Method and system for accurate measurement of deep structures with distorted shapes

Abstract

A method and system are presented herein for estimating values ​​of geometric parameters characterizing the in-plane distortion shape of high aspect ratio semiconductor structures based on X-ray scatterometry measurements. The parameterized geometric model captures the scattering signature of the in-plane non-elliptical distortion of the hole shape. Increasing the number of independent parameters employed to describe the in-plane shape of the hole structure improves the fit of the model to the actual shape of the high aspect ratio structure. In one aspect, the geometrically parameterized measurement model includes two or more degrees of freedom to characterize the in-plane shape of the measured structure. In some embodiments, the geometric model includes a closed curve with three or more degrees of freedom. In some embodiments, the geometric model includes a piecewise collection of two or more conic sections. The independent geometric model parameters are expressed as a function of depth to capture the shape changes through the structure.

Description

[Technical Field]

[0001] The embodiments described relate to measurement systems and methods, and more particularly to methods and systems for improved measurement of deep semiconductor structures fabricated by repeated lithography and etching process steps. [Background technology]

[0002] Cross-reference of related applications This patent application claims priority to U.S. Provisional Application No. 63 / 147,758, entitled “Accurate Modelling of Lithographic and Etch Shapes using Distorted Ellipses,” filed on 10 February 2021 under Section 119 of the U.S. Patent Act, the subject matter of which is incorporated herein by reference in its entirety.

[0003] Semiconductor devices, such as logic and memory elements, are typically manufactured by a series of processing steps applied to a sample. Various features and multiple structural levels of the semiconductor device are formed by these processing steps. For example, lithography is one semiconductor manufacturing process that involves creating patterns on a semiconductor wafer. Additional examples of semiconductor manufacturing processes include, but are not limited to, chemical mechanical polishing, etching, deposition, and ion implantation. Multiple semiconductor devices may be manufactured on a single semiconductor wafer and then separated into individual semiconductor devices.

[0004] Measurement processes are used at various steps in semiconductor manufacturing to detect defects on wafers and promote higher yields. Several measurement-based techniques, including scatometry and reflectometry implementations and related analytical algorithms, are widely used to characterize the critical dimensions, film thickness, components, and other parameters of nanoscale structures. X-ray scatometry techniques offer the potential for high throughput without the risk of sample destruction.

[0005] Traditionally, optical scattometry limiting dimension (SCR) measurements are performed on targets consisting of thin films and / or iterative periodic structures. During device manufacturing, these films and periodic structures generally represent the geometry and material structure of the actual device, or an intermediate design. As devices (e.g., logic and memory devices) move toward smaller nanometer-scale dimensions, characterization becomes more difficult. Devices incorporating complex three-dimensional geometry and materials with diverse physical properties contribute to the difficulty of characterization. For example, modern memory structures are often high-aspect-ratio three-dimensional structures that make it difficult for light radiation to penetrate to the lower layers. While optical measurement tools utilizing infrared-visible light can penetrate many layers of translucent materials, longer wavelengths that provide good penetration depth do not provide sufficient sensitivity to small anomalies. In addition, the increasing number of parameters required to characterize complex structures (e.g., FinFETs) leads to increased parameter correlation. As a result, the parameters characterizing the target are often inseparable from the available measurements.

[0006] In one example, longer wavelengths (e.g., near-infrared) have been used in attempts to overcome transmission problems in 3D flash devices that utilize polysilicon as one of the alternating materials within the stack. However, the mirror-like structure of 3D flash inherently causes a decrease in light intensity as illumination penetrates deeper into the film stack. This leads to sensitivity loss and correlation problems with depth. In this situation, optical SCDs can only successfully extract a reduced set of measurement dimensions with high sensitivity and low correlation.

[0007] In another example, opaque, high-k materials are increasingly used in modern semiconductor structures. Light emission often cannot penetrate layers constructed from these materials. As a result, measurements using thin-film scatometry tools such as elliptic polarimeters or reflectometers are becoming increasingly difficult.

[0008] In response to these challenges, more complex optical measurement tools have been developed. For example, tools have been developed that offer multiple illumination angles, shorter illumination wavelengths, a wider range of illumination wavelengths, and more complete information acquisition from reflected signals (e.g., measuring multiple Müller matrix elements in addition to more conventional reflectance or polarization analysis signals). However, these techniques do not reliably overcome the fundamental challenges associated with measuring many advanced targets (e.g., complex 3D structures, structures smaller than 10 nm, structures using opaque materials) and measurement applications (e.g., line edge roughness and line width roughness measurements).

[0009] While optical methods can provide non-destructive tracking of process variables between process steps, periodic calibration by destructive methods is necessary to maintain accuracy in the face of process drift that optical methods cannot independently distinguish.

[0010] Atomic force microscopes (AFM) and scanning tunneling microscopes (STM) can achieve atomic resolution, but they can only explore the surface of a sample. In addition, AFM and STM microscopes require long scanning times. Scanning electron microscopes (SEM) achieve an intermediate level of resolution, but they cannot penetrate structures to sufficient depth. Therefore, high aspect ratio pores cannot be adequately characterized. Furthermore, the required sample input negatively impacts imaging performance. X-ray reflectometers also suffer from transmission problems that limit their effectiveness when measuring high aspect ratio structures.

[0011] To overcome the transmission depth problem, conventional imaging techniques such as TEM and SEM are used in conjunction with destructive sample transmission techniques such as focused ion beam (FIB) processing, ion milling, blanket, or selective etching. For example, while transmission electron microscopy (TEM) can achieve high resolution levels and explore arbitrary depths, TEM requires destructive sections of the sample. Several iterations of material removal and measurement generally provide the information needed to measure critical measurement parameters across the entire three-dimensional structure. However, these techniques require sample destruction and long process times. Because the measurement results are available long after the process is complete on the wafer being measured, the complexity and time required to complete these types of measurements introduce significant errors due to etching drift and measurement steps. Thus, the measurement results are affected by bias from further processing and delayed feedback. In addition, these techniques require numerous iterations, which introduce registration errors. In summary, device yield is negatively impacted by the long destructive sample preparation required for SEM and TEM techniques.

[0012] In general, there are numerous process monitoring methods that combine optical, acoustic, and electron beam tools. These techniques can directly measure devices, measure specially designed targets, or measure specific monitor wafers. However, they cannot cost-effectively and timely measure the parameters of targets in high aspect ratio structures, resulting in low yields, especially in the memory sectors of wafers.

[0013] Transmission small-angle X-ray scatterometry (T-SAXS) systems using hard X-ray energy levels (>15 keV) photons have shown promise in addressing challenging measurement applications. Various aspects of the application of SAXS technology to the measurement of limit dimensions (CD-SAXS) and overlays (OVL-SAXS) are described in: 1) U.S. Patent No. 7,929,667 by Zhuang and Fielden (titled "High-brightness X-ray metrology"), 2) U.S. Patent Publication No. 2014 / 0019097 by Bakeman, Shchegrov, Zhao, and Tan (titled "Model Building And Analysis Engine For Combined X-Ray And Optical Metrology"), 3) U.S. Patent Publication No. 2015 / 0117610 by Veldman, Bakeman, Shchegrov, and Mieher (titled "Methods and Apparatus For Measuring Semiconductor Device Overlay Using X-Ray Metrology"), and 4) U.S. Patent Publication No. 2016 / 0202193 by U. Hench, Shchegrov, and Bakeman (titled "Measurement System Optimization For X-Ray Based The following documents are described in U.S. Patent Publication No. 2017 / 0167862 (titled "X-Ray Metrology For High Aspect Ratio Structures") by Dziura, Gellineau, and Shchegrov, and U.S. Patent Publication No. 2018 / 0106735 (titled "FullBeam Metrology for X-Ray Scatterometry Systems") by Gellineau, Dziura, Hench, Veldman, and Zalubovsky, the contents of which are incorporated herein by reference in their entirety. The aforementioned patent documents are assigned to KLA-Tencor Corporation in Milpitas, California, USA.Furthermore, U.S. Patent No. 9,606,073 to Mazor et al., entitled "X-ray scatterometry apparatus," describes various aspects of the application of SAXS technology to semiconductor structures, the contents of which are incorporated herein by reference in their entirety.

[0014] SAXS is also being applied to material characterization and other non-semiconductor-related applications. Exemplary systems are commercialized by several companies, including Xenocs SAS (www.xenocs.com), Bruker Corporation (www.bruker.com), and Rigaku Corporation (www.rigaku.com / en). Bruker and Rigaku offer small-angle X-ray scatterometry systems and wide-angle X-ray scatterometry systems named "Nanostar" and "Nanopix," respectively. These systems include adjustable sample-detector distances.

[0015] Research on CD-SAXS measurements of semiconductor structures is also documented in scientific literature. Most research groups employ high-brightness X-ray synchrotron light sources, but these sources are unsuitable for use in semiconductor manufacturing facilities due to their large size and cost. An example of such a system is described in the paper "Intercomparison between optical and x-ray scatterometry measurements of FinFET structures" by Lemaillet, Germer, and Kline et al., Proc. SPIE, v.8681, p.86810Q (2013), the contents of which are incorporated herein by reference in their entirety. More recently, a group at the National Institute of Standards and Technology (NIST) has begun research using a small, bright X-ray source similar to that described in U.S. Patent No. 7,929,667. This research is described in the paper entitled "X-ray scattering critical dimensional metrology using a compact x-ray source for next generation semiconductor devices" (J. Micro / Nanolith. MEMS MOEMS 16(1),014001 (Jan-Mar 2017)), and the contents of each of these documents are incorporated herein by reference in their entirety.

[0016] X-ray scatometry is an indirect method for measuring the physical properties of a sample under test. In most cases, the physical properties of a sample cannot be directly determined using the raw measurement signal. Instead, a measurement model is employed that estimates the values ​​of one or more parameters characterizing the structure under test based on the raw measurement signal. Generally, the raw measurement signal (e.g., detected intensity, I) is used. meas To determine the physical properties of a sample based on these factors, a physics-based measurement model or a machine learning-based measurement model is required.

[0017] In some examples, a physics-based measurement model is created that attempts to predict raw measurement signals based on assumed values of one or more model parameters. As shown in Equation (1), the measurement model includes parameters related to the measurement tool itself, such as system parameters (P system ), and parameters related to the sample being measured. When solving for the target parameters, some sample parameters are treated as fixed values (P spec-fixed ), and other target sample parameters are floating values (P spec-float ), i.e., they are resolved based on the raw measurement signal. I model =f(P system ,P spec-fixed ,P spec-float )(1)

[0018] System parameters are parameters used to characterize the measurement tool (e.g., X-ray scatterometry). Exemplary system parameters include the angle of incidence (AOI), azimuth angle (Az), illumination wavelength, etc. Sample parameters are parameters used to characterize the sample (e.g., materials and geometric parameters that characterize the structure being measured). For a thin film sample, exemplary sample parameters include refractive index, dielectric function tensor, nominal layer thickness of the entire layer, layer order, etc. For a CD sample, exemplary sample parameters include geometric parameter values related to different layers, refractive indices related to different layers, etc. For measurement purposes, system parameters and many sample parameters are treated as known fixed value parameters. However, the values of one or more sample parameters are treated as unknown, target floating parameters.

[0019] In some examples, the values of the target floating parameters are solved by an iterative process (e.g., regression) that provides the best fit between the theoretical predicted values and the experimental data. The value of the target unknown floating parameter is varied, the model output value (e.g., I model ) is calculated, and the raw measurement data I meas is iteratively processed until a set of sample parameter values is determined that provides a close enough match between the model output value and the experimentally measured value.Compare with. In some other examples, the floating parameters are solved by searching through a library of pre-computed solutions to find the closest match.

[0020] In recent papers, many geometric distortions observed in deep channel holes fabricated by state-of-the-art semiconductor manufacturing equipment have been reported. Such deep channel holes are common structural elements of NAND and DRAM memory devices. As an example, there is a paper by Shuo Huang et al. titled "Plasma etching of high aspect ratio features in SiO2 using Ar / C4F8 / O2 mixtures: A computational investigation", the content of which is incorporated herein by reference in its entirety.

[0021] When attempting to etch a relatively deep cylindrical hole in a device, various types of distortions occur. Some of the distortions include variations in critical dimension (CD) and variations in the orientation of the hole profile as a function of height. Other distortions include in-plane shape distortions that result in a virtually non-elliptical in-plane hole shape. FIG. 1 is a partial view of a figure (FIG. 21) that exists in the aforementioned paper titled "Plasma etching of high aspect ratio features in SiO2 using Ar / C4F8 / O2 mixtures: A computational investigation".

[0022] FIG. 1 depicts five images 10A-E of the hole structure. Each image is a horizontal slice of the hole structure at a different depth from the surface of the hole structure. The hole structure is fabricated through a material layer of an anti-reflective coating. Image 10A shows the shape of the hole etched through the photoresist material. Images 10B-E show the shape of the hole etched through the silicon oxide material. As depicted in FIG. 1, the shape of the hole structure changes as a function of depth, and more importantly, the actual in-plane hole shape is non-elliptical, especially at greater depths within the hole structure.

[0023] As depicted in FIG. 1, non-elliptical distortion tends to increase as a function of hole depth. Near the top of the etched hole structure, the hole is well approximated as a circle. However, near the bottom of the etched hole structure, the hole is highly distorted such that a simple elliptical shape cannot accurately approximate the hole shape.

[0024] Typically, measurement models associated with measurements based on X-ray scatterometry describe the shape of etched holes in a semiconductor device as a simple ellipse. At any given height within the device, the ellipse is described by an eccentricity e along a nominal principal axis (e.g., the x-axis or y-axis), a center position (e.g., x0 and y0), and a rotation θ with respect to the x-axis. These parameters vary as a function of height z perpendicular to the wafer surface. In this way, the model captures in-plane shape variations as a function of the height of the structure.

[0025] A simple description of an elliptical shape centered at the origin with principal axes described by orthogonal x and y is given by equations (2A)-(2B). x = r cos θ (2A) y = r sin θ (2B) As shown in equations (2A)-(2B), the traditional model of an ellipse is a closed curve with two degrees of freedom, i.e., two independent parameters that define the shape of a curve in a two-dimensional plane. In this example, the two degrees of freedom are the nominal radius r and the eccentricity e.

[0026] Unfortunately, an elliptical shape model such as the model shown in equations (2A)-(2B) does not account for non-elliptical in-plane distortion and cannot parameterize deviations from the nominal elliptical shape in a meaningful way. In current implementations of measurement models, when non-linear components are present, systematic errors occur in the signal residuals. In some examples, these systematic errors bias the solution of parametric regression. This reduces the reproducibility, consistency, and accuracy of the estimated values of the parameters of interest of the structure (e.g., critical dimensions).

Prior Art Documents

[0027] [Patent Document 1] U.S. Patent No. 10352695 [Overview of the project] [Problems that the invention aims to solve]

[0028] In summary, the continuous reduction in feature size and the increasing depth of many semiconductor structures impose challenging requirements on measurement systems. Therefore, improved measurement systems and methods are desired to measure high aspect ratio structures while maintaining high device yield. [Means for solving the problem]

[0029] A method and system for estimating the values ​​of geometric parameters characterizing the in-plane strain shape of high aspect ratio semiconductor structures based on X-ray scantometry are presented herein. In some embodiments, one or more structures under measurement are deep-hole structures fabricated by a series of lithography and etching steps.

[0030] In some examples, scantometry-based measurement tools are sensitive to small variations in hole geometry across the entire depth of the hole structure, including in-plane and non-elliptic distortions of the hole shape. Measurement models include parameterized geometric models that capture the scattering signature of distorted shapes. By increasing the number of parameters employed to describe the in-plane geometry of the hole structure, the resulting geometric models fit the actual shape of the hole structure, including shape errors induced by non-ideal lithography and etching processes.

[0031] In one embodiment, a scatometry-based measurement model includes a geometrically parameterized measurement model having two or more degrees of freedom to characterize the in-plane shape of the measured structure.

[0032] In some embodiments, the geometric model employed to characterize the in-plane shape of a measured hole structure at a specific depth includes a closed curve having three or more degrees of freedom, i.e., three or more independent parameters define the shape of the curve in a two-dimensional plane.

[0033] In some embodiments, the geometric model employed to characterize the in-plane shape of a measured hole structure at a particular depth is a closed curve comprising a piecewise collection of two or more conical cross-sections, e.g., elliptical, parabolic, or hyperbolic cross-sections. In these embodiments, each conical cross-section is defined by at least one independent parameter, i.e., each conical cross-section has at least one degree of freedom. Furthermore, a piecewise collection of conical cross-sections defined by the sum of two or more independent parameters, i.e., two or more degrees of freedom, is employed to describe the in-plane shape of one or more hole structures at a particular depth.

[0034] Generally, the independent parameters of a geometric model describing in-plane hole shape are expressed as a function of depth through the structure. In this way, the geometric model captures the actual changes in the in-plane shape of a semiconductor device, treated as a function of depth. In a typical patterning process, holes in the resist are nearly circular, and the strain is minimal or very small. However, there are limits to the process control of the etching tool. As a result, the strain changes with depth as the etching process transfers the lithographic pattern to the semiconductor layer. The ability to accurately describe the change in shape as a function of depth leads to a more accurate fit between the model and measured data, and consequently, an improved estimate of the shape parameter values.

[0035] In another embodiment, process modifications are determined based on measurements of the parameter in question, and the modifications are communicated to the process tool to change one or more process control parameters of the process tool (e.g., lithography tool, etching tool, deposition tool, etc.).

[0036] The above is a summary and therefore inevitably includes simplifications, generalizations, and omissions of details, and as a result, those skilled in the art will understand that the summary is merely illustrative and not in any way restrictive. Other aspects of the devices and / or processes described herein, features of the invention, and advantages will become apparent in the non-restrictive detailed description provided herein. [Brief explanation of the drawing]

[0037] [Figure 1] Figure 1 shows images of multiple horizontal slices of the hole structure at different depths from the wafer surface. [Figure 2] Figure 2 shows an exemplary wafer measurement system 100 for monitoring the etching process based on X-ray scatterometry measurements of semiconductor structures placed on a wafer. [Figure 3] Figure 3 shows an exemplary model building and analysis engine 180. [Figure 4] Figure 4 shows the shape described by a linear circular function and a quadratic circular function in one embodiment. [Figure 5] Figure 5 shows a three-dimensional plot of the signal error in the detector in one embodiment. [Figure 6] Figure 6 shows a shape described by a linear circular function, a quadratic circular function, and a piecewise collection of four conical cross-sections in one embodiment. [Figure 7A] Figure 7A is an isometric view of a typical 3D flash memory device that is the subject of measurements as described herein. [Figure 7B] Figure 7B is a top view of a typical 3D flash memory device that is the subject of measurements as described herein. [Figure 7C] Figure 7C is a cross-sectional view of a typical 3D flash memory device subjected to measurements as described herein. [Figure 8]Figure 8 shows an exemplary wafer processing system 200 for monitoring the etching process based on reflected X-ray scatterometry measurements of semiconductor structures placed on a wafer. [Figure 9] Figure 9 is a flowchart illustrating method 300 for measuring high aspect ratio structures based on small-angle X-ray scatterometry measurements. [Modes for carrying out the invention]

[0038] Next, background examples and some embodiments of the present invention will be referred to in detail, examples of which are shown in the accompanying drawings.

[0039] This specification presents a method and system for estimating the values ​​of geometric parameters characterizing high aspect ratio semiconductor structures based on X-ray scatometry measurements. More specifically, the measurement model employed to perform scatometry-based measurements includes a parameterized geometric model that characterizes the in-plane distorted shape of one or more hole structures under measurement.

[0040] The latest small-angle X-ray scatterometry (SAXS) measurement tools are sensitive to small changes in hole geometry across the entire depth of a hole structure, including in-plane and non-elliptic distortions of the hole shape. Traditionally, simple ellipses have been used as the basic shape function to describe the shapes of many semiconductor devices, such as DRAM structures and 3D NAND structures, at any given height within the structure. Unfortunately, the in-plane geometry of many real devices is significantly distorted from simple elliptic shapes. As a result, conventional measurement models cannot capture the scattering signatures of such distorted shapes present in the measurement signals generated by the latest SAXS measurement tools.

[0041] By employing a geometric model that more accurately captures the actual distorted in-plane shape of nominal circular etch and lithography features, the measurement accuracy of hole geometry structures is significantly improved across their entire depth. Accurately modeling the in-plane shape of hole geometry structures improves the fit between modeled and measured signals, along with more stable parametric tracking, for deep structures such as DRAM and VNAND memory. As described herein, by increasing the number of independent parameters employed to describe the in-plane shape of hole structures, the resulting geometric models better fit the actual shape of the hole structures, including shape errors induced by non-ideal lithography and etching processes.

[0042] Generally, X-ray scatterometry measurements of high aspect ratio structures are performed at one or more steps in the manufacturing process flow. Exemplary process steps include etching, deposition, and lithography. Measurements are performed quickly and with sufficient accuracy to enable yield improvements in the ongoing semiconductor manufacturing process flow. High aspect ratio structures contain sufficient overall scattering volume and material contrast to efficiently scatter incident X-rays. The collected scattered X-rays allow for accurate estimation of the target structural parameters of the device under measurement. The X-ray energy is high enough to penetrate the silicon wafer and process gases in the optical path while minimizing signal contamination.

[0043] Device yield at advanced semiconductor manufacturing nodes continues to struggle, particularly with complex, high-aspect-ratio (depth, 3D) structures. Real-time monitoring and process control based on X-ray scatometry enables process control of high-aspect-ratio structures in a cost-effective manner compared to conventional destructive methods such as SEM and TEM.

[0044] X-ray scantometry accurately estimates the structural parameters of high-aspect-ratio structures at high throughput without damaging the sample during measurement. Since the measurement sensitivity is not significantly affected by the penetration depth, it can accurately measure structures located deep within the vertical stack of the measured semiconductor structure.

[0045] In one embodiment, the SAXS-based measurement model includes a geometrically parameterized response model of one or more structures under measurement. The geometrically parameterized response model characterizes the in-plane shape of the structure measured by a geometric model with two or more degrees of freedom. In some embodiments, one or more structures under measurement are deep hole structures fabricated by a series of lithography and etching steps.

[0046] In some embodiments, the geometric model employed to characterize the in-plane shape of a measured hole structure at a particular depth includes a closed curve having three or more degrees of freedom, i.e., three or more independent parameters define the shape of the curve in a two-dimensional plane. An example of a curve with three degrees of freedom is a curve defined by a cubic function, e.g., a cubic spline. In some other embodiments, the geometric model employed to characterize the in-plane shape of a measured hole structure at a particular depth includes a closed curve comprising a piecewise collection of two or more conical sections, e.g., elliptic, parabolic, and hyperbolic sections. In these embodiments, each conical section is defined by at least one independent parameter, i.e., each conical section has at least one degree of freedom. Furthermore, a piecewise collection of conical sections defined by the sum of two or more independent parameters, i.e., two or more degrees of freedom, is employed to describe the in-plane shape of one or more hole structures at a particular depth. For example, a piecewise collection of four different elliptic curves may have up to eight independent parameters describing the closed curve, i.e., two independent parameters describing each of the four elliptic curves. A cone section is an example of a curve defined by a quadratic function, such as a quadratic function.

[0047] A closed curve containing a piecewise collection of curves defines its shape as a collection of open curves, where each endpoint of each open curve connects to the endpoint of another curve in the collection, forming a continuous closed curve. In some embodiments, the gradient at one or more endpoints is smooth, i.e., the linear spatial derivative at the point where the two curves connect is the same for both curves.

[0048] Figure 2 shows an exemplary wafer measurement system 100 for performing X-ray scatterometry measurements of semiconductor structures placed on a wafer. In the embodiments depicted, the measurement system is a transmission small-angle scatterometry (T-SAXS) measurement system. In some embodiments, the measured values ​​of the parameter 122 of interest are provided as feedback for controlling manufacturing process tools, such as etching process tools, lithography process tools, deposition tools, etc.

[0049] The wafer measurement system 100 includes a vacuum chamber 104 containing a vacuum environment 103. The semiconductor wafer 101 is placed inside the vacuum chamber 104. The wafer 101 is mounted on a wafer chuck 105 and positioned for X-ray scanning by a wafer stage 140.

[0050] In some embodiments, the wafer stage 140 moves the wafer 101 in the XY plane by combining rotational and translational motion (e.g., translational motion in the X direction and rotational motion around the Y axis) to position the wafer 101 with respect to illumination provided by X-ray scatometry. In some other embodiments, the wafer stage 140 positions the wafer 101 with respect to illumination provided by X-ray scatometry by combining two orthogonal translational motions (e.g., motion in the X and Y directions). In some embodiments, the wafer stage 140 is configured to control the position of the wafer 101 with respect to illumination provided by X-ray scatometry with six degrees of freedom. Generally, the sample positioning system 140 may include, but is not limited to, a goniometer stage, a hexapod stage, an angular stage, and a linear stage, any suitable combination of mechanical elements to achieve desired linear and angular positioning performance.

[0051] In some embodiments, the wafer measurement system 100 does not include a wafer stage 140. In these embodiments, a wafer transport robot (not shown) positions the wafer 101 on a wafer chuck 105 in a vacuum chamber 104. The wafer 101 is transported from the wafer transport robot onto the electrostatic wafer chuck 105, which is adapted to the vacuum environment 103. In these embodiments, the measurement performed by the X-ray scatterometry is limited to the portion of the wafer 101 within the field of view of the X-ray scatterometry after the wafer 101 has been clamped onto the wafer chuck 105. In this sense, the wafer stage 140 is optional. To overcome this limitation, the wafer measurement system 100 includes multiple X-ray scatterometry systems, each capable of measuring different regions of the wafer 101.

[0052] As shown in Figure 2, the optical elements of the X-ray scatometry are located outside the vacuum chamber 104. However, in some other embodiments, the optical elements of the X-ray scatometry are located inside the vacuum chamber 104.

[0053] In the depicted embodiments, the SAXS measurement system includes an X-ray illumination subsystem 125, which includes an X-ray illumination source 110, a focusing optical element 111, a beam divergence control slit 112, an intermediate slit 113, and a beam shaping slit mechanism 120. The X-ray illumination source 110 is configured to generate X-ray emission suitable for T-SAXS measurement. In some embodiments, the X-ray illumination source 110 is configured to generate wavelengths from 0.01 nanometer to 1 nanometer. Generally, any suitable high-brightness X-ray illumination source capable of generating high-brightness X-rays at a radiant flux level sufficient to enable high-throughput in-line measurement may be intended to supply X-ray illumination for T-SAXS measurement. In some embodiments, the X-ray source includes a variable monochromator, which allows the X-ray source to deliver X-ray emission at different selectable wavelengths.

[0054] In some embodiments, one or more X-ray sources emitting radiation with photon energies greater than 15 keV or greater than 17 keV are used to ensure that the X-ray sources supply light at wavelengths that allow sufficient penetration into the entire device, as well as the wafer substrate and any intervening elements. Intervening elements may include one or more windows (e.g., windows made of beryllium, sapphire, diamond, etc.). Intervening elements also include structures in the path of scattered X-rays between the wafer 101 and the detector 119 (such as elements of the wafer chuck 105, load port, or stage 140). There is no risk of excessive contamination of the scattered signal by penetration through structural plastic materials. Signal contamination can be minimized by employing openings or windows through structural elements of the wafer chuck 105, stage 140, or load port. For example, the X-ray spot on the wafer may be as small as 50-200 micrometers. For elements located close to the wafer, the aperture size required to minimize contamination of the order of scattering is minimal. However, the required aperture size increases with increasing distance from the wafer, due to the finite scattering angle related to the desired scattering order.

[0055] An example of an X-ray source is an electron beam source configured to irradiate a solid or liquid target to stimulate X-ray emission. A method and system for generating high-intensity liquid metal X-ray illumination is described in U.S. Patent No. 7,929,667 issued to KLA-Tencor Corp. on April 19, 2011, which is incorporated herein by reference in its entirety.

[0056] As a non-limiting example, the X-ray illumination source 110 may include any of the following: a particle accelerator source, a liquid anode source, a rotating anode source, a stationary solid anode source, a microfocus source, a microfocus rotating anode source, a plasma-based source, and an inverse Compton source. In one example, an inverse Compton source available from Lyncen Technologies, Inc., Palo Alto, California (USA) may be considered. The inverse Compton source has the additional advantage of being able to generate X-rays over a range of photon energies, thereby allowing the X-ray source to deliver X-ray emission at different selectable wavelengths.

[0057] In some embodiments, the computing system 130 transmits a command signal 137 to the X-ray illumination source 110, causing it to emit X-ray radiation at a desired energy level. The energy level is changed to obtain measurement data that contains more information about the high aspect ratio structure being measured.

[0058] The X-ray illumination source 110 generates X-ray emission over a source region having a finite lateral dimension (i.e., a non-zero dimension perpendicular to the beam axis). The focusing optical element 111 focuses the source emission onto a measurement target located on the sample 101. The finite lateral source dimension results in a finite spot size 102 on the target, defined by a line 117 entering from the edge of the source. In some embodiments, the focusing optical element 111 includes an elliptical focusing optical element.

[0059] The beam divergence control slit 112 is located in the beam path between the focusing optical element 111 and the beam shaping slit mechanism 120. The beam divergence control slit 112 limits the divergence of illumination provided to the sample under measurement. An additional intermediate slit 113 is located in the beam path between the beam divergence control slit 112 and the beam shaping slit mechanism 120. The intermediate slit 113 provides additional beam shaping. However, the intermediate slit 113 is generally optional.

[0060] The beam shaping slit mechanism 120 is located in the beam path in front of the sample 101. In some embodiments, the beam shaping slit mechanism 120 includes a plurality of independently operated beam shaping slits. In one embodiment, the beam shaping slit mechanism 120 includes four independently operated beam shaping slits. These four beam shaping slits effectively block portions of the incoming beam 115, generating an illumination beam 116 with a box-shaped illumination cross section.

[0061] Generally, an X-ray optical element shapes the X-ray emission and directs it toward the sample 101. In some examples, the X-ray optical element includes an X-ray monochromator to monochromatize the X-ray beam incident on the sample 101. In some examples, the X-ray optical element uses a multilayer X-ray optical element to collimate or focus the X-ray beam toward the measurement area 102 of the sample 101 to a divergence of less than 1 millirad. In these examples, the multilayer X-ray optical element also functions as a beam monochromator. In some embodiments, the X-ray optical element includes one or more X-ray collimating mirrors, X-ray diaphragms, X-ray beam stops, refractive X-ray optical elements, diffractive optical elements such as zone plates, Montell optical elements, spectroscopy X-ray optical elements such as oblique incidence ellipsoidal mirrors, polycapillary optical elements such as hollow capillary X-ray waveguides, multilayer optical elements or systems, or any combination thereof. Further details are provided in U.S. Patent Publication No. 2015 / 0110249, which is incorporated herein by reference in its entirety.

[0062] In some embodiments, the X-ray illumination source 110, the focusing optical element 111, the slits 112 and 113, or any combination thereof, are maintained in a controlled atmospheric environment (e.g., a gas-purged environment). However, in some embodiments, the optical path lengths between and within these elements are long, and X-ray scattering in the air noises the image on the detector. Therefore, in some embodiments, the X-ray illumination source 110, the focusing optical element 111, and any of the slits 112, 113 are maintained in a local vacuum environment. In the embodiment shown in Figure 2, the focusing optical element 111, the slits 112, 113, and the beam shaping slit mechanism 120 are maintained in a controlled environment (e.g., a vacuum) within a vacuum-sealed flight tube 118. The illumination beam 116 passes through a window 122 at the end of the flight tube 118 before being incident on a window 106 of the vacuum chamber 104. In some embodiments, the flight tube 118 is integrated with the vacuum chamber 104.

[0063] After being incident on the wafer 101, the scattered X-ray emission 114 exits the vacuum chamber 104 through the window 107. In some embodiments, the optical path length (i.e., the collected beam path) between the vacuum chamber 104 and the detector 119 is long, and X-ray scattering in the air noises the image on the detector. Therefore, in a preferred embodiment, most of the collected beam path length between the vacuum chamber 104 and the detector 119 is maintained in a local vacuum environment separated from the environment by a vacuum window (e.g., vacuum window 124). In some embodiments, the vacuum chamber 123 is integrated with the vacuum chamber 104, which has a window separating the vacuum environment 103 from the vacuum environment maintained within the vacuum chamber 123. In some embodiments, the X-ray detector 119 is maintained in the same local vacuum environment as the beam path length between the vacuum chamber 104 and the detector 119. For example, as shown in Figure 2, the vacuum chamber 123 maintains the local vacuum environment surrounding the detector 119 and most of the beam path length between the vacuum chamber 104 and the detector 119.

[0064] In some other embodiments, the X-ray detector 119 is maintained in a controlled atmospheric environment (e.g., a gas-purged environment). This may be advantageous for removing heat from the detector 119. However, in these embodiments, it is preferable to maintain the majority of the beam path length between the vacuum chamber 104 and the detector 119 in a local vacuum environment within the vacuum chamber. Generally, the vacuum window can be made of any suitable material that is substantially transparent to X-ray radiation (e.g., Kapton, beryllium, etc.).

[0065] The X-ray detector 119 collects X-ray radiation 114 scattered from the sample 101 and generates an output signal 135 that exhibits the characteristics of the sample 101 that are sensitive to the incident X-ray radiation, according to the T-SAXS measurement method. In some embodiments, the scattered X-rays 114 are collected by the X-ray detector 119 while the sample positioning system 140 positions and orients the sample 101 to generate angle-resolved scattered X-rays.

[0066] In some embodiments, the T-SAXS system has a high dynamic range (e.g., 10 5 It includes one or more photon counting detectors having a (super) aspect ratio. In some embodiments, a single photon counting detector detects the position and number of photons to be detected.

[0067] In some embodiments, the X-ray detector resolves the energies of one or more X-ray photons and generates a signal for each X-ray energy component that indicates the properties of the sample. In some embodiments, the X-ray detector 119 includes one of the following: a CCD array, a microchannel plate, a photodiode array, a microstrip proportional counter, a gas-filled proportional counter, a scintillator, or a fluorescent material.

[0068] In this configuration, X-ray photon interactions within the detector are determined by energy, in addition to pixel position and count. In some embodiments, X-ray photon interactions are determined by comparing the energy of the X-ray photon interaction with a predetermined upper threshold and a predetermined lower threshold. In one embodiment, this information is communicated to a computing system 130 via an output signal 135 for further processing and storage (e.g., to memory 190).

[0069] In a further embodiment, the T-SAXS system is used to determine the properties of a sample (e.g., structural parameter values) based on one or more diffraction orders of scattered light. As depicted in Figure 2, the system 100 includes a computing system 130 employed to acquire a signal 135 generated by a detector 119, characterize the sample based at least partially on the acquired signal, and store the determined parameters of the subject 122 in memory (e.g., memory 190). In some embodiments, the computing system 130 is configured as a process control measurement engine that uses a measurement model to directly estimate the values ​​of one or more parameters of the subject based on wafer scatterometry measurements.

[0070] In another embodiment, T-SAXS-based measurements involve determining the dimensions of a sample by inverse solving a predetermined measurement model using the measured data. The measurement model includes several (approximately 10) adjustable parameters representing the geometry and optical properties of the sample, as well as the optical properties of the measurement system. The inverse solving method includes, but is not limited to, model-based regression, tomography, machine learning, or any combination thereof. In this form, the target profile parameters are estimated by finding values ​​in the parameterized measurement model that minimize the error between the measured scattered X-ray intensity and the modeled result.

[0071] In some embodiments, the measurement model is an electromagnetic model of the measurement (e.g., a Born wave model) that generates an image representing scattering from the target object being measured. The modeled image can be parameterized by process control parameters (e.g., etching time, etching gradient, etching selectivity, deposition rate, etc.). The modeled image can also be parameterized by structural parameters of the measured high-aspect-ratio structure (e.g., height, diameter at different heights, hole alignment relative to other structures, straightness of hole structures, concentricity of hole structures, thickness of deposited layers as a function of depth, uniformity of deposited layers across or between different hole structures, etc.).

[0072] The measured scattering images are used to estimate the values ​​of one or more parameters of the subject by performing an inverse analysis. In these examples, the inverse method solves for the values ​​of process parameters, geometric parameters, or both, that produce a model scattering image that is the closest to the measured image. In some examples, the space of scattering images is searched using a measurement model that employs regression analysis methods (e.g., gradient descent). In some examples, a pre-calculated library of images is generated, and the library is searched to find the values ​​of one or more parameters of the subject that yield the best match between the modeled image and the measured image.

[0073] In some other examples, the measurement model is trained by a machine learning algorithm to associate many samples of scattered images with known process conditions, geometric parameter values, or both. In this way, the trained measurement model maps the measured scattered images to estimates of process parameters, geometric parameters, or both. In some examples, the trained measurement model is a signal-response measurement (SRM) model that defines a direct functional relationship between the actual measurements and the parameters under consideration.

[0074] In general, any of the trained models described herein are implemented as neural network models. In other examples, any of the trained models may be implemented as linear models, nonlinear models, polynomial models, response surface models, support vector machine models, decision tree models, random forest models, deep network models, convolutional network models, or other types of models.

[0075] In some examples, any of the trained models described herein may be implemented as a combination of models. Further descriptions of model training and the use of trained measurement models for semiconductor measurement are provided in U.S. Patent Publication 2016 / 0109230 by Pandev et al., the contents of which are incorporated herein by reference in their entirety.

[0076] To improve the accuracy and precision of measured parameter values, it is desirable to perform measurements over a wide range of incident and azimuth angles. This approach reduces correlations between parameters by expanding the number and diversity of datasets available for analysis and including various large-angle out-of-plane directions. For example, in normal orientations, T-SAXS can resolve the critical dimensions of a structure, but is largely unaffected by the sidewall angles and heights of the structure. However, by collecting measurement data over a wide range of out-of-plane angular directions, the sidewall angles and heights of the structure can be resolved. In another example, measurements performed over a wide range of incident and azimuth angles provide sufficient resolution and transmission depth, characterizing high-aspect-ratio structures throughout the entire depth.

[0077] Measurements of the intensity of the diffracted radiation as a function of the X-ray incidence angle relative to the wafer surface normal are collected. The information contained in multiple diffraction orders is usually unique among the model parameters under consideration. Therefore, X-ray scattering yields estimations of the target parameter values ​​with small errors and reduced parameter correlations.

[0078] Figure 3 illustrates an exemplary model building and analysis engine 180 implemented by the computing system 130. As depicted in Figure 3, the model building and analysis engine 180 includes a structural model building module 181 that generates a structural model 182 of the measured structure of a sample. In some embodiments, the structural model 182 also includes the material properties of the sample. The structural model 182 is received as input to a T-SAXS response function building module 183. The T-SAXS response function building module 183 generates a T-SAXS response function model 184 based at least partially on the structural model 182. In some embodiments, the T-SAXS response function model 183 is based on the X-ray form factor.

[0079]

number

[0080] Here, F is the form factor, q is the scattering vector, and ρ(r) is the electron density of the sample in spherical coordinates. Next, the X-ray scattering intensity is given by equation 2.

[0081]

number

[0082] The T-SAXS response function model 184 is received as input to the fitting analysis module 185. The fitting analysis module 185 compares the modeled T-SAXS response with the corresponding measurement data 135 to determine the geometric and material properties of the sample.

[0083] In some cases, fitting modeled data to experimental data is achieved by minimizing the chi-squared value. For example, in T-SAXS measurements, the chi-squared value can be defined as follows:

[0084]

number

[0085] Here,

number

number

[0086] Equation (5) assumes that the uncertainties associated with different channels are uncorrelated. In cases where the uncertainties associated with different channels are correlated, the covariance between the uncertainties can be calculated. In such cases, the chi-squared value of the T-SAXS measurement can be expressed as follows:

[0087]

number

[0088] Here, V SAXS is the covariance matrix of the uncertainty of the SAXS channel, and T is the transpose matrix.

[0089] In some embodiments, the fitting analysis module 185 resolves at least one sample parameter value by performing a fitting analysis of the T-SAXS measurement data 135 using the T-SAXS response model 184. In some embodiments,

number

[0090] As previously stated herein, fitting T-SAXS data is achieved by minimizing the chi-squared value. However, in general, fitting T-SAXS data may be achieved by other functions.

[0091] Fitting T-SAXS measurement data is advantageous for any type of T-SAXS technique that provides sensitivity to the geometric and / or material parameters of the subject. Sample parameters can be deterministic (CD, SWA, etc.) or statistical (rms height of sidewall roughness, roughness correlation length, etc.), as long as a suitable model describing the interaction between the T-SAXS beam and the sample is used.

[0092] Generally, the computing system 130 is configured to access model parameters in real time using real-time critical dimensioning (RTCD), or the computing system 130 may access a library of pre-calculated models for determining the value of at least one sample parameter value associated with the sample 101. Generally, several forms of CD engines may be used to evaluate the difference between the assigned CD parameter of the sample and the measured CD parameter associated with the sample. Exemplary methods and systems for calculating sample parameter values ​​are described in U.S. Patent No. 7,826,071, issued to KLA-Tencor Corp. on 2 November 2010, which is incorporated herein by reference in its entirety.

[0093] In some embodiments, the model building and analysis engine 180 improves the accuracy of measured parameters by any combination of feedside-way analysis, feedforward analysis, and parallel analysis. Feedside-way analysis refers to taking multiple datasets from different regions of the same sample and passing common parameters determined from the first dataset to the second dataset for analysis. Feedforward analysis refers to taking datasets from different samples and passing common parameters to subsequent analyses using a stepwise, accurate parameter copy feedforward approach. Parallel analysis involves applying a nonlinear fitting method to multiple datasets in parallel or simultaneously, with at least one common parameter being coupled during fitting.

[0094] Multiple-tool and structural analysis refers to feedforward, feedside-way, or parallel analysis based on regression, lookup tables (i.e., "library" matching), or other fitting procedures for multiple datasets. Exemplary methods and systems for multiple-tool and structural analysis are described in U.S. Patent No. 7,478,019, issued to KLA-Tencor Corp. on January 13, 2009, which is incorporated herein by reference in its entirety.

[0095] In another embodiment, one or more SAXS systems are configured to measure multiple different regions of a wafer. In some embodiments, a wafer uniformity value associated with each target parameter is determined based on measurements of each target parameter across the wafer.

[0096] In some embodiments, multiple measurement systems are integrated with a process tool, and the measurement systems are configured to simultaneously measure different regions across the wafer during the process. In some embodiments, a single measurement system integrated with a process tool is configured to continuously measure multiple different regions of the wafer during the process.

[0097] In some embodiments, the SAXS-based measurement methods and systems for semiconductor devices described herein are applied to the measurement of memory structures. These embodiments enable limit dimension (CD), film, and composition measurements of periodic and planar structures.

[0098] Scalculometry measurements as described herein can be used to determine the properties of various semiconductor structures. Examples of structures include, but are not limited to, low-dimensional structures such as FinFETs, nanowires, or graphene, sub-10nm structures, lithographic structures, through-substrate vias (TSVs), memory structures such as DRAM, DRAM 4F2, FLASH, MRAM, and high-aspect-ratio memory structures. Example structural properties include, but are not limited to, geometric parameters such as line-edge roughness, line-with-thickness roughness, pore size, pore density, sidewall angle, profile, critical dimensions, pitch, thickness, and overlay, as well as material parameters such as electron density, composition, grain structure, morphology, stress, strain, and elemental identification. In some embodiments, the measurement target is a periodic structure. In some other embodiments, the measurement target is non-periodic.

[0099] In some examples, the limit dimensions, thickness, overlay, and material properties of high aspect ratio semiconductor structures, including but not limited to spin-transition torque random access memory (STT-RAM), three-dimensional NAND memory (3D-NAND) or vertical NAND memory (V-NAND), dynamic random access memory (DRAM), three-dimensional flash memory (3D-FLASH), resistive random access memory (Re-RAM), and phase-change random access memory (PC-RAM), are measured using the T-SAXS measurement system described herein.

[0100] In some cases, the measurement model is implemented as an element of the SpectraShape® limit dimension measurement system, available from KLA-Tencor Corporation in Milpitas, California, USA. In this way, the measurement model is generated and prepared for use immediately after the system collects the scattered images.

[0101] In several other examples, the measurement model is performed offline by a computing system implementing AcuShape® software, for example, available from KLA-Tencor Corporation in Milpitas, California, USA. The resulting model can be incorporated as an element of the AcuShape® library, which is accessible by the measurement system performing the measurement.

[0102] In some embodiments, the geometric models employed to characterize the in-plane shape of a measured hole structure at a particular depth include a closed curve with three or more degrees of freedom, i.e., three or more independent parameters define the shape of the curve in a two-dimensional plane, examples of which are illustrated by equations (7A) to (7B). As shown in equations (7A) to (7B), a model of the in-plane hole shape, for example, a quadratic circular function, includes four degrees of freedom, i.e., four independent parameters employed to determine the shape described by the function. The four independent parameters are the nominal radius r, the primary eccentricity e, and the secondary eccentricity e. x , second-order eccentricity e y It includes.

[0103]

number

[0104] Increasing the parameterization of the in-plane shape to four degrees of freedom may slightly distort the shape described by closed curves.

[0105] Figure 4 illustrates a shape described by an elliptic function and a quadratic circular function as an example. Plot line 210 illustrates a shape described by an elliptic function illustrated by equation (2A)-(2B) having two degrees of freedom, r and e, which are non-zero. Plot line 211 illustrates a shape described by an elliptic function with r, e and e y The shape described by the quadratic circular function in the y-direction, as illustrated by equations (7A) to (7B) which have non-zero values, is shown. As depicted in Figure 4, the shape described by the quadratic circular function is distorted in the y-direction.

[0106] Figure 5 shows a three-dimensional plot of signal error in a detector in one embodiment. In the example depicted in Figure 5, the signal error is determined as a weighted difference between the measured signal in the detector and the signal modeled using an elliptic function model of the hole shape. The weighting is a logarithmic function. The logarithmic function normalizes the error signal across the entire detector. Generally, this emphasizes strongly scattered errors, i.e., signals far from the detector center, and does not emphasize lower-order errors, i.e., errors close to the detector center. As depicted in Figure 5, significant errors exist at lower scattering orders, but also at higher scattering orders where valuable shape information tends to be located. These errors indicate a low degree of agreement between the intensity measured in the detector and the intensity modeled.

[0107] In some other embodiments, the geometric model employed to characterize the in-plane shape of a hole structure measured at a particular depth includes a piecewise collection of two or more conical cross-sections, such as elliptical, parabolic, and hyperbolic cross-sections.

[0108] In one embodiment, the distorted ellipse described by the quadratic circular function given by equations (7A)-(7B) is closely approximated by a piecewise collection of four pure elliptic quarters, each having its own radius and elliptic parameters. These linear parameters can be determined by parametric constants that weight the nonlinear terms in equations (7A)-(7B), as shown in equations (8A)-(8B), (9A)-(9B), (10A)-(10B), and (11A)-(11B).

[0109] For the northeast quadrant, that is, for θ from 0 to 90 degrees measured counterclockwise from the x-axis

number

[0110] For the northwest quadrant, that is, for θ from 90 to 180 degrees measured counterclockwise from the x-axis.

number

[0111] For the southwest quadrant, that is, for θ from 180 to 270 degrees measured counterclockwise from the x-axis.

number

[0112] For the southeast quadrant, that is, for θ from 270 to 360 degrees measured counterclockwise from the x-axis.

number

[0113] Figure 6 shows a shape described by an elliptic function, a quadratic circular function, and a piecewise collection of four conical cross-sections in one embodiment. Plot line 230 shows a shape described by a linear circular function given by equation (2A)-(2B) with two degrees of freedom, r and e, having non-zero values. Plot line 231 shows r, e and e y The shape described by the quadratic circular function in the y direction, illustrated by equations (7A) to (7B) which have non-zero values, is illustrated. Plot lines 232A-D show a piecewise collection of four elliptic cross-sections, each described by a different linear elliptic function. Plot line 232A is described by equations (8A) to (8B), plot line 232B is described by equations (9A) to (9B), plot line 232C is described by equations (10A) to (10B), and plot line 232D is described by r, e, e x , and e y The non-zero values ​​of are described by equations (11A) to (11B). As shown in Figure 6, the shape described by the piecewise set of elliptic functions closely matches the shape described by the quadratic circular function. Furthermore, the shape described by the piecewise set of elliptic functions is distorted by about 20% compared to the shape described by the linear circular function.

[0114] Generally, independent parameters describing in-plane hole shape are expressed as a function of depth through the structure, i.e., in the z-direction, to describe the actual change in the in-plane shape of a semiconductor device, which is treated as a function of depth. In a typical patterning process, holes in the resist are nearly circular, and the strain is minimal or very small. However, there are limits to the process control of the etching tool. As a result, the strain changes with depth as the etching process transfers the lithography pattern to the semiconductor layer. The ability to accurately describe the change in shape as a function of depth leads to a more accurate fit between the model and the measured data, and consequently, an improved estimate of the shape parameter values.

[0115] The parameterizations employed to describe elliptic curves described above are those of radius, linear eccentricity, and quadratic eccentricity parameters. However, other parameterizations can generally be employed to describe elliptic curves with the same results, and such parameterizations are intended within the scope of this patent document. As a non-limiting example, parameterization using major and minor axis parameters is intended within the scope of this patent document.

[0116] The aforementioned parameterization of elliptic distortion is explained by a second-order elliptic term, but generally, higher-order terms are intended within the scope of this patent document. However, the mathematics of power approximations and etching physics suggest that the magnitude of the parameterization's contribution to describing the shape decreases proportionally to the order of the error. Therefore, second-order distortion captures more variation in in-plane shape than higher-order distortions, etc.

[0117] In general, any translation and rotation of any in-plane shape employed to describe a measured structure is contemplated within the scope of this patent document. As a non-limiting example, the elliptic axis can be rotated using Given's rotation. Similarly, the elliptic axis may be shifted so that the ellipse is positioned at any location in the Cartesian plane.

[0118] In general, higher-order strains can be equivalently described by any number of conical sections, such as a piecewise collection of elliptical sections. For example, three or six elliptical sections can be used to describe a triangular hole structure. In another example, eight elliptical sections may be used to represent a quadrilateral or octagonal hole structure. Thus, within this patent document, it is intended that any number of conical sections greater than one may be adjacent to each other to describe the shape of a measured structure.

[0119] The advantage of a piecewise collection of conical sections for approximating the shape of a measured structure is that all curves have known analytical formulas for intersections with other conical sections or linear curves. The available analytical solutions are compatible with calculations already performed by AcuShape® software, available from KLA-Tencor Corporation in Milpitas, California, USA.

[0120] Conversely, the difficulty arising from the use of quadratic or higher-order descriptions of curves is that calculating the intersections between quadratic or higher-order curves requires numerical solutions, which increases the computational burden of the model.

[0121] Generally, a measurement target is characterized by an aspect ratio, which is defined as the maximum height dimension of the measurement target (i.e., the dimension perpendicular to the wafer surface) divided by the maximum lateral dimension (i.e., the dimension aligned with the wafer surface). In some embodiments, the measurement target under measurement has an aspect ratio of at least 20. In some embodiments, the measurement target has an aspect ratio of at least 40.

[0122] Figures 7A–7C show isometric, top, and cross-sectional views, respectively, of a typical 3D flash memory device 195 measured by the method described herein. The total height (or equivalent depth) of the memory device 195 ranges from 1 to several micrometers. The memory device 195 is a vertically fabricated device. Vertically fabricated devices such as the memory device 195 essentially rotate conventional planar memory devices by 90 degrees, oriented the bit lines and cell strings vertically (perpendicular to the wafer surface). To provide sufficient memory capacity, numerous alternating layers of different materials are deposited on the wafer. This requires a patterning process that works well to a depth of several microns for structures with a maximum lateral spread of 100 nanometers or less. As a result, aspect ratios of 25:1 or 50:1 are not uncommon.

[0123] In another embodiment, process modifications are determined based on measurements of the parameters in question (e.g., limit dimensions, overlay, height, sidewall angle, etc.), and the modifications are communicated to the process tools to change one or more process control parameters of the process tools (e.g., lithography tools, etching tools, deposition tools, etc.). In some embodiments, SAXS measurements are performed while the process is being carried out on the measured structure, and the process control parameters are updated. In some embodiments, SAXS measurements are performed after a particular process step and the process control parameters associated with that process step have been updated for future device processing by that process step. In some embodiments, SAXS measurements are performed after a particular process step and the process control parameters associated with a subsequent process step have been updated for processing the measured device or other devices by the subsequent process step.

[0124] In some examples, the values ​​of measurement parameters determined based on the measurement methods described herein can be communicated to the etching tool to adjust the etching time to achieve a desired etching depth. Similarly, etching parameters (e.g., etching time, diffusion rate, etc.) or deposition parameters (e.g., time, concentration, etc.) can be included in the measurement model to provide active feedback to the etching tool or deposition tool, respectively. In some examples, modifications to process parameters determined based on measured device parameter values ​​can be communicated to the process tool. In one embodiment, the computing system 130 determines the values ​​of one or more parameters of interest during the process based on measurement signals 135 received from the measurement system 101. Furthermore, the computing system 130 communicates control commands to the process controller based on the determined values ​​of one or more parameters of interest. The control commands cause the process controller to change the state of the process (e.g., stop the etching process, change the diffusion rate, etc.). In one example, the control command causes the process controller to stop the etching process when a desired etching depth is measured. In another example, the control command causes the process controller to change the etching rate to improve the measured wafer uniformity of the CD parameter.

[0125] Figure 2 shows a transmission-type SAXS measurement system, but generally, a reflection-type SAXS measurement system can be used to measure features as described herein.

[0126] Figure 8 shows an exemplary wafer measurement system 200 for X-ray scatterometry measurement of semiconductor structures. In some embodiments, the measured values ​​of the parameter 222 of interest are provided as feedback for controlling manufacturing process tools, such as etching process tools, lithography process tools, deposition tools, etc.

[0127] The wafer measurement system 200 includes a vacuum chamber 204 containing a vacuum environment 203 and a reflective X-ray scattermeter. The semiconductor wafer 201 is placed inside the vacuum chamber 204. The wafer 201 is mounted on a wafer chuck 205 and positioned relative to the vacuum chamber 204 and the X-ray scattermeter by a wafer stage 240.

[0128] In the illustrated embodiment, the SAXS measurement system includes an X-ray illumination source 210 configured to produce X-ray emission suitable for reflective SAXS measurement, similar to the description of illumination source 110 with reference to Figure 2.

[0129] In some examples, the computing system 130 transmits a command signal 237 to the X-ray illumination source 210, causing the X-ray illumination source 210 to emit X-ray radiation at a desired energy level. By changing the energy level, measurement data is obtained that includes detailed information about the high aspect ratio structure being measured.

[0130] The irradiation beam 216 passes through the window 206 of the vacuum chamber 204 and irradiates the sample 201 on the measurement spot 201. After entering the wafer 201, the scattered X-rays 214 exit the vacuum chamber 204 through the window 207. In some embodiments, the optical path length (i.e., the collection beam path) between the vacuum chamber 204 and the detector 219 is long, and X-ray scattering in the air noises the image on the detector. Therefore, in a preferred embodiment, most of the collection beam path length between the vacuum chamber 204 and the detector 219 is maintained in a local vacuum environment.

[0131] The X-ray detector 219 collects X-ray radiation 214 scattered from the sample 201 and generates an output signal 235 that describes the characteristics of the sample 201 that are sensitive to the incident X-ray radiation, according to the reflected SAXS measurement method. In some embodiments, the scattered X-rays 214 are collected by the X-ray detector 219 while the sample positioning system 240 positions and orientations the sample 201 to generate angle-resolved scattered X-rays according to a command signal 239 communicated from the computing system 230 to the sample positioning system 240.

[0132] In a further embodiment, a computing system 230 is used to determine the properties (e.g., structural parameter values) of the wafer 201 based on one or more diffraction orders of scattered light. As shown in Figure 8, the system 200 includes a computing system 230 used to acquire a signal 235 generated by a detector 219, to determine the properties of the sample based at least in part on the acquired signal, and to store instructions 222 of the determined values ​​of the parameter of interest in memory (e.g., memory 290). In some embodiments, the computing system 230 is configured as a process control measurement engine that directly estimates the values ​​of one or more parameters of interest based on scattermetric measurements of the wafer during the process using a measurement model such as those described herein.

[0133] Figure 9 shows a method 300 for performing a measurement of a high aspect ratio structure in at least one novel aspect. Method 300 is suitable for implementation by a measurement system such as the SAXS measurement system shown in Figures 2 and 8 of the present invention. In one embodiment, it is recognized that the data processing block of Method 300 can be performed via a pre-programmed algorithm executed by one or more processors of computing system 130, computing system 230, or any other general-purpose computing system. It is recognized herein that the specific structural aspects of the measurement system shown in Figures 2 and 8 should be interpreted as illustrative only and not as limiting.

[0134] In block 301, a certain amount of X-ray illumination light is directed towards a measurement spot that contains one or more structures fabricated on a semiconductor wafer.

[0135] In block 302, the amount of X-ray light reflected from or transmitted through the semiconductor wafer is detected in accordance with the amount of X-ray illumination light.

[0136] In block 303, based on the detected X-ray intensity, the values ​​of one or more parameters of one or more objects related to a geometrically parameterized response model of one or more structures are determined. The geometrically parameterized response model characterizes the in-plane shape of one or more structures using a geometric model with two or more degrees of freedom.

[0137] In further embodiments, the system 100 includes one or more computing systems 130 used to perform measurements of a semiconductor structure based on scattermetric measurement data collected according to the method described herein. The one or more computing systems 130 can be communicatively coupled to one or more detectors, active optical elements, process controllers, etc. In one embodiment, the one or more computing systems 130 are configured to receive measurement data related to scattermetric measurements of the structure of the wafer 101.

[0138] It should be recognized that one or more steps described throughout this disclosure may be performed by a single computer system 130 or, alternatively, by multiple computer systems 130. Furthermore, different subsystems of system 100 may include computer systems suitable for performing at least some of the steps described herein. Accordingly, the foregoing description should not be construed as a limitation on the invention, but merely as an example.

[0139] Furthermore, the computer system 130 can be communicatively coupled to the spectrometer in any manner known in the art. For example, one or more computing systems 130 may be coupled to a computing system associated with the scattermeter. In another example, the scattermeter may be directly controlled by a single computer system coupled to the computer system 130.

[0140] The computer system 130 of system 100 may be configured to receive and / or acquire data or information from subsystems of the system (e.g., a scattermeter) via a transmission medium which may include a wired portion and / or a wireless portion. In this way, the transmission medium may function as a data link between the computer system 130 and other subsystems of system 100.

[0141] The computer system 130 of system 100 may be configured to receive and / or acquire data or information (e.g., measurement results, modeling inputs, modeling results, etc.) from other systems via a transmission medium which may include wired and / or wireless portions. In this way, the transmission medium may function as a data link between the computer system 130 and other systems (e.g., the memory-equipped system 100, external memory, or other external systems). For example, the computing system 130 may be configured to receive measurement data from a storage medium (i.e., memory 132 or external memory) via the data link. For example, scatter images acquired using the scattermeter described herein may be stored in a persistent or semi-persistent memory device (e.g., memory 132 or external memory). In this regard, scattermetric images can be imported from onboard memory or an external memory system. Furthermore, the computer system 130 can send data to other systems via the transmission medium. For example, measurement models or estimated parameter values ​​determined by the computer system 130 may be communicated and stored in external memory. In this regard, measurement results can be exported to another system.

[0142] The computing system 130 may include, but is not limited to, a personal computer system, a mainframe computer system, a workstation, an image computer, a parallel processor, or any other device known in the art. Generally, the term “computing system” can be broadly defined to encompass any device having one or more processors that execute instructions from a memory medium.

[0143] Program instructions 134 that implement methods such as those described herein may be transmitted via a transmission medium such as a wired, cable, or wireless transmission link. For example, as shown in Figure 1, program instructions 134 stored in memory 132 are transmitted to processor 131 via bus 133. Program instructions 134 are stored in a computer-readable medium (e.g., memory 132). Exemplary computer-readable mediums include read-only memory, random-access memory, magnetic or optical disks, or magnetic tapes. A computing system 230 including elements 231-234 is similar to a computing system 130 including elements 131-134, respectively, as described herein.

[0144] As described herein, the term “limit dimension” includes any limit dimension of a structure (e.g., bottom limit dimension, middle limit dimension, top limit dimension, side wall angle, grid height, etc.), limit dimensions between any two or more structures (e.g., distance between two structures), and displacements between two or more structures (e.g., overlay displacement between structures with grids stacked on top of each other). Structures may include three-dimensional structures, patterned structures, overlay structures, etc.

[0145] As described herein, the terms “limit dimension application” or “limit dimension measurement application” include any limit dimension measurement.

[0146] As described herein, the term “measurement system” includes any system used at least partially to characterize a sample in any aspect, including measurement applications such as limit dimensional measurement, overlay measurement, focus / dose measurement, and composition measurement. However, such terms in the art do not limit the scope of the term “measurement system” as described herein. In addition, a measurement system may be configured for the measurement of patterned wafers and / or unpatterned wafers. A measurement system may be configured as an LED inspection tool, an edge inspection tool, a back inspection tool, a macro inspection tool, or a multimode inspection tool (including data from one or more platforms simultaneously), and any other measurement or inspection tool that benefits from the calibration of system parameters based on limit dimensional data.

[0147] Various embodiments of semiconductor measurement systems are described herein, which may be used to measure samples within any semiconductor process tool (e.g., inspection system or lithography system). The term “sample” is used herein to mean a wafer, reticle or any other sample which may be processed (e.g., printed or inspected for defects) by means known in the art.

[0148] As used herein, the term “wafer” generally refers to a substrate formed from a semiconductor or non-semiconductor material. Examples include, but are not limited to, single-crystal silicon, gallium arsenide, and indium phosphide. Such substrates may be commonly found and / or processed within semiconductor manufacturing facilities. In some cases, a wafer may consist of only a substrate (i.e., a bare wafer). Alternatively, a wafer may consist of one or more layers of different materials formed on the substrate. One or more layers formed on the wafer may be “patterned” or “unpatterned.” For example, a wafer may consist of multiple dies having repeating pattern features.

[0149] "Reticle" may refer to a reticle at any stage of the reticle manufacturing process, or a finished reticle that may or may not be released for use within a semiconductor manufacturing facility. A reticle or "mask" is generally defined as a substantially transparent substrate on which substantially opaque regions are formed to constitute a pattern. The substrate may include, for example, a glass material such as amorphous SiO2. The pattern on the reticle may be transferred to the resist by being placed on a resist-coated wafer during the exposure step of a lithography process.

[0150] One or more layers formed on a wafer may be patterned or unpatterned. For example, a wafer may contain multiple dies, each having repeating pattern features. The formation and processing of such layers of material may ultimately result in a finished device. Many different types of devices may be formed on a wafer, and the term wafer as used herein is intended to include wafers on which any type of device known in the art is fabricated.

[0151] In one or more exemplary embodiments, the described functions may be implemented in hardware, software, firmware, or any combination thereof. If the functions are implemented in software, they may be stored on a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable medium includes both computer storage media and communication media, and these include any media that facilitate the transfer of computer programs from one location to another. Storage media may be any available media that can be accessed by a general-purpose or dedicated computer. Examples, but not limited to, such computer-readable media may comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage, or any other media, which may be used to transmit or store desired program code means in the form of instructions or data structures, and may be accessed by a general-purpose or dedicated computer, or a general-purpose or dedicated processor. Any connection is correctly referred to as computer-readable medium. For example, if software is transmitted from a website, server, or another remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of media. As used herein, disks (disk and disc) include compact discs (CDs), laser discs, optical discs, digital multipurpose discs (DVDs), floppy disks, and Blu-ray discs, in which case a disk typically copies data magnetically, while a disc optically copies data by laser. The above combinations must also be included within the scope of computer-readable media.

[0152] Although certain embodiments are described above for teaching purposes, the teachings in this patent document have general applicability and are not limited to those specific embodiments. Accordingly, various modifications, adaptations, and combinations of various features of the described embodiments may be carried out without departing from the scope of the invention as stated in the claims.

Claims

1. The steps include providing a certain amount of X-ray illumination light to a measurement spot which includes one or more structures fabricated on a semiconductor wafer, A step of detecting the amount of X-ray light reflected from or transmitted through the semiconductor wafer in response to the amount of X-ray illumination light, A step of determining the values ​​of one or more parameters related to a geometrically parameterized response model of one or more structures based on the detected X-ray light intensity, wherein the geometrically parameterized response model characterizes the in-plane shape of the one or more structures using a geometric model having two or more degrees of freedom. Includes, The geometric model includes a closed curve comprising a piecewise collection of two or more conical cross-sections that characterize the in-plane, non-elliptical distortion of the hole shape as one or more structures, wherein the two or more conical cross-sections include at least one of an elliptical cross-section, a parabolic cross-section, or a hyperbolic cross-section.

2. The method according to claim 1, The method is characterized in that the step of determining the values ​​of one or more target parameters includes a fitting analysis of the detected X-ray light quantity using the geometrically parameterized response model.

3. The method according to claim 1, The method is characterized in that the geometric model includes a closed curve having a shape defined on a two-dimensional plane by three or more independent parameters.

4. The method according to claim 1, The method is characterized in that the two or more conical cross-sections include a plurality of elliptical cross-sections, each of which is described by an independent radius and elliptic parameter.

5. The method according to claim 1, The method is characterized in that the two or more conical cross-sections include a plurality of parabolic cross-sections, each described by two independent parameters.

6. The method according to claim 1, The method is characterized in that one or more of the structures include a three-dimensional NAND structure or a dynamic random access memory (DRAM) structure.

7. The method according to claim 1, The method is characterized in that the value of the parameter of one or more targets is determined in a process step of the manufacturing process flow of the one or more structures, the instruction for the value of the parameter of one or more targets is transmitted to a manufacturing tool, and the manufacturing tool adjusts the value of one or more process control parameters of the manufacturing tool in the process step.

8. The method according to claim 1, A method characterized in that the independent values ​​of the geometric model change as a function of the depth to one or more structures during measurement.

9. The method according to claim 1, The method is characterized in that the aforementioned amount of X-ray illumination light is directed towards the measurement spot at multiple incident angles, azimuth angles, or both.

10. The method according to claim 1, The method is characterized in that the aforementioned amount of X-ray illumination light is directed to the measurement spot at multiple different energy levels.

11. An illumination source configured to provide a certain amount of X-ray illumination light to a measurement spot which includes one or more structures fabricated on a semiconductor wafer, A detector configured to detect the amount of X-ray light reflected from or transmitted through the semiconductor wafer in response to the amount of X-ray illumination light, A computing system configured to determine the values ​​of parameters for one or more objects based on a fitting analysis of the detected X-ray light intensity and the geometrically parameterized response model of one or more structures, wherein the geometrically parameterized response model characterizes the in-plane shape of the one or more structures using a geometric model having two or more degrees of freedom. A system comprising a geometric model comprising a closed curve comprising a piecewise collection of two or more conical cross-sections that characterize the in-plane, non-elliptical distortion of one or more hole shapes as structures, wherein the two or more conical cross-sections include at least one of an elliptical cross-section, a parabolic cross-section, or a hyperbolic cross-section.

12. The system according to claim 11, The system is characterized in that the geometric model includes a closed curve having a shape defined on a two-dimensional plane by three or more independent parameters.

13. The system according to claim 11, The system is characterized in that the two or more conical cross-sections include a plurality of elliptical cross-sections, each of which is described by an independent radius and elliptic parameter.

14. The system according to claim 11, The system is characterized in that the two or more conical cross-sections include a plurality of parabolic cross-sections, each described by two independent parameters.

15. The system according to claim 11, The system is characterized in that one or more of the structures include a three-dimensional NAND structure or a dynamic random access memory (DRAM) structure.

16. The system according to claim 11, The system is characterized in that the value of one or more target parameters is determined in a process step of the manufacturing process flow of one or more structures, the instruction for the value of one or more target parameters is transmitted to a manufacturing tool, and the manufacturing tool adjusts the value of one or more process control parameters of the manufacturing tool in the process step.

17. The system according to claim 11, A system characterized in that the independent values ​​of the geometric model change as a function of the depth to one or more structures during measurement.

18. An illumination source configured to provide a certain amount of X-ray illumination light to a measurement spot which includes one or more structures fabricated on a semiconductor wafer, A detector configured to detect the amount of X-ray light reflected from or transmitted through the semiconductor wafer in response to the amount of X-ray illumination light, A non-temporary computer-readable medium that, when executed by one or more processors, stores instructions causing the one or more processors to determine the values ​​of parameters of one or more objects based on a fitting analysis of the detected X-ray light quantity and a geometrically parameterized response model of one or more structures, wherein the geometrically parameterized response model characterizes the in-plane shape of one or more structures using a geometric model having two or more degrees of freedom, A system comprising a geometric model comprising a closed curve comprising a piecewise collection of two or more conical cross-sections that characterize the in-plane, non-elliptical distortion of one or more hole shapes as structures, wherein the two or more conical cross-sections include at least one of an elliptical cross-section, a parabolic cross-section, or a hyperbolic cross-section.

Images