Techniques for reducing electromagnetic interference effects in charged particle microscopes

Abstract

Embodiments of the present disclosure partially improve the performance of a charged particle beam system during imaging and / or microanalysis by allowing the system and / or user to account for the effects of electromagnetic interference on beam direction and / or shape. Techniques for identifying, tracking and / or correcting electromagnetic interference induced beam drift, as well as techniques for locating defects in integrated circuit devices, are described herein.

Description

Cross - Reference to Related Applications

[0001] This application claims priority to U.S. Provisional Application No. 63 / 481,739, filed on January 26, 2023, with the title "Techniques for Reducing Electromagnetic Interference Effects in a Charged Particle Microscope", the entire content of which is incorporated herein by reference. Technical Field

[0002] Embodiments of the present disclosure relate to charged particle microscope systems and algorithms and methods for their operation. In particular, some embodiments relate to performing electronic failure analysis on nanostructured integrated circuits. Background Art

[0003] Integrated circuit (IC) testing involves measuring individual transistors or groups of transistors, referred to as "devices under test" or DUTs, in a semiconductor wafer or a portion of a wafer (e.g., a diced wafer). The feature sizes of such transistors have fallen below the resolution limit of photon optical systems, and thus charged particle beam systems are needed to resolve device features. A typical IC test regime involves applying periodic signals to the circuit at increasing powers to examine the circuit's performance under different operating conditions and to determine when the circuit fails to operate to specification.

[0004] During the generation of charged particle beam imaging data of an active integrated circuit, electrical activity on the integrated circuit (IC) creates electric and magnetic fields that interfere with the electron beam within the imaging system. Typically, the integrated circuit is coupled to a chip tester that tests the integrated circuit through a periodic test cycle. Activity on the integrated circuit draws device current that typically varies with the test cycle, and such device current generates magnetic fields that may inadvertently deflect the electron beam from its intended target. These unintended beam deflections may be synchronized periodically with the tester cycle, causing the image to shift and blur, making imaging and IC voltage data acquisition difficult or impossible.

[0005] In addition, testing may rely on correlating secondary electron signals with positions on the sample surface. This correlation provides information about the sample area, which in turn allows identification of the specific device to be probed by the electron beam based on the timing relative to the scan signal. Thus, electromagnetic (EM) interference during the test signal cycle may reduce the accuracy of a given test. For example, when electromagnetic (EM) interference causes a beam deflection, the electrons making up the beam may probe a different position on the sample surface than the position indicated by the scan pattern. Therefore, techniques and systems are needed to identify, track, and correct beam deflections in scanning electron microscope images, such as as part of an electron - induced device alteration (EIDA) process. Summary of the Invention

[0006] Aspects of the present disclosure include a method for locating defects in a sample. The method includes directing a charged particle beam towards the sample. The method includes generating drift information of the charged particle beam in one or more directions, the drift being caused at least in part by an electromagnetic field near the sample. The method further includes using the drift information to locate a defect in the sample.

[0007] In some embodiments of the method, generating the drift information includes generating a sequence of images of the sample, where the image sequence includes a plurality of sample images, and using the image sequence to generate the drift information. At least a portion of the images may depict a feature of the sample. Generating the drift information may include tracking a feature of the sample in the image sequence and generating offset data describing the movement of the feature in the image sequence. Locating the defect may include using the image sequence and pass / fail data of the sample to generate location data describing the location of the defect in the sample. The drift information may include a drift vector, and locating the defect may further include using the drift vector to generate a correction vector and using the correction vector to generate a scan signal for the charged particle beam, thereby directing the charged particle beam to the defect location. The location data may describe the pass or fail state of image pixels in the image sequence.

[0008] In some embodiments of the method, locating a defect in the sample includes identifying a defective device in the sample with reference to a sample map. The sample map may include a schematic description of the device. The defect may correspond to a malfunction in the device. The map may include computer-aided design data describing one or more devices in an integrated circuit.

[0009] In some embodiments, the electromagnetic field may be excited by a transient electrical signal applied to at least a portion of the sample. The sample may include a device under test (DUT). The transient signal may include a signal configured to operate the DUT under pass-fail boundary conditions as part of a device perturbation test. The transient electrical signal may include a segment of a periodic voltage. Locating the defect may include performing a binary search on a segment of the signal to identify a malfunction in the DUT.

[0010] Aspects of the present disclosure include a method for processing beam drift artifacts in charged particle microscope images. The method may include determining an acquisition window that constitutes at least a portion of an integrated circuit test cycle. The acquisition window may include a time of interest (TOI) of the test cycle. The method may include generating detector data for a charged particle microscope system. The detector data may describe the sample and may correspond to at least a portion of the acquisition window. The method may further include generating a deflection vector for the time of interest. The deflection vector may describe an offset of the charged particle beam caused by an electromagnetic field near the charged particle beam.

[0011] In some embodiments, the method may further include identifying a location on the surface of the sample using a deflection vector. Identifying the location may include modifying detector data using the deflection vector. The method may further include generating frequency information for the location on the surface using the detector data at the pixel level.

[0012] In some embodiments of the method, generating the deflection vector includes generating a sequence of images of the sample, where the sequence of images includes a plurality of sample images, and using the sequence of images to generate the deflection vector. At least a portion of the images may depict a feature of the sample. Generating the deflection vector may further include tracking the feature of the sample in the sequence of images and generating offset data that describes the movement of the feature in the sequence of images. The frequency information may describe the operating frequency of the device at that location. The detector data may include data at a plurality of time points at that location. The method may further include generating waveform data using the detector data. The waveform data may describe the operating voltage of the device at a plurality of time points.

[0013] In some embodiments of the method, the electromagnetic field may be generated by a transient electrical signal applied to at least a portion of the sample. The transient electrical signal may include a segment of a periodic voltage.

[0014] Aspects of the present disclosure include a method for reducing beam drift artifacts in a charged particle microscope image. The method may include generating a sequence of images of a sample using a charged particle beam. Electromagnetic interference may cause drift of the charged particle beam. The method may include generating drift information for the charged particle beam in one or more directions using the sequence of images. The drift information may describe the beam drift caused by electromagnetic interference. The method may include generating beam deflection instructions that describe beam drift correction. The method may further include modifying the scan mode of the charged particle microscope using the beam deflection instructions.

[0015] In some embodiments, generating the sequence of images may include generating an image by incrementing the spot position of the beam on the surface of the sample at the pixel level and coordinating detector data for the spot position with a transient electrical signal applied to at least a portion of the sample. Generating the sequence of images may include coordinating the beam blanking circuit of the charged particle microscope with the transient electrical signal, the beam blanking circuit configured to block a portion of the beam and generate one or more charged particle pulses, and generating detector data when irradiating a carrier with one or more charged particle pulses at the location, the detector data describing the time steps of the transient electrical signal.

[0016] In some embodiments, at least a portion of the images depict a feature of the sample. Generating the drift information may include tracking the feature of the sample in the sequence of images and generating offset data that describes the movement of the feature in the sequence of images.

[0017] In some embodiments, the sample may include a device under test (DUT), wherein at least a portion of the DUT is subjected to a transient electrical signal. The method may further include generating detector data using a modified scan pattern on a transistor or group of electrically coupled transistors of the DUT. The detector data may include secondary electron detector data. The method may further include generating frequency data for a plurality of time points of the transient electrical signal using the detector data. The method may further include generating voltage data for a plurality of time points of the transient electrical signal using the detector data. The method may further include generating waveform data for a given integrated circuit device of the sample using the voltage data. The method may further include generating timing data for the integrated circuit device of the sample using a modified beam scan pattern. The timing data may include the clock speed of the integrated circuit device.

[0018] Aspects of the present disclosure include a charged particle beam system that includes a charged particle source. The system may include computing circuitry operably coupled to the charged particle source. The system may further include one or more media storing machine-readable instructions that, when executed by the computing circuitry, cause the system to perform operations including performing the operations of one or more of the foregoing aspects of the various embodiments, individually or in combination.

[0019] The following description and figures will elaborate in detail various technical features and corresponding advantages of identifying, tracking, and / or correcting beam offsets in scanning electron microscope images, such as as part of an electron-induced device alteration (EIDA) process. In an illustrative example, a method of processing beam drift artifacts in a charged particle microscope image may include determining an acquisition window. The determination of the acquisition window may be at least partially based on a transient current signal that forms at least a portion of an integrated circuit test cycle. The acquisition window may include the time of interest (TOI) of the test cycle. The method may include determining a deflection vector. The deflection vector may describe a characteristic offset of the charged particle beam caused by a periodic current signal during the acquisition window. The method may further include modifying the beam position on the surface of the device under test (DUT) using the deflection vector.

[0020] The method of the present disclosure may include generating an image sequence of a sample using a charged particle microscope. The image sequence may include a plurality of sample images acquired periodically at different time points. The sample may include carriers that generate transient electromagnetic interference. The method may include determining drift of the beam in one or more directions using the image sequence. The method may include generating beam deflection instructions that describe correction of periodic drift of the beam. The method may include modifying the scan pattern of the charged particle microscope using the beam deflection instructions.

[0021] The method of the present disclosure may include locating a defect in a sample. A method may include directing a charged particle beam towards the sample. The method may include generating drift information of the charged particle beam in one or more directions. The drift may be caused at least in part by an electromagnetic field near the sample. The method may further include using the drift information to locate a defect in the sample.

[0022] The terms and expressions employed are used as descriptive terms and not of limitation, and in using such terms and expressions it is not intended to exclude any equivalents of the features shown and described or portions thereof, but on the contrary, it should be recognized that various modifications are possible within the scope of the subject matter claimed. Accordingly, it should be understood that although the subject matter claimed of the present disclosure has been specifically disclosed by way of embodiments and optional features, those skilled in the art may make modifications and variations to the concepts disclosed herein, and such modifications and variations are considered to be within the scope of the present disclosure as defined by the appended claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The foregoing aspects and many of the attendant advantages of this disclosure will become more readily appreciated when considered in conjunction with the following detailed description taken in conjunction with the accompanying drawings.

[0024] Figure 1 is a schematic diagram of an exemplary integrated circuit test system in accordance with some embodiments of the present disclosure.

[0025] Figure 2 is a block diagram flowchart of an example process of interrogating a device under test using a charged particle beam in accordance with some embodiments of the present disclosure.

[0026] Figures 3A - 3H is a schematic diagram of an example image sequence of a device under test in accordance with some embodiments of the present disclosure.

[0027] Figures 4A - 4C is in accordance with some embodiments of the present disclosure from Figures 3A - 3H is a chart of example offset data and corresponding correction vectors in a Cartesian coordinate space generated from the image sequence of.

[0028] Figures 5A - 5C is a chart of example test signal data and corresponding offset vectors at a given time of interest (TOI) in accordance with some embodiments of the present disclosure.

[0029] Figures 6A - 6C is a schematic diagram of an example technique of perturbing an integrated circuit device using a charged particle beam in accordance with some embodiments of the present disclosure.

[0030] Figure 7 is a schematic diagram of an example technique for detecting a defect in a device under test in accordance with some embodiments of the present disclosure.

[0031] Figures 8A - 8D FIG. is a schematic diagram of an exemplary technique for locating a defect in a device under test based on a reference device under test map according to some embodiments of the present disclosure.

[0032] Figure 9 FIG. is a block flow diagram of data processing operations of an exemplary technique for interrogating a device under test with a charged particle beam according to some embodiments of the present disclosure.

[0033] In the drawings, unless otherwise noted, the same reference numerals represent the same components in each view. Where appropriate, not all instances of an element need be labeled to reduce clutter in the drawings. The drawings are not necessarily to scale, but rather focus on illustrating the described principles. DETAILED DESCRIPTION

[0034] Although the exemplary embodiments have been illustrated and described, it should be understood that various changes may be made therein without departing from the spirit and scope of the present disclosure. In the following paragraphs, embodiments of charged particle beam systems, components, and methods for identifying, tracking, and / or correcting beam offsets in scanning electron microscope images are described, for example, as part of an electron-induced device alteration (EIDA) process. For simplicity of description, the embodiments of the present disclosure focus on scanning electron microscopes and related instruments (SEM). Thus, the embodiments are not limited to such instruments, but are contemplated for analytical instrument systems where sample analysis may be complicated by electromagnetic field disturbances near the sample. In one exemplary example, electromagnetic fields near or originating from a sample may cause a charged particle beam to deflect or otherwise distort, and thus imaging, accuracy, guiding the beam to a desired location on the sample, and / or operation of the scanning beam may benefit from the techniques of the present disclosure. The techniques of the present disclosure are contemplated to include applications to the following instruments: transmission electron microscopes (TEM), scanning transmission electron microscopes (STEM), STEM in SEM, electron beam microanalysis instruments, and / or other instruments configured to generate image data based on signals produced by the interaction of a sample with a charged particle beam (such as ions or electrons).

[0035] The embodiments of the present disclosure partially improve the performance of a charged particle beam system during imaging and / or microanalysis by enabling the system and / or user to account for the effects of electromagnetic interference on the beam direction and / or shape. For example, accurately probing a sample containing nanostructured features (such as integrated circuit devices like transistors) using an electron beam depends on the precise positioning of the beam spot. In this way, electromagnetic fields that cause beam distortion or deflection may introduce errors and have an adverse impact on the accuracy of the probing technique.

[0036] The techniques described herein can improve the quality and accuracy of charged particle microscope images and / or detector data generated during testing of a device under test (DUT) (e.g., by imaging one or more regions of the DUT). Typically, a chip tester tests an IC in a repeatable manner, where the test cycle is repeated in a loop that includes multiple iterations. Although the current consumption of the IC may vary at different times during the test cycle, the current draw can repeat in a predictable manner between cycles and correspond to different times. During testing, the IC may generate electromagnetic interference, which can distort the SEM imaging process. For example, the magnetic field generated by the current within the IC may generate electromagnetic interference (EMI), which can deflect the electron beam used by the SEM for imaging. In this way, an unexpected beam shift may also repeat with each test cycle.

[0037] Charged particle beam techniques for evaluating DUT performance may be affected by electromagnetic interference that causes beam deflection. For example, such interference may significantly reduce the precision and accuracy of certain techniques that correlate spatial information (e.g., using beam scan data) with secondary electron detector data, such as electron beam perturbation techniques.

[0038] To this end, embodiments of the present disclosure include coordinating imaging of a portion of the IC with a given portion of the test cycle running on the IC. For example, by coordinating an imaging sequence (e.g., a beam scan pattern) with a segment of a periodic current signal, the beam shift can be coordinated with the time of interest of the test cycle. By coordinating the imaging with the test cycle, a static image of the IC can be generated. Without eliminating the unexpected beam shift, by synchronizing the SEM imaging or voltage probing with a given portion of the test cycle, the dynamic effect of current consumption on the beam shift can be reduced or eliminated, thereby giving the image temporal stability.

[0039] The method of the present disclosure also includes determining the degree of beam shift for a given portion of the test cycle. By coordinating the SEM imaging process with the tester cycle that causes the unexpected beam shift, the beam shift that occurs at or near the time of interest within the test cycle can be determined. The degree of beam shift can precisely position the probe beam on the device (e.g., a transistor) in the presence of the beam shift without interrupting or otherwise modifying the test cycle.

[0040] Figure 1 is a schematic diagram of an exemplary integrated circuit test system in accordance with some embodiments of the present disclosure. Exemplary system 100 includes instruments 105, instrument computing devices (IPCs) operatively coupled to each other via one or more networks 120 110 and client computing device 115. Exemplary system 100 is configured to interrogate an IC device known as the device under test (DUT) 125 using test component 130, which is electrically coupled to elements of DUT 125 via a controller (also referred to as test apparatus 135). By applying time-varying electronic signals to elements of DUT 125 (referred to as a "test cycle" or "test mode"), performance characteristics of the circuit components of DUT 125 can be derived as part of a quality control and failure analysis technique for ICs manufactured according to a given IC design.

[0041] Instrument 105 includes a test section 140 in which test component 130 is disposed, including DUT 125 and electronic components for driving a test cycle (such as test apparatus 135), vacuum components for isolating DUT 125 from the atmosphere, and a thermal management system for removing heat from DUT 125 during testing. Coupled to test section 140 is a charged particle column 145. Charged particle column 145 can be an ion beam (such as a focused ion beam (FIB)) column or an electron beam column (such as part of a scanning electron microscope). In some embodiments, instrument 105 includes an FIB column and an electron beam column, where one of the charged particle sources is coupled to test section 140 at an angle to charged particle column 145.

[0042] Charged particle column 145 can generate a charged particle beam 147 and can focus charged particle beam 147 onto region 127 of DUT 125. Region 127 can include one or more conductive features as described in more detail with reference to FIG. 3, which can be in an electrically active or inactive state depending on one or more transient electrical signals (also referred to as test mode, test signal, test cycle, etc.) applied to the DUT. The interaction of charged particle beam 147 with DUT 125 produces one or more detectable signals, which can be received by one or more detectors 155 operably coupled to test section 140 and are configured to generate detector data based at least in part on measurements of the signals. In an illustrative example, detector 155 can include a secondary electron detector, a backscattered electron detector, a photon detector, an imaging sensor (such as a CCD), etc. Compared to a typical scanning electron microscope (SEM), test section 140 can omit sample manipulation tools, such as interlocks, sample stages, etc., at least in part because DUT 125 can be removably coupled to test component 130, which can be disposed on a stage, bracket, or other holding component that enables electrical and thermal coupling to test section 140 (such as coupling to test apparatus 135). Charged particle beam 147 can be directed to DUT 125 in a variety of operating modes, including but not limited to imaging mode, line scan mode, spot mode, and / or pulse mode.

[0043] To this end, the charged particle column 145 may include electronic components and electron optical components to manipulate the shape and / or direction of the charged particle beam 147. For example, a beam blanker 150 disposed in the column may be configured to apply an electric field and / or a magnetic field to the entire path of the charged particle beam 147. A control electronic component 151 operably coupled to the beam blanker 150 may apply a time-varying voltage to the electrodes of the beam blanker 150, such that the electric field may reversibly deflect the charged particle beam 147 into the beam blanker. The operation of the beam blanker may allow the charged particle column 145 to direct charged particle pulses to the DUT 125. In some embodiments, a pulse includes as few as 1 charged particle to about 1000 charged particles, including the physical significance fractions of the recited ranges and their subranges. For example, a pulse may include 3 charged particles, 5 charged particles, 10 charged particles, 15 charged particles, etc. In some embodiments, a pulse may span multiple cycles of the test signal.

[0044] The charged particle column 145 may include one or more steering components 153. The steering component 153 may include components configured to generate an asymmetric electromagnetic field. For example, a set of steering elements (such as electromagnetic coils, electrostatic steering plates, etc.) may be operably coupled to a control electronic component 151 configured to apply a voltage to one or more elements. When the charged particles of the charged particle beam 147 pass through the component 153, the asymmetric field can controllably deflect the beam 147, which is a common practice when generating SEM or other forms of detector data (such as in STEM or SEM instruments). As described in more detail in reference Figures 4A - 7 as described in more detail. In some embodiments, the steering component 153 is controlled by a control circuit different from the control circuit used to control the beam blanker 150..

[0045] In some embodiments, the test component 130 is electrically coupled to the components of the test section 140 via a coupler 165, and one or more test cards 170 can be driven through the coupler. The test card 170 can encode a test cycle protocol and can interface with the DUT 125 to input and output signals to and from the DUT 125 and relay the signals to other constituent elements of the exemplary system 100 (such as, the client PC 115 and / or the IPC 110).

[0046] Computing devices 110 and 115 can be general-purpose machines (e.g., laptop computers, tablets, smartphones, servers, etc.) configured to operate instrument 105 or otherwise interact with it. Instrument 105, in turn, can include electronic components that form part of a dedicated computing device, including control circuitry configured to drive test cycles, operate test component 130, control electron beam column 145, and operate the vacuum system and thermal management system. In an illustrative example, test component 130 can be driven by a chip tester that operates independently of instrument 105. The operation of test component 130 can be coordinated with the operation of instrument 105. For example, the chip tester can generate a trigger signal at the start of each test cycle and transmit it to instrument 105. In turn, instrument 105 can respond to the trigger signal by, at least in part, detecting signals at a given location in the DUT. In another example, instrument 105 can be equipped with dedicated control electronics to coordinate the operation of instrument 105 (such as the operation of detector 155 and / or blanker 150) with the operation of test component 130. IPC 110 can be a machine with software configured to interface with instrument 105 and allow a user of instrument 105 to perform tests on DUT 125. Similarly, client PC 115 can be configured to control one or more systems of instrument 105 (e.g., interface with instrument 105 via IPC 110 and / or via network 120) to perform tests on DUT 125.

[0047] In some embodiments, instrument 105, IPC 110, and / or client PC 115 are located in different physical locations and are coupled via network 120 and / or other means (such as direct connection or via a wireless connection (e.g., near-field radio)). Network 120 can include a public network (e.g., the Internet) and / or a private network (e.g., an intranet or local area network). In some embodiments, IPC 110 and / or client PC 115 are configured to operate instrument 105 autonomously (e.g., without human intervention) or semi-autonomously (e.g., with limited human intervention, such as starting a test, identifying a sample, and / or confirming an automated analysis result). In this way, exemplary system 100 can be configured to operate with manual control and / or autonomously as part of a scalable IC characterization system for the automated testing of ICs.

[0048] Exemplary system 100 can include more and / or alternative components than those shown. For example, instrument 105 can be operably coupled to one or more external components (such as a signal generator, a data acquisition system, a power supply system, thermal management, etc.). Such components can be housed, for example, in a cabinet that is physically separate from instrument 105, but can be operably coupled to charged particle column 145, test section 140, detector 155, etc. via electrical connectors and / or fluid handling connectors.

[0049] Figure 2 is a block diagram flowchart of an example process 200 for interrogating a device under test using a charged particle beam, in accordance with some embodiments of the present disclosure. One or more operations of the exemplary process 200 may be performed by a computer system in communication with other systems, including but not limited to a characterization system, a network infrastructure, a database, and a user interface device. In some embodiments, at least one subgroup of the operations described Figure 2 is performed automatically (e.g., without human intervention) or pseudo-automatically (e.g., with human initiation or limited human intervention). In an illustrative example, the operations of applying a test signal, directing a charged particle beam towards a DUT (e.g., Figure 1 DUT 125) and generating detector data may be performed automatically, and the system (e.g., Figure 1 exemplary system 100) is configured to generate visualization data presenting one or more output data forms for interpretation by a human user.

[0050] Embodiments of the present disclosure allow for the generation of images during and / or throughout a tester cycle (e.g., images of a device under test (DUT) or DUT region during and / or throughout a tester cycle). Video recording may be used to determine beam deflections during the tester cycle, such that deflections may be tracked and / or corrected. For example, correction may be implemented in a scan control circuit by providing an offset correction synchronized with the tester cycle. In this way, an electron beam may be positioned at a specific location on an active integrated circuit, and under the influence of a perturbation field generated by the integrated circuit, the electron beam will remain stationary at that location. Additionally, or alternatively, correction may be achieved by applying an offset to the detector data that takes into account interference effects. Although the exemplary process 200 is described as a series of operations, it should be understood that at least some operations may be omitted, repeated, parallelized, combined, and / or reordered. In some embodiments, there are additional operations before and / or after the operations of the exemplary process 200 that are omitted for clarity of illustration. For example, operations include calibrating an electron source, aligning and correcting aberrations of a charged particle beam, introducing a DUT sample into a vacuum system, calibrating the system, etc. In another example, as part of determining one or more failure modes of a DUT (e.g., a perturbation test procedure), a test pattern of a time-varying voltage signal is applied to an integrated circuit component. Referring to the exemplary process 200, the operation of the instrument may be coordinated with the operation of a test component (e.g., Figure 1 test component 130) as part of generating data describing the electrical activity of an IC component of the DUT, from which defect information and / or other information may be derived.

[0051] At operation 205, the exemplary process 200 includes directing a charged particle beam towards a sample (e.g., Figure 1of the DUT 125). As described in more detail in reference Figure 1 As described in more detail in Figure 1 , the charged particle beam (e.g., Figure 1 charged particle beam 147) can be an electron beam, but can also include ions, neutral particles, etc. Operation 205 can form part of a broader imaging and / or microanalysis procedure. Operation 205 can be part of an electron fault analysis procedure for semiconductor integrated circuit quality control, as described in more detail in reference

[0052] . For example, operation 205 can be part of a frequency mapping, waveform acquisition, and / or electron-induced device alteration (EIDA) procedure. These specific examples are described in more detail in Examples 1-3 below, but are not limiting of the application examples of the exemplary process 200. To this end, operation 205 can include focusing the charged particle beam onto the sample. Operation 205 can include pulsing the charged particle beam or otherwise strobing it. Finally, operation 205 can include operating in a spot mode, directing the beam to a given position on the sample surface and holding it stationary for a period of time; in a line scan mode, the beam scans the sample surface in a linear pattern; and / or in an image scan mode, the beam scans the entire sample surface in a two-dimensional pattern.

[0053] In operation 210, the exemplary process 200 includes determining drift in one or more directions. As described in reference Figures 3A - 4CAs described in more detail below, the drift can be determined using sample images generated during a test cycle. Movement of one or more features of the sample surface (such as edges or other trackable aspects) can be used to generate deflection data as a function of the sample number. The sample number, in turn, can be associated with increments, time intervals, time points, time steps, etc. in the test signal, from which the deflection at the time of interest (TOI) can be determined.

[0054] In some embodiments, when the acquisition window and / or TOI have been described or are otherwise known (e.g., provided by an external client communicating with the system or user), corresponding image frames can be generated by coordinating the operation of the charged particle beam instrument with the test signal (e.g., by signal locking), and the deflection parameters of the charged particle beam within a particular TOI or acquisition window can be determined therefrom. The selection of the image frames can be at least partially based on information derived from the test signal. For example, when the test signal includes a time-varying voltage component and / or a time-varying current component (e.g., described with power as a function of time), a time can be selected from the detector data at which the corresponding power or any of its components is relatively low (e.g., below a threshold), and an image generated when EMI is absent or relatively absent can be acquired, such that the image serves as a reference state. Similarly, a time can be selected from the detector data at which the corresponding power or any component is relatively high, and an image generated when EMI is present or relatively present can be acquired.

[0055] In an illustrative example technique of a circuit device, a rectangular printed circuit board of length L and width W, having a conductive copper layer covering its surface, has a current “I” flowing through the copper layer. The current is characterized by a 1-ampere square wave. In the absence of current flow on the surface of the printed circuit board, an electron beam is applied to generate a first contamination spot. In the presence of current along the length of the rectangular printed circuit board, a magnetic field is induced, which deflects the path of the incident electron beam and causes the beam spot position to shift in two spatial directions (e.g., “x-y” coordinates) on the surface of the circuit board. A second contamination spot is generated in the presence of current flow, and the distance from the first contamination spot is determined to be approximately 900 nanometers, which represents the spatial magnitude of the deflection vector. In addition, the angle relative to a reference axis is determined, thereby determining the deflection vector. Reversing the direction of the current can cause an equivalent reversal of the beam offset direction (e.g., corresponding to a reversal of the polarity of the induced magnetic field). Switching the direction of the current in a periodic manner can cause a periodic offset of the image, thereby introducing blurring artifacts in the image. During the current switching, a series of images are generated at 1-microsecond intervals.

[0056] In some embodiments, the deflection vector can be determined without generating sample surface contamination. One or more images can be generated in the absence of EMI and one or more images can also be generated in the presence of EMI, and various image processing techniques can be applied to determine the deflection vector. For example, one or more features present in the image can be tracked to generate the deflection vector. Similarly, with the aid of images generated in the absence or relative absence of EMI and images generated in the presence or relative presence of EMI, image convolution techniques can be used to generate the deflection vector. The convolution algorithm can be independent of the DUT tester loop parameters, for example when the image frequency is known. In some embodiments, a fiducial or other marker can be tracked, and its movement can be used to determine the deflection vector.

[0057] At operation 215, exemplary process 200 includes determining an acquisition window. When voltage probe data is to be acquired from a given location of the IC at a time of interest (TOI) within a test mode, image acquisition can be coordinated with the test mode so as to collect image data during an acquisition window that includes the TOI (e.g., centered on the TOI). As described in more detail in reference Figures 5A - 5C the coordination of imaging and the test mode can include strobing or otherwise pulsing the charged particle beam (e.g., Figure 1 charged particle beam 147 is blanked by beam blanker 150).

[0058] When the duration of the acquisition window is relatively short, beam offset during imaging can be reduced or negligible, and an image of the device with minimal distortion can be generated. The image generated during the acquisition window can be used to determine the beam positioning for IC voltage probing, such as defect localization as described in reference Figures 6A - 8D which is one example.

[0059] To this end, the range, amplitude, or other parameters of an unintended beam offset that occurs periodically or non-periodically over time (“t”) (e.g., according to a non-periodic test signal) can be determined in a given coordinate system (e.g., Cartesian “x” and “y” coordinates). With these parameters, the system can use the offset signals generated in the scan control logic to cancel the unintended beam offset. This process can be represented in the form of an algorithm that includes operations for determining x(t) and y(t), where 0 ≤ t ≤ T0, and x(t) and y(t) respectively describe the beam offset in the x and y scan directions at time t due to an external perturbing electromagnetic field with a period of T0. In combination with the above-described embodiments regarding the reference deflection vector, x(t) and y(t) can be components of a time-varying deflection vector r(t). x(t), y(t), r(t), etc., as in reference Figures 4A - 4Bas described in more detail below. Within the length scale of the surface being inspected, spatial variations in EMI can render x(t), y(t), and / or r(t) uniform over the surface of a region of interest (ROI) (e.g., Figure 1 region 127). Thus, beam deflection can be consistent over the image, line scan, and / or sampling locations on the ROI surface.

[0060] In some embodiments, the IC can be covered by a metal heat sink through which a hole can be formed for electron beam imaging and / or processing of the DUT. Experimental data shows that the skin depth of the metal cover plate can effectively shield high-frequency magnetic field variations from the DUT current. In this way, high-frequency effects can be limited while low-frequency magnetic field variations affect the SEM beam. Advantageously, when imaging or probing such shielded DUTs, the acquisition window including the TOI is extended relative to the unshielded DUT, thanks to the relatively reduced high-frequency EMI effects.

[0061] At operation 220, exemplary process 200 includes generating corrected detector data. The corrected detector data generally describes an image of the sample surface, frequency information (e.g., the operating frequency of a device at a given location in the DUT), waveform information (e.g., the frequency, voltage, or other signal sampled over time for a given device), device status information, timing data, etc., where the effects of EMI are attenuated or eliminated by at least partially accounting for beam drift caused by EMI. In one example, by using a deflection vector, corrections can be applied to precisely position a charged particle beam (e.g., an electron beam, an ion beam, a neutral beam, etc.) on a given structure, surface, material, etc. of the DUT for imaging and / or probing, as described above with reference to operation 205. In another example, the timing data can include the clock speed of at least a portion of the DUT.

[0062] Deflection information can be used to generate counteracting beam deflection commands, thereby reducing or eliminating known periodic disturbances in a SEM imaging system. Advantageously, the system can be configured to counteract known periodic unintended beam offsets. For a SEM image that would otherwise be blurred due to periodic beam offsets, it can be acquired without the influence of such disturbances. In the presence of external disturbances (e.g., from the device itself), the electron beam of the SEM system can still be directed to a given location or ROI on the device; such external disturbances act on the electron beam to deflect it from that location or ROI. In this way, the process of correcting the influence of EMI on a charged particle microscope can include modifying the beam deflection logic to correct x(t) and y(t) (e.g., updating the beam deflection commands) to counteract the unintended beam offsets caused by the perturbing electromagnetic fields. After counteracting the disturbances caused by the perturbing fields, the process can include generating secondary electron images, waveform data, device status data, and / or frequency data, in which the disturbance artifacts are reduced or negligible (e.g., the periodic disturbances have been removed from the measurements), and / or voltage probing can be performed without interrupting the test cycle.

[0063] Reducing beam offset artifacts can include inserting a time-varying counteracting beam offset during SEM image scanning or in the point-mode beam signal. For example, a predefined scan period can be modified to accommodate the beam offset. Advantageously, the techniques described herein counteract unintended beam offsets during image generation with relatively little or no delay, thereby generating static images that are substantially free of image blur caused by unintended beam offsets, and / or stable voltage signals that are substantially free of noise caused by time-varying EMI.

[0064] Embodiments of the present disclosure also allow a charged particle system to correct for periodic EMI perturbations from environmental and / or external sources. Examples include, but are not limited to, correcting 50 Hz and 60 Hz line interference, thereby enabling images to be acquired without considering aligning data acquisition with the 50 Hz or 60 Hz line rate. In this way, the techniques described herein can be more widely applied to periodic EMI signals that interfere with STEM, SEM, or FIB imaging. In an illustrative example, SEM image artifacts caused by an active integrated circuit (e.g., the DUT) generating a magnetic field when powered on can be corrected.

[0065] At operation 225, exemplary process 200 includes locating a defect in a sample. As described in more detail below with reference to Example 1, locating a defect in a sample can include one or more sub-operations for detecting, identifying a fault in, and / or mapping the sample to a schematic or other representation of an integrated circuit. In Example 1, this technique is referred to as "electron-induced device alteration" (EIDA), similar to laser-assisted device alteration (LADA), although with significantly improved spatial resolution. One disadvantage of the EIDA technique relative to the LADA technique is the sensitivity of electrons to EMI. Without the techniques of the present disclosure, EMI during testing would significantly eliminate the spatial resolution gain of EIDA relative to LADA.

[0067] Figures 3A - 3H is a schematic illustration of an exemplary sequence of images of a device under test (DUT) according to some embodiments of the present disclosure. These figures depict a microelectronic device 305, such as an integrated circuit device or device component, imaged using a scanning electron microscope. The exemplary sequence is labeled with timestamp references t1 to t8 and does not correspond to a specific frame rate. Instead, the illustrations making up the exemplary sequence illustrate the effects that electromagnetic (EM) interference can have on a charged particle beam (e.g., Figure 1 charged particle beam 147), at least in part by deflecting the beam, thereby causing an offset in the secondary electron image generated by the interaction of the beam with device 305. For a transient electromagnetic field, t1 to t8 can correspond to times on the order of tens of nanoseconds, hundreds of nanoseconds, microseconds, or longer, at least in part based on the dynamic characteristics of the beam, detector hardware, etc.

[0068] Figure 3A The image shown represents the undisturbed state of the DUT, labeled t1, where device 305 is located substantially at the center of the frame. The image includes a reference marker 310, which can be associated with a given pixel in the detector or other reference that is stationary relative to the surface image. The relative movement of device 305 and / or other image features relative to reference marker 310 can be used to generate offset information, as described in more detail in references Figure 2 and Figures 4A - 5C below.

[0069] Figures 3B - 3H shows the reversible effect of electromagnetic interference on secondary electron image data, where the image of device 305 is offset in the vertical ("y") and lateral ("x") dimensions in response to a transient electromagnetic field near the sample. This offset is visually apparent, as shown by the movement of device 305 partially out of the image frame in Figures 3D - 3F and then back into the frame in Figure 3G . Figure 3H shows the beam gradually returning to the undisturbed state of t1 in Figure 3A .

[0070] Periodic test signals can cause repetitive offsets such as Figures 3A - 3H shown. When the frequency of the test signal or the time between pulses in an aperiodic signal is shorter than the characteristic relaxation time of the charged particle beam (during which the charged particle beam is at least partially deflected), the detector data can include offsets and blurring of the x-y coordinates. This blurring can be at least partially attributed to the repetitive partial relaxation of the beam between the initial and deflected states. In some cases, the blurring can be at least partially based on oscillations of EMI on a time scale shorter than the detector bandwidth used to generate the image sequence.

[0071] Figures 4A - 4C is a graph of example offset data and corresponding correction vectors in a Cartesian coordinate space generated from a secondary electron image sequence (e.g., Figures 3A - 3H the image sequence). As described in more detail in reference Figures 1 - 2 and Figures 5A - 5C , by coordinating the pulses of the charged particle beam with the test signal, the offset vector can be determined and used to account for and / or correct the interference effects of EMI on the charged particle beam and detector data. Referring to the image sequence of Figures 3A - 3H , the pulsed beam method allows sampling of a single state (e.g., one of t1 to t8). The three marked states S1 to S3 correspond to Figure 4A and 4B the sampling states of the offset data used in Figure 4C to generate the corresponding correction vectors.

[0072] Generating image data can include coordinating the pixel clock with the test cycle signal (e.g., synchronizing the pixel clock with the start trigger of the test cycle). In this way, detector data can be generated on a pixel-by-pixel basis, causing the beam to dwell on one or more test mode cycles for each pixel in the frame. This method can be implemented using a pulsed beam (e.g., for sampling the test cycle) or a non-pulsed beam (e.g., generating data at the detector bandwidth). When the detector bandwidth is lower than the test mode bandwidth, this method can also be understood as sampling the test cycle signal. The advantage of this method is that it is faster than using a pulsed beam to acquire image data for individual pixels. For example, the pulsed method involves sampling a particular time point over multiple cycles of the test mode, such that generating a single image is a geometric function of the number of pixels and the number of time points. In this way, using a non-pulsed beam can pulse once per pixel per test cycle, thereby generating a full frame of images in the same length of time used to capture a single frame of images. Sub-sampling can be used to reduce the charging effect caused by the pixel dwell time. For example, two images can be generated using a non-pulsed beam starting from different pixels (e.g., adjacent pixels) with a dwell time of half the test mode length.

[0073] Figures 4A - 4B The data shown corresponds to an image sequence (sampling number) from which x-y offset information (in pixels in the x and y dimensions respectively) is generated. As referenced Figure 2 and Figures 3A - 3H As described in more detail in

[0074] Figure 4C For a correction vector for a given state Si, the individual components for each dimension (e.g., x and y in a two-dimensional Cartesian space) in an “n”-dimensional space can be used as components of an n-dimensional space vector for determination. The offset vector describes the magnitude of beam deflection in a given state, and the correction vector describes the magnitude of correction for attenuating the beam deflection, and the two are opposite (e.g., the dot product of the two vectors is zero). Thus, the correction vector can be determined by multiplying the offset vector for a given state element-wise by negative one (−1).

[0075] As referenced Figure 2 in more detail in Figure 1 the control circuit 151 of Figures 5A - 5C can be used by a control circuit (e.g.,

[0076] Figures 5A - 5C to modify the scanning pattern of a charged particle beam. In this way, the offset can be counteracted or even eliminated. As referenced Figures 5A - 5B in more detail in Figures 5A - 5BThe x-axis is represented in arbitrary units. Advantageously, the sampling time demonstrates the applicability of the disclosed technology to different time scales. For example, the typical time scale for a single cycle of the DUT test signal is in the microsecond range, millisecond range, or even up to several tens of milliseconds. In some embodiments, the lower limit of the test cycle duration corresponds to a cycle frequency of approximately 500 kHz, or approximately 2 microseconds. However, the frequency of the test pattern signal can reach the GHz range, and thus downsampling is a useful method for reducing the amount of data and addressing the bandwidth limitations of the columns, detectors, and signal processing hardware (e.g., of the instrument system 100 in Figure 1 ). In some embodiments, the provided signal processing electronics and detector components have a wide enough bandwidth to handle the unsampled data (e.g., the amount of detector data and DUT voltage data is substantially equal).

[0077] The acquisition window 500 can be defined as the window in which detector data is used to determine offset information, as described in more detail in Figures 2 - 4C . In some embodiments, the acquisition window 500 is received from the client system (e.g., as an input parameter). However, in certain cases, the position and duration of the acquisition window 500 can be determined as part of the test program, as described in more detail in Figure 2 operation 215 of Figure 9 . To identify and / or locate faults in the DUT, the acquisition window 500 can be searched, for example, by the binary tree search described in Example 2 and

[0078] The acquisition window 500 can be defined such that the measurement procedure includes one or more TOIs. In this way, the performance of the DUT can be analyzed at the TOI while the detector data and the detection performance of the charged particle beam are substantially unaffected by EMI. For example, the offset data and correction data generated for the acquisition window 500 can allow the derivation of a transfer function F(s) that can be applied to the pilot circuit and / or secondary electron detector data applicable to a given state (e.g., applying an offset to the scan signal to account for the EMI effect at a given TOI). When the transfer function is applied to data generated without scan signal correction, the detector information can be retained and the reference marker is shifted with the detector data by a state-dependent offset 510 (e.g., the offset 510 can vary with the sampling time and / or relative position in the test signal). Example 1: Electron-Induced Device Changes

[0079] Figures 6A - 6CSchematic diagram of an example technique for perturbing an integrated circuit device using a charged particle beam according to some embodiments of the present disclosure. Utilizing the spatial resolution of the charged particle beam, individual devices (such as transistors) and small groups of transistors can be modulated between a passing state and a failing state, at least in part by injecting charge from the beam. In Figure 6A a graph of pass-fail data is shown with paired axes, where the horizontal axis represents inverse frequency and the vertical axis represents the common voltage Vcc. In this example data, the inverse frequency is a dependent variable because the frequency is determined by the test signal at a given TOI. The test signal can be a transient electrical signal including at least one segment of periodic voltage alternation (such as an AC signal, a square wave signal, etc.).

[0080] When the device is irradiated with a charged particle beam (such as an electron beam), the voltage can be modulated to move the device from a passing state (hollow circle “ο”) to a failing state (solid circle “·”) or vice versa. In some cases, the frequency can also be varied, for example as a method of determining the failure point of a given device in a test cycle. The setup of the EIDA analysis involves connecting the device to a test excitation source, as described in more detail in reference Figure 1 Then the test parameters of the operating voltage and device speed are adjusted to place the DUT at the boundary state of pass-fail or fail-pass transition. It is very useful to use a tester Shmoo plot to select appropriate operating conditions. The effect of irradiating an individual device in the DUT is to change a given device from a passing state to a failing state, or from a failing state to a passing state. The result of the transition can be measured by the test system by changing the DUT output signal.

[0081] Figures 6B - 6C Shown is an example circuit diagram, which shows PMOS and NMOS transistors coupled in an example CMOS inverter logic gate circuit. The charge injection of the charged particle beam will have different effects on the NMOS and PMOS transistors. For the NMOS, the transistor will change from an “off” state to an “on” state. However, for the PMOS, the effect is to reduce the transistor threshold voltage. As the charge injection level (such as proportional to the beam current) increases, the effect on the PMOS transistor can be enhanced proportionally. This effect is to speed up or slow down the speed of the device under test. Thus making EIDA a suitable technique for determining critical timing paths in semiconductor circuits. In Figures 6B - 6C the example of Figure 6B the electron beam will cause the device to switch from a passing state to a failing state in Figure 6C and from a failing state to a passing state in

[0082] The EIDA technique relies on spatial resolution to alter the functionality of specific transistors in an integrated circuit, demonstrating the importance of: 1) the spatial location information of transistors in a sample; 2) the precise positioning of the beam spot on the transistor; 3) the stability of the beam under EMI during a test cycle. Advantageously, the techniques described herein allow EIDA to be performed with significantly reduced or negligible EMI effects.

[0083] Figure 7 is a schematic diagram of an example technique for detecting defects in a device under test (e.g., DUT 125) according to some embodiments of the present disclosure. The surface 700 of the DUT 125 has an exposed area 705 that is irradiated with a charged particle beam 147 (e.g., Figure 1 the charged particle beam 147). The beam 147 is used to perform a pass / fail analysis on the circuit 710 of the transistor device 715 that makes up the DUT 125. By irradiating multiple positions on the surface 700 in a pixel-by-pixel manner 720 (e.g., making measurements at multiple positions along a line in a line scan mode), as described in more detail in reference Figure 2 defective devices on the surface 700 can be identified and located, and this method has higher precision compared to laser-based techniques (LADA) and electronic techniques without controlled EMI effects.

[0084] In an illustrative example, a fail determination is made for the position 725 of the region 705. These determinations can be based on an analysis of the responses generated, such as a pass / fail analysis plot indicating "fail" and / or "pass", as described in more detail in reference Figure 6A -C. In device perturbation (DP), the DUT 125 can be placed at the boundary between pass and fail (e.g., at a specific temperature, voltage, or frequency), and the electron beam can be used for device perturbation. In one example of this technique, when the chip is operating under pass / fail boundary conditions, the electron beam scans the region 705 (e.g., in a raster pattern). When the electron beam hits a device and causes the chip to change from pass to fail (or vice versa), the position of the boundary device is recorded in the position data. The position data can be generated using scan information, detector data, and / or DUT output data. By correlating the position of the beam (e.g., in a scan mode) with the pass-fail data, the relative position in the scan mode (e.g., represented as a fraction of time, etc.) can be determined, and this can then be correlated with a position on the surface of the DUT 125 (e.g., a point on a reference surface or other spatial registration markers).

[0085] In some embodiments, the electron beam scans the region 705, and the resulting detector data and DUT 125 output information can be used for mapping (e.g., by electron beam signal image mapping (ESIM) or electron beam logic state imaging (ELSI)). As described in reference Figure 2As described above, such data can be generated at the per-pixel level (e.g., by stepwise scanning at multiple resolvable points), while repeating the test signal and operating the DUT 125 at the pass-fail boundary. When the test pattern cycle is coordinated with the electron pulse (e.g., using the beam blanker described in reference Figure 1 ), the DP technique can include monitoring the pass-fail and / or fail-pass transitions of the DUT 125 (e.g., as a function of the time steps in the test signal). For non-destructive measurements, the energy of the electron beam can be below the threshold energy for changing the structure of the DUT 125, which depends on the material and parameters (e.g., beam current and beam energy).

[0086] The DP method allows the SEM image to be correlated with the defect map by coordinating the beam deflection signal with the output signal of the DUT 125. A composite image can be generated to show the location of defective devices, critical timing devices, or other features relevant to failure analysis and timing analysis in the DUT 125. A resolution of approximately 5 nanometers can be achieved in the absence of EMI, while EMI significantly affects the performance of typical DP. For this reason, the techniques of the present disclosure allow DP data to be generated at the device-level resolution, as described in more detail in reference Figures 8A - 8D .

[0087] Figures 8A - 8D FIG. is a schematic diagram of an exemplary technique for locating defects in a DUT according to a map of the device under test (DUT) in some embodiments of the present disclosure. The techniques of the present disclosure include coordinating spatial information, DUT pass-fail and / or timing data, and imaging data as part of localizing failure analysis and / or timing data to one or more devices at the functional level. As described in more detail in reference Figures 2 - 5C , EMI affects the localization of spatial information generated from the interaction of the charged particle beam with the DUT device. Beam deflection and distortion affect imaging, failure analysis, timing analysis, and waveform / frequency measurements. By correcting the effect of EMI on the beam direction and / or shape, the EIDA technique can be used to correlate the DUT schematic / map information with the DP information described in reference Figure 7 .

[0088] As Figure 8AAs shown, the secondary electron image data 800 of the DUT 125 region 705 for DP analysis may include details showing one or more devices 715 (such as fins, transistors, etc.), which can be modified by a charged particle beam and / or whose operation can be changed by a charged particle beam. The secondary electron image data 800 can be associated with the DP data 805 at least in part by generating an array that correlates the DP data 805 with the secondary electron image data 800. For example, a binary mask is used, where each pixel (or subset of pixels) of the secondary electron image data 800 is associated with a true value 810 (such as value = 1) when a state transition occurs (such as from pass to fail or vice versa), and with a false value (such as value = 0) when no state transition occurs. An example of the binary mask method is as Figure 8B shown in

[0089] By coordinating (such as superimposing, convolving, etc.) the DP data 805 with the secondary electron image data 800, a measurement information map with spatial resolution can be generated. In Figure 8C the example, the region of the secondary electron image data 800 corresponding to the true value 810 of the DP data 805 is occluded as an intuitive method for identifying the defective region of the DUT 125. The secondary electron data 805 can in turn be mapped to schematic information, such as the CAD data or other design specifications of the DUT 125, so that the DP data 805 can be associated with the map of the DUT 125. An example is as Figure 8D shown in

[0090] Figure 8DThe DP data 805 is represented as a shaded cell "heat map" 815 with gray scale grading, where lighter shading 820 represents a weaker DP signal and darker shading 825 represents a stronger DP signal. By correlating the DP heat map 815 data with the accurate and precise spatial information derived from the EMI-corrected SEM beam position, one or more defective devices 830 can be identified from among the multiple devices 715 in the circuit 710. In an illustrative embodiment, cells in the heat map 815 above a given value 835 (e.g., a failure density as a function of surface area above a certain threshold) can be correlated with a failure in the DUT. Thus, localizing the defective devices 830 in the sample can include identifying devices 710 located at relatively high failure density positions. It will be appreciated that the success of the EIDA technique depends at least in part on the accurate spatial information of the secondary electron data 800, the DP data 805, and the schematic information. The accurate localization of each data type in turn depends on the accurate information of the beam spot position on the surface of the DUT 125. The variable and unpredictable effects of EMI on the beam spot position pose a significant challenge to the accurate and precise spatial correlation of different data types, making it difficult to implement EIDA and other techniques described herein. Advantageously, by identifying, tracking, accounting for, and / or correcting the effects of EMI on the beam, the challenges faced in accurately and precisely localizing or otherwise characterizing / analyzing the failures of the DUT 125 can be addressed.

[0091] In Figure 8D the illustration, the defective device 830 is represented by diagonal shading. In some embodiments, Figure 8D the schematic of Figure 8D can be represented as visualization data formatted to be presented on a computer display (e.g., using the IPC 110 and / or the client computing device 115). To this end, the processes and methods of the present disclosure include generating visualization data that includes at least some of the information shown in Example 2: Search Techniques

[0092] Figure 9 is a block diagram flowchart of an example process 900 for interrogating a device under test using a charged particle beam in accordance with some embodiments of the present disclosure. One or more operations of the exemplary process 900 can be performed by a computer system in communication with other systems, including but not limited to a characterization system, a network infrastructure, a database, and a user interface device. In some embodiments, reference is made to Figure 9At least one subgroup of the described operations is performed automatically (e.g., without human intervention) or pseudo - automatically (e.g., with human initiation or limited human intervention). In an illustrative example, the operations of defining a test window, steering a charged particle beam towards a DUT (e.g., Figure 1 DUT 125) and generating detector data can be performed automatically, and the system (e.g., Figure 1 exemplary system 100) is configured to generate visualization data displaying one or more output data forms for interpretation by a human user.

[0093] Although the exemplary process 900 is described as a series of operations, it should be understood that at least some of the operations can be omitted, repeated, parallelized, combined, and / or reordered. In some embodiments, there are additional operations before and / or after the operations of the exemplary process 900, which are omitted for clarity of illustration. For example, the operations include calibrating the electron source, aligning and correcting aberrations of the charged particle beam, introducing the DUT sample into the vacuum system, calibrating the system, etc. In another example, as part of determining one or more failure modes of a DUT (e.g., perturbation test procedure), a test pattern of a time - varying voltage signal is applied to an integrated circuit component. Referring to the exemplary process 900, the operation of the instrument can be coordinated with the operation of a test component (e.g., Figure 1 test component 130) as part of generating data describing the electrical activity of the IC component of the DUT, from which defect information and / or other information can be derived.

[0094] The constituent operations of the exemplary process 900 are described using an exemplary binary search algorithm, and a time - of - interest (TOI) can be identified through this algorithm. However, in some embodiments, other and / or alternative search algorithms are applied. For example, binary search can be supplemented and / or replaced by interpolation search, exponential search, ternary search, interval search, etc. In this way, by using the test signal output of the DUT, faults can be isolated through the TOI of a given signal. By coordinating the operation of the charged particle beam system with the DUT test signal output, the system of the present disclosure can locate defects in the DUT. Advantageously, as described in more detail in reference to the foregoing figures (e.g., Figure 2 ), techniques for tracking and / or correcting EM - induced beam deflection and / or distortion are capable of precisely and accurately correlating time information (e.g., fault timing information) with spatial information (e.g., beam spot position).

[0095] At operation 905, exemplary process 900 includes defining a test window. The test window describes a portion of a test pattern (e.g., one or more signals for DUT fault analysis) for implementing a fault search algorithm. In some cases, the test window can be defined relative to the duration of the test pattern or a portion of the test pattern. When the entire test pattern is used for the test window, the search algorithm interrogates the entire test pattern. In the case of a binary search of the entire test pattern, the test window can be defined as half of the total duration of the test pattern. For a ternary search, the test window can be defined as one-third of the total duration of the test pattern. For other search algorithms, the test window can be defined according to a search strategy that reduces the time, data volume, and / or number of operations required to identify faults in the test data.

[0096] At operation 910, exemplary process 900 includes generating test data. The test data can include detector data (e.g., secondary electron detector data) and / or DUT output data such that fault information generated during DUT testing can be coordinated with detector data generated by the charged particle beam system as part of correlating spatial information with time information and signal performance information. As part of the search algorithm of exemplary process 900 (operation 910), detector data can be generated during the test window by irradiating a portion of the DUT with a charged particle beam, e.g., as part of a device perturbation technique. For portions of the test cycle outside of the test window, the charged particle beam can be blocked or otherwise interrupted (e.g., by redirecting the beam to a beam collector). In this way, the operations of exemplary process 900 can be repeatedly performed for different test windows to search the entire test pattern.

[0097] As with the techniques described previously in Figures 1 - 8D the background, the operation of the charged particle beam system can be coordinated with the operation of the DUT test system in such a way as to address the latency and / or bandwidth limitations of one or more data generation subsystems (e.g., detectors, communications, etc.). For example, strobing techniques including phase matching can be applied to align the operation of the charged particle beam system with the DUT test system in order to reliably interrogate a particular portion of the test cycle over multiple iterations of the test cycle.

[0098] At decision 915, exemplary process 900 includes determining whether a fault has occurred during a test window. The fault may be reflected in test data, including detector data (such as voltage / contrast information in secondary electron data) and / or test output data (such as output signals using the DUT). The output of decision 915 can be used to iterate one or more operations of exemplary process 900 as part of a search tree. Advantageously, using a search tree can improve the performance of exemplary process 900 to more quickly and economically (with reference to data volume and computational resource requirements) identify and / or locate faults in the DUT.

[0099] At operation 920, exemplary process 900 includes defining a new test window. For each iteration of operation 920, the level of the test window can be incremented, which is consistent with the progression from one level of the search to the next. Regardless of the type of search algorithm implemented, each level can correspond to a shorter test window duration, so the new test window can be shorter than the previous test window defined in previous iterations of operation 905 and / or operation 920.

[0100] Additionally, or alternatively, the starting point of the test window in the test pattern can be different from the starting point of the previous test window. In an example of a binary search of the entire test pattern, level one (R1) corresponds to a test window duration of approximately half of the test pattern duration. The R1 test window can substantially cover the first half of the test pattern. Subsequent iterations of level two (R2) correspond to approximately half of the R1 test window duration. The R2 test window can be substantially continuous with the R1 test window so that the search algorithm interrogates portions of the test pattern multiple times, thereby attempting to reduce or avoid errors introduced by signal drift. For the new test window output by operation 920, additional test data can be generated by at least partially repeating operation 910. To this end, the operation of the charged particle system can be modified (such as by changing the operation of the beam control circuit) to irradiate a portion of the DUT during the new test window.

[0101] In another example, after a fault is detected, exemplary process 900 can include comparing the level of the test window with a termination level criterion of the search tree (such as a power of 2 described for a binary search) at decision 925. If the current level is below the termination level, the exemplary process can include iterating operations 920, 910, 915, and 925 to reduce the duration of the test window, thereby using more precise techniques of operations 925 and 930 during this time. In the case where a fault is detected and the level is below the termination level criterion, operation 920 can include defining a new test window within the current test window. Conversely, when no fault is detected, operation 920 can include defining a new test window outside and / or substantially continuous with the current test window.

[0102] At operation 930, exemplary process 900 includes narrowing the width of a charged particle pulse guided by a charged particle beam system. In this context, the pulse width can be correlated with the time resolution of detector data by beam parameters, including beam current. The width of the pulse corresponds to the duration of the interaction of the charged particles with the DUT, from which detector data can be generated. In this way, narrowing the pulse width can improve the time resolution of detector data, but at the expense of signal strength and integration time.

[0103] At operation 935, exemplary process 900 includes searching a test window. Searching the test window can include one or more techniques for probing individual time steps in a test pattern by scanning the test window based on at least partially incrementing a phase offset (e.g., using phase matching techniques). Each time step can correspond to the time resolution of the charged particle beam system, and the time steps are substantially continuous (e.g., with some minimal overlap to reduce / eliminate drift errors). Advantageously, using the search algorithm of exemplary process 900 can improve the performance of the fault detection technique by limiting the amount of data, increasing the search speed, and restricting the high-resolution search to test windows of shorter duration (e.g., referring to the hierarchical values described above).

[0104] In the foregoing description, various embodiments have been described. For purposes of explanation, specific configurations and details have been set forth in order to provide a thorough understanding of the embodiments. However, it will also be apparent to those skilled in the art that these embodiments can be practiced without specific details. Additionally, well-known features may have been omitted or simplified in order not to obscure the described embodiments. Although the example embodiments described herein are centered around a charged particle beam system, particularly a scanning electron microscope system, these are meant to be non-limiting illustrative embodiments. The embodiments of the present disclosure are not limited to such embodiments, but rather are intended for analytical instrument systems through which a wide variety of samples can be analyzed to determine aspects such as electron response characteristics, manufacturing quality, and / or timing performance.

[0105] Some embodiments of the present disclosure include a system having one or more data processors and / or logic circuits. In some embodiments, the system includes a non-transitory computer-readable storage medium containing instructions that, when executed on the one or more data processors and / or logic circuits, cause the one or more data processors and / or logic circuits to perform some or all of one or more of the methods disclosed herein and / or some or all of one or more processes and workflows. Some embodiments of the present disclosure include a computer program product tangibly embodied in a non-transitory machine-readable storage medium, the computer program product including instructions configured to cause one or more data processors and / or logic circuits to perform some or all of one or more of the methods disclosed herein and / or some or all of one or more processes.

[0106] The terms and expressions employed herein have been used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it should be recognized that various modifications are possible within the scope of the claims. Accordingly, it should be understood that although the present disclosure includes particular embodiments and optional features, those skilled in the art may make modifications and variations to the concepts disclosed herein, and such modifications and variations are considered to be within the scope of the appended claims.

[0107] When a term is used without a definite definition, it should be understood that, unless the term has a special and / or specific meaning in the field of charged particle microscopy systems or other related fields, the term refers to the ordinary meaning of the word. The terms “about” or “substantially” are used to indicate a deviation from the stated property, where the deviation has little or no effect on the corresponding function, property, or attribute of the described structure. In the illustrated examples where a dimensional parameter is described as “substantially equal” to another dimensional parameter, the term “substantially” is intended to reflect that the two parameters being compared may be unequal within an admissible limit (such as a manufacturing tolerance or a confidence interval inherent in the operation of the system). Similarly, in cases where a geometric parameter (such as alignment or angular orientation) is described as “about” perpendicular, “substantially” perpendicular, or “substantially” parallel, the terms “about” or “substantially” are intended to reflect that the alignment or angular orientation may differ from the precisely stated condition within an admissible limit (e.g., not precisely perpendicular). For numerical values (such as diameter, length, width, etc.), the term “about” can be understood to describe a deviation of up to ±10% from the stated value. For example, a dimension of “about 10 mm” can describe a dimension from 9 mm to 11 mm. This description provides exemplary embodiments and is not intended to limit the scope, applicability, or configuration of the present disclosure. On the contrary, the subsequent description of the exemplary embodiments will provide those skilled in the art with an enabling description for implementing various embodiments. It should be understood that various changes may be made to the function and arrangement of the elements without departing from the spirit and scope set forth in the appended claims. Specific details are given in the description to provide a thorough understanding of the embodiments. However, it should be understood that the embodiments may be practiced without these specific details. For example, specific system components, systems, processes, and other elements of the present disclosure may be shown schematically or omitted from the illustrations to avoid obscuring the embodiments with unnecessary details. In other instances, well-known circuits, processes, components, structures, and / or techniques may be shown without unnecessary detail.

Claims

1. A method for locating a defect in a sample, the method comprising: Directing a charged particle beam towards the sample; Generating drift information of the charged particle beam in one or more directions, the drift being caused at least in part by an electromagnetic field near the sample; And Using the drift information to locate a defect in the sample.

2. The method according to claim 1, wherein generating the drift information comprises: Generating a sequence of images of the sample, wherein the sequence of images comprises a plurality of sample images; And Using the sequence of images to generate the drift information.

3. The method according to claim 2, wherein at least a part of the images depicts a feature of the sample, and wherein generating the drift information comprises: Tracking the feature of the sample in the sequence of images; And Generating offset data describing the movement of the feature in the sequence of images.

4. The method according to claim 2, wherein locating the defect comprises: Using the sequence of images and pass / fail data of the sample to generate position data describing the location of the defect in the sample.

5. The method according to claim 4, wherein the drift information comprises a drift vector, and wherein locating the defect further comprises: Using the drift vector to generate a correction vector; and Using the correction vector to generate a scan signal for the charged particle beam to direct the charged particle beam to the defect location.

6. The method according to claim 4, wherein the position data describes the pass state or fail state of image pixels in the sequence of images.

7. The method according to claim 1, wherein locating a defect in the sample comprises identifying a defective device of the sample with reference to a sample map, wherein the sample map comprises a schematic description of the device, and the defect corresponds to a fault in the device.

8. The method according to claim 7, wherein the map comprises computer-aided design data describing one or more devices of an integrated circuit.

9. The method according to claim 1, wherein the electromagnetic field is generated by a transient electrical signal applied to at least a part of the sample.

10. The method according to claim 9, wherein the sample comprises a device under test (DUT), and the transient signal comprises a signal configured to operate the DUT at a pass-fail boundary as part of a device perturbation test.

11. The method according to claim 9, wherein the transient electrical signal comprises a segment of periodic voltage.

12. The method according to claim 9, wherein locating the defect comprises performing a binary search on the signal segment to identify a fault in the DUT.

13. A charged particle beam system, the system comprising: A charged particle source; A computing circuit operably coupled to the charged particle source; And One or more media storing machine-readable instructions that, when executed by the computing circuit, cause the system to perform operations, the operations comprising: Directing a charged particle beam towards the sample; Generating drift information of the charged particle beam in one or more directions, the drift being caused at least in part by an electromagnetic field near the sample; and Using the drift information to locate a defect in the sample.

14. The system according to claim 13, wherein generating the drift information comprises Generating a sequence of images of the sample, wherein the sequence of images comprises a plurality of sample images; and Using the sequence of images to generate the drift information.

15. The system according to claim 14, wherein at least a part of the images depicts a feature of the sample, and wherein generating drift information comprises: tracking the feature of the sample in the image sequence; and generating offset data that describes the movement of the feature in the image sequence.

16. The system according to claim 14, wherein locating a defect comprises: using the image sequence and the pass / fail data of the sample to generate position data that describes the location of the defect in the sample.

17. The system according to claim 16, wherein the drift information includes a drift vector, and wherein the positioning defect further includes: generating a correction vector using the drift vector; and generating a scan signal for the charged particle beam using the correction vector to direct the charged particle beam to the location of the defect.

18. The system according to claim 16, wherein the position data describes the pass state or the fail state of the image pixels in the image sequence.

19. The system according to claim 13, wherein locating a defect in the sample comprises identifying a defective device of the sample with reference to a sample map, wherein the sample map comprises a schematic description of the devices, and the defect corresponds to a malfunction in the device.

20. The system according to claim 19, wherein the map comprises computer-aided design data that describes one or more devices of an integrated circuit.

21. A method for processing beam drift artifacts in charged particle microscope images, the method comprising: determining an acquisition window that constitutes at least a part of an integrated circuit test cycle, wherein the acquisition window includes the time of interest (TOI) of the test cycle; generating detector data for a charged particle microscope system, the detector data depicting the sample and corresponding to at least a part of the acquisition window; and generating a deflection vector for the TOI, wherein the deflection vector describes the offset of the charged particle beam caused by the electromagnetic field near the charged particle beam.

22. The method according to claim 21, wherein generating the deflection vector comprises: generating an image sequence of the sample; and generating the deflection vector using the image sequence.

23. The method according to claim 22, wherein at least a part of the images depicts a feature of the sample, and wherein generating the deflection vector further comprises: tracking the feature of the sample in the image sequence; and generating offset data that describes the movement of the feature in the image sequence.

24. The method according to claim 21, further comprising identifying a location on the surface of the sample using the deflection vector.

25. The method according to claim 24, wherein identifying the location comprises modifying the detector data using the deflection vector.

26. The method according to claim 24, further comprising generating frequency information for the location on the surface using the detector data at the pixel level.

27. The method according to claim 26, wherein the frequency information describes the operating frequency of the device at the location.

28. The method according to claim 24, wherein the detector data includes data at multiple time points at the location, and the method further comprises generating waveform data using the detector data.

29. The method according to claim 28, wherein the waveform data describes the operating voltage of the device at the location at multiple time points.

30. The method according to claim 21, wherein the electromagnetic field is generated by a transient electrical signal applied to at least a part of the sample.

31. The method according to claim 30, wherein the transient electrical signal comprises a segment of periodic voltage.

32. A charged particle beam system, the system comprising: a charged particle source; a computing circuit operably coupled to the charged particle source; and one or more media storing machine-readable instructions that, when executed by the computing circuit, cause the system to perform operations, the operations comprising: determining an acquisition window that forms at least a part of an integrated circuit test cycle, wherein the acquisition window includes a time of interest (TOI) of the test cycle; generating detector data for a charged particle microscopy system, the detector data describing a sample and corresponding to at least a part of the acquisition window; and generating a deflection vector for the TOI, wherein the deflection vector describes an offset of the charged particle beam caused by an electromagnetic field near the charged particle beam.

33. The system according to claim 32, wherein generating the deflection vector comprises: generating a sequence of images of the sample; and using the sequence of images to generate the deflection vector.

34. The system according to claim 33, wherein at least a part of the images depicts a feature of the sample, and wherein generating the deflection vector further comprises: tracking the feature of the sample in the sequence of images; and generating offset data describing the movement of the feature in the sequence of images.

35. The system according to claim 32, wherein the operations further comprise identifying a location on the surface of the sample using the deflection vector.

36. The system according to claim 35, wherein identifying the location comprises modifying the detector data using the deflection vector.

37. The system according to claim 35, wherein the operations further comprise generating frequency information for the location on the surface using the detector data at a pixel level.

38. The system according to claim 37, wherein the frequency information describes an operating frequency of a device at the location.

39. The system according to claim 35, wherein the detector data includes data at a plurality of time points for the location, and the operations further comprise generating waveform data using the detector data, the waveform data describing an operating voltage of a device at the location at the plurality of time points.

40. The system according to claim 32, wherein the electromagnetic field is generated by a transient electrical signal applied to at least a part of the sample, the transient electrical signal comprising a segment of periodic voltage.

41. A method for reducing beam drift artifacts in a charged particle microscopy image, the method comprising: generating a sequence of images of a sample using a charged particle beam, wherein electromagnetic interference causes drift of the charged particle beam; using the sequence of images to generate drift information of the charged particle beam in one or more directions, the drift information describing beam drift caused by electromagnetic interference; generating beam deflection instructions describing beam drift correction; and using the beam deflection instructions to modify a scanning pattern of the charged particle microscope.

42. The method according to claim 41, wherein generating the sequence of images comprises: generating an image by incrementing a beam spot position on the surface of the sample at a pixel level; and coordinating with a transient electrical signal applied to at least a part of the sample to generate detector data for the beam spot position.

43. The method according to claim 41, wherein at least a part of the images depicts a feature of the sample, and wherein generating the drift information comprises: Tracking the features of a sample in an image sequence; and Generating offset data that describes the movement of the features in the image sequence.

44. The method according to claim 41, wherein the sample includes a device under test (DUT), and at least a portion of the DUT is subjected to a transient electrical signal.

45. The method according to claim 44, further comprising generating detector data on one transistor or a group of electrically coupled transistors of the DUT using a modified scan pattern.

46. The method according to claim 45, wherein the detector data includes secondary electron detector data, and the method further comprises generating frequency data at multiple time points of the transient electrical signal using the detector data.

47. The method according to claim 45, wherein the detector data includes secondary electron detector data, and the method further comprises generating voltage data at multiple time points of the transient electrical signal using the detector data.

48. The method according to claim 47, further comprising generating waveform data for a given integrated circuit device of the sample using the voltage data.

49. The method according to claim 41, further comprising generating timing data for an integrated circuit device of the sample using a modified beam scan pattern.

50. The method according to claim 49, wherein the timing data includes the clock speed of the integrated circuit device.

51. A charged particle beam system, the system comprising: A charged particle source; A computing circuit operably coupled to the charged particle source; and One or more media storing machine-readable instructions that, when executed by the computing circuit, cause the system to perform operations, the operations including: Generating an image sequence of a sample using a charged particle beam, wherein electromagnetic interference guides the drift of the charged particle beam; Generating drift information of the charged particle beam in one or more directions using the image sequence, the drift information describing the beam drift caused by electromagnetic interference; Generating beam deflection instructions that describe beam drift correction; and Modifying the scan pattern of the charged particle microscope using the beam deflection instructions.

52. The system according to claim 51, wherein generating the image sequence includes: At the pixel level, generating an image by incrementing the beam spot position on the sample surface; and Coordinating with a transient electrical signal applied to at least a portion of the sample to generate detector data for the beam spot position.

53. The system according to claim 51, wherein at least a portion of the image depicts a feature of the sample, and wherein generating the drift information includes: Tracking the features of the sample in the image sequence; and Generating offset data that describes the movement of the features in the image sequence.

54. The system according to claim 51, wherein the sample includes a device under test (DUT), and at least a portion of the DUT is subjected to a transient electrical signal.

55. The system according to claim 54, wherein the operations further include generating detector data on one transistor or a group of electrically coupled transistors of the DUT using a modified scan pattern.

56. The system according to claim 55, wherein the detector data includes secondary electron detector data, and the operations further include generating frequency data at multiple time points of the transient electrical signal using the detector data.

57. The system according to claim 55, wherein the detector data includes secondary electron detector data, and the operation further includes generating voltage data at a plurality of time points of the transient electrical signal using the detector data.

58. The system according to claim 57, wherein the operation further includes generating waveform data for a given integrated circuit device of the sample using the voltage data.

59. The system according to claim 51, wherein the operation further includes generating timing data for the integrated circuit device of the sample using the modified beam scan pattern.

60. The system according to claim 59, wherein the timing data describes the integrated circuit device, including the clock speed of the integrated circuit device.

Images