Multi-electron beam imaging device with improved performance

Abstract

An electron beam imaging device, comprising: a sample holder, the sample holder is used to hold a sample; an electron source, the electron source is used to generate a precursor electron beam; an aperture plate, the aperture plate includes an array of holes, the array of holes is used to generate an electron beam array from the precursor electron beam; an electron beam column, the electron beam column is used to guide the electron beam array onto the sample, wherein the electron beam column is configured to have a length L less than 300 mm.

Description

[0001] The present invention relates to an electron beam imaging device, comprising:

[0002] - a sample holder for holding a sample;

[0003] - an electron source for generating a precursor electron beam;

[0004] - an aperture plate comprising an array of apertures for generating an array of electron beams from the precursor electron beam;

[0005] - an electron beam column for directing the electron beam array onto the sample.

[0006] The invention also relates to a method of using such an imaging device.

[0007] Examples of electron beam imaging equipment include electron microscopes (EM) and electron beam lithography imaging systems (used, among other things, to write integrated circuit patterns onto photoresist-coated semiconductor substrates). Electron microscopy is a well-known and increasingly important technique for imaging microscopic objects. Historically, the basic genus of electron microscopes has evolved into many well-known equipment species, such as transmission electron microscopes (TEMs), scanning electron microscopes (SEMs), and scanning transmission electron microscopes (STEMs), and has also evolved into various subspecies, such as so-called "dual beam" equipment (e.g., FIB-SEMs), which additionally employ a "processing" focused ion beam (FIB), thereby allowing supporting activities such as, for example, ion beam milling or ion beam induced deposition (IBID). More specifically:

[0008] - In an SEM, the emission of "auxiliary" radiation from a sample is accelerated, for example in the form of secondary electrons, backscattered electrons, X-rays and cathodoluminescence (infrared, visible and / or ultraviolet photons) by irradiating the sample with a scanning electron beam; one or more components of this emitted radiation are then detected and used for image accumulation purposes.

[0009] In a TEM, the electron beam used to illuminate the sample is chosen to have a sufficiently high energy to penetrate the sample (for this purpose, the sample will generally be thinner than in the case of SEM samples); an image can then be created using the transmitted electrons emitted from the sample. When such a TEM is operated in scanning mode (hence the name STEM), the image is accumulated during the scanning movement of the illuminating electron beam.

[0010] The SEM can also be used in "transmission mode," for example, when using relatively thin samples and relatively high incident beam energies. Such tools are often called "TSEMs" (transmission SEMs), and typically have a relatively basic imaging system (e.g., a single lens and deflector) positioned between the sample and a post-sample detector.

[0011] It should be noted that, in addition to imaging, performing (local) surface modifications (e.g., grinding, etching, deposition, etc.), and conducting spectroscopic analysis, electron microscopes can also have other functions, such as examining diffraction patterns, studying ion channeling / ion backscattering (Rutherford Backscattering Spectrometry), etc. Note that in electron microscopy, the beam impinging on the sample is usually called the "probe."

[0012] In a lithographic imaging system (e.g., a wafer stepper / wafer scanner), a beam of actinic radiation is used to pattern a layer of energy-sensitive material (photoresist) that has been provided (e.g., spun on) on the surface of a substrate (e.g., a semiconductor wafer). Traditionally, the actinic beam has comprised a broad beam of photons (e.g., from a mercury lamp or a laser) that passes through a mask / reticle and later sensitizes its pattern to energy. However, other types of lithographic imaging systems utilize charged particles, such as so-called "direct-write" electron beam tools, which track one or more electron beams over the energy-sensitive layer according to the desired pattern.

[0013] In all cases, the electron beam imaging equipment will include at least the following components:

[0014] - an electron source comprising, for example, a field emission gun (FEG; Schottky or cold FEG) or a LaB6 thermal source.

[0015] - an illuminator (electron beam column; electron light column) which is used to manipulate the "raw" radiation beam from the source and perform certain operations on it, such as focusing, aberration reduction, cropping (using an aperture), filtering, etc. It usually includes one or more (charged particle) lenses and may also include other types of (particle) optical components. If necessary, the illuminator can be provided with a deflector system that can be called upon to make its outgoing beam perform a scanning motion across the sample being studied.

[0016] - A sample holder on which the sample being studied can be fixed and positioned (e.g., tilted, rotated). If necessary, this holder can be moved to affect the scanning motion of the sample relative to the beam. Generally, such a sample holder will be connected to a positioning system. When designed for holding cryogenic samples, the sample holder may include means for maintaining the sample at cryogenic temperature, for example, using a suitably connected cryogen tank.

[0017] For example, a transmission-type instrument such as a (S)TEM or TSEM would additionally include:

[0018] - an imaging system, which essentially absorbs electrons transmitted through the sample (plane) [or mask plane, in lithography] and directs (focuses) them onto a target, such as a detection / imaging device, a spectroscopic apparatus, a resist-coated substrate [in lithography], etc. As with the illuminator mentioned above, the imaging system may also perform other functions, such as aberration mitigation, cropping, filtering, etc., and it will typically include one or more charged particle lenses and / or other types of electron optical components.

[0019] Conventional electron beam imaging devices utilize a single imaging beam. However, in recent years, multi-beam device designs have begun to emerge in which the precursor beam from an electron source is converted into an array of beamlets by positioning a multi-aperture plate within the footprint of the precursor beam; such an aperture plate may also be referred to as an aperture lens array (ALA) because the apertures therein have a lensing effect (when used in conjunction with complementary electrodes). Multi-beam electron microscopes are described, for example, in the doctoral thesis of one of the inventors: "196 Beams in a Scanning Electron Microscope", Ali Mohammadi-Gheidari, TU Delft, The Netherlands, November 2013 (incorporated herein by reference):

[0020] https: / / repositorv.tudelft.nl / islandora / obiect / uuid:98869173-12db- 49cf-9466-0341cf6ad845? collection=research

[0021] Furthermore, multi-beam lithography imaging systems are developed and marketed, for example, by Mapper Lithography B.V., a company in Delft, the Netherlands. The basic idea behind using several primary beams simultaneously—rather than just a single beam—is that it can significantly increase throughput, as the use of a beam array allows for "parallel imaging." However, the simultaneous use of multiple imaging beams (beamlets, beamlets, component beams) also presents various technical issues and challenges that need to be satisfactorily addressed. One area where such issues arise is imaging / patterning quality, particularly when relatively high image brightness is desired.

[0022] The object of the present invention is to provide an improved multi-electron beam imaging apparatus. More specifically, the object of the present invention is that such an imaging apparatus should be able to achieve more satisfactory imaging performance than existing multi-electron beam designs, especially at relatively high beam currents.

[0023] These and other objects are achieved by an imaging device as set forth in the opening paragraph, characterized in that the electron beam column is configured to have a length L of less than 300 mm by comprising the features as defined in claim 1. The column length L is defined herein as the distance between the electron source and the sample.

[0024] The various insights that led to / supported the present invention can be stated as follows:

[0025] - A fundamental problem with electron beam imaging devices (as opposed to photon imaging, for example) is that the beam electrons can influence each other via Coulomb interactions. Such interactions can slightly change the trajectory of the electrons, causing their position relative to the optical axis to deviate from the original / nominal value. The cumulative effect of many such interactions is the (statistical) phenomenon of "beam spreading", whereby, for example, the nominal circular footprint of the beam in a given plane (e.g. the sample plane or the detector plane) becomes expanded, and a peripheral "halo" of deviated electrons is acquired. This results in a blurred image, but also a reduction in image intensity (because the same number of electrons is now spread over a larger footprint). For a given required resolution, this Coulomb beam spreading sets a fundamental limit on the maximum beam current that can be allowed in the beam, and therefore on the throughput that can be achieved in a single beam imaging device.

[0026] Multi-beam electron imaging devices aim to increase throughput. A common assumption underpinning the use of multi-beam devices is that electrons in a given beamlet will only (or primarily) experience Coulomb interactions with other electrons in its own beamlet, and will experience no (or almost no) Coulomb interactions with electrons in other beamlets, provided that these beamlets do not cross each other. Based on this assumption, one can expect to increase the total current in the column essentially arbitrarily by appropriately increasing the number of beamlets, provided that these beamlets have no (or almost no) crossovers (i.e., where the beams cross each other).

[0027] - The inventors investigated this Coulomb interaction extensively. Assuming that beam crossings in electron beam column designs are a "breeding ground" for such Coulomb interactions, they set out to test column designs in which the amount of such crossings was reduced or essentially eliminated; surprisingly, the inventors found that although the magnitude of the above-mentioned blurring / beam spreading problem was reduced in such designs, it still remained to a large extent. Additional extensive analysis unexpectedly revealed that the interactions between electrons in different beamlets are larger than previously assumed. This can be illustrated using the following (non-relativistic) example. Consider a beamlet carrying a beam current of I = 1 nA at a beam potential of V = 10 kV. The average longitudinal (Z) distance between the electrons in this beamlet is d z =ev z / I=10mm,use v z =√(2eV / m)=6x 10 7 m / s (where e is the elementary electron charge, v zis the electron velocity, and m is the electron mass). This far exceeds the typical lateral spacing between beamlets, which indicatively ranges from about 10-500 μm. Therefore, it is more likely that a given electron's nearest neighbor will be in a different beamlet than in the same beamlet. Thus, it now appears that Coulomb interactions occur between all electrons in all beamlets, with only a relatively small dependence on whether the individual beamlets intersect one another or whether they travel independently along the column.

[0028] - Important for the present invention, the inventors have found that the key factor influencing the achievable image resolution is the length L of the electron beam column; in fact, the inventors have found that:

[0029] ·According to ~L 2 , the effects of electron-electron interactions scale to a good approximation. This can be roughly explained on the basis of:

[0030] o Longer columns provide a linearly longer time for the interaction to occur; and

[0031] o Longer columns provide linearly longer "arms" to exploit the effects of this interaction.

[0032] These two linear effects combine to produce the quadratic effect of L.

[0033] Shorter L values ​​lead to lower magnification values, which allows the aperture plate / ALA to have larger holes at a smaller hole pitch - resulting in greater transmission; such aperture plates / ALAs are relatively easy to manufacture compared to aperture plates with smaller holes. Relatedly, shorter L values ​​lead to smaller beam opening angles at the aperture plate / ALA; in a column design without additional (intervening) beam-defining holes, this will simultaneously lead to smaller beam opening angles at the sample level, which in turn will generally result in reduced aberrations. An alternative explanation is provided:

[0034] ο The (spatial) magnification from source to sample roughly scales with L.

[0035] o The angular magnification from source to sample scales inversely with the (spatial) magnification; thus, for a fixed opening angle at the source, a larger L provides a larger opening angle in the "probe" that strikes the sample, and correspondingly larger aberrations in the probe-forming optics.

[0036] o These aberrations can be reduced by limiting the beam diameter (e.g., by using a smaller hole in the aperture plate / ALA), but this will correlate to a reduction in probe current, which is undesirable and also reduces the manufacturability of the aperture plate / ALA.

[0037] Therefore, steps were taken to develop reduced columns, and for L values ​​below about 300 mm, the above-mentioned blurring / beam spread problem was brought within acceptable tolerances. The solution of the present invention is particularly advantageous because it does not itself require a reduction in beam current (which would reduce throughput by the same amount) to bring the Coulomb interaction problem under control.

[0038] It should be noted that reducing the column length L in this manner is somewhat counterintuitive. This is because a typical column will contain a multiplicity of relatively bulky particle optical components that need to be positioned and aligned within tight tolerances, and this has traditionally led to relatively long / roomy column designs. Therefore, abandoning such roomy designs in favor of tighter / more compact configurations is generally viewed as undesirable. For reference purposes, it should be noted that the multibeam SEM described in the aforementioned doctoral thesis has a column length of approximately 500 mm—a column length that is 66% greater than the upper limit proposed in the present invention.

[0039] In a preferred embodiment of the present invention, the column length L satisfies L<200 mm. The inventors have found that the "Coulomb blur" BC (inverse resolution) at the sample level scales according to the following formula:

[0040] B C ~IL 2 / V4 / 3

[0041] Where I is the (accumulated) beam current, V is the beam potential (accelerating voltage), and L is the column length. Thus, reducing L allows I to be increased (for comparable resolution) or allows resolution to be further enhanced (for a given value of I). Although L values ​​in the range of 200-300 mm generally provide good results, L values ​​below 200 mm have been found to be more satisfactory.

[0042] As defined herein, the electron beam column is configured to have a length L of less than 300 mm by comprising:

[0043] - a single individual beam intersection plane, wherein each of said electron beams forms an intermediate image of said electron source;

[0044] - a single common beam crossing plane, where the electron beams in the array cross each other (or in other words: cross the optical axis of the electron beam column).

[0045] Conventional electron beam columns (both in single beam devices and in multi-beam SEMs such as described in the aforementioned doctoral thesis) tend to have multiple beam crossings in their design. In addition to the objective lens, such a configuration may, for example, wish to utilize various other optical elements upstream of the objective lens, such as a focusing lens, a variable field diaphragm in the column core (with which the beam current and hence image brightness can be conveniently adjusted), additional magnifiers, etc. The incorporation of such optical elements inevitably leads to the presence of associated beam crossings (individual and / or common). As mentioned above, because such crossings tend to exacerbate Coulomb interactions, the inventors have developed a "spartan" column design that limits the number / type of crossings to a practical minimum. Furthermore, this spartan design also makes it easier to achieve relatively short columns. It should be noted that the single individual beam crossing planes and the single common beam crossing plane are parallel planes, meaning that these planes do not coincide. In other words, the single common beam crossing plane is parallel to the single individual beam crossing plane.

[0046] In one embodiment, the individual electron beams are defocused when crossing each other in a single common beam crossing plane. This reduces the Coulomb interactions between the electrons in the individual beamlets.

[0047] In a particular embodiment of the invention, the electron beam column comprises:

[0048] - a correction lens positioned downstream of the aperture plate;

[0049] An objective lens situated between the correction lens and the sample.

[0050] As specified herein, a (synthetic / compound) corrective lens may, for example, comprise:

[0051] An accelerating stage for accelerating electrons from the relatively low potential (typically about 1-2 kV) at which the aperture plate / ALA operates to the relatively higher potential (typically about 10-20 kV) at which the main column operates. The relatively low potential of the aperture plate / ALA is advantageous because it facilitates its construction; on the other hand, the relatively high potential of the main / lower column is advantageous because the effects of Coulomb interactions are reduced at higher potentials and because the resolution of the "probe" is improved due to the shorter wavelength of the electrons and the reduced color blurring at higher potentials. The accelerator stage may, for example, comprise a series of paired accelerator electrodes in a "chain" that provide a cascaded electrostatic field configuration in which electrons are accelerated stepwise.

[0052] A collimating lens, which is used to collimate / focus the beamlets. This collimating lens is ideally positioned so that the plane of the (intermediate) image of the source is in / near the center of this lens, to avoid defocusing / blurring these intermediate images (e.g., by off-axis chromatic aberration). Furthermore, this lens will typically be positioned relatively close (e.g., within about 30 mm) to the aperture array / ALA, to prevent the array of beamlets emanating from the aperture array / ALA from becoming large enough in the collimating lens to produce spherical aberration effects, thereby significantly distorting the pattern of the beamlets.

[0053] It should be noted that the aforementioned doctoral thesis describes a specific correction lens comprising four electrodes configured such that:

[0054] - The first three electrodes mainly collimate (but also accelerate);

[0055] - The fourth electrode mainly accelerates (but also collimates).

[0056] The designated objective is used to image and reduce the plane of the (intermediate) image array with the source (on the sample). To minimize aberrations, the collimating lens is ideally fine-tuned so that all beamlets pass through the so-called "comma-free" point of the objective. The coma point of the lens is usually close to the core of the lens; as known to those skilled in the art, when an oblique beam passes through this coma point, the lens will not refract an oblique beam (such as the beamlet in question) with coma.

[0057] Referring to the discussion in the previous paragraph, for example, a column design can be used where:

[0058] - the individual beam intersection plane is located in / proximal to the correction lens;

[0059] - The common beam crossing plane is located in / proximal to the objective lens.

[0060] For example, see Figure 3 .

[0061] The present invention is applicable to scanning type imaging devices, in which images are accumulated during a relative (lateral) scanning motion of the beam array and the sample using a scanning assembly. This relative scanning motion can be achieved in different ways. For example:

[0062] (i) In one possible ("mechanical scanning") setup, the scanning assembly employed comprises an actuator system for generating a scanning motion of the sample holder.

[0063] (ii) In an alternative ("beam scanning") setup, the scanning assembly employed includes a beam deflector system for deflecting the beam array relative to the sample. For example, in multi-beam STEM:

[0064] - a first beam deflector system upstream of the sample can be called upon to produce a scanning motion of the beam array relative to the sample;

[0065] - A second beam deflector system downstream of the sample and upstream of the imaging detector can be used to nullify the scanning motion produced by the first beam deflector system (so called "de-scanning").

[0066] An advantage of method (i) is that it tends to be less complex / complex than method (ii) - although it will require the use of a sample holder stage capable of precisely traversing the scan path; this need not present a technical hurdle, as complex scanning stages are already used in, for example, photolithography and are available in many different implementations. Beam scanning is the more traditional approach used for single-beam EM (such as SEM and STEM), although the present invention will require modification to multiple beams. Needless to say, a hybrid of methods (i) and (ii) could also be chosen, employing both types of scanning, if desired.

[0067] Regarding the cardinality of the beam array (the number of component beams) used in the present invention, it should be emphasized that this is a freely selectable choice. On the one hand, the greater the number of beams, the greater the throughput enhancement that can be achieved; on the other hand, a larger number of beams will increase the complexity of the overall system; furthermore, as mentioned above, since electrons in different beams do interact with each other, a larger number of beams will also tend to increase Coulomb blurring. A skilled person will be able to find a compromise between these competing effects and select a number that is appropriate for a given scenario / use case. For example, a system can be constructed using 196 beams in a 14 x 14 array—although many other beam multiplicities / configurations are also possible.

[0068] The invention will now be described in detail with reference to exemplary embodiments and the accompanying schematic drawings, in which:

[0069] Figure 1 A longitudinal cross-sectional elevation view is given of an embodiment of a charged particle imaging apparatus (in this case, an electron microscope) embodying the present invention.

[0070] Figure 2 demonstrated how an array of electron beams can be generated from a single precursor beam.

[0071] Figure 3 A specific electron beam column according to an embodiment of the present invention is shown.

[0072] In the figures, corresponding parts are indicated by corresponding reference numerals where relevant.

[0073] Example 1

[0074] Figure 1(not to scale) is a highly schematic diagram of an embodiment of an electron beam imaging apparatus in which the present invention is utilized; more specifically, it shows an embodiment of an SEM - however, in the context of the present invention it could also be, for example, a STEM or a lithography imaging system. The imaging apparatus (microscope) 1 comprises (among other things):

[0075] An electron source 3 (such as, for example, a Schottky emitter, a cold FEG or a LaB6 filament) which generates an electron beam B propagating along a particle-optical axis B′.

[0076] - an electron beam column 5 comprising various lenses 11, 13 for directing / focusing the electron beam B and a deflection unit 15 for performing beam deflection / scanning of the beam B. The column 5 may also comprise an aberration corrector, a field stop and the like.

[0077] - A vacuum chamber 7 comprising a sample holder 17 and an associated stage / actuator 19 for holding / positioning a sample S. The vacuum chamber 7 (and also the column 5 / source 3) is evacuated using a vacuum pump (not depicted). By means of a voltage source 21, the sample holder 17 or at least the sample S can be biased (floated) to a potential relative to ground if necessary.

[0078] By scanning the electron beam B over the sample S, output radiation (including, for example, X-rays, infrared / visible / ultraviolet light, secondary electrons and / or backscattered electrons) is emitted from the sample S. The detectors 23, 27 can be selected from a variety of possible detector types that can be used to detect different types / morphologies of this output radiation. In the apparatus depicted here, the following detector selections have been made:

[0079] The detector 23 is a segmented electron detector comprising a plurality of independent detection segments (e.g. quadrants) arranged around a central opening 25 (encompassing the optical axis B′). For example, such a detector can be used to study the angular dependence of the (second / backscattered) electrons emitted from the sample S.

[0080] The detector 27 is, for example, an X-ray detector which can be used to record X-rays emitted from the sample S and thereby perform energy dispersive X-ray spectroscopy (EDX). Alternatively, it can be, for example, a cathodoluminescence detector.

[0081] Alternatively / supplementarily, a backscattered electron detector such as that described in co-pending European patent application EP18176596.7 may be used. Since the detected output radiation is position dependent (due to the scanning motion), the information obtained from the detectors 23, 27 will also be position dependent and can therefore be used to assemble an image of a map of the detector outputs essentially as a function of the scan path position on the sample S. The signals from the detectors 23, 27 are transmitted along a control line (bus) 29', processed by a controller 29 and displayed on a display unit 31. Such processing may include operations such as combining, integrating, subtracting, pseudo-coloring, edge enhancement and other processing known to those skilled in the art. In addition, automatic recognition processes (e.g., as used for particle analysis) may be included in such processing.

[0082] Those skilled in the art will appreciate various modifications and alternatives to this "basic" setup, including but not limited to:

[0083] - using dual primary beam materials - for example an electron beam for imaging and an ion beam for processing (or in some cases, imaging) the sample S;

[0084] - Using a controlled environment at the sample S - for example maintaining a pressure of a few mbar (as used in so-called ambient SEMs) or by admitting gases, such as etching or precursor gases.

[0085] Essential to the present invention is the fact that the single primary electron beam used in conventional SEMs is replaced here by a multi-beam array. Figure 2 Shows how to use Figure 1 Relatively minor modifications in generate such a multi-beam array. A precursor (synonymous progenitor; initial) electron beam Bo leaves the electron source 3, traverses a series of electrodes 3′ and impinges on an aperture plate / ALA 9 containing a plurality of holes 9′; for example, such a structure 9 can be manufactured by etching an array of small holes in a silicon wafer using thin film (MEMS) technology (see, for example, the above-mentioned doctoral thesis). The aperture plate / ALA 9 is located within the electron beam column 5 and will typically be positioned relatively close to the electron source 3, for example, at a spacing of about 5-10 mm downstream therefrom. As a result of the impact on the ALA 9, the precursor beam Bo is converted into a plurality / array B of sub-beams / beamlets / component beams B″ having the same geometrical configuration as the plurality of holes 9″ used to generate them. This beam array B then follows its course along the axis B′ through the column 5, which directs it onto the sample S (see Figure 1 ). In more detail:

[0086] - The well plate / ALA comprises an array of 14 x 14 wells 9', wherein each well has a diameter of approximately 10 μm and a pitch / spacing of approximately 20 μm.

[0087] - The electrode 3' generates a potential variation at each aperture 9' so that each aperture 9' acts as an electrostatic microlens. In this way, each beamlet B" is focused onto the (intermediate) image array of the source 3 downstream of the aperture plate / ALA 9.

[0088] - In principle, only one electrode 3' is needed to cooperate with the aperture plate / ALA 9 to produce this lens effect; however, using more than one electrode 3' offers more possibilities in terms of "fine tuning".

[0089] The distance between the source 3 and the individual apertures 9' (electrostatic microlenses) of the aperture plate / ALA 9 increases slightly with increasing distance from the optical axis / cylindrical axis B'. If all apertures / microlenses 9' have the same focusing strength, this will result in a variation in the image distance of the (intermediate) image of the source 3, so that these images will not lie in a common plane perpendicular to the optical axis / cylindrical axis B', but on a curved surface (essentially a portion of a sphere). To prevent this curvature of this plane and the (intermediate) source image, a correction electrode can be used, the dimensions of which are such that its field primarily affects the focusing strength of the outer microlenses. By appropriately fine-tuning the potential on this correction electrode, the aforementioned curvature can be mitigated / eliminated. Such a correction electrode may be included in the illustrated set of electrodes 3′, for example as described in the journal article by Y. Zhang and P. Kruit, “Design of a high brightness multi-electron-beam source,” Proc. Seventh Int. Conf. on Charged Particle Optics, Physics Procedia 1 (2008), pp. 553-563, Elsevier pub. (incorporated herein by reference). Alternatively, a correction electrode of this type may be located downstream of the ALA 9, for example as described in U.S. Pat. No. 8,598,545 (incorporated herein by reference).

[0090] In the specific context of the present invention, the length L of the column 5 (between the electron emitting structure (tip) of the source 3 and the (top) surface of the sample S) is reduced relative to the prior art multi-beam imaging system and has a value L<300 mm, preferably L<200 mm.

[0091] Example 2

[0092] Figure 3 A specific electron beam column according to an embodiment of the present invention is shown. Here:

[0093] - Lens 11 is a (compound) corrective lens as described above. Note that the individual beams in the lenses (planes) cross (intermediate source images).

[0094] - Lens 13 is an objective lens as described above. Note that the common beams in the lenses (planes) cross. Here, the individual beams are defocused when they cross each other.

[0095] Individual beam crossings (intermediate source images) are also present on the surface of the sample S, but no other crossings exist (particularly between lens 11 and lens 13). In order not to unnecessarily clutter the drawing, the scanning coils 15 have been omitted from this description.

Claims

1. An electron beam imaging device, comprising: - a sample holder, the sample holder being used to hold a sample; - an electron source for generating a precursor electron beam; - an aperture plate comprising an array of apertures for generating an array of electron beams from the precursor electron beam; - an electron beam column for directing the electron beam array onto the sample, Characterized in that the electron beam column is configured to have a length L less than 300 mm by comprising: - a single individual beam intersection plane, in which each of said electron beams forms an intermediate image of said electron source; and - a single common beam crossing plane, where the electron beams in the array cross each other.

2. The apparatus according to claim 1, wherein L < 200 mm.

3. The apparatus according to claim 1 or 2, wherein the electron beam column comprises: - a correction lens arranged downstream of the orifice plate; - an objective lens situated between the correction lens and the sample.

4. The apparatus according to claim 3, wherein: - the individual beam intersection plane is located substantially within the correction lens; and / or - said common beam crossing plane is located substantially within said objective.

5. A method of using an electron beam imaging device, the electron beam imaging device comprising: - a sample holder, the sample holder being used to hold a sample; - an electron source for generating a precursor electron beam; - an aperture plate comprising an array of apertures for generating an array of electron beams from the precursor electron beam; - an electron beam column for directing the electron beam array onto the sample, Characterized in that the electron beam column is configured to have a length L of less than 300 mm by providing: - a single individual beam intersection plane, in which each of said electron beams forms an intermediate image of said electron source; and - a single common beam crossing plane, where the electron beams in the array cross each other.

Images