Device for imaging and processing a sample with a focused particle beam

Abstract

Device (400, 800) for imaging and processing a sample (190) with a focused particle beam (250), comprising: a. at least one particle source (110) configured to generate a particle beam in an ultra-high vacuum environment; b. at least one sample chamber (170) for receiving the sample (190), which is set up for imaging the sample (190) in a high vacuum environment and for processing the sample (190) in a low vacuum environment; c. at least one column (430, 830) arranged in a high vacuum environment and comprising at least one particle optical component (175) configured to form a focused particle beam (250) from the particle beam and direct it towards the sample (190); d. at least one detection unit (350) arranged within the at least one column (430, 830) and configured to detect particles (390) emanating from the sample (190); e. at least a gas conduit system (180) terminating at the outlet of the focused particle beam (250) from the column (430, 830), and configured to provide at least one process gas (870) locally on the sample (190) at a pressure such that the focused particle beam (190) can induce a particle beam-induced local chemical reaction for processing the sample (190); f. at least one pressure control unit (450, 810) through which the particle beam and the particles (390) emanating from the sample (190) pass, and which is configured to limit a pressure increase caused by processing the sample (250) at the at least one detection unit (350) to a factor of 10 or less, preferably to a factor of 5 or less, more preferably to a factor of 3 or less, and most preferably to a factor of 2 or less, without impeding access of the particles (390) emanating from the sample (190) to the at least one detection unit (350); and g. at least one element from the group: - a magnetic prism (310) or - a magnetic obstacle, wherein the at least one element is arranged in the at least one column (430, 830) and is configured to direct the particles (390) emanating from the sample (190) towards the at least one detection unit (350).

Description

1. Technical field

[0001] The present invention relates to a device for imaging and processing a sample with a focused particle beam, in particular with an electron beam. 2. State of the art

[0002] Advances in nanotechnology enable the production of components with ever smaller structural elements. Tools are needed to visualize and manipulate the chip structures of microscopic or nanoscopic components, enabling them to image and modify these structures.

[0003] Microscopes are powerful tools for imaging nanostructures. In microscopes, a particle beam typically interacts with a sample to be analyzed and / or processed. Microscopes that use massive particles, such as electrons, to scan a sample have a high diffraction-limited resolution when imaging nanostructures by scanning the sample with the particle beam, due to the small de Broglie wavelength of the particles in their particle beam. Currently, electron beams can be focused to diameters in the single-digit nanometer range.

[0004] Scanning electron microscopes typically operate in a vacuum, i.e., under high-vacuum (HV) conditions, so that their electron beam is not scattered by interactions with molecules on its way from the electron source to the sample, thus preventing the beam from spreading out. The generation of electrons for an electron beam, for example by field emission at a Schottky cathode, takes place in an ultra-high vacuum (UHV) environment.

[0005] Scanning electron microscopes (SEM), ion beam microscopes (FIB), or particle beam microscopes in general can be modified to allow for local sample manipulation in addition to imaging. For this purpose, a process gas or precursor gas is applied to the sample and activated by the focused particle beam, causing a local etching or deposition reaction. A minimum quantity or concentration of precursor gas particles on the sample surface is necessary for the focused particle beam to initiate and sustain a local chemical reaction. Typically, depending on the process gas used, the reaction rate increases with increasing local gas concentration at the reaction site, making a high local process gas concentration desirable.

[0006] However, introducing a process gas into a SEM or FIB microscope contradicts its high-voltage environment. Some of the resulting difficulties are briefly outlined below.

[0007] Some of the process gas and / or its reaction products can penetrate the optical or particle-optical system, hereinafter also referred to as the particle beam column or simply column, and cause temporary or permanent damage to various components, such as the particle beam source and / or the particle detector(s). A SEM features a particle beam column in the form of an electron beam column, or simply column.

[0008] The focused particle beam is scattered by the particles of the process gas and / or its reaction products. This causes the particle beam to spread out ("beam skirt") and reduces the lateral spatial resolution of the local chemical reaction. Furthermore, the resulting beam spread makes controlling the local chemical reaction more difficult.

[0009] Furthermore, gas discharges in areas with high electric field strengths, such as in an electro-optical lens near the sample, can damage the lens and / or its power supply.

[0010] The process gas or precursor gas can contain hazardous gases. From a safety perspective, the frequent changing of gas cylinders in a SEM or FIB system caused by a high gas flow is undesirable.

[0011] To satisfy these conflicting requirements, various solutions have been developed. For example, US 2005 / 0199086A1 describes a specially shaped gas piping system that allows a process gas to be directed symmetrically around a charged particle beam onto a sample being processed. US 6872956B2 describes a cascade pump arrangement with a backing pump, two turbomolecular pumps, and an ion getter pump. US patent 9070533B2 explains the insertion of a process bell with a gas supply line, which is placed on the sample, into the vacuum chamber of a SEM to minimize the amount of process gas required.

[0012] US patent 8,921,811 B2 describes a process cell with its own sample holder, a gas inlet and outlet, and an opening for a charged particle beam. The process cell reduces the amount of gas required for sample processing. Aside from the additional effort involved, its use is limited to small samples.

[0013] In a further embodiment for limiting the pressure increase caused by the process gas and / or its reaction products in the column, the column of the device has one or more openings to the high-vacuum environment of the microscope. To further protect the sensitive components, such as optical or particle-optical components, apertures, ion getter pumps, which are essential for maintaining a continuous high vacuum, and in particular the Schottky cathode used for field emission, a pressure-stage tube is used in the prior art. This tube divides the microscope column into two pressure-stage sections. The pressure-stage tube is a long (e.g., 15 mm) thin (e.g., 1 mm in diameter) tube that allows the collimated particle beam to pass through.The pressure-stage tube significantly reduces the flow conductance of the precursor gas particles and / or their reaction products in the molecular flow pressure range (for example, to 0.01 1 / s for nitrogen). The fraction of the process gas that can pass through the pressure-stage tube is approximately 0.1%. This remaining fraction is removed from the system by an ion getter pump located upstream of the Schottky cathode.

[0014] However, this approach still has some drawbacks. To ensure that the pressure-stage tube effectively decouples the lower part of the column from the upper part of the apparatus, it should be as long as possible and have a small diameter. These requirements, however, lead to the adsorption of particles from the process gas and / or its reaction products, or the accumulation of other contaminants from the column, which impairs the beam quality of the particle beam passing through the pressure-stage tube. For example, these can cause stigma and / or drift of the particle beam. These effects can significantly reduce the process stability of the microscope.

[0015] Due to the small diameter of the pressure-stage tube, the detector(s) used to image the sample must be mounted below it. The particles emanating from the sample, which are used for detection, have a significantly larger solid angle than the primary focused particle beam directed at the sample. The limited placement of the detector(s) within the microscope restricts the types of detectors that can be used, as they must be able to withstand high concentrations of corrosive gases. Furthermore, there is a space constraint beneath the microscope objective; its resolution can be maximized by minimizing the distance between the objective and the sample.

[0016] The impact of particles with low kinetic energy on the sample is preferred to minimize damage induced by particle bombardment and to limit the lateral extent of the local chemical reaction. However, especially at low impact energies of the particle beam particles on the sample, it is desirable to detect all particles emanating from the sample, particularly those emitted at a small polar angle relative to the primary particle beam direction. For this, however, it is necessary to spatially separate the particles traveling essentially antiparallel to the particle beam. The long, narrow pressure-stage tube described above prevents the passage of a significant portion of the particles emanating from the sample.However, shortening and / or widening the pressure stage tube leads to a pressure increase in the area above the modified pressure stage tube, in which, for example, ion getter pumps can no longer operate.

[0017] Poor high-voltage (HV) conditions prevail in the section of the column below the pressure-stage tube. This means that the particle beam is exposed to interaction with gas particles for a considerable portion of its path to the sample. The resulting scattering of the particles in the particle beam leads to a widening of its focus diameter.

[0018] German patent application DE 10 2006 043 895 A1 relates to a method for producing an object with miniaturized structures, comprising: (a) processing the object by supplying reaction gas while simultaneously directing an electron beam onto a processing area to deposit or remove material; and inspecting the object by scanning its surface with the electron beam and directing the generated backscattered and secondary electrons to an energy selector, detecting backscattered electrons passing through the energy selector, and generating an electron image of the scanned area as a function of the detected backscattered electrons; and examining the generated electron image and deciding whether to deposit or remove further material. The patent application also relates to an electron microscope and a processing system configured for carrying out the method.

[0019] Patent application US 2020 / 0363350A1 describes a technology for evaluating a property of a pattern formed within a sample from two-dimensional sample information. A pattern evaluation system described in the application includes a computer subsystem that performs a process of evaluating a property of a pattern by reading a program from memory that stores the program for evaluating a property of a pattern formed within a sample. The computer subsystem performs: a process of capturing an image of the sample; a process of extracting a signal waveform from the image; a process of calculating a feature magnitude in a predefined region of the signal waveform; a process of comparing the feature magnitude to a reference value; and a process of evaluating the pattern based on the comparison result.

[0020] The present invention therefore addresses the problem of providing a device that makes it possible to improve the imaging and processing of a sample with a focused particle beam. 3. Summary of the invention

[0021] According to one embodiment of the present invention, this problem is at least partially solved by the subject matter of independent claim 1 of the present application. Exemplary embodiments are described in the dependent claims.

[0022] In one embodiment, a device for imaging and processing a sample with a focused particle beam comprises: (a) at least one particle source configured to generate a particle beam in an ultra-high vacuum environment; (b) at least one sample chamber for receiving the sample, configured to image the sample in a high vacuum environment and to process the sample in a low vacuum environment; (c) at least one column arranged in a high vacuum environment, comprising at least one particle-optical component configured to form a focused particle beam from the particle beam and direct it onto the sample; (d) at least one detection unit arranged within the at least one column, configured to detect particles emanating from the sample;(e) at least one gas piping system terminating at the outlet of the focused particle beam from the column, and configured to supply at least one process gas locally on the sample at a pressure such that the focused particle beam can induce a particle beam-induced local chemical reaction for processing the sample; and (f) at least one pressure control unit through which the particle beam and the particles emanating from the sample pass, and configured to limit a pressure increase caused by processing the sample at the at least one detection unit to a factor of 10 or less, preferably to a factor of 5 or less, more preferably to a factor of 3 or less, and most preferably to a factor of 2 or less, without impeding access of the particles emanating from the sample to the at least one detection unit.

[0023] By continuously operating the detection unit in a (at least) high-vacuum (HV) environment, it is virtually exposed to corrosive gases that can be generated during sample processing. This elimination of the boundary condition broadens the range of detectors that can be used to image the sample. Furthermore, contamination of the detection unit, for example by adsorption of gas particles onto its detector surface, is effectively prevented. In addition, the consistent HV conditions protect the sensitive optical or particle-optical components of the column from potential contamination or even damage.

[0024] Simultaneously, a pressure control unit according to the invention allows essentially all particles emanating from the sample unrestricted access to the detection unit. This enables realistic imaging of the sample. Moreover, the beam quality of a focused particle beam is not impaired when passing through the pressure control unit. A reduction in process stability can thus be avoided. An inventive device therefore allows the processing of a sample with a high local process gas concentration and simultaneously prevents contamination or even damage to the sensitive particle-optical components, thereby enabling high-quality imaging of the sample.

[0025] A fine vacuum (FV) encompasses a pressure range from 1 mbar or 100 Pa to 10 -3 mbar or 0.1 Pa. The HV range follows the FV range and covers the pressure range of 10 -3 mbar up to 10-8 mbar. Even lower pressures occur in ultra-high vacuum (UHV), namely 10 -8 mbar up to 10 -11 mbar. Starting from atmospheric pressure or standard pressure (1013.25 mbar, 298 K), the mean free path of the gas particles in the region of the flow velocity (FV) reaches values ​​that exceed the dimensions of typical vacuum vessels, and an initially viscous gas flow transitions via a Knudsen flow into the region of molecular gas flow. In the region of molecular flow, the gas particles no longer interact with each other, but only with the walls of the vacuum vessel. In a device according to the invention, pressure conditions prevail everywhere—except at the location where the sample is processed—under which a molecular flow of gas particles develops.

[0026] The particles in a particle beam can be massless (m0 = 0 kg), such as photons, or they can be particles with a rest mass (m0 > 0 kg), such as electrons, ions, atoms, or molecules. Currently, electron beams are preferred. Due to their low rest mass, electrons have a small de Broglie wavelength, which allows them to be focused to a small spot diameter. Furthermore, irradiating a sample with electrons—unlike bombardment with ions—causes little to no damage to the sample. The term particle-optical component encompasses a component that can combine massless particles, such as photons, and / or massless particles, such as electrons, into a focused particle beam.

[0027] Particles emitted from the sample can be photons, electrons, and / or ions. Preferably, the particles emitted from the sample include secondary (SE) electrons and / or backscattered (BE) electrons. Additionally, photons emitted by the sample can be used for analysis if required.

[0028] The type of particle striking the sample can be the same as that detected by the detection unit. However, it is also possible that a first type of particle strikes the sample and a second type of particle is detected by the detection unit, where the first and second particle types are different. For example, the first type of particle can include photons and / or ions, and the second type of particle can include electrons.

[0029] At least one column can exert a pressure < 10 in the area of ​​at least one detection unit without performing a processing process. -5mbar, preferably < 3·10 -6 mbar, preferably < 10 -6 mbar, and most preferably < 3·10 -7 exhibit mbar.

[0030] A device according to the invention enables a detection unit to operate continuously in a high-voltage environment. This broadens the range of detectors that can be used in the detection unit.

[0031] In the area of ​​at least one particle source of at least one column, a pressure increase due to processing the sample < 10 -8 mbar, preferably < 5·10 -9 mbar, preferably < 10 -9 mbar, and most preferably < 10 -10 mbar remain.

[0032] At least one column can have a vacuum pump connection and / or can have at least one pressurised bypass connection to the sample chamber.

[0033] Each sample chamber has at least one connection for a vacuum pump. Without the introduction of a process gas, the vacuum pump reduces the pressure in the sample chamber to a pressure in the high-voltage range, typically around 10 bar. -5 mbar up to 10 -7 mbar. By introducing the process gas and initiating a local chemical reaction that generates reaction products of the process gas and the sample, the pressure in the sample chamber increases locally. A portion of the unreacted process gas and the generated reaction products is removed from the sample chamber via the vacuum pump connection.

[0034] The at least one vacuum pump connection and / or the at least one pressurised bypass connection to the sample chamber can be located in the lower part of the at least one column.

[0035] Since the local chemical reaction takes place in the immediate vicinity of the focused particle beam exiting the column, some of the process gas and the generated reaction products can penetrate the column of the device. The partially reactive or highly reactive particles of the process gas, as well as the partially corrosive reaction products, must be removed from the column as quickly as possible to prevent contamination or even damage to the sensitive particle-optical components located within the column. For this purpose, the column has at least one connection for a vacuum pump. This vacuum pump connection is positioned as close as possible to the particle beam exiting the column to prevent the process gas and / or its reaction products from penetrating into the upper parts of the column.Instead of using its own vacuum pump, the column's vacuum pump connection can be connected to a pressurized bypass connection of the sample chamber and pumped through the sample chamber's vacuum pump.

[0036] When the lower part of the column is mentioned below, this refers to the part of the column where the focused particle beam exits. The upper part of the column refers to the part where the particles from the particle beam source enter the column. In this convention, the direction of the particle beam runs from top to bottom through the column of the device.

[0037] The device can further include at least one turbomolecular pump that pumps the at least one sample chamber. A turbomolecular pump can also be connected to the vacuum pump port of the sample chamber. A turbomolecular pump requires pressures in the inlet area at which molecular flow of the gas particles is already present. This type of vacuum pump can generate pressures in the ultra-high vacuum (UHV) range.

[0038] The sample may include a photolithographic mask.

[0039] Currently, photolithographic masks typically have lateral dimensions of 152 mm × 152 mm. To process a photomask, the sample chamber of a device according to the invention must be large enough to accommodate it. Furthermore, the sample holder of the sample chamber has sliding elements that can move a photomask across its active area, typically 142 mm × 142 mm.

[0040] The processing of photomasks, or more generally of samples, can involve the local removal of material. In this case, the process gas includes at least one etching gas, such as xenon difluoride (XeF₂) or nitrosyl chloride (NOCl). Sample processing can also involve the local deposition of material onto the sample. For this, the process gas includes at least one precursor gas. The precursor gas can be a metal carbonyl, such as dicobalt octocarbonyl (CO₂(CO)₈) or TEOS (tetraethyl orthosilicate, C₈H₈). 20 The process gas may include O4Si). Additionally, the process gas may include an additive gas that facilitates etching of the sample and / or deposition of material onto the sample. An additive gas could, for example, include oxygen (O2) or ammonia (NH3).

[0041] The at least one gas piping system can be set up to provide the at least one process gas locally on the sample at a pressure in the range of 1 mbar to 0.001 mbar, preferably 0.6 mbar to 0.003 mbar, more preferably 0.3 mbar to 0.006 mbar, and most preferably from 0.1 mbar to 0.01 mbar.

[0042] The minimum local pressure of the process gas required for the focused particle beam to trigger a local chemical reaction depends on the specific reaction. Below a reaction-specific minimum gas concentration, the local chemical reaction ceases. Conversely, a high local pressure of the process gas accelerates the chemical reaction and is therefore advantageous from a process economy perspective. However, a high local pressure of the process gas, and consequently of its reaction products, leads to a temporary drop in the high-voltage (HV) condition within the column and the associated adverse effects of contamination or degradation of the column's sensitive optical and particle-optical components, as mentioned above.

[0043] The at least one particle source can be configured to generate a particle beam with a current of 0.1 pA to 10 nA, preferably 0.3 pA to 3 nA, more preferably 1 pA to 1 nA, and most preferably 3 pA to 0.3 nA.

[0044] The at least one particle-optical component can be configured to focus the particle beam to a spot diameter < 5 nm, preferably < 2 nm, more preferably < 1 nm, and most preferably < 0.8 nm.

[0045] The spot diameter is defined as R 50 , i.e., the radius within which 50% of the beam intensity lies. The radius of the minimum area in which a focused particle beam can process a sample by triggering a particle-beam-induced chemical reaction is approximately three to four times larger.

[0046] The at least one detection unit may comprise a scintillation counter, in particular an Everhart-Thornley detector, and / or a semiconductor detector, in particular a direct electron detector.

[0047] A device according to the invention can further comprise at least one element from the group: a magnetic prism, a magnetic baffle and a Wien filter, wherein the at least one element is arranged in the at least one column and is configured to direct the particles emanating from the sample onto the at least one detection unit.

[0048] The use of a magnetic prism to direct the particles emanating from the sample onto the detection unit enables, in particular, the detection of particles whose trajectories are essentially antiparallel to the direction of the particle beam. This portion of the particles emanating from the sample cannot be detected by an in-lens detector, for example, because it has an opening for the particle beam directed at the sample to pass through. To increase the lateral resolution of a particle beam, especially an electron beam, the landing energy of the particles—that is, the kinetic energy with which they strike the sample surface—is reduced as much as possible. This reduces the solid angle at which the particles exit the sample. However, this also means that the proportion of particles leaving the sample antiparallel to the focused particle beam increases.By enabling the detection unit of a device according to the invention to detect these particles, it increases the image quality of the images taken from the sample.

[0049] A device according to the invention can further comprise at least one electrode arranged at the outlet of the at least one column, and configured to scan the focused particle beam across the sample. The at least one electrode can comprise an octupole electrode.

[0050] Furthermore, a device according to the invention can have a charge compensation grid that is arranged at the outlet of at least one column below the gas piping system.

[0051] Furthermore, the column of a device according to the invention can include a beam guide tube (liner tube) inserted in its lower part in the region of the particle beam exit, and which is designed to prevent or reduce contamination of the optical components by the process gas and its reaction products. The beam guide tube is typically in the form of a metal tube, such as an aluminum tube, and can have a diameter of a few millimeters, for example, 4 mm to 5 mm. An electrostatic potential can be applied to the metal tube, which, in combination with the charge compensation grid, generates an electric field that reduces the kinetic energy with which the particles of the focused particle beam strike the sample.By applying a voltage to the metal tube, the particles of the focused particle beam can be slowed down to a kinetic energy in the range of 100 eV to 1200 eV, preferably 130 eV to 1000 eV, more preferably 160 eV to 800 eV, and most preferably 200 eV to 600 eV.

[0052] On the one hand, reducing the landing energy of the particles in the focused particle beam reduces the area in which the particles cause a local chemical reaction. On the other hand, reducing the kinetic energy in the region where the introduction of the process gas creates high pressure or a high particle density increases the probability of the focused particle beam particles scattering off the molecules of the process gas or its reaction products. This impairs the (lateral) control of the local chemical reaction.

[0053] Due to its large diameter, the jet guide tube only acts to a limited extent as a pressure stage; typically, it can reduce the pressure by about a factor of one.

[0054] The at least one pressure control unit can comprise at least one element from the group: a differentially pumped pressure stage located in the at least one column, and at least one aperture located above the at least one gas piping system at the outlet of the focused particle beam from the at least one column.

[0055] Both versions of the pressure control unit ensure that no significant drops in high pressure (HV) conditions occur at the detection unit's position within the column. This means that the detection unit and the other optical components in the column above it are largely protected from corrosive gas particles. Simultaneously, virtually no particles emanating from the sample are obstructed on their way to the detection unit, thereby improving the image quality of the sample through the focused particle beam.

[0056] The term "essentially" here means that individual particles are scattered at a large angle away from the beam axis of the particle beam by a charge compensation grid, so that they cannot enter the opening of the column. Furthermore, by applying a negative electrostatic potential (for example, 20 V to 200 V) to the charge compensation grid, charged particles with a kinetic energy insufficient to overcome this potential barrier can be prevented from leaving the sample.

[0057] The at least one differentially pumped pressure stage can be arranged in the direction of the particle beam in the region of a rear focal plane of an objective lens of the at least one column.

[0058] Positioning the differentially pumped pressure stage at this jet position is advantageous because the objective lens of the device creates a jet waist in the steel envelope of the particles emanating from the sample at this point. The differentially pumped pressure stage reduces the proportion of process gas and its reaction products that can enter the upper part of the column by approximately two orders of magnitude. Without the differentially pumped pressure stage, one to two percent of the gas particles entering the column from below can reach the upper part of the column. The differentially pumped pressure stage reduces this proportion to approximately 0.02%.

[0059] A device according to the invention can further comprise a turbomolecular pump for pumping a chamber of at least one differentially pumped pressure stage.

[0060] Furthermore, the column can have a vacuum pump connection located in the upper part of the column. A device according to the invention can include an ion getter pump for pumping the vacuum pump connection of the upper part of the column.

[0061] Without the differentially pumped pressure stage, the pressure at this point in the column would be too high for an ion getter pump. It would overheat and fail. Using a turbomolecular pump instead of an ion getter pump to pump the upper part of the column can achieve the required pressure level of < 10 -6 mbar in the upper part of the column cannot be ensured due to the complex column structure and problems with the vibration decoupling of the turbomolecular pump.

[0062] The chamber of the at least one differentially pumped pressure stage can have a pressure bypass connection to the sample chamber for pumping the inlet area of ​​the chamber of the at least one differentially pumped pressure stage.

[0063] This configuration makes it possible to eliminate the need for a vacuum pump and thus simplifies the design of a device according to the invention.

[0064] At least one differentially pumped pressure stage can be arranged in the direction of the particle beam upstream of the vacuum pump connection of the column.

[0065] This arrangement ensures that the majority of the process gas and its reaction products entering the column are removed from it via the column's vacuum pump connection.

[0066] The inlet area of ​​the at least one differentially pumped pressure stage can comprise a pressure stage tube with a pressure gauge of 1 mm to 3 mm, preferably 1.3 mm to 2.7 mm, more preferably 1.6 mm to 2.4 mm, and most preferably 1.9 mm to 2.1 mm, and with a length in the range of 5 mm to 25 mm, preferably 7 mm to 183 mm, more preferably 8 mm to 14 mm, and most preferably 9 mm to 11 mm.

[0067] Due to this dimensioning, the pressure stage pipe in the inlet area of ​​at least one differentially pumped pressure stage has a molecular flow conductance of 0.10 1 / s.

[0068] The outlet area of ​​the at least one differentially pumped pressure stage can have a pressure stage tube with a pressure gauge of 2 mm to 4 mm, preferably 2.3 mm to 3.7 mm, more preferably 2.6 mm to 3.4 mm, and most preferably 2.9 mm to 3.1 mm, and with a length in the range of 20 mm to 36 mm, preferably 23 mm to 33 mm, more preferably 26 mm to 30 mm, and most preferably 27 mm to 29 mm.

[0069] As a result of this dimensioning, the pressure stage pipe in the outlet region of the at least one differentially pumped pressure stage exhibits a molecular flow conductance of 0.12 1 / s. This numerical value, as well as other flow conductances specified in this application, refer to nitrogen gas. If other gases are used to investigate the pressure conditions in the device, the molecular flow conductances change proportionally to the mass number of the gas used relative to the mass number of nitrogen.

[0070] A chamber of the at least one differentially pumped pressure stage can have a height in the range of 5 mm to 30 mm, preferably 6 mm to 20 mm, more preferably 7 mm to 15 mm, and most preferably 8 mm to 12 mm. Furthermore, the chamber of the at least one differentially pumped pressure stage can have a width of 10 mm to 30 mm, preferably 13 mm to 27 mm, more preferably 16 mm to 24 mm, and most preferably 19 mm to 21 mm. In addition, the chamber can have a length of 50 mm to 200 mm, preferably 70 mm to 150 mm, more preferably 90 mm to 120 mm, and most preferably 95 mm to 110 mm.

[0071] A greater chamber height in the differentially pumped pressure stage increases its molecular flow conductance towards the pump, thus directly resulting in a larger pump cross-section. This effectively decouples the upper part of the column from the lower part in terms of pressure. On the other hand, a greater chamber height lengthens the path of the particles emanating from the sample to the detection unit, and consequently increases the beam expansion.

[0072] The chamber of the differentially pumped pressure stage can have a molecular flow conductance that is 10 times, preferably 15 times, more preferably 18 times, and most preferably 20 times greater than the molecular flow conductance of the outlet part of the differentially pumped pressure stage.

[0073] Furthermore, the chamber of at least one differentially pumped pressure stage can exhibit the potential of the metal tube that is inserted into the column at the beam outlet.

[0074] At least one aperture can have an adjustable aperture opening.

[0075] By installing an orifice directly above the gas inlet area, the volume that needs to be filled with process gas to carry out the local chemical reaction can be significantly reduced. This minimizes the required amount of process gas. This has the advantage that the gas cylinder(s) of a device according to the invention only need to be changed infrequently. Furthermore, the reduced volume to be filled with process gas shortens the time required for the necessary pressure build-up and release at the reaction site. In addition, an orifice installed directly above the gas piping system reduces the portion of the process gas and its reaction products that can enter the column of the device and must be pumped out.

[0076] Furthermore, adjusting the orifice aperture ensures that essentially all particles exiting the sample can pass through the orifice, while simultaneously minimizing the proportion of process gas and its reaction products that can enter the column. Moreover, adjusting the orifice aperture allows its size to be adapted to the distance between the orifice and the sample surface. Finally, an orifice at the column outlet reduces the path that the focused particle beam with high gas concentration must traverse. This reduces the probability of particle scattering by the gas particles and prevents an increase in the focal spot on the sample surface.

[0077] At least one aperture can include at least one piezo actuator designed to adjust the aperture opening.

[0078] The aperture can have any shape. Simple geometric shapes, such as rectangles or squares, are preferred, as their openings can be changed by simple piezo actuator arrangements.

[0079] The at least one piezo actuator can change the area of ​​the aperture by a factor of 1.1, preferably by a factor of 1.2, more preferably by a factor of 1.5, and most preferably by a factor of 2.0.

[0080] A device according to the invention can further comprise a power supply which is configured to apply an electrostatic potential to the at least one aperture.

[0081] Applying an electrostatic potential to the aperture can eliminate or at least reduce distortion of the electric field between the liner tube and a charge compensation grid. The power supply can apply an electrostatic potential in the range of 20 V to 1000 V, preferably 50 V to 600 V, more preferably 100 V to 400 V, and most preferably 150 V to 300 V to the at least one aperture.

[0082] The aperture of the aperture can be larger, preferably by a factor of 1.5, more preferably by a factor of 1.8, and most preferably by a factor of 2.0, than the distance of the aperture from a sample surface.

[0083] The aperture can cover a range from 100 µm to 3000 µm, preferably 130 µm to 2000 µm, more preferably 160 µm to 1000 µm, and most preferably 200 µm to 600 µm.

[0084] By using a factor of 2 for the ratio of orifice diameter to distance from the sample surface, the proportion of process gas entering the column can be reduced by a factor of 10 for small distances between the orifice and the sample surface. This reliably prevents an ion getter pump, which pumps the upper part of the column, from being overloaded.

[0085] The distance between the aperture and the sample surface can range from 80 µm to 1000 µm, preferably 100 µm to 800 µm, more preferably 150 µm to 700 µm, and most preferably 200 µm to 600 µm.

[0086] A charge compensation grid can have a grid opening of 10 µm to 50 µm, preferably 15 µm to 45 µm, more preferably 20 µm to 40 µm, and most preferably 25 µm to 35 µm.

[0087] The charge compensation grid can have a distance to the sample surface that is half the size of a grid opening of the charge compensation grid.

[0088] The device may further include a computer system comprising at least one non-volatile storage medium. The computer system may be configured to control the device during the imaging of the sample and / or during the processing of the sample.

[0089] A computer program can include instructions that cause a computer system to image and / or process a sample. In particular, the computer system can include instructions that set a working distance of the column from a sample and / or control the adjustment of the aperture. 4. Description of the drawings

[0090] In the following detailed description, currently preferred embodiments of the invention are described with reference to the drawings, wherein Fig. 1 presents a schematic section through a device for imaging with an electron beam and for processing a sample with the electron beam and at least one process gas according to the prior art; Fig. 2 an enlarged section of a part of the column of the device Fig. 1 reproduces; Fig. 3 presents a schematic section through a device in which a detector or detection unit is arranged in the upper part of the column downstream of a pressure stage, i.e. upstream; Fig. 4 the device of Fig. 3 illustrated after installation of a differentially pumped pressure stage; Fig. 5 a simulation of a beam envelope of the electrons emerging from a sample along the column of the device of the Fig. 4 presented, which is generated by an electron beam striking the sample surface with low kinetic energy; Fig. Figure 6 in the upper part of the diagram schematically shows the molecular gas flow through the pressure stage of the device. Fig. Figure 3 illustrates, and the lower part of the image schematically shows, the molecular gas flows into the differentially pumped pressure stage of the device. Fig. 4 illustrated in and out of this; Fig. 7 in the upper part of the image, the lower part of the column of the Fig. 3 reproduced and in the lower part of the image shows the beam fanning of the electrons of a focused particle beam as a result of scattering by the particles of a process gas introduced into the sample chamber; and Fig. 8 in the upper section, the installation of a baffle in the outlet of the column of the device Fig. Figure 2 shows and illustrates in the lower part of the image the resulting reduction in the beam dispersion of the primary focused particle beam. 5. Detailed description of preferred embodiments

[0091] Preferred embodiments of devices according to the invention are described below. Two embodiments of a device according to the invention are explained in detail using the example of a scanning electron microscope (SEM). However, devices according to the invention are not limited to the use of a massive particle beam in the form of an electron beam. Rather, they can be used for any particle beam that uses particles in the form of bosons or fermions. Furthermore, the use of devices according to the invention is explained using the example of imaging and processing a photolithographic mask. This, however, does not represent a limitation. Rather, devices according to the invention can be used for imaging and processing any sample.For example, the devices described in this application can be used to image and modify chip structures or semiconductor structures on wafers, MEMS (micro electromechaical systems) and / or PICs (photonic integrated circuits) using a particle beam or a particle beam-induced processing process.

[0092] The Fig. Figure 1 shows a schematic section through a device 100 for imaging and processing a sample 190 with a focused electron beam as an example of a focused particle beam. The particle source 110 or electron source 110 has a Schottky field emitter 105 from which, in a strong electric field in an ultra-high vacuum (UHV) environment (1·10 -10 mbar ≤ p ≤·3·10 -9 mbar) electrons are released. The electron source 110 has a vacuum connection 115, to which typically an ion getter pump (in the Fig. (1 not shown) can be connected. The electron beam passes through the aperture 120 and enters the upper part 125 of the column 130 of the device 100.

[0093] In the Fig. As in all subsequent figures, the particle source 110 is flanged to the column 130 at the upper end of the upper part 125. The particle beam enters the column 130 at the upper end and exits it at its lower outlet 187. In this convention, the direction of the particle beam is from top to bottom. Upstream means a direction opposite to the beam direction, i.e., upwards, and downstream denotes the direction of the particle beam, i.e., downwards.

[0094] The upper part 125 and the lower part 135 of the column 130 contain in the example the Fig. 1. The electron-optical components for focusing and directing the electron beam onto the sample 190. The upper part 125 and the lower part 135 of the column 130 are separated from each other by a pressure-stage tube 140. The pressure-stage tube 140 is designed such that the electron beam can pass through it, but the upper part 125 of the column 130 is largely shielded from pressure variations in the lower part 135 of the column 130. A long, thin pressure-stage tube 140 is advantageous for this purpose. Gas particles can adsorb on the inner wall of a long, thin pressure-stage tube 140. This can degrade the beam quality of the electron beam passing through the pressure-stage tube 140. This can have negative effects on the imaging performance of the electron beam of the device 100. In addition, the distorted beam profile of the focused electron beam can negatively affect the quality of a processing operation.

[0095] In the upper part 125 of column 130, an ion getter pump, connected to the vacuum port 145 and located in the Fig. 1 is not shown, the pressure is approximately 1·10 -7 mbar (typically 5·10 -8 mbar up to 5 x 10 -7 reduced by mbar.

[0096] However, the opening of the pressure stage tube 140 is not large enough for the majority of the electrons 192 emerging from the sample 190 to pass through it. The device 100 of Fig. 1 therefore has two so-called in-lens detectors 150, 160, which are arranged in the lower part 135 of the column 130. However, varying pressure conditions prevail in the column 130 below the pressure stage tube 140. These are caused by the introduction of a process gas through the gas piping system 180 into the process chamber 170. Without the introduction of a process gas through the gas piping system 180, the pressure level in the lower part 135 of the column 130 is in the range of 10 -5 mbar up to 10 -6 mbar, i.e., exhibits stable HV conditions.

[0097] During a processing operation in which the focused electron beam initiates a local chemical reaction on the surface 197 of a sample 190, the pressure level in the lower part 135 of the column 130 rises to 10 -2 mbar up to 10 -4mbar. Detectors 150 and 160 are thus exposed to a considerable gas concentration. The gases in the lower part 135 of column 130 comprise a significant amount of unreacted process gas and reaction products of the process gas. These typically reactive gases have a high corrosive potential, which can contaminate or damage detectors 150 and 160, as well as the other electro-optical components housed in the lower part 135 of column 130, such as the objective lens 175.

[0098] The lower part 135 of column 130 has a vacuum pump connection 155. The lower part 135 of column 130 can be pumped via this connection, for example using a turbomolecular pump (in the Fig. (1a not shown), evacuate. Alternatively, as in the Fig. Figure 1 schematically shows that the vacuum pump connection 155 forms a pressurized bypass connection to the sample chamber 170. The dashed horizontal line 172 illustrates the upper end of the sample chamber 170. The sample chamber 170 itself has a vacuum pump connection 165 through which it can be pumped. For this purpose, for example, a turbomolecular pump (in the Fig. 1 not reproduced) are used.

[0099] The lower part 135 of the column 130 of the device 100 has, in addition to the detectors 150 and 160, at least one electro-optical objective 175 or a lens 175 that focuses the electron beam onto the sample 190. The part of the column 130 in which the lens 175 is arranged is shown in diagram 200 of the Fig. Figure 2 is shown again enlarged. At the outlet of the focused electron beam 250 – as an example of a focused particle beam 250 – from the column 130, the column has an octupole electrode 185, which allows the focused electron beam 250 to be rasterized or scanned across the sample 190. The gas line system 180 also terminates in the octupole electrode 185. The gas line system 180 supplies the process gas in the area of ​​the point of impact 260 of the focused electron beam 250 on the sample 190.

[0100] By opening the gas pipeline system 180, the pressure in the area of ​​carrying out a local chemical reaction can be reduced to 10 -1 mbar up to 10 -3 mbar, i.e., rising into the range of the FV. The schematic representation of the Fig. Figure 2 illustrates that the majority of the unreacted particles of the process gas, as well as the locally generated reaction products, can penetrate into column 130 of the device 100.

[0101] To ensure the imaging quality of the electron beam, particularly at low electron landing energies, the components in column 130 along the electron beam path are at an electrical potential that corresponds to that of the electrons in column 130 up to its outlet 187. To guarantee this requirement even at the narrow outlet 187 of the electron beam from column 130, a replaceable metal tube 220 is typically inserted into the outlet 187 of column 130. This is referred to as a liner tube 220 in English and a beam guide tube 220 in German. The beam guide tube 220 is generally made of a non-magnetic and corrosion-resistant material and has a diameter of 4 mm to 5 mm.The beam guide tube 220 limits the diameter of the column outlet 187 in a defined manner and thereby has the positive side effect of counteracting contamination of the electro-optical, generally particle-optical components, which are housed in the lower part 135 of the column 130.

[0102] The column 130 of the exemplary device 100 of the Fig. 1 and Fig. Figure 2 features a charge compensation grid 195, whose function is to minimize the effects of electrostatic charging of the sample 190. To meet this requirement, the distance between the charge compensation grid 195 and the sample surface 197 is chosen to be very small. Typical values ​​for this distance are < 70 µm. A sample 190 in the form of an electrical insulator, such as a photolithographic mask 190, can become charged as a result of irradiation with a focused electron beam 250. A photolithographic mask often has an electrically insulating quartz substrate.

[0103] The device 300 of the Fig. Figure 3 schematically shows a column 330 in which a detection unit 350 or a detector 350 can operate at a lower pressure level compared to the detectors 150, 160 of the Fig. 1. The detector 350 can be a scintillation detector 350, such as an Everhart-Thornley detector 350, and / or comprise a semiconductor detector, in particular a direct electron detector. In the Fig. 3 and in the following figures, the sample chamber 170 of the Fig. 1 is not shown for the sake of simplicity.

[0104] The column 330 has a pressure stage 370 with a pressure stage tube 380 at the transition from its lower part 335 to its upper part 325. In order to protect the detection unit 350 from exposure to reactive gases, it is – unlike the detectors 150 and 160 of the device 100 – Fig. 1 - installed upstream in the upper part 325 of column 330. The opening of the pressure stage tube 380 of pressure stage 370 is dimensioned such that essentially all electrons 390 emanating from sample 190 can reach the detector 350. The molecular flow conductance of pressure stage 370 is in the range of 0.12 1 / s (1 / s stands for liters per second) and is thus about thirteen times greater than the molecular flow conductance of pressure stage tube 140 of column 130 of device 100.

[0105] The dotted line indicates in the Fig. Figure 3 illustrates the beam envelope 395 of the electron beam 390 emanating from the sample 190, or the electron distribution 395 generated by the sample. In the Fig. In the exemplary device 300 shown, the pressure stage tube 380 has a diameter of 3 mm and a length of 28 mm. The pressure stage 380 does not impede either the primary electron beam 250 emanating from the electron source 110 and focused on the sample 190, nor the electron distribution 395 generated by it on its way to the detection unit 350.

[0106] However, during a processing process of sample 190, pressure stage 370 allows dips in the HV environment in the overlying part 325 of column 330 to pressure levels > 10 -5 mbar to which an ion getter pump connected to the vacuum pump port 345 (in the Fig. (3 not shown) cannot process. The vacuum pump connection 345 serves to evacuate the upper part 325 of the column 330 and corresponds to the vacuum pump connection 145 of the device 100. Fig. 1.

[0107] If a vacuum pump capable of handling these pressure levels, such as a turbomolecular pump, is connected to the vacuum port 345 instead of an ion getter pump, the required residual gas pressure level < 10 is not achieved in the upper part 325 of the column 330. -7 mbar is reached. This pressure is required, firstly, to reliably protect the electro-optical and particle-optical components in the upper part 325 of column 330, and secondly, to adjust the pressure level in the upper part 325 of column 330 to the UHV level of the electron source 110. During the imaging of sample 190, detector 350 "sees" a significantly lower pressure level than detectors 150 and 160; however, this pressure drops to an unacceptable level during processing. More than 2% of the reactive process gas or its reaction products can overcome pressure stage 370 and penetrate into the upper part 325 of column 330.

[0108] The device 400 of the Fig. 4 corresponds to device 300 of the Fig. 3, into which, however, a differentially pumped pressure stage 450 is additionally installed. This is installed downstream below the pressure stage 370 of the device 300 in the column 330. This means that the outlet area 370 of the differentially pumped pressure stage 450 corresponds to the pressure stage 370 of the device 300. Fig. 3. The pressure stage pipe 380 of the differentially pumped pressure stage 450 is the pressure stage pipe 380 of the pressure stage 370 of the Fig. 3.

[0109] The differentially pumped pressure stage 450 has an inlet region 470 formed by a pressure stage tube 480. The opening of the pressure stage tube 480 can be dimensioned somewhat smaller than that of the pressure stage tube 380, since the beam envelope 395 of the electrons 390 exiting the sample widens in the opposite direction to the beam direction of the primary focused particle beam 250. This will be shown below with reference to the Fig. 5 explained in detail. The one in the Fig. The exemplary differential pressure stage 450 shown here has a pressure stage tube 480 with an opening diameter of 2 mm and a length of 10 mm. The pressure stage tube 480 thus has a molecular flow conductance of approximately 0.1 l / s.

[0110] Two conflicting requirements are placed on the chamber 410 of the differentially pumped pressure stage 450. Firstly, it should have the highest possible molecular flow conductance to decouple the inlet section 470 as effectively as possible from the outlet section 370 of the differentially pumped pressure stage 450, ensuring that a large proportion of the gas particles flowing into chamber 410 via the inlet section 470 exits chamber 410 through its vacuum pump connection 465. For this purpose, the height of chamber 410 should be as large as possible. However, this requirement lengthens the path of the electrons 390 exiting sample 190 to detector 350 and therefore spreads their beam envelope 395. Simulations have shown a good compromise with a chamber height of approximately 10 mm. Furthermore, chamber 410 has a width of 20 mm and a length of 100 mm.

[0111] The vacuum pump port 465 of the differentially pumped pressure stage 450 can, for example, be pumped with a turbomolecular pump. If, for example, a turbomolecular pump with a pumping capacity of 10 l / s is connected to the vacuum port 465, the proportion of the process gas and its reaction products that can penetrate through the differentially pumped pressure stage 450 into the upper part 425 of the column 430 is in the range of 0.02%. This remaining proportion can easily be removed via the vacuum pump port 345 of the upper part 425 of the column. This means that the pressure level in the upper part 425 of the column 430 is so low that the vacuum pump port 345 can be safely pumped, for example, by an ion getter pump.

[0112] After the installation of the differentially pumped pressure stage 450, the column 430 of the device 400 fulfills the two opposing requirements. The sensitive electro-optical components of the column 430, and in particular the detection unit 350, are reliably protected against vacuum dips and thus against contamination. The electrons 390 leaving the sample 190 are not obstructed on their way to the detection unit 350.

[0113] Diagram 500 of the Fig. Figure 5 presents a simulation of the variation in the diameter of the beam envelope 395 of the electrons 390 emanating from the sample 190 (x-axis) as a function of the distance from the sample surface (y-axis). The electrons 390 emanating from the sample 190 are generated by a primary focused electron beam 250, whose electrons have a low kinetic energy of approximately 200 eV on the sample surface 197. The beam envelope 395 exhibits a beam waist 550 with a diameter of approximately 1 mm about 65 mm above the sample surface 197. This waist is caused by the imaging effect of the electro-optical objective 175. The waist 550 of the beam envelope 395 marks the rear focal plane of the objective 175.

[0114] Within a distance of approximately 25 mm to 105 mm, the diameter of the beam envelope 395 is less than 2 mm, illustrated by the dashed line 510. At this distance from the sample surface 197, the pressure stage tube 480 of the differentially pumped pressure stage 450 allows unrestricted entry of the electrons 390 emanating from the sample 190 into the differentially pumped pressure stage 450. Up to a distance of approximately 140 mm from the sample surface 197, the diameter of the beam envelope 395 is less than 3 mm. Fig. This is symbolized by the dashed line 520. This means that the electrons 390 emanating from the sample 190 can pass unimpeded through the pressure stage tube 380 of the differentially pumped pressure stage 450 if its distance to the sample surface 197 is less than 140 mm. Based on simulations of the paths of the electrons 390 emanating from the sample 190, the design of the differentially pumped pressure stage 450 can be optimized and its best possible placement in the column 430 of the device 400 can be determined.

[0115] The upper part of image 600 of the Fig. Figure 6 illustrates the molecular flow of the gas particles within column 330 of the Fig. 3 of its lower part 335 into its upper part 325 via the pressure stage 370. The following are some basic equations for estimating molecular flows across the pressure stage 370 and into or out of the differentially pumped pressure stage 450. The molecular flow conductance for nitrogen, given in l / s (liters per second), in a tube with a length L [cm] and a diameter d [cm], is given under the condition of molecular flow by the equation: C = 12.1·d 3 / L (see, for example: Handbook of Vacuum Technology, ISBN 978-3-658-13386-5).

[0116] The molecular flow Q (with the unit mbar·l / s) is driven by a pressure difference or pressure gradient Δp. The proportionality constant is the molecular flow conductance C introduced above: Q = C · Δp. This equation is equivalent to the fundamental electrical equation: I = (1 / R) · U.

[0117] The molecular flow conductance for the pressure stage pipe 380 of pressure stage 370 was calculated above to be C 380 A pressure difference of 10 is given as 0.12 1 / s. -3 The pressure in mbar between the lower part 335 and the upper part 325 of the column 330 results in a molecular current or molecular flow of Q. 380 = 0.12 l / s · 10 -3 mbar = 1.2·10 -4 mbar·l / s.

[0118] The lower part of image 650 of the Fig. Figure 6 illustrates the molecular flows in the differentially pumped pressure stage 450. A molecular gas flow Q flows through the pressure stage pipe 480 of the differentially pumped pressure stage 450. 480 into chamber 410 of the differentially pumped pressure stage 450. There it branches into the molecular streams or molecular flows Q. 465 and Q 380, which exit chamber 410 via the vacuum port 465 and the pressure stage tube 380. The molecular flux 465 or molecular flow 465 is proportional to the pumping speed S (in l / s) of a vacuum pump at a pressure p in chamber 410 of the differentially pumped pressure stage 450: Q 465 = S · p 410 .

[0119] The molecular flow conductance of the pressure stage pipe 480 of the differentially pumped pressure stage 450 was calculated above with C 480 The flow rate is specified as 0.1 l / s. At a pressure difference of 10 -3 A molecular flow Q results between the lower part 435 of the column 430 and the chamber 410 at a pressure of mbar. 480 from the lower part 435 of the column 430 into the chamber 410 of the differentially pumped pressure stage 450: Q 480 = 0.1 l / s · 10 -3 mbar = 10 -4mbar·l / s. With a pumping speed of S = 10 l / s of the vacuum pump connected to the vacuum port 465 of the differentially pumped pressure stage 450, an approximate pressure of P is established in its chamber 410: 410 = Q 480 / S = 10 -4 mbar·l / s / (10 l / s) = 10 -5 mbar.

[0120] As stated above, the pressure stage pipe 380 at the outlet of the differentially pumped pressure stage 450 has a molecular flow conductance of C 380 = 0.12 1 / s. A pressure difference of 10 -5 mbar drives a molecular flow Q 480 = 0.12 l / s · 10 -5 mbar = 1.2·10 -6 mbar·l / s. This means that the differentially pumped pressure stage 450 reduces the molecular flow in the upper part 425 of the column 430 by two orders of magnitude Q. 380 / Q 480 = 1.2·10 -4 mbar·l / s / (1.2·10 -6mbar·l / s) = 100. Only about one percent of the gas particles flowing into chamber 410 of the differentially pumped pressure stage 450 leave it via the outlet area 370.

[0121] The electrical equivalent of the differentially pumped pressure stage 450 is a voltage divider. By changing the load resistance R L (corresponding to 1 / C 463 ) small compared to resistance R2 (corresponding to 1 / C 480 When this is done, the current (the molecular gas flow) flows largely through the load resistance R. L and no longer via resistor R2.

[0122] The upper part of image 700 of the Fig. 7 reproduces again the lower part 335 of the column 330 of the device 300 of the Fig. 3. The section of the dotted circle 750 is shown in the lower part of the diagram 755. Fig. 7 shown again enlarged. As already mentioned in the context of the Fig. As explained in Figure 3, the introduction of a process gas 770 through the gas piping system 180 onto the sample surface 197 leads to a significant pressure increase in the outlet region of the focused particle beam 250 or electron beam 250 from the column 330. The locally increased concentration of gas particles results in the scattering of the electrons from the focused electron beam 250 by the gas particles of the process gas 770, causing an undesired beam spread, which is illustrated by the cone 760 in Figure 755. Depending on the set gas pressure and the kinetic energy of the focused particle beam 250, up to 50% of the electrons can be scattered once or several times in the outlet region 750. This scattering increases the lateral dimensions of a local chemical reaction initiated by the focused electron beam 250 on the process gas 770 in a largely unpredictable manner.The threshold at which the concentration of the process gas and the concentration of electrons are no longer sufficient to maintain a local chemical reaction depends on a number of parameters. Under these conditions, the local processing operation carried out on sample 190 is difficult to control.

[0123] The molecular gas flow entering column 335 is proportional to the process gas flow 780 of the gas piping system 180. For column 330 of the in the Fig. In the device 300 shown, approximately 1.5% of the particles of the process gas 770 or the reaction products of the process gas 770 can penetrate into the lower part 335 of the column 330.

[0124] Figure 755 further shows the charge compensation grid 195 of the Fig. 1. In the operation of the device 300, this is typically located a few tens of micrometers above the sample surface 197. The charge compensation grid 195 can be grounded to minimize the effects of charging the sample surface 197 by the focused electron beam 250. Furthermore, the potential of the charge compensation grid 195, in combination with the potential of the beam guide tube 220, which has the potential of the electro-optical objective 175, serves to generate an electric field (in the Fig. 7 not shown), in which the focused electron beam 250 is slowed down to a predetermined landing energy.

[0125] Furthermore, a voltage U2 in the range of -20 V to -200 V can be applied to the charge compensation grid 195. The electric field thereby generated between the charge compensation grid 195 and the sample surface represents an energy barrier or energy filter for the electrons exiting the sample 190. Only electrons with a kinetic energy higher than the energy barrier can leave the sample 190 and enter the column 330.

[0126] The upper part of image 805 of the Fig. 8 presents the lower part 335 of column 300 of the Fig. 3. An additional aperture 810 is inserted into the outlet of column 330 in the area of ​​the octupole electrode 185. The area 850 at the outlet of column 830 is shown again enlarged in the lower part of the diagram 855.

[0127] The orifice 810 is inserted into the electrode 185 above the end of the gas piping system 180. The orifice 810 and the sample 190 form a type of pressure chamber without side walls. The two-sided limitation of the volume of the process gas 870 optimizes the required quantity of the process gas 870. To allow the focused electron beam 250 to strike the sample surface 197 and to enable the electrons 390 exiting the sample 190 to enter the lower part 835 of the column 830, the orifice 810 has an opening 820 with a diameter 825. The opening 820 of the orifice 810 determines the largest angle at which the electrons 390 exiting the sample 190 can leave the sample 190.

[0128] The aperture diameter 825 of the aperture 810 can, for example, cover a range from 200 µm to 2000 µm.

[0129] The aperture diameter of 825 or the opening diameter of 825 can be adjusted using one or more piezo actuators located in the Fig. The distance 840 between the sample surface 197 and the aperture 810 can be varied. Depending on the distance 840 between the sample surface 197 and the aperture 810, the diameter 825 of the aperture 810 can be chosen to be just large enough so that the entry of the electrons 390 emitted from the sample 190 into the column 830 is not impeded. This simultaneously ensures that the proportion of process gas flow and molecular flow of reaction products, which would otherwise also enter the column 830, is minimized. If the distance 840 is chosen to be half the diameter of the aperture 825, the aperture 810 has an opening angle of 90° for the electrons 390.

[0130] The distance 830 of the aperture 810 from the sample surface 197 is typically less than or equal to about one millimeter. Preferred distances 840 are currently in the range of 100 µm to 300 µm. By raising or lowering the sample 190 using a sample holder (in the Fig. (8 not shown) to a numerical value within a working distance of the device 800, the distance 840 between aperture 810 and the sample surface 197 can be set.

[0131] With an aperture diameter 825 of 2 mm and a distance 840 of 1 mm, approximately 1.5% of the process gas 870 or its reaction products penetrate the lower part 835 of the column 830. With an aperture 825 of 400 µm and a distance of 200 µm, this proportion decreases by an order of magnitude to 0.15%. Halving these two dimensions further reduces the proportion of gas penetrating the column 830 to approximately 0.05%.

[0132] Furthermore, the aperture 810 effectively shortens the high-pressure gas path traveled by the electrons of the focused electron beam 250 to the sample surface 197, to distances < 1 mm. Under these conditions, the focused electron beam 250 experiences only very slight beam spreading 860.

[0133] Furthermore, an electrostatic potential can be applied to the aperture 810, which in the Fig. Figure 8 is illustrated by U1. This largely prevents distortion of the electric field between the beam guide tube 220 and the charge compensation grid 195 by the aperture 810.

[0134] The orifice 810, with appropriate dimensioning of its distance 840 to the sample surface 197 and its aperture 825, effectively prevents variations in the pressure level in the upper part 825 of the column 830 due to a processing operation of the sample 190. Thus, the orifice 810, in combination with the pressure stage 370 of the column 830, effectively protects the sensitive electro-optical components in the upper part 825 of the column 830 from the influence of reactive particles of the process gas 870 and its reaction products, for example, the detector 350. By minimizing the proportion of the process gas 870 and its reaction products, the orifice 810 also prevents contamination and / or damage to the components located in the lower part 835 of the column 830, such as the lens 175.

[0135] Naturally, it is also possible to combine the aperture 810 of the device 800 with the differentially pumped pressure stage 450 of the device 400.

Claims

[1] Device (400, 800) for imaging and processing a sample (190) with a focused particle beam (250), comprising: a. at least one particle source (110) configured to generate a particle beam in an ultra-high vacuum environment; b. at least one sample chamber (170) for receiving the sample (190), which is set up for imaging the sample (190) in a high vacuum environment and for processing the sample (190) in a low vacuum environment; c. at least one column (430, 830) arranged in a high vacuum environment and comprising at least one particle optical component (175) configured to form a focused particle beam (250) from the particle beam and direct it towards the sample (190); d. at least one detection unit (350) arranged within the at least one column (430, 830) and configured to detect particles (390) emanating from the sample (190); e. at least a gas conduit system (180) terminating at the outlet of the focused particle beam (250) from the column (430, 830), and configured to provide at least one process gas (870) locally on the sample (190) at a pressure such that the focused particle beam (190) can induce a particle beam-induced local chemical reaction for processing the sample (190); f. at least one pressure control unit (450, 810) through which the particle beam and the particles (390) emanating from the sample (190) pass, and which is configured to limit a pressure increase caused by processing the sample (250) at the at least one detection unit (350) to a factor of 10 or less, preferably to a factor of 5 or less, more preferably to a factor of 3 or less, and most preferably to a factor of 2 or less, without impeding access of the particles (390) emanating from the sample (190) to the at least one detection unit (350); and g. at least one element from the group: - a magnetic prism (310) or - a magnetic obstacle, wherein the at least one element is arranged in the at least one column (430, 830) and is configured to direct the particles (390) emanating from the sample (190) towards the at least one detection unit (350). [2] Device (400, 800) according to claim 1, wherein the at least one column (430, 830) has a vacuum pump connection (355) and / or has at least a pressure bypass connection (355) to the sample chamber (170). [3] Device (400, 800) according to one of the preceding claims, wherein the sample (190) comprises a photolithographic mask. [4] Device (400, 800) according to one of the preceding claims, wherein the at least one gas piping system (180) is arranged to provide at least one process gas (870) locally on the sample (190) at a pressure in the range of 1 mbar to 0.001 mbar, preferably 0.6 mbar to 0.003 mbar, more preferably 0.3 mbar to 0.006 mbar, and most preferably 0.1 mbar to 0.01 mbar. [5] Device (400, 800) according to one of the preceding claims, wherein the at least one detection unit (350) comprises a scintillation counter, in particular an Everhart-Thornley detector, and / or a semiconductor detector, in particular a direct electron detector. [6] Device (400, 800) according to one of the preceding claims, further comprising at least one Wien filter arranged in the at least one column (430, 830) and configured to direct the particles (390) emanating from the sample (190) onto the at least one detection unit (350). [7] Device (400, 800) for imaging and processing a sample (190) with a focused particle beam (250) comprising: a. at least one particle source (110) configured to generate a particle beam in an ultra-high vacuum environment; b. at least one sample chamber (170) for receiving the sample (190), which is set up for imaging the sample (190) in a high vacuum environment and for processing the sample (190) in a low vacuum environment; c. at least one column (430, 830) arranged in a high vacuum environment and comprising at least one particle optical component (175) configured to form a focused particle beam (250) from the particle beam and direct it towards the sample (190); d. at least one detection unit (350) arranged within the at least one column (430, 830) and configured to detect particles (390) emanating from the sample (190); e. at least a gas piping system (180) terminating at the outlet of the focused particle beam (250) from the column (430, 830), and configured to supply at least one process gas (870) locally on the sample (190) at a pressure such that the focused particle beam (190) can induce a particle beam-induced local chemical reaction for processing the sample (190); and f. at least one pressure control unit (450, 810) through which the particle beam and the particles (390) emanating from the sample (190) pass, and which is configured to limit a pressure increase caused by the processing of the sample (250) at the at least one detection unit (350) to a factor of 10 or less, preferably to a factor of 5 or less, more preferably to a factor of 3 or less, and most preferably to a factor of 2 or less, without impeding access of the particles (390) emanating from the sample (190) to the at least one detection unit (350), wherein the at least one pressure control unit (450, 810) comprises at least one element from the group: - at least one differentially pumped pressure stage (450) arranged in the at least one column (430, 830), wherein the at least one differentially pumped pressure stage (450) is arranged in the beam direction of the particle beam (250) in the region of a rear focal plane of an objective lens (175) of the at least one column (430, 830), or - at least one aperture (810) which is arranged above the at least one gas conduit system (180) at the outlet of the focused particle beam (250) from the at least one column (430, 830). [8] Device (400, 800) according to one of the preceding claims, wherein the at least one column (430, 830) in the area of ​​the at least one detection unit (350)) exerts a pressure < 10 -5 mbar, preferably < 3·10 -6 mbar, preferably < 10 -6 mbar, and most preferably < 3·10 -7 has mbar. [9] Device (400, 800) according to claim 7 or 8, further comprising a turbomolecular pump for pumping a vacuum port (465) of a chamber (410) of the at least one differentially pumped pressure stage (450). [10] Device (400, 800) according to claim 7 or 8, wherein the chamber (410) of the at least one differentially pumped pressure stage (450) has a pressure bypass connection (465) to the sample chamber (170) for pumping the chamber (410) of the at least one differentially pumped pressure stage (450). [11] Device (400, 800) according to one of claims 7-10, wherein the at least one differentially pumped pressure stage (450) is arranged in the direction of the particle beam upstream of the vacuum pump connection (355) of the column (430, 830). [12] Device (400, 800) according to one of claims 7-11, wherein the inlet area (470) of the at least one differentially pumped pressure stage (450) has a pressure stage tube (480) with a pressure gauge of 1 mm to 3 mm, preferably 1.3 mm to 2.7 mm, more preferably 1.6 mm to 2.4 mm, and most preferably 1.9 mm to 2.1 mm and with a length in the range of 5 mm to 25 mm, preferably 7 mm to 18 mm, more preferably 8 mm to 14 mm, and most preferably 9 mm to 11 mm. [13] Device (400, 800) according to one of claims 7-12, wherein the outlet area (370) of the at least one differentially pumped pressure stage (450) has a pressure stage tube (380) with a pressure gauge of 2 mm to 4 mm, preferably 2.3 mm to 3.7 mm, more preferably 2.6 mm to 3.4 mm, and most preferably 2.9 mm to 3.1 mm and with a length in the range of 20 mm to 36 mm, preferably 23 mm to 33 mm, more preferably 26 mm to 30 mm, and most preferably 27 mm to 29 mm. [14] Device (400, 800) according to claim 7, wherein the at least one aperture (810) has an adjustable aperture opening (820). [15] Device (400, 800) according to claim 7 or 14, wherein the at least one aperture (810) comprises at least one piezo actuator configured to adjust the aperture opening (820). [16] Device (400, 800) according to one of claims 7, 14 or 15, further comprising a power supply configured to apply an electrostatic potential to the at least one aperture (810). [17] Device (400, 800) according to one of claims 7 or 14-16, wherein the aperture opening (820) of the aperture (810) is larger, preferably by a factor of 1.5, more preferably by a factor of 1.8, and most preferably by a factor of 2.0, than a distance (830) of the aperture opening (820) from a sample surface (197). [18] Device (400, 800) according to one of claims 14-17, wherein the aperture (820) comprises a range from 100 µm to 3000 µm, preferably 130 µm to 2000 µm, more preferably 160 µm to 1000 µm, and most preferably 200 µm to 600 µm. [19] Device (400, 800) according to the preceding claim, wherein a charge compensation grid (195) has a distance to the sample surface (197) that is half the size of a grid opening of the charge compensation grid (195).

Images