080] The checker sensor pattern shown in this embodiment is generally useful in both raster scanning and step-and-repeat inspection systems. In the former, which is raster scanning, the sample being inspected moves in the y-direction (see Figure 9a), and the image of the sample moves along the y-axis relative to the image sensor. The overlap of the image sensors in the x-direction creates a gapless swath of the sample image in the inspection area along the x-direction. In a step-and-repeat inspection system, the sensor positions can be selected so that there is no gap in the inspection area after each of the two resulting inspection steps.

[0081] The embodiments described herein are also effective in reducing distortion-induced blurring in scanning inspection systems. In some embodiments, the imaging system includes a scanning subsystem configured to scan the sample with energy directed onto the sample by an illumination subsystem, the illumination subsystem having a substantially non-curved field of view (FV) on the sample, and a pre-selected shape of the sensor die being curved. The scanning and illumination subsystems may be further configured as described herein.

[0082] Figure 10 illustrates the advantages of such an embodiment by showing optical distortion and its effect on the optical resolution of the scanning system. The “perfect” grid and the curved grid in image space, projected onto object space, result in the inversion of distortion of the curved grid 1000 and the perfect grid 1002, respectively. A “perfect” grid is defined herein as a grid that is substantially free of curvature across its entirety. In other words, the term “perfect” grid as used herein is defined as a grid that is substantially flat or has only a negligible amount of non-flatness across its entirety. Scanning imaging systems with relatively wide FOVs typically exhibit considerable field curvature, as shown by the curved grid 1000. Just as a perfect grid in object space corresponds to a distorted grid in image space, the reverse is also true; that is, when a perfect pixel array is projected onto the object space in which the sample under inspection is held, a curved grid 1000 is generated.

[0083] As the sample is scanned across the optical system and sensor, a defect of interest (DOI) located at the edge of the image plane may exhibit smearing across multiple pixels. For example, if light from defects 1004 and 1006 on the sample (not shown in Figure 10) is scanned across the curved grid 1000 at swaths 1008 and 1010, respectively, the light from the defects will be scanned across different parts of the grid with different curvatures. In particular, defect 1004 is scanned across the center of the image plane where there is little to no horizontal distortion, while defect 1006 is scanned across the edge of the image plane where there is relatively large horizontal distortion. Therefore, even if defects 1004 and 1006 all have the same characteristics and are illuminated with light of the same characteristics, the output signals generated by the sensor for these defects may differ. For example, as shown in Figure 10, image 1012 of defect 1004 is substantially different from image 1014 of defect 1006 due to light smearing from defect 1006 across multiple pixels at the edge of the image plane, caused by distortion of the image plane.

[0084] In one such embodiment, the sensor is configured as a TDI sensor. For example, in an imaging system configured for inspection as described herein, a TDI sensor is typically used, which accumulates optical signals so that the image is scanned across the entire pixel array. This results in a larger effective point spread function (PSF), and therefore lower resolution and signal-to-noise ratio (SNR). Conversely, assembling an image sensor on a curved surface simulates the distortion of the system. Thus, such a sensor projected onto an object surface corresponds to a nearly perfect grid, represented by the perfect grid 1002, which improves resolution and SNR away from the center of the image plane. For example, as shown in Figure 10, the image 1016 of defect 1004 is substantially the same as the image 1018 of defect 1006, thanks to the substantially perfect grid 1002, which eliminates the smear of light from defect 1006 across multiple pixels at the edge of the image plane. Therefore, the embodiments described herein enable further significant improvements to systems based on the TDI scanning architecture.

[0085] One embodiment of the imaging system is shown in Figure 11. The imaging system 1100 comprises and / or can be combined with a computer subsystem, for example, a computer subsystem 1102, and / or one or more computer systems 1104, which may be configured as described further herein. This imaging system is based on the flip-chip sensor embodiment described herein and may be configured for a variety of applications such as inspection or measurement.

[0086] The imaging systems described herein generally comprise at least an energy source, a sensor, and a scanning subsystem. The energy source is configured to generate energy which is directed onto the sample by the illumination subsystem. The sensor is configured to detect energy from the sample and generate an output corresponding to the detected energy. The scanning subsystem is configured to change the position on the sample to which energy is directed and energy from the sample is detected. In one embodiment, as shown in Figure 11, the energy directed onto the sample is light, and therefore the imaging system is configured as a light-based imaging system.

[0087] In the embodiment of the imaging system shown in Figure 11, the imaging system includes an illumination subsystem configured to direct light onto the sample 1106. The energy source includes at least one light source, for example, light source 1108. The illumination subsystem is configured to direct light onto the sample at one or more angles of incidence, which may include one or more oblique angles and / or one or more normal angles. For example, as shown in Figure 11, light from light source 1108 is directed onto the sample 1106 at an oblique angle of incidence, passing through optical element 1110 and then lens 1112. The oblique angle of incidence may include any preferred oblique angle of incidence, which may vary depending, for example, the characteristics of the sample and the process being performed on the sample.

[0088] The illumination subsystem can be configured to direct light onto the sample at various angles of incidence from time to time. The imaging system can be configured to modify one or more properties of one or more elements of the illumination subsystem so that, for example, light can be directed onto the sample at angles of incidence different from those shown in Figure 11. In such an example, the imaging system can be configured to move the light source 1108, optical element 1110, and lens 1112 so that light is directed onto the sample at various oblique angles of incidence or normal (or near-normal) angles of incidence.

[0089] The imaging system may, in some cases, be configured to direct light onto the sample at multiple incident angles simultaneously. The illumination subsystem may, for example, include multiple illumination channels, one of which may include the light source 1108, optical element 1110, and lens 1112 shown in Figure 11, and another illumination channel (not shown) may be configured differently or be the same, include similar elements, or include at least a light source and, optionally, one or more other components as further described herein. When such light is directed onto the sample simultaneously with other light, one or more properties (e.g., wavelength, polarization, etc.) of the light directed onto the sample at different incident angles may differ, thereby allowing the sensor to distinguish between the light resulting from the illumination of the sample at different incident angles.

[0090] In another example, the imaging system may comprise only one light source (e.g., light source 1108 shown in Figure 11), and the light from the light source may be separated into different optical paths by one or more optical elements (not shown) of the illumination subsystem (e.g., based on wavelength, polarization, etc.). Each of the light from the different optical paths may then be directed to the sample. Multiple illumination channels may be configured to direct light to the sample simultaneously or at different times (e.g., when different illumination channels are used to continuously illuminate the sample). In another example, the same illumination channel may be configured to direct light to the sample with different characteristics at different times. Optical element 1110 may be configured, for example, as a spectral filter, and the characteristics of the spectral filter can be changed in various different ways (e.g., by swapping one spectral filter with another) so that light of different wavelengths can be directed to the sample at different times. The illumination subsystem may have any other preferred configuration known in the art for directing light with different or the same characteristics to the sample continuously or simultaneously at different or the same angle of incidence. The illumination subsystem may be configured so that light enters the sample from below (not shown in Figure 11) and passes through the sample before being received by the sensor.

[0091] The light source 1108 may include a narrowband light source such as a laser, or a plasma source such as an EUV or broadband plasma (BBP) light source. Thus, the light generated by the light source and directed towards the sample may include narrowband or broadband light. The light source may also include a laser design known in the art and configured to generate light of any suitable wavelength. The laser may be configured to generate monochromatic or nearly monochromatic light. The laser may thus be a narrowband laser. The light source may also include a polychromatic light source that generates light at multiple individual wavelengths or wavelength bands.

[0092] Light from the optical element 1110 can be focused onto the sample 1106 by the lens 1112. Although the lens 1112 is shown as a single refractive optical element in Figure 11, the lens 1112 may actually comprise several refractive, diffracting, and / or reflective optical elements in combination to focus the light from the optical element onto the sample. The illumination subsystem shown in Figure 11 and described herein may comprise other suitable optical elements (not shown). Examples of such optical elements include, but are not limited to, polarizing components, spectral filters, spatial filters, reflective optical elements, apodizers, beam splitters, apertures, and any other suitable optical elements known in the art. The imaging system may also be configured to change one or more elements of the illumination subsystem based on the type of illumination to be used for imaging.

[0093] The imaging system may also include a scanning subsystem configured to direct light onto the sample, detect light from the sample, change its position on the sample, and scan the sample as much as possible. The imaging system may include, for example, a stage 1114 on which the sample 1106 is positioned during imaging. The scanning subsystem may include any suitable mechanical and / or robotic assembly (including the stage 1114) that can be configured to move the sample so that light can be directed to various positions on the sample and detected from various positions. The imaging system may additionally or alternatively be configured so that one or more optical elements of the imaging system perform some scanning of light on the sample, thereby directing light to various positions on the sample and detecting from various positions. When light is scanned on the sample, the light may scan the sample in any suitable manner, such as a meandering or helical path.

[0094] The imaging system further comprises one or more detection channels. At least one of the detection channels includes a sensor configured to detect light from a sample as the imaging system illuminates the sample, and to produce an output corresponding to the detected light. For example, the imaging system shown in Figure 11 comprises two detection channels, one formed by a collector 1116, an element 1118, and a sensor 1120, and the other formed by a collector 1122, an element 1124, and a sensor 1126. The two detection channels are configured to collect and detect light at different collection angles, as shown in Figure 11. In some cases, both detection channels are configured to detect scattered light, and the detection channels are configured to detect light scattered from the sample at various angles. However, one or more detection channels may be configured to detect another type of light from the sample (e.g., reflected light).

[0095] As further shown in Figure 11, both detection channels are shown to be positioned in the plane of the paper, and the illumination subsystem is also shown to be positioned in the plane of the paper. Thus, in this embodiment, both detection channels are positioned on the plane of incidence (e.g., at the center of the plane of incidence). However, one or more detection channels may be positioned off-plane. For example, a detection channel formed by collector 1122, element 1124, and sensor 1126 may be configured to collect and detect light scattered off-plane. Such a detection channel may therefore be commonly referred to as a “side” channel, and such a side channel may be located at the center of a plane substantially perpendicular to the plane of incidence.

[0096] Figure 11 shows an embodiment of an imaging system having two detection channels, but the imaging system may have various numbers of detection channels (e.g., only one detection channel or two or more detection channels). In such an example, the detection channel formed by the collector 1122, element 1124, and sensor 1126 may form one side channel as described above, and the imaging system may have an additional detection channel (not shown) formed as another side channel located on the opposite side of the incident plane. The imaging system may then have a detection channel comprising the collector 1116, element 1118, and sensor 1120, positioned at the center of the incident plane and configured to collect and detect light at a scattering angle normal to or close to the normal of the surface of the sample. This detection channel may then be commonly referred to as the “top” channel, and the imaging system may also have two or more side channels configured as described above. The imaging system may therefore comprise at least three channels (i.e., one top channel and two side channels), each of which comprises its own collector, each collector configured to collect light at a scattering angle different from that of each of the other collectors.

[0097] Furthermore, as described above, each of the detection channels provided in the imaging system may be configured to detect scattered light. Thus, the imaging system shown in Figure 11 may be configured for dark-field (DF) imaging of a sample. However, the imaging system may further, or alternatively, include detection channels configured for bright-field (BF) imaging of a sample. In other words, the imaging system may include at least one detection channel configured to detect light specularly reflected from the sample. The imaging systems described herein may therefore be configured for DF only, BF only, or both DF and BF imaging. Although each collector is shown as a single refractive optical element in Figure 11, each collector may comprise one or more refractive optical elements and / or one or more reflective optical elements. The collector shown in Figure 11 may also be configured as an embodiment of the camera lens subsystem or tube lens subsystem described herein, or may be replaced by an embodiment of the camera lens subsystem or tube lens subsystem.

[0098] Sensors provided in one or more detection channels may be configured according to any of the embodiments described herein. The output generated by each of the sensors provided in each of the detection channels of the imaging system may be an image signal, image data, or any other suitable output known in the art. In addition, although each of the detection channels is shown in Figure 11 as comprising a single sensor, each of the detection channels may comprise multiple sensors, as shown in Figure 9a and further described herein. Furthermore, the various detection channels provided in the imaging system may comprise various embodiments of sensors described herein. Sensor 1120 may be configured to have a pre-selected shape, for example, different from sensor 1126.

[0099] It should be noted that Figure 11 is presented herein to illustrate an overall embodiment of an imaging system configuration, which may comprise one or more of the sensor embodiments described herein. The imaging system configuration described herein may, obviously, be modified to optimize the performance of the imaging system, as is typically done when designing commercial imaging systems. The imaging systems described herein may also be implemented using existing systems, such as the 29xx / 39xx series of tools commercially available from KLA Corp. in Milpitas, California (for example, by adding the functions described herein to an existing inspection system). In some such imaging systems, the sensors described herein may be provided as optional elements of the imaging system (for example, in addition to other existing sensors in the imaging system). Alternatively, the imaging systems described herein may be designed "from scratch" to implement a completely new imaging system.

[0100] The computer subsystem 1102 may be coupled to the sensor of the imaging system in any preferred manner (for example, via one or more transmission media, which may include “wired” and / or “wireless” transmission media) so that the computer subsystem can receive the output generated by the sensor. The computer subsystem 1102 may be configured to perform several functions, including steps and functions further described herein, with or without the sensor output. Thus, the steps described herein may be performed “on the tool” by the computer subsystem, which is coupled to or part of the imaging system. The computer system 1104 may additionally or alternatively perform one or more of the steps described herein. Thus, one or more of the steps described herein may be performed “off the tool” by a computer system that is not directly coupled to the imaging system. The computer subsystem 1102 and the computer system 1104 may be further configured as described herein.

[0101] Computer subsystem 1102 (and other computer subsystems described herein) may also be referred to herein as computer systems. Each of the computer subsystems or systems described herein may take various forms, including personal computer systems, image computers, mainframe computer systems, workstations, network appliances, internet appliances, or other devices. The term “computer system” may be broadly defined to encompass any device having one or more processors that execute instructions from a memory medium. A computer subsystem or system may also have any suitable processors known in the art, such as parallel processors. In addition, a computer subsystem or system may include a computer platform having a high-speed processing unit and software, which may be a standalone tool or a networked tool.

[0102] If a system comprises multiple computer subsystems, different computer subsystems can be coupled to one another so that images, data, information, instructions, etc., can be transmitted between them. Computer subsystem 1102 can be coupled to computer system 1104 by any suitable transmission medium, which may include any suitable wired and / or wireless transmission medium known in the art, as shown by the dashed lines in Figure 11. Two or more such computer subsystems can also be efficiently coupled by a shared computer-readable storage medium (not shown).

[0103] The imaging system is described above as comprising an optical or light-based energy source; however, in another embodiment, the energy source is configured as an electron beam source. In such an imaging system, the energy directed towards the sample includes electrons, and the energy detected from the sample includes electrons. In one such embodiment shown in Figure 11a, the imaging system comprises an electron column 1128 which can be coupled to a computer subsystem 1130. The computer subsystem 1130 may be configured as described above. Such an imaging system may also be coupled to one or more other computer systems in the same manner as described above and shown in Figure 11.

[0104] The electron column comprises an electron beam source 1132 configured to generate electrons focused onto a sample 1134 by one or more elements 1136, as also shown in Figure 11a. The electron beam source may comprise, for example, a cathode source or an emitter tip, and one or more elements 1136 may comprise, for example, a gun lens, an anode, a beam limiting aperture, a gate valve, a beam current selective aperture, an electrostatic or magnetic objective lens suitable for imaging charged particles, and a scanning subsystem, and all of the elements 1136 may comprise any such elements known in the art.

[0105] Electrons returning from the sample can be focused to the sensor 1140 by one or more elements 1138. The one or more elements 1138 may comprise, for example, a camera lens subsystem or a tube lens subsystem, which may be configured as described herein. The sensor 1140 may be configured according to any of the embodiments described herein. The sensor 1140 may also be replaced by a sensor array, such as the one shown in Figure 9a and further described above.

[0106] The electronic column may comprise any other suitable elements known in the art. In addition, the electronic column may further be configured as described in U.S. Patent No. 8,664,594 granted to Jiang et al. on April 4, 2014, U.S. Patent No. 8,692,204 granted to Kojima et al. on April 8, 2014, U.S. Patent No. 8,698,093 granted to Gubbens et al. on April 15, 2014, and U.S. Patent No. 8,716,662 granted to MacDonald et al. on May 6, 2014, which are incorporated herein by reference as if they were described in their entirety.

[0107] The electron column is shown in Figure 11a as being configured such that electrons are directed towards the sample at an oblique angle of incidence and scattered from the sample at another oblique angle; however, the electron beam may be directed towards the sample at any preferred angle and scattered from the sample at any preferred angle. The imaging system may also be configured to generate output with respect to the sample using multiple modes (e.g., different illumination angles, acquisition angles, etc.), as will be further described herein. The multiple modes of the imaging system may have different output generation parameters for the imaging system.

[0108] The computer subsystem 1130 may be coupled to the sensor 1140 described above. The sensor detects electrons returning from the surface of the sample, thereby forming an image of the sample (or other output of the sample). The computer subsystem 1130 may be configured to perform one or more functions on the output generated by the sensor 1140, functions which may be performed as further described herein. The computer subsystem 1130 may be configured to perform any additional steps described herein. The system including the imaging system shown in Figure 11a may be further configured as described herein.

[0109] It should be noted that Figure 11a is presented herein to illustrate an overall alternative embodiment of the configuration of an imaging system, which may comprise one or more of the sensor embodiments described herein. The imaging system configuration shown in Figure 11a, like the imaging system shown in Figure 11, may be modified to optimize the performance of the imaging system, as is typically done when designing commercial systems. In addition, the imaging systems described herein may be realized using existing systems, such as commercially available tools from KLA (for example, by adding the sensors described herein to an existing system). The sensors described herein may be provided in some such systems as an optional element of the system (for example, in addition to the existing sensors in the system). Alternatively, the imaging systems described herein may be designed "from scratch" to realize an entirely new imaging system.

[0110] Although the imaging system is described above as comprising an optical or electron beam energy source, the imaging subsystem may also comprise an ion beam energy source. Such an imaging system may be configured as shown in Figure 11a, except that the electron beam source may be replaced with any suitable ion beam source known in the art. The imaging system may also include any other suitable ion beam imaging systems, such as those provided in commercially available focused ion beam (FIB) systems, helium ion microscopy (HIM) systems, and secondary ion mass spectrometry (SIMS) systems.

[0111] The imaging system may be configured to have multiple modes, as further mentioned above. A “mode” is generally defined by the parameter values ​​of the imaging system used to generate an output relating to a sample. Thus, if the modes are different, at least one value of the imaging parameters of the imaging system (excluding the position on the sample where the output is generated) may be different. For example, in the case of an optical imaging system, different wavelengths of light may be used in different modes. The modes may have different wavelengths of light directed at the sample (for example, by using different light sources, different spectral filters, etc., for different modes), as further described herein. In another embodiment, different illumination channels may be used in different modes. For example, as mentioned above, the imaging system may have multiple illumination channels. Thus, different illumination channels may be used for different modes.

[0112] Multiple modes may also differ in illumination and / or acquisition / detection. For example, as further described above, an imaging system may have multiple sensors. Thus, one of the sensors may be used for one mode, and another of the sensors may be used for another mode. Furthermore, modes may differ from one another in multiple techniques described herein (for example, different modes may have one or more different illumination parameters and one or more different detection parameters). Multiple modes may also have different viewpoints, meaning they have one or both of different incident angles and acquisition angles, which is achievable as further described above. An imaging system may be configured to scan a sample in different modes in the same scan or different scans, for example, depending on its ability to scan the sample simultaneously using multiple modes.

[0113] The imaging systems described herein may, in some cases, be configured as inspection systems. However, the imaging systems described herein may also be configured as other types of semiconductor-related quality control systems, such as defect review systems and measurement systems. For example, in the embodiments of the imaging systems described herein and shown in Figures 11 and 11a, one or more parameters may be modified to achieve different imaging capabilities depending on the intended application. In such an example, the imaging system shown in Figure 11 may be configured to have a higher resolution when it is to be used for defect review or measurement rather than inspection. The embodiments of the imaging systems shown in Figures 11 and 11a describe, in other words, a somewhat general-purpose, variety of configurations of imaging systems that can be adjusted in several ways that will be obvious to those skilled in the art, to produce imaging systems with various imaging capabilities that are generally suitable for a variety of applications.

[0114] The imaging system is configured, as mentioned above, to direct energy (e.g., light, electrons) towards the physical version of the sample and / or to scan the energy over the physical version of the sample, thereby generating an actual image of the physical version of the sample. The imaging system can thus be configured as a “real” imaging system rather than a “virtual” system. However, the storage medium (not shown) and the computer system 1104 shown in Figure 11 can be configured as a “virtual” system. The storage medium and the computer system, in particular, are not part of the imaging system 1100 and have no ability to handle the physical version of the sample, but can be configured using the stored sensor outputs as a virtual inspector that performs functions such as inspection, a virtual measurement system that performs functions such as measurement, a virtual defect review tool that performs functions such as defect review, and so on. Systems and methods configured as “virtual” systems are described in U.S. Patent No. 8,126,255 granted to Bhaskar et al. on 28 February 2012, U.S. Patent No. 9,222,895 granted to Duffy et al. on 29 December 2015, and U.S. Patent No. 9,816,939 granted to Duffy et al. on 14 November 2017, which are incorporated herein by reference as if they were described in their entirety. Embodiments described herein may further be configured as described in those patents. For example, computer subsystems described herein may further be configured as described in those patents.

[0115] In one embodiment, the imaging system includes a computer subsystem configured to determine information about a sample based on outputs generated by sensors. The imaging system shown in Figure 11 may include, for example, a computer subsystem 1102 and / or a computer system 1104, and the imaging system shown in Figure 11a may include a computer subsystem 1130. These computer subsystems or systems may be coupled to one or more sensors of the imaging system described above so that the computer subsystems or systems receive outputs generated by sensors. The information to be determined, and the way in which the outputs generated by one or more sensors are used for information determination, may vary depending on the process being performed on the sample. The step of determining information may be performed by the computer subsystem using an algorithm or method, such as one of those further described herein, or any other suitable algorithm or method known in the art.

[0116] In another embodiment, the imaging system includes a computer subsystem configured to detect defects in a sample based on an output generated by a sensor. The output generated by the sensor can be used for defect detection in much the same way as any other image. In other words, the output generated by the sensor described herein is not specific to any defect detection algorithm or method, and defect detection using the output can be performed using any suitable defect detection algorithm or method known in the art. Defect detection can be performed, for example, by generating a difference image by subtracting a criterion from the output and using a threshold on the difference image. Any pixel in the difference image with a value exceeding the threshold can be identified as a defect, and all other pixels cannot be identified as defects. Of course, this is perhaps the simplest method by which defect detection can be performed and is included herein simply as a non-limiting example.

[0117] In some embodiments, detecting defects in a sample may therefore involve generating or determining information about the sample, which may include information about any defects detected in the sample. In such cases, the information may include, for example, the type of defect detected, the location of the detected defect, and any other arbitrary information about the defect generated by the defect detection method or algorithm and / or computer subsystem, relating to one or more of the sample image, the sample, the imaging system, and the design relating to the sample. The information determined by the computer subsystem may further, or alternatively, include any preferred defect attributes (other than the location of the reported defect), such as classification, size, and shape (other than the location of the reported defect), which can be determined from the consistency of the output with respect to other information about the sample, such as the output and / or design data described herein. Such information may be output and / or stored by the computer subsystem, as further described herein.

[0118] Unlike inspection processes, defect review processes generally involve re-examining individual locations on a sample where defects are detected. Imaging systems configured for defect review may generate sample images as described herein, which can be input into computer subsystems as described herein for one or more defect review functions, such as defect re-detection, defect attribute determination, defect classification, and defect root cause identification. The computer subsystems may also be configured to use any suitable defect review method or algorithm, used on any suitable defect review tool, to determine information about defects or samples from sensor output, potentially combined with other arbitrary information determined by the defect review process or from sensor output, for defect review applications.

[0119] In some embodiments, the imaging system may be configured for measuring a sample. In one such embodiment, the information includes measurements of one or more structures formed on the sample. The imaging system described herein may, for example, be configured as a measurement tool, and the sensor output generated by such a measurement tool can be used to determine the measurement information of the sample. The measurement information may include any measurement information of interest, which may vary depending on the structure on the sample. Examples of such measurement information include, but are not limited to, critical dimensions (CD), such as line width, and other dimensions of the structure on the sample. The computer subsystem may also be configured, for measurement applications, to use any suitable measurement method or algorithm used on any suitable measurement tool to determine information about the sample from the sensor output, in combination with any other arbitrary information, which may be determined by the measurement process or from the sensor output.

[0120] The computer subsystem may also be configured to generate results containing determined information, which may include any of the results or information described herein. The results of the information determination may be generated by the computer subsystem in any preferred manner. All embodiments described herein may be configured to store the results of one or more steps of the embodiment in a computer-readable storage medium. The results may include any of the results described herein and may be stored in any manner known in the art. The results containing the determined information may have any preferred form or format, such as a standard file type. The storage medium may include any storage medium described herein or any other preferred storage medium known in the art.

[0121] After the results are stored, these results can be accessed from a storage medium, used in any of the embodiments of the methods or systems described herein, formatted for display to a user, and used in another software module, method, or system, etc., to perform one or more functions on the sample or another sample of the same type. The results generated by the computer subsystem may include, for example, information about any defects detected on the sample, such as the location of the bounding box of the detected defect; information about the classification of the defect, such as the detection score, type label or ID; attributes of any defect determined from any of the following, such as an image; predicted structural measurements, dimensions, shape, etc., of the sample; or any such preferred information known in the art. This information may be used by the computer subsystem or another system or method to perform additional functions on the sample and / or detected defects, such as defect review or sampling of defects for other analysis, or determination of the root cause of the defects.

[0122] Such functions include, but are not limited to, changes to processes such as manufacturing processes or steps that have been or will be performed on a sample in a feedback or feedforward manner. The computer subsystem may be configured, for example, to determine one or more changes to processes that have been performed on a sample and / or will be performed on a sample based on the determined information. Changes to processes may include any preferred changes to one or more parameters of the process. In such an example, it is preferable that the computer subsystem determines such changes so that defects in other samples on which the revised process is performed can be reduced or prevented, defects in the sample can be corrected or eliminated in another process performed on the sample, or defects can be compensated for in another process performed on the sample. The computer subsystem may determine such changes in any preferred manner known in the art.

[0123] Such changes can then be transmitted to a semiconductor manufacturing system (not shown) or a storage medium (not shown) accessible to both the computer subsystem and the semiconductor manufacturing system. The semiconductor manufacturing system may or may not be part of the embodiments of the system described herein. For example, the imaging subsystem and / or computer subsystem described herein may be coupled to the semiconductor manufacturing system via one or more common elements, such as a housing, power supply, sample handling device or mechanism. The semiconductor manufacturing system may include any semiconductor manufacturing system known in the art, such as lithography tools, etching tools, chemical-mechanical polishing (CMP) tools, and deposition tools.

[0124] Each of the embodiments of the systems described above may be combined together to form a single embodiment.

[0125] Further embodiments relate to a non-temporary computer-readable medium for storing program instructions executable on a computer system for performing a computer-implemented method for determining information about a sample. One such embodiment is shown in Figure 12. The non-temporary computer-readable medium 1200 has, in particular, program instructions 1202 executable on a computer system 1204, as shown in Figure 12. The computer-implemented method may include any step of any method described herein.

[0126] Program instructions 1202 relating to an algorithm that implements the methods described herein may be stored in a computer-readable medium 1200. The computer-readable medium may be a storage medium such as a magnetic or solid disk, a magnetic tape, or any other suitable non-temporary computer-readable medium known in the art.

[0127] The algorithm may be implemented using any of the following methods, including, in particular, procedure-based techniques, component-based techniques, object-oriented techniques, and embodiments of neural network architectures. Program instructions may be implemented using preferred programming frameworks and languages ​​known in the art, such as C, C++, or Python, and may be executed on on-premises, remote, or centrally managed computing systems, or combinations thereof. Custom accelerators may be implemented individually or in combination as needed within application-specific integrated circuit devices (ASIC chips), custom-configured field-programmable gate arrays (FPGAs), or graphics processing units (GPUs).

[0128] The computer system 1204 may be configured according to any of the embodiments described herein.

[0129] Further variations and alternative embodiments of various aspects of the present invention will be apparent to those skilled in the art in consideration of this description. For example, sensors, imaging systems, and methods for forming sensors are provided. This description should therefore be interpreted as illustrative only and is intended to teach those skilled in the art a general way of carrying out the present invention. It should be understood that the embodiments of the present invention illustrated and described herein should be interpreted as preferred embodiments at present. Elements and materials may be replaced with those illustrated and described herein, parts and processes may be reversed, and some features of the present invention may be used separately, all of which will be apparent to those skilled in the art in reading this description of the present invention. Modifications to the elements described herein may be made without departing from the spirit and scope of the present invention, as described in the following claims.

Claims

1. circuit board and One or more components attached to the aforementioned substrate, A sensor die comprising a thin back surface and an energy-sensitive element configured to detect the energy irradiating the thin back surface of the sensor die, Separated heat conduction structure, A method for manufacturing a sensor equipped with the following: A flip-chip process forms the separated heat conduction structure between the front surface of the sensor die and the substrate, thereby bonding the sensor die to the substrate, and the thin back surface of the sensor die having a pre-selected shape, the pre-selected shape being defined by a curved shape or a higher-order polynomial, and at least a portion of the separated heat conduction structure electrically connects the sensor die to one or more components. The shape of one or more of the separated heat conduction structures is modified based on the pre-selected shape before the sensor die is bonded to the substrate in the flip-chip process. A method for manufacturing a sensor, comprising applying pressure to the sensor die through the flow chamber by a gas flow that controls the pressure in the flow chamber in contact with the sensor die, thereby conforming the shape of the sensor die to one or more modified shapes of the separated heat conduction structures.

2. A method for manufacturing a sensor according to claim 1, characterized in that, before bonding the sensor die to the substrate in the flip-chip process, the separated heat conduction structures are formed on the substrate, and the shape of one or more of the separated heat conduction structures is deformed such that the combination of the separated heat conduction structures has substantially the same shape as the pre-selected shape.

3. A method for manufacturing a sensor according to claim 1, characterized in that the pre-selected shape is determined before the flip-chip process, and the shape of one or more of the separated heat conduction structures formed on the substrate is changed based on the pre-selected shape before the sensor die is bonded to the substrate in the flip-chip process.

4. A method for manufacturing a sensor according to claim 1, characterized in that the surface of the substrate on which the separated heat conduction structure is formed has a shape different from the shape selected in advance.

5. A method for manufacturing a sensor according to claim 1, characterized in that the surface of the substrate on which the separated heat conduction structure is formed has a shape determined based on the pre-selected shape.

6. A method for manufacturing a sensor according to claim 1, characterized in that the substrate is made of a ceramic material.

7. A method for manufacturing a sensor according to claim 1, characterized in that the substrate is formed of a material selected such that residual stress at the joint is minimized based on the thermal expansion coefficient of the material, which is determined by the size of the sensor die and the pre-selected shape.

8. A method for manufacturing a sensor according to claim 1, characterized in that the separated heat conduction structure is formed of a material selected such that residual stress at the joint is minimized based on the reflow temperature of the material, which is determined by the size of the sensor die and the pre-selected shape.

9. A method for manufacturing a sensor according to claim 1, further comprising an underfill material formed around the separated heat conduction structure between the front surface of the sensor die and the substrate.

10. A method for manufacturing a sensor according to claim 9, characterized in that the underfill material is configured to stabilize the solder joint and maintain the shape of the sensor die when the sensor die is exposed to a vacuum.

11. A method for manufacturing a sensor according to claim 9, characterized in that the underfill material includes a resin containing dispersed particles made of a dielectric material having high thermal conductivity.

12. A method for manufacturing a sensor according to claim 1, further comprising thermally conductive and conductive vias formed in the substrate, wherein at least a portion of the separated thermal conductive structure is connected to one or more components, thereby connecting the sensor die to one or more components.

13. A method for manufacturing a sensor according to claim 1, characterized in that one or more components are configured to perform one or more functions in response to the output generated by the energy-sensitive element, corresponding to the detected energy.

14. A method for manufacturing a sensor according to claim 1, characterized in that the energy-sensitive element is further configured to detect deep ultraviolet light.

15. A method for manufacturing a sensor according to claim 1, characterized in that the energy-sensitive element is further configured to detect vacuum ultraviolet light.

16. A method for manufacturing a sensor according to claim 1, characterized in that the energy-sensitive element is further configured to detect extreme ultraviolet light.

17. A method for manufacturing a sensor according to claim 1, characterized in that the energy-sensitive element is further configured to detect X-rays.

18. A method for forming a sensor, The steps include forming a separate heat conduction structure on a substrate, A step of changing the shape of the separated heat conduction structure based on a pre-selected shape of the thin back surface of the sensor die, the step of deforming the shape of one or more of the separated heat conduction structures such that the combination of the separated heat conduction structures has substantially the same shape as the pre-selected shape, The steps include: joining the front surface of the sensor die to the substrate via the separated heat conduction structure, thereby shaping the thin back surface of the sensor die into the pre-selected shape; and ensuring that at least a portion of the separated heat conduction structure electrically connects the sensor die to one or more components attached to the substrate. Includes, By controlling the pressure in the flow chamber that contacts the sensor die, pressure is applied to the sensor die through the flow chamber, thereby conforming the shape of the sensor die to one or more modified shapes of the separated heat conduction structures. The aforementioned pre-selected shape is defined by a curved shape or a higher-order polynomial. A method characterized in that the sensor die comprises an energy-sensitive element configured to detect energy irradiating the thin back surface of the sensor die.

Images