Device and method for measuring the height profile of an object

Abstract

The present invention relates to an optical device and system for sensing the surface profile of an object surface (30) of an object by interferometric distance measurement, comprising: a beam splitter (3) for splitting a light beam (84) of a light source (7) into a first sub-beam (410) and a second sub-beam (420); a beam splitter (4) for splitting each sub-beam (410, 420) of the light source (7) into a reference beam (470, 480) and a measuring beam (430, 440); and a mirror (5) for reflecting the two reference beams (470, 480), wherein each measuring beam (430, 440) is guided to a measuring area of ​​the object surface for reflection and is guided to the beam splitter (4) as an object beam (450, 460) after reflection, and each reference beam (470, 480) is reflected at the mirror and guided to the beam splitter (4) as a mirror beam (490, 500). The object beam (450, 460) and the mirror beam (490, 500) each interfere after striking the beam splitter (4) and are each sent as an evaluation beam (520, 510) to a detector unit (8) for evaluation, wherein the two evaluation beams (520, 510) have a phase difference. The system further comprises a light source (7) for generating a monochromatic light beam (84), a detector unit (8) for detecting the evaluation beams (520, 510) and a signal evaluation unit (9) for evaluating the detected evaluation beams (520, 510) and determining the surface profile.

Description

Technical Field

[0001] The invention relates to a device, a system and a method for measuring the surface profile or height profile on a planar object, in particular on a wafer in the semiconductor industry. The device comprises a beam splitter for splitting a monochromatic light beam and directing the sub-beams to the object. The system comprises an optical device, a detector unit, a light source and a signal evaluation unit for determining the surface profile. Background Art

[0002] In many industries, the surface profile of flat products is measured optically using imaging methods. For example, in the semiconductor industry, the surface of wafers is measured. Wafers are slices made of semiconductor, glass or ceramic materials. In certain applications, wafers are often inspected over the entire surface or at least in large local areas. Circuits ("chips") are manufactured on wafers, which later fulfill a variety of purposes in electronic devices for computing technology, sensor technology, control and monitoring.

[0003] In order to achieve electrical contact of the circuit, in the conventional process, wires are guided from the chip housing to the chip and fixed there to the metal contact surface. With the continuous development of miniaturization and the increasing integration level, contacting by applying solder balls to the contact points on the chip has become a state of the art. This is done when the chip is still in the wafer complex, that is, when the wafer has not yet been cut to separate the individual chips. The contact points can thus be distributed over the entire chip surface ("ball grid array"). Contact is achieved by inserting the chip into a housing with corresponding contact surfaces, where the solder is melted by heat and pressure and the contact is established. The uniform height of the solder contact balls is of decisive significance for successful contacting according to this method.

[0004] For this method, the current trend is to move away from solder balls and towards copper cylinders, which only provide a very thin solder cap for making the electrical contact. As a result, the contact structures can be made smaller and placed more densely. Since copper is not as ductile as solder, the permissible height tolerances of this new method are significantly reduced.

[0005] It is therefore necessary to check the height of the contact structures (solder balls or copper pillars) more precisely than before. For this purpose, fast and accurate measuring methods are required in order to be able to measure the height of the contact structures precisely and as comprehensively as possible. A complete measurement of all contact points is considered particularly advantageous, since even a single deviating contact point can render the finished component unusable. Therefore, particular emphasis is placed on high measuring speeds. Typically, the diameter of current contact structures is a few dozen μm, with a strong tendency to decrease. The aspect ratio (height to diameter) is typically around 1.

[0006] At the same time, methods are currently being developed for stacking a number of chips to form a composite. Here, the use of raised contact structures is almost completely eliminated. These methods are being investigated for integration in a wafer composite and for individual chips, which are each placed on the chip still located in the wafer composite. For these methods, too, accurate and fast profile measurements are highly required in order to be able to guarantee the high requirements for the planarity of the surface in the low nanometer range.

[0007] Typical dimensions of the contact structures of these methods are diameters in the range of a few μm and profile heights of a few nm to tens of nm.

[0008] Similar tasks have to be solved in other structures and in other industries. These tasks are, for example, measuring the height of conductor tracks and other structures on electronic boards (printed circuit boards) or measuring various mechanical-electrical system components (MEMS) such as acceleration sensors, pressure sensors and velocity sensors or gyroscopes.

[0009] All of these applications have some common features. All require the rapid testing of large numbers of usually similar test samples such as printed circuit boards, wafers, displays and similar objects. The use of sensors to generate profile and / or height data of the structures being probed is also common in these applications. In addition, the applications considered here do not necessarily require 3-dimensional images with high resolution in the lateral direction. In some cases, a single numerical value (such as the cap height of the contact ball and cylinder discussed above) is sometimes sufficient for inspection.

[0010] Prior art

[0011] Conventional height measurement methods use point sensors, line sensors or surface sensors. The advantage of surface sensors based on deflectometry or white light interferometers, for example, is that a surface can be sensed as a whole. In deflectometry, the surface to be measured is used as a mirror through which a regularly structured light source is viewed. Changes in the height of the object are reflected in the deformation of the mirror image and can be extracted from it. This method is very fast, but its lateral resolution is limited and is therefore not or only conditionally suitable for the described task.

[0012] In white light interferometry, a series of receptions must be performed at different distances from the object or the interferometer mirror. This is generally too slow for full-surface inspection of the wafer. Line sensors use, for example, triangulation, stereometry or chromatic confocal imaging. Each of these methods suffers from measurement artifacts when transitioning to small contact points, making precise measurement difficult or even impossible, or they are too slow. The conventional application variants of point sensors require very precise positioning on the contact structure. Since the lateral positioning of the contact point itself is of the order of magnitude of the resolution of the sensor, this is time-consuming and increasingly prone to errors. Summary of the invention

[0013] The object of the invention is to propose an improved system and an improved device for sensing the contour of an object surface of an object by interferometric distance measurement, in particular also for measuring the height of small contact structures with high precision in a short time.

[0014] This object is achieved by an optical device having the following features, a system having the following features and a method having the following features.

[0015] According to one aspect of the present invention, the optical device includes a beam splitter for splitting a light beam of a light source into a first sub-beam and a second sub-beam, a beam splitter for splitting each sub-beam of the light source into a reference beam and a measuring beam, and a mirror for reflecting the two reference beams. Each measuring beam is guided to a measuring area on the surface of an object for reflection, and after reflection, is redirected to the beam splitter as an object beam. Each reference beam is reflected on the mirror and guided to the beam splitter as a mirror beam. The mirror is designed so that the two reflected reference beams, that is, the mirror beams, have a phase difference. The object beam and the mirror beam interfere with each other after hitting the beam splitter and are sent to the detector unit as evaluation beams for evaluation. The two evaluation beams have a phase difference when they hit the detector unit.

[0016] According to the invention, the device comprises a detector unit for detecting an evaluation light beam and a signal evaluation unit for evaluating the detected evaluation light beam and ascertaining the surface contour.

[0017] To determine the surface profile, each measuring beam is directed to a measuring area on the object surface for reflection and, after reflection, is directed to a beam splitter as an object beam. Each reference beam is reflected on a mirror and directed to a beam splitter as a mirror beam. The respective object beam and the respective mirror beam each interfere after impinging on the beam splitter and are each sent to a detector unit as an evaluation beam for evaluation.

[0018] The signal evaluation unit determines the intensities of the two evaluation beams and determines the height profile of the surface using the intensities of the two evaluation beams. For the height profile, the influence of the reflectivity of the surface on the measured intensity is compensated by calculation. Thus, the correct profile can be determined for object surfaces that are not 100% reflective and for object surfaces that have different reflectivity in the area to be measured.

[0019] According to another aspect of the present invention, a system for sensing the surface profile of an object by interferometric distance measurement comprises a light source for generating a monochromatic light beam and the above-mentioned optical device.

[0020] According to the invention, the two evaluation beams impinging on the detector unit have a phase difference from which the surface profile of the object surface of the object can be derived. In arrangements that have been based on interferometric distance measurement to date, so-called light traps are used to absorb at least one sub-beam after the splitting of a beam into two sub-beams. The known arrangements are designed such that a non-interfering beam and an interfering beam are fed to the detector arrangement. From the evaluation of the two different beam types, the height profile can thus be inferred.

[0021] However, it is known within the scope of the invention that the two interfering light beams can also be evaluated by a detector unit and the height profile can be sensed from the measurement result by a signal evaluation unit. The detector unit has two detectors, preferably sensors or cameras. The advantage of this arrangement according to the invention is that no optical traps (light traps) need to be used. A light trap is an optical element that completely absorbs the radiation that occurs. In practice, this can only be produced very complicatedly and with great difficulty. If almost all of the radiation cannot be absorbed, interference effects will occur, which can influence the measurement and falsify the result.

[0022] It is further known within the scope of the invention that when two interfering light beams are used, they must have a phase difference in order to be able to be evaluated. Preferably, the phase difference is not equal to an integer multiple of the wavelength used.

[0023] The advantage of the device and system according to the invention is that by using two evaluation beams with a (preferably fixed) phase offset or phase difference, objects whose surfaces do not reflect 100% of the incident light can also be measured. Thus, in particular, surfaces whose surface sections have different reflectivities can also be measured. This is the case, for example, with objects of different thicknesses or different materials. In particular, wafers of the semiconductor industry have locally a variety of different material combinations, some of which also form semi-transparent multilayers. Such thicknesses and material combinations can be measured precisely and quickly by the device according to the invention and the system according to the invention.

[0024] According to the invention, a reference beam is generated from each of the two sub-beams. In a preferred embodiment, the phase difference is generated in that each reference beam (the first reference beam from the first sub-beam and the second reference beam from the second sub-beam) strikes a mirror, for example with two mirror segments or split mirrors or mirrors, and is reflected there, wherein the mirrors or mirror segments have different positions. Preferably, the mirror surfaces are arranged parallel to each other, wherein the positions of the mirror surfaces are different in the direction of the surface normal. The mirrors therefore have an offset in the direction of the surface normal. As a result, differences in the propagation times between the reference beams or the reflected so-called mirror beams occur, which leads to phase differences.

[0025] In contrast to optical traps, in which a large part of the radiation is absorbed, mirrors reflect a large part of the radiation. At least 30% of the radiation is reflected, preferably at least 50%, very preferably at least 80%, particularly preferably at least 95%.

[0026] In a preferred embodiment, a mirror is used which comprises two parallel mirror segments, the mirror surfaces of which are also offset in the normal direction. Alternatively, the mirror segments themselves can also be mirrors. The mirror surface can also be applied, for example vapor deposited, on an optical element, for example a transparent block, a glass block or an acrylic glass block.

[0027] In a particularly preferred embodiment, the beam splitter and the mirror of the device according to the invention are integrated in an optical element. For example, the optical element can be composed of a transparent material or glass block. For example, the optical element can include two transparent blocks, each of which has a triangular base, preferably an equilateral and right-angled base.

[0028] One of the blocks can be constructed in such a way that the side faces have an offset on the sides of the triangular plan, wherein the outer sides are parallel but offset in the direction of their normals. Mirrors or mirror segments can be mounted on these two offset faces, which are preferably of equal size and preferably oriented in parallel. Preferably, the mirrors are vapor deposited or applied on the transparent block and its offset faces.

[0029] If two transparent blocks are connected together so that their long sides (corresponding to the sides at the hypotenuse of the triangular base) are close together and in contact, a beam splitter can be formed. For example, the two transparent blocks can be glued together. One or both long sides can have a semi-transparent mirror.

[0030] In a preferred embodiment, the light source comprises a laser producing monochromatic light. If multiple light sources are used, the light beams of these light sources are preferably bundled together, each light source comprising a laser.

[0031] In a preferred embodiment, a plurality of light sources are used, each of which generates a monochromatic light beam. Preferably, the light beams of the respective light sources have different wavelengths. Preferably, the wavelength of no light source is an integer multiple of the wavelength of another light source. Therefore, the light beams of the respective light sources have different wavelengths, which wavelengths are preferably not an integer multiple of the wavelength of another light beam.

[0032] In a preferred embodiment with multiple light sources, the light beams of each light source are bundled in a beam coupler so that the emitted light beam includes the light beams of all light sources. For example, different light sources can generate light beams of different colors.

[0033] In a preferred embodiment, the beam splitter comprises a semi-transparent mirror which allows a portion of the impinging beam to pass as a reference beam and reflects another portion of the impinging beam as a measuring beam towards the object surface of the object, thereby deflecting it. Preferably, the beam is split 50% into a reference beam and a measuring beam.

[0034] Preferably, the detector unit of the system according to the invention comprises a time-delay integration camera, a so-called TDI camera, preferably a TDI line scan camera. In a preferred embodiment, a TDI color camera or a TDI color line scan camera is used. Particularly preferably, the detector unit has a TDI multi-channel line scan camera, which can be color or monochrome.

[0035] When the light beam of the light source or light sources is split into two sub-beams, two cameras or line scan cameras, preferably TDI cameras, are preferably used in the detector unit. Due to the overall arrangement consisting of the two sub-beams and their respective splitting into the reference beam and the measuring beam and the subsequent superposition of the reflected mirror beam or object beam to form the evaluation beam, there are again two beams, each of which is guided separately to a camera.

[0036] In another preferred embodiment, an optical element is provided for separating the bundled light beams, in particular the bundled evaluation light beam. The optical element is arranged in such a way that the separation takes place before the light beams impinge on the detector unit and its camera. The separation of the bundled light beams, for example the bundled evaluation light beam, can take place before or after the two evaluation light beams are separated and directed to the respective cameras. The bundled light beams can also be separated after the evaluation light beams have been separated and directly before the light beams enter the respective cameras. In this case, the separation is advantageous when using cameras that are sensitive to the individual light beams and their wavelengths. When using a multi-channel line scan camera or a TDI color camera, this separation and this optical element for separation can be omitted.

[0037] According to the present invention, in order to further achieve the object: the light source continuously illuminates the object to be measured with a light beam having at least one monochromatic wavelength;

[0038] The relative movement between the combined detector unit (so-called camera arrangement / interferometer arrangement) and the object takes place continuously and without interruption;

[0039] A fixed, appropriately structured reference mirror produces a phase shift between the two sub-beams of the light source;

[0040] using a detector of a detector unit, such as a camera sensor of an interferometer device, to simultaneously measure the intensity of two sub-beams of light (evaluation beam) which are each formed by interference of light reflected by the object (object beam) with light reflected at a reference mirror (mirror beam) at all used wavelengths;

[0041] the measured intensity of the first sub-beam interference signal (first evaluation beam) is corrected at each wavelength by means of the measured intensity of the second sub-beam interference signal (second evaluation beam) corrected by a fixed reference phase shift; and

[0042] The distance to the object is calculated from the corrected interference signal intensity of the first sub-beam interference signal (first evaluation beam).

[0043] In this arrangement, a line sensor is preferably used as detector for interferometric measurement. Thus, wide strips of an object can be measured continuously. One aspect of the invention is to correct for the influence of different surface materials of the object to be measured. In order to extend the measuring range to the range required for the profile height to be measured, the invention proposes to use multiple wavelengths simultaneously in the system according to the invention.

[0044] It is known within the scope of the invention that from the intensity measurement of the signal modulated by the interferometry method, the distance to the measured object can be clearly ascertained with very high resolution. However, this only applies if the surface of the measured object reflects the light used 100% or if the reflectivity is known for each measuring point. If the object to be measured has semi-transparent layers, then for absolute distance measurement their thickness and the optical properties (refractive index and absorption coefficient) must also be known exactly for each measuring point at the wavelength used. For relative profile measurement, these values ​​may not be known exactly, but are limited by the fact that the thickness and material must be exactly the same at all measuring points, or their variations would limit the accuracy of the measurement.

[0045] Therefore, for the purpose of profiling the surface of an object, in particular height measurement on a wafer, interferometry is not very suitable, since the intensity of the measured interference signal also depends on the material of the structure to be measured and its environment. For example, the wafer to be measured has locally a combination of different materials, part of it consists of semi-transparent multilayers, the thickness and material variations of which make precise and fast measurements with one or a few wavelengths impossible. Without knowledge of the local reflectivity of the reflected light (evaluation beam) and the phase shift caused by the semi-transparent layer, it is not possible to make sufficiently accurate statements about the distance.

[0046] In order to correct the influence of local material-related variations in the intensity of the reflected light, according to the invention an interferometric measurement is carried out simultaneously with at least two sub-beams which differ by a fixed phase shift.

[0047] If two sub-beams are used, the accuracy of the measurement can already be significantly improved, since the influence of reflectivity can be eliminated. This is sufficient for many tasks, for example if there are no semi-transparent layers on the measured object or their influence is negligible.

[0048] Using more than two beamlets can improve the robustness of the measurement by increasing the redundancy of the data.

[0049] In the following, the correction for a single detector element and a single wavelength λ1 is explained with reference to height measurement on a wafer, for example, and without limiting the generality. In order to realize the profile measurement at high speed, a line scan camera with a plurality of detection lines is advantageously used. The signal Si arriving at the interferometer can be expressed as

[0050]

[0051] Here, q represents the intensity of the light source; R represents the surface of the measured object at wavelength λ i γ is the instrument constant, which is derived from the detector sensitivity, absorption and reflection losses of the optics; and z ri and z s represents the distance between the interferometer splitter mirror and the reference mirror or the surface of the measurement object, and φ represents the phase shift, which can be obtained by a semi-transparent layer on the surface of the measurement object if necessary.

[0052] For the case of a negligible phase shift φ~0, the combination of the two sub-beams results in a ratio of the measured intensities of i=1.2.

[0053]

[0054] This can be transformed in the following way

[0055] Δz r =z r1 -z r2 ,Δz=z r1 -z s and

[0056] become:

[0057]

[0058] In order to find out the actual distance to search for the surface of the measuring object, Δz must be determined by solving the transcendental equation. ris a known variable (or can be ascertained by calibration measurements in which a plane mirror is substituted for the object being measured), the surface profile can be ascertained as long as the surface profile of the object being measured does not have any vertical steps greater than half the wavelength of the light used.

[0059] When only one wavelength is used for interferometric measurements and the solutions according to the invention proposed so far, the working range of the measured distances is still limited to half the wavelength of the light used. Outside this range, the relationship between intensity and distance becomes unclear, since it is not possible to directly ascertain how many cycles the path difference between the reference beam and the object beam is from the signal intensity alone. Conventional interferometers solve this problem by ascertaining the reference distance during initialization and subsequently counting the measured intensity cycles continuously. However, this is not suitable for the use according to the invention of line interferometers, since a reference better than half a wavelength parallel to the interferometer line must be provided and there should be no steps on the measured object that are more than half a wavelength high. However, the latter does not exist, at least when measuring contact structures on wafers.

[0060] If, on the other hand, two or more different wavelengths are used, the working range can be extended, since different wavelengths have different periodicities and can therefore be measured unerringly by the signal combination of the wavelengths used in the larger range, so that measurements can be made unambiguously within this working range without the need for starting a reference and cycle counting. The basic principle of this method is described in "Multi-Wavelength Interferometry for Length Measurements Using Diode Lasers" by K. Meiners-Hagen et al. The directly detectable working range A for the two wavelengths λ1 and λ2 is.

[0061]

[0062] If not only the total intensity of the synthetic wavelength Λ is considered, but also the two wavelengths λ1 and λ2 themselves are considered separately, the working range A can be extended to multiples of the wavelengths λ1 and λ2 (exact phase fractional method).

[0063] However, the method proposed in the above-mentioned publication cannot be used in this way for height measurement on a wafer, since the intensity of the measured interferometric measurement signal also depends on the material of the structure being measured and its environment. The above-mentioned method according to the invention solves this problem.

[0064] By means of two phase-shifted interferometric measurement signals measured simultaneously, interferometric measurements of different wavelengths can be corrected individually. Material dependencies are thus eliminated. Thus, within a well-defined working range as shown in the above-mentioned publication, the distance of each point recorded with the line sensor arrangement can be calculated individually. Thus, an exact distance image of the surface can be formed. By using a line sensor with a high clock rate, very high measuring speeds can be achieved.

[0065] With the device and system according to the invention, contours with a height of several hundred μm can be sensed with a resolution of less than 0.1 μm. This is particularly sufficient to meet many of the requirements when measuring semiconductor chips and wafers mentioned at the beginning. Applications are, for example, measuring the height of contact structures (solder balls or cylinders), measuring the height of conductor circuits on the uppermost wiring surface of a chip, or determining the flatness on a reconstructed wafer (a wafer obtained by embedding previously separated chips in a carrier material, usually epoxy resin). Similar tasks also occur in other structures and in other industries. These tasks are, for example, the measurement of structures on electronic boards and printed circuit boards, or the measurement of various mechanical-electrical system components (MEMS), such as pressure sensors, speed sensors or gyroscopes.

[0066] With appropriate illumination of the measurement area on the surface of the object and the use of suitable line or surface detectors, this arrangement allows simultaneous measurement at multiple points. Unlike white light interferometry, for example, the measurement result is already obtained from a single measurement at different wavelengths without having to change the distance to the object or the detector. This significantly speeds up the measurement.

[0067] Preferably, a recording device is provided for recording an object in the form of a disk or other flat object with a flat object surface. In the recording device, the flat object to be measured is positioned just in such a way that the object surface is illuminated by the measuring light in the desired manner.

[0068] Lasers can be used as radiation sources, since they provide sufficiently monochromatic light with high radiation intensity. Lasers are particularly suitable as radiation sources for the present invention because they have a high radiation intensity in a small wavelength range. However, one or more spectrally broadband radiation sources can also be used, which are equipped with corresponding spectral filters. Using multiple spectral filters, multiple wavelengths can also be generated from the radiation source.

[0069] In particular, the invention may provide that the radiation source generates light of wavelengths where none of the wavelengths is an integer multiple of the wavelength of any other radiation source. For example, the wavelength is not twice the wavelength of another radiation source. The smaller the difference in wavelengths, the greater the operating range. However, in practice it must be taken into account that the radiation beam is never completely monochromatic and always has a bandwidth which must be taken into account when choosing the wavelength.

[0070] When using different wavelengths, the radiation can be combined using spliced ​​waveguides or dichroic mirrors. Spliced ​​waveguides are waveguides that are fused together at a certain part or are guided close to each other so that the light passes from one waveguide into another. In a dichroic mirror, only the light of a certain wavelength range is reflected, while the light of at least one other wavelength range is transmitted. The dimensions of such mirrors can be set for use at different angles. The most common angle of use is 45°. Thus, in a suitable arrangement, the emerging reflected light and the transmitted light can have the same propagation direction.

[0071] The phase difference of the beam required for the arrangement (device and system) can be caused by the difference in the path covered by the light beam and the transmission or reflection on different media.

[0072] One or more prisms, gratings or other spectrally dispersive elements can be used to separate the beams of different wavelengths. Spatial separation enables the signals to be received simultaneously, which allows for high measurement speeds. However, temporal separation can also be achieved by performing measurements consecutively at different wavelengths, which makes longer measurement times acceptable.

[0073] A particularly preferred configuration of the invention provides that each detector device comprises a multi-channel time delay integration (TDI) line scan camera. These cameras comprise a plurality of TDI blocks, each of which is sensitive to a specific wavelength range. Thus, the TDI arrangement can be used to produce a high signal-to-noise ratio, and at the same time all wavelengths used can be measured simultaneously. This enables particularly fast and very precise measurements. However, in the same way, a single-channel TDI line scan camera can also be used in a so-called surface mode, i.e. like a normal surface scan camera, for which the different wavelengths used are distributed over the sensor area by means of a dispersive element (e.g. a prism). This enables large widths of TDI detection lines, e.g. 16384 pixels, to be used for high throughput, and at the same time 48, 96 or 256 lines of the TDI can be used to simultaneously and individually capture different wavelengths. In this arrangement, the measuring speed is low due to the low clock frequency in this surface readout mode. This also applies to normal surface scan cameras that can also be used.

[0074] In particular, it can be provided that the optical device for generating interference comprises a semi-transparent mirror, which reflects a portion of the radiation (measurement beam) at a first surface in the direction of the object surface and allows another portion of the radiation (reference beam) to pass. The reflected radiation can then be combined with the already transmitted beam. This can be achieved, for example, in that the transmitted beam (reference beam) is reflected back onto itself (as a mirror beam) at a mirror and coupled into the reflected beam (object beam) at a semi-transparent mirror. Such an arrangement is known as a Michelson interferometer. However, other arrangements can also be used, such as the so-called Mach-Zehnder interferometer, in which the irradiated light is split and the sub-beams are recombined after reflection at the sample surface.

[0075] The method according to the invention for sensing the surface profile of an object surface (30) of an object by interferometric distance measurement comprises a plurality of steps, such as emitting monochromatic light by a light source in the direction of an optical device; splitting the light beam of the light source into a first sub-beam and a second sub-beam by a beam splitter; preferably splitting each sub-beam by a beam splitter into a reference beam and a measuring beam; directing the reference beam to a mirror; reflecting the reference beam on the mirror and forming a mirror beam with a phase shift. The mirror beam is directed to the beam splitter and reflected at the beam splitter. The measuring beam is directed to a measuring area on the object surface and reflected there, thereby forming an object beam directed to the beam splitter. In a further step, the mirror beam and the object beam interfere with each other to form an evaluation beam with a phase shift. One step provides for detecting the evaluation beam, preferably by means of a detector unit, preferably by means of a signal evaluation unit, to determine the surface profile of the object surface. In this case, two measured intensity values ​​of the evaluation beam reflected from the same position on the object surface are used to compensate for the influence of the reflectivity of the beam at the object surface when determining the surface profile. In this way, the correct surface profile can be determined, on the one hand, for object surfaces whose reflection is not 100% and, on the other hand, for object surfaces which have different reflection coefficients in the region to be measured.

[0076] Preferably, a monochromatic test image is created for each wavelength using the method based on the determined reflection coefficient of the object surface. The reflection coefficient is calculated from the previously determined height profile and the measured signal intensity of at least one light beam (evaluation beam). However, it is particularly preferred that the signal intensities of both evaluation beams are used.

[0077] In the following, the method is explained using a wafer as an example without limiting the generality. However, the method can also be used in the same way for other planar objects or test objects to measure surface profiles and contours.

[0078] Preferred configurations of the present invention are described above. It is understood that the above-mentioned features and those still to be explained below can be used not only in each given combination, but also in other combinations or individually without departing from the scope of the present invention. In particular, the method of the device and system for sensing the surface profile of the object surface of an object by interferometric distance measurement can be implemented according to the embodiments described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Below, exemplary embodiments of the present invention are described with reference to the accompanying drawings. In the drawings:

[0080] Figure 1 shows a schematic diagram of a system according to the invention for interferometric distance measurement using three laser light sources;

[0081] Figure 2 A schematic diagram showing method steps for setting up the system and interferometric distance measurement;

[0082] Figure 3 by Figure 1 A magnified detail of shows a detailed schematic diagram of the path length covered by the first sub-beam of light and the associated directions;

[0083] Figure 4 by Figure 1 A zoomed-in detail of the image shows a detailed schematic of the path length covered by the second sub-beam of light and the associated directions;

[0084] Figure 5 Shows the Figure 1 A schematic diagram of a first embodiment of a system according to the invention; and

[0085] Figure 6 Shows the Figure 1 Schematic diagram of a second embodiment of the system according to the invention. DETAILED DESCRIPTION

[0086] Figure 1 An embodiment of a system 1 according to the invention is shown, comprising an optical device 2, at least one light source 7, a detector unit 8 and a signal evaluation unit 9. The optical device 2 comprises a beam splitter 3, a beam splitter 4 and a mirror 5. The mirror 5 has two mirror segments 6, which in turn are formed in the form of two mirrors 60, 70. The beam splitter 3 is preferably formed as a beam shaping optical device 80. The beam splitter 4 is preferably a semi-transparent mirror 90, whose semi-transparency is preferably 50%. The at least one light source 7 preferably comprises three laser sources 110, 120, 130, as shown.

[0087] Hereinafter, the structure of the system 1 and the optical path in the system 1 will be described by taking a wafer as an example of a measurement object. In the system 1, a wafer 20 or other planar object whose surface profile is to be sensed moves sequentially relative to a camera (detector unit 8).

[0088] In the currently preferred embodiment, the camera is stationary and the object passes under the camera. For this purpose, a receiving and transport unit 10 is preferably provided which supports and moves the wafer 20. The movement is preferably carried out in a direction perpendicular to the surface normal of the object surface 30 of the wafer 20. In another exemplary embodiment, the camera can move. In an alternative exemplary embodiment, the relative movement can be divided between the camera and the object, so that, for example, the camera preferably performs a movement in the axial direction, while the object can preferably move in a direction perpendicular thereto. The movement between the camera and the object is preferably continuous.

[0089] Figure 1 The exemplary embodiment in FIG. 1 uses three lasers as light sources 7 and two TDI multi-channel line scan cameras (TDI=Time Delay Integration) as sensors or detectors of the detector unit 8. However, other numbers of light sources can also be used. Likewise, a combination of a broadband light source with a narrowband filter can also be used as an alternative. Likewise, a simple multi-channel line sensor (without TDI method) or an area scan camera or a group of several line scan cameras can be used instead of a TDI sensor.

[0090] exist Figure 1 In order to simplify the description, the beam deviation caused by the beam passing through the plane parallel plate is omitted. Figure 1 In the exemplary embodiment of the invention, the light of three laser sources 110, 120 and 130 is combined by a fiber coupler 100. In the fiber coupler 100, the three optical fibers 170, 180, 190 supplied by the laser sources are spliced ​​together so that the light of the laser sources is emitted as a common polychromatic light beam 84. The transmission via the optical conductor is only used to disconnect the laser from the actual filming system. Instead of the optical conductor transmission, the laser beam can also be directly redirected to the filming unit via suitable optical devices, which can reduce losses but requires more adjustment work. Similarly, the combination of the three light beams of the three laser modules can also be achieved by a dichroic mirror.

[0091] By means of the beam shaping optical device 80, the light beam 84 is shaped into a parallel light beam consisting of two beam halves, so-called sub-beams 410 and 420, whose cross section is adapted to the recording plane. The beam halves 410 and 420 are arranged around the beam center axis 400. The beam halves do not necessarily have to be adjacent to each other, that is, the central area between them can remain unused. The splitting is preferably symmetrical, but not necessarily symmetrical. In this case, all wavelength components are distributed uniformly over the entire cross section, in particular between the beam halves (sub-beams) 410 and 420. The light beam (consisting of the sub-beams 410, 420) is partially deflected by a 50% semi-transparent mirror 90 as sub-beams (measurement beams 430 and 440) through a lens 50 to the wafer surface 40 to be measured. The other 50% of the light beam (sub-beams 410, 420) passes through the semi-transparent mirror 90 as sub-beams (reference beams) 470 and 480.

[0092] The reference beams 470 and 480 impinge on the two reference mirrors 60 and 70, where they are reflected. The resulting sub-beams (mirror beams 490 and 500) are reflected again by 50% on the semi-transparent mirror 90.

[0093] The sub-beams diverted to the wafer surface 40, i.e. the measuring beams 430 and 440, are focused on a common imaging area by means of a lens 50. Due to reflection on the wafer surface 40, they are reflected back by means of the lens 50 as sub-beams (object beams) 450 and 460. In this case, the first object beam 450 comprises the reflected light of the first measuring beam 430, and the second object beam 460 comprises the reflected light of the second measuring beam 440. Half of the light from the object beams 450 and 460 respectively passes through a semi-transparent mirror 90. The light from the first object beam 450 then interferes with the light of the first mirror beam 490 reflected on the semi-transparent mirror 90 to form a light beam, i.e. a first evaluation beam 520.

[0094] The intensity of the first evaluation beam 520 is now modulated by interference of the light waves as a function of the distance difference between the semitransparent mirror 90 and the mirror 60 and the distance difference between the semitransparent mirror 90 and the wafer surface 40. The intensity modulation takes place independently of one another for all three included wavelengths.

[0095] Likewise, an interferometric measurement signal is generated by the light of the second object beam 460. After passing through the semi-transparent mirror 90, the light from the second object beam 460 interferes with the light of the second mirror beam 500 reflected on the semi-transparent mirror 90 to form a light beam, i.e., a second evaluation beam 510.

[0096] The intensity of the second evaluation beam 510 is now modulated by mutual interference of the light waves as a function of the distance difference between the semitransparent mirror 90 and the mirror 70 and the distance difference between the semitransparent mirror 90 and the wafer surface 40. The intensity modulation takes place independently of one another for all three included wavelengths. Figure 3 and Figure 4 Shows Figure 1 A zoomed-in detail of the image is shown to illustrate the formation of interference and the path differences that need to be taken into account.

[0097] The two evaluation beams 510 and 520 differ in intensity at the three wavelengths used. The intensity difference at each wavelength itself is completely determined by the phase shift produced on the mirrors 60 and 70. In the exemplary embodiment shown, the phase shift is produced by the different distances of the two mirrors 60 and 70 from the semi-transparent mirror 90. However, the phase shift can also be produced in other ways, for example, by applying a transparent coating to one of the mirror segments 6 or mirror segments (e.g. mirror 70).

[0098] The second evaluation beam 510 is redirected by the mirror 260 as a second detection beam 540 to the prism 270. The prism 270 is used to spectrally split the light from the second detection beam 540 into three wavelength components λ1, λ2 and λ3. Without limiting the generality, other suitable dispersive elements, such as gratings, can be used instead of the prism 270. The spectral components split from the second detection beam 540 in this way are focused by the first barrel optical device 200 onto the TDI multi-channel line scan camera 210. When a TDI multi-channel line scan camera is used and a color filter is fixedly installed in front of the sensor block, the spectral splitting of the beam can also be omitted.

[0099] By recording the intensity of light impinging on each pixel of the line scan camera 210, three signals i 1,1 、i 2,1 and i 3,1 , whose intensity is modulated by the height of the profile at each location on the wafer surface 40, and can therefore be used for interferometric determination of the surface profile. Here, the first index represents the wavelength and the second index represents the camera 210 providing the signal.

[0100] Since the mirror 260 only blocks the path of the second evaluation beam 510, the first evaluation beam 520, which is intensity modulated by interference, hits the prism 250 unimpeded as the first detection beam 530. The prism 250 is used to spectrally split the light from the first detection beam 530 into three wavelength components λ1, λ2 and λ3. Without limiting the generality, other suitable dispersive elements, such as gratings, can also be used instead of the prism 250. The spectral components split from the first detection beam 530 are focused by the second barrel optical device 230 onto the TDI multi-channel line scan camera 220. By recording the light intensity that hits each pixel of the line scan camera 220, three signals i are generated. 1,2 、i 2,2 and i 3,2 , whose intensity is modulated by the surface profile of the wafer surface 40 and can be used for interferometric determination of the surface profile.

[0101] Figure 5 An alternative embodiment of the system according to the invention is shown, in which the splitter mirror 90 and the two mirrors 60, 70 (reference mirrors) are replaced by a prism 600 made of two blocks bonded to one another and provided with a partially reflecting layer on the 45° faces of the bond. This layer forms a semi-transparent mirror 610, which splits the two sub-beams 410, 420, just as according to Figure 1 The details of the individual beams and the optical paths are as follows: Figure 1 For clarity, Figure 5 No longer displayed in.

[0102] according to Figure 1 In the embodiment of the present invention, two mirrors 60, 70 as reference mirrors are arranged on the outer surface of the prism 600. For this purpose, for example, the mirroring part can be vapor-deposited on the outer surface, or the mirror surface can be applied or glued. By appropriately configuring the prism 600, the necessary offset can be generated on the outer surface, so that the mirrors 60, 70 have the same Figure 1 Same offset as shown.

[0103] After the mirror beams 490, 500 and the object beams 450, 460 interfere in the prism 600 and form the evaluation beams 520 and 510, these beams are bundled by the bundling optical element 620 and redirected to the optical element 700. The optical element 700 preferably has a rectangular or square base surface. Mirrored portions or mirrors 730, 740 are arranged on both sides facing the bundling optical element 620, which reflect the impinging beams. The mirrors 730, 740 can also be designed as independent mirrors or reflective elements; however, they must then be aligned or adjusted.

[0104] The impinging beam is deflected at the optical element 700. At the mirrors 730, 740, the two evaluation beams 520 and 510 are separated from each other and directed to the cameras 210, 220. At the mirror 730, the evaluation beam 520 is reflected into a first detection beam 530, which is deflected to the first camera 210 by the deflection mirror 710. At the mirror 740, the second evaluation beam 510 is deflected to the deflection mirror 710 as a second detection beam 540, so that the second detection beam 540 impinges on the second camera 210. The signals received by the cameras 210 and 220 are processed in the signal evaluation unit 9, so that the contour of the wafer surface 40 of the wafer 20 is sensed.

[0105] In the embodiment shown here, the detection beams 530 , 540 are also focused so that they impinge on the cameras 210 , 220 at one point.

[0106] Figure 6 An alternative embodiment of the system 1 according to the invention is shown, which is similar to the embodiment according to Figure 5 The structure of the embodiment shown is corresponding. Figure 5 and Figure 6 The two embodiments differ only in the use of optical elements which redirect the two evaluation beams 510, 520 into detection beams 530, 540. Figure 6 , using a flatter optical element 750, so there is no Figure 5 Right-angle beam guidance is achieved as provided in the embodiments of the present invention. Figure 5 The evaluation beam is deflected by 90° toward two deflection mirrors 710 , 720 , which are arranged at an angle of 45°, so that the detection beams 530 , 540 are directed to the two cameras 210 , 220 after impinging at an angle of 90°.

[0107] according to Figure 6 The optical element 750 and the two mirrors 730 and 740 achieve a non-90° deflection because the mirrors 730 and 740 are arranged at the non-right-angled side surfaces of the rhombus-shaped optical element 750. The two deflection mirrors 710 and 720 are correspondingly arranged to deflect and preferably focus the detection beam to a point on the respective cameras 210 and 220. This achieves a significantly more compact measurement structure, so that the system can be smaller and more compact, and the accuracy of the measurement can also be improved.

[0108] certainly, Figure 5 and Figure 6 The prism 600 can preferably also be used according to Figure 1 In the embodiment of the present invention, the semi-transparent mirror 90 and the mirrors 60 and 70 are replaced. Figure 1 The optical device 2 can also be used according to Figure 5 ,6 In the embodiment of .

[0109] Similarly, according to Figure 1 The detection device includes cameras 210, 220 and prisms 275, 270, barrel optical devices 200, 230 and mirror 260, and the detection device can be Figure 5 , 6 The detection device can be replaced by the beam-forming optical element 620, the optical element 700 or 750, the mirrors 710, 720, 730 and 740 and the cameras 210, 220. Vice versa.

[0110] Signal evaluation

[0111] The following is the signal i obtained for the three wavelengths λ1, λ2 and λ3 x 、r x and q x (x=1, 2, 3) calculates the surface profile for explanation, which is only for one pixel of the camera sensor of the line scan cameras 210 and 220. It will be appreciated that such calculations can be performed for each pixel of the cameras 210, 220. Therefore, depending on the size and arrangement of the sensor, multiple height points of the profile can be ascertained simultaneously.

[0112] For example, for a line scan camera with 16,384 points per row that is currently available, this means that there are 16,384 height values ​​for each readout cycle of the camera. In addition, in the arrangement described here, the line scan cameras 210, 220 can be continuously moved relative to the wafer 20. Depending on the cycle speed of the camera, a number of rows with height information per unit time are obtained accordingly. Therefore, at the clock frequency of the aforementioned multi-channel TDI line scan camera (for example, 100kHz), more than 1600 Mio. height values ​​can be obtained per second. Different manufacturers (such as Vieworks and Dalsa Teledyne) all provide such cameras. The use of such a TDI multi-channel line scan camera is a particularly suitable variant because very high measurement speeds can be achieved. These cameras contain multiple (usually 4) TDI blocks in one camera, which can be operated and read out simultaneously. When using such a camera, the wavelengths used for a row on the wafer surface are photographed in sequence. This means that while the first TDI block records the line range at λ1, the second TDI block detects the signal at λ2 and the third TDI block detects the signal at λ3.

[0113] In principle, this temporal offset is not important for the following calculations. It is only necessary to distribute and evaluate the acquired signal graph in a temporally staged manner (corresponding to the spatial offset of the TDI blocks). With this arrangement, a high signal quality (correspondingly a high resolution and a robust measurement) can be combined particularly well with a high speed. An alternative arrangement will be described below.

[0114] Figure 2 The process of measurement and preparation is explained. In order to correctly determine the height value, the measurement must be prepared by dark signal measurement and determination of the transfer functions of the optical system and sensor system.

[0115] In the dark signal measurement, the signal d is measured at each camera pixel y of the two line scan cameras 210 and 220 when the light source is off. The so-called dark noise of the camera is thus determined, which represents the offset of each further measurement and is subtracted from the signal. This is done for the two line scan cameras (sensor arrangements) 210 (index z=1) and 220 (index z=2).

[0116] d xyz At wavelength x(λ x ) is the signal value read out at pixel y of sensor z at the time (5)

[0117] In order to determine the optical and electrical transfer functions of the arrangement (system 1), a bright signal measurement h is performed with a known planar object. For this purpose, the wafer surface 40 is replaced by a planar reference with known reflection properties. Since the signal h at each sensor pixel is unambiguously determined by the intensity of the light source (signal value q), the transfer function M, the reflectivity r of the reference, the path difference l of the two interfering light beams (the first object beam 450 and the first mirror beam 490, impinging on the sensor (line scan camera) 220, z=2; or the second mirror beam 500 and the second object beam 460, impinging on the sensor (line scan camera) 210, z=1) z The transmission function M can be determined as a function of the path difference for each wavelength x, each pixel y and the two sensors z by determining the dark signal d and the dark signal d, as long as the values ​​of h, q, r and d are known. The transmission function is usually different for each wavelength x, each pixel y and each camera z. The transmission function is determined by the sensitivity of the individual pixels, the illumination through the optics, the material properties, the coatings and the imaging errors.

[0118] In order to check the output intensity of each laser light source 110, 120, 130 and include it in the calculation as a correction value or reference value, the signals q1, q2 and q3 of the monitor diodes 140, 150, 160 which are usually installed in each laser module (on the side of the laser away from the output end) can be used directly.

[0119] The signal h of the bright reference measurement is:

[0120] h xyz,href (l z )= q xy,href *M xyz (l z )*r x,ref +d xyz (6)

[0121] in:

[0122] h xyz,href (l z ) as the path difference l z The measured value of the reference measurement at the wavelength x at pixel y of sensor z (interferometer, line scan cameras 210 [z=1] and 220 [z=2]) as a function of .

[0123] q xyz,href The reference measured irradiance intensity at pixel y at wavelength x.

[0124] M xyz (l z ) as the path difference l z The transmission function of a function of pixel y at sensor z at wavelength x.

[0125] r x,ref Reflectance of a reference measurement (known material) at wavelength x.

[0126] d xyz Dark signal value (no excitation light) read at wavelength x at pixel y of sensor z.

[0127] Note: Although "dark = no light", the dark signal value is "wavelength dependent" because different sensor pixels are used for different wavelengths - they can have different dark value values.

[0128] Here, the parameter l z Represents the path difference between the two interfering beams, namely the first object beam 450 and the first mirror beam 490, impinging on the line scan camera 220, z=2; or the second object beam 460 and the second mirror beam 500, impinging on the line scan camera 210, z=1.

[0129] Figure 3 The ratio of the light beam pair of the first object beam 450 and the first mirror beam 490 is shown. It can be seen that the path difference l2 is exactly 2x(z mirror2 -z sample ), because the path difference z differenzwill cancel itself out, since each sub-beam (the first measuring beam 430 and the second sub-beam 420) passes through the path difference exactly once. mirror2 Here, they are provided with the index 2 because the two reference mirrors 60 , 70 are at different distances from the splitting mirror 90 of the beam splitter 4 , without restricting the general applicability, in order to establish the phase shift according to the invention.

[0130] Similar ratios apply to the light beam pair of the second object beam 460 and the second mirror beam 500, as Figure 4 Here, the two sub-beams (the first sub-beam includes the second sub-beam 420 and the second object beam 460; the second sub-beam includes the second reference beam 480 and the second mirror beam 500) that later interfere with each other just pass through the path difference z differenz 2 times, so the path difference is canceled out again, and the path difference l1 to be considered is exactly equal to 2x(z mirror1 -z sample ).

[0131] In order to determine the relationship between the wafer surface 40 and the sensor arrangement z sample (which corresponds to the distance of the optical device 2), a slightly different approach can be used to define and use the transfer function. The chosen approach is also used for the two sensor line scan cameras 210 and 220.

[0132] On the one hand, the transfer function M(l) can be decomposed into non-interference factors M max and interference effects. The factor M is obtained by determining the transfer function M(l) only at its maximum max In this case, the intensity modulation caused by the interference can be ascertained directly from the values ​​of the measurement run. This first approach will be described in more detail below.

[0133] On the other hand, the intensity modulation caused by interference can be included in the transmission function M(l). In this case, M(l) is ascertained as a function of the path difference of the two beamlets (first and second sub-beam) and the distance to the wafer surface in a reference measurement over the entire operating range. During the measurement, the distance difference to the wafer surface is ascertained by comparing the transmission function value with the bright signal measurement which was performed as a reference. This comparison is performed separately for three sets of wavelengths x=1,2,3 and two sensors z=1,2, and a search is performed for distances that match all three wavelengths and both sensors.

[0134] The transfer function is applicable to sensor line scan cameras 210 and 220 (where the / 2 in the denominator is obtained by differentiating z mirror -z sample 2 times the running time).

[0135]

[0136] in:

[0137] M xyz (l z ) as the path difference l z The transmission function of a function of pixel y at sensor z at wavelength x.

[0138] Maximum value of the transmission function at wavelength x at pixel y of sensor z → structural interference = path difference l z = wavelength λ x An integer multiple of .

[0139] z mirror,z The distance between the splitter mirror 90 and the mirrors 60 and 70 (refer to Figure 4 ).

[0140] z sample The distance between the splitting mirror 90 and the wafer surface 40 (refer to Figure 4 ).

[0141] If the maximum signal value is determined for each of the three wavelengths x = 1, 2, 3, the M for each wavelength can be found max .

[0142]

[0143] in:

[0144] l z f or hmax(x) The path difference when the signal at sensor z is maximum, i.e., l z is the wavelength λ x An integer multiple of .

[0145] Determine M for each wavelength max , for example by continuously changing the distance z of the reference piece sample , so that a complete wavelength cycle is run for all three wavelengths x=1, 2, 3 and two sensors z=1, 2. Here, it is also possible to sample，y The phase shift Δz of the two mirrors 60 and 70 is determined by the position (at which the maximum occurs) r =z mirror1，y -z mirror2，y How big is the y at each pixel.

[0146] In a measurement operation of an unknown wafer surface 40 to be detected, the sensor signals i of the camera sensors 210 [z=1] and 220 [z=2] are now simultaneously received. z1 、i z2 and i z3, and the output intensity of the laser light sources 110 , 120 , 130 is received by means of built-in monitor diodes 140 , 150 , 160 .

[0147] For the signals of the sensors, line scan cameras 210 and 220, the following applies:

[0148] i xyz = q xy *M xyz (l z )*r xy,wafer +d xyz (9)

[0149]

[0150] in:

[0151] i xyz Interferometric measurement value at pixel y of sensor z at wavelength x.

[0152] q xy The intensity of the irradiated light measured at wavelength x at pixel y.

[0153] Maximum value of the transmission function at wavelength x at pixel y of sensor z → Structural interference = path difference l z = wavelength λ x An integer multiple of .

[0154] r xy,wafer The reflectance of wafer 20 at wavelength x at pixel y.

[0155] d xyz Dark signal value (no excitation light) read out at wavelength x at pixel y of sensor z.

[0156] Using the relationship Δz ry =z mirror1,y -z mirror2,y and the transformation Δz y =z mirror1,y -z sample,y , the equations for the two sensors can be simplified to:

[0157]

[0158] and

[0159]

[0160] Due to the profilometry of the wafer surface 40, ie, sensing the surface profile of the wafer, Δ zy It is sufficient, and the distance difference Δz can be obtained from the predicted measurement ryTo find out the transfer function M max , so the pair of signal equations (11), (12) defines for each pixel y a system of 6 equations (for the parameters: x = 1, 2, 3 and z = 1, 2) for the 4 unknown variables:

[0161] – Three reflection coefficients r of the wafer surface xy,wafer 40,

[0162] – Search distance difference Δz y , which gives a profile 40 of the wafer surface.

[0163] By appropriate choice of wavelengths λ1, λ2 and λ3 for x = 1, 2, 3, a working range of 0.5 mm can be easily established, tripling the two intensity measurements i xy1 and i xy2 Explicitly assigned to the path difference Δz y , which is sufficient for various contour measurement tasks. This is mentioned at the beginning by K.Meiners-Hagen, R. It is described in the publication of F. Pollinger and A. Abou-Zeid. In the arrangement disclosed therein, for example, wavelengths of 532 nm, 632 nm and 780 nm are used to achieve a working range of 0.6 mm with a clear distance difference assignment.

[0164] To simplify the mathematical evaluation, the equations can be combined into:

[0165]

[0166] and with

[0167]

[0168] composition:

[0169]

[0170] Therefore, the task for each pixel y is reduced to determining λz from the remaining 3 equations y , the equations must be satisfied simultaneously. By combining, the intensity q of the laser source 1y ,q 2y and q 3y is no longer directly included in the calculation, and the reflection coefficient r of the wafer surface 40 xy,wafer Nor does it have to be clearly identified.

[0171] If Δz y Expressed in units of half wavelength, the Δz for each of the three wavelengths λx is y It can be expressed as the integer part δx and remainder f x The sum is:

[0172]

[0173] By finding out the Δz yx The smallest integer part of the average deviation from the corresponding mean is three times δ x , from Δz yx Find the path difference Δz in the search y .

[0174] Use reference contours to perform height measurements of repetitive structures.

[0175] An important application of height profile measurement shown here is the measurement of the height of contact structures on the wafer. Here, only the height of the structures is relevant in order to ensure that all contact structures actually have electrical contact during the subsequent insertion and contacting of the chip in the housing or during the three-dimensional integration of multiple chips, and that no structure is so large that it causes mechanical damage to the chip during installation. To this end, it should be determined whether the vertices of all contact structures lie in one plane and have sufficiently small tolerances.

[0176] In order to accomplish this task quickly and reliably, the present invention proposes to perform high-resolution reference measurements on several samples of the contact structure, thereby establishing a reference model of the contact structure, and for conventional measurements on similar wafers, only so many measurement points need to be taken to ensure that some of them are located on the upper side of the contact structure.

[0177] Such a reference model can be created using the same profilometry method described above. To this end, in a first embodiment of the invention, the lens 50 can be configured as an interchangeable lens with different magnifications. In this case, reference measurements can be performed with a high magnification. For example, a reference model can be generated from a number of measured contacts by averaging or median calculations. Conventional measurements can be performed with a smaller magnification so that enough measurement points can be identified on the contact surface to enable a reliable comparison with the reference model.

[0178] In another alternative, the lens 50 may be a fixed lens with a fixed magnification. In this case, the reference measurement may be performed by multiple measurements with a small lateral offset. This again provides a denser coverage of measurement points for the contact being measured. The reference model may again be generated from multiple measured contacts by averaging or median calculations.

[0179] The advantage of comparing with a reference model for the pinpoint determination is that the density of measuring points can be selected to be small. It is not always necessary to use the vertices themselves as measuring points. Instead, it is sufficient to use a few points, for example 5 to 10 points on the contact surface. These points can be located at any position on the contact. In this case, the vertex height is determined by matching the model height to the measuring points. This allows working with a smaller density of measuring points and with a higher yield overall, and still obtaining a reliable measurement of the vertex height of the contact.

[0180] For inspection

[0181] After determining the height profile of the surface, the value Δz y is known. Therefore, from the intensity data i of one of the channels xy1 The reflection coefficient r can be found according to formula (11) xywafer The value of .

[0182]

[0183] Thus, by combining the pixel values ​​of all y-positions calculated in this way, the corresponding wavelength λ x Create a monochromatic image of the corresponding wavelength.

[0184] To improve the image quality, the intensity values ​​of two channels (z=1, 2) can also be used according to formula (18), and an average image can be created.

[0185]

[0186] These images can be used for other tasks, such as searching for defects or checking the lateral alignment of the data received.

[0187] Embodiment Variations

[0188] Alternative embodiments are intended to be part of the present invention but do not limit the generality.

[0189] Therefore, in the described invention, using TDI technology is just to improve signal-to-noise ratio, which is not necessary for the principle of the present invention. Therefore, for simpler requirements, the arrangement of three independent line sensor cameras or three independent single-channel TDI line sensor cameras can also be selected to replace the TDI multi-channel camera line sensor (line camera 210 and 220). The further spatial splitting required for this embodiment can be realized by increasing the corresponding distance between prism 250 or 270 and lens barrel optics 230 or 200. For this reason, lens barrel optics 230 or 200 also can be designed to 3 independent optical devices.

[0190] Alternatively, a conventional area scan camera or a TDI single-channel line scan camera operating in an area readout mode may be used. In this arrangement, the clock frequency is correspondingly reduced to, for example, a clock frequency of 1 kHz (for the area readout mode of a TDI single-channel line scan camera), and more than 16 Mio. height values ​​per second can still be obtained.

[0191] It is understood that when an area scan camera or a TDI single-channel line scan camera is used in the area readout mode, the three wavelengths λ1, λ2 and λ3 are imaged on the camera sensor as different lines x=1, 2, 3. The signal patterns obtained when observing the same point on the wafer surface 40 at the three wavelengths are assigned to each other by a spatial assignment of the areas of the camera sensor. For the above-mentioned basic evaluation, it is not important whether the assignment is performed spatially (for an area sensor) or in a time-phased manner (for a line sensor).

[0192] It should be noted that for lateral high-resolution applications (e.g. in the range of a few μm), by using multiple cameras, a lateral and rotational correction of the received signal map is required anyway, since it is difficult to mechanically adjust the entire arrangement to an offset of the cameras of less than 1 μm. This mathematical correction can usually be achieved by receiving a reference pattern from which the exact position on the wafer observed by each pixel can be ascertained.

[0193] In a further embodiment variant, the number of wavelengths used can be adapted to the required working range. For particularly small working ranges, an embodiment with only one wavelength is already possible, or for small ranges, with two wavelengths. With only one wavelength, two "sensor arms" are used according to the invention to ascertain the reflectivity of the sample, so that, in contrast to conventional multi-wavelength interferometry, profiles with alternative or unknown materials can be measured. For larger working ranges or to improve reliability by redundancy of the measurement, an extension to more than 3 wavelengths is appropriate, which can be achieved in particular by means of the above-mentioned multi-block TDI camera (a camera with 7 TDI blocks has already been described).

[0194] In a further embodiment variant, the combination of two interferometric measuring sensors (line scan cameras 210 , 220 ) presented here can also be divided into two consecutive measuring embodiments or implemented in two measuring heads to be used in sequence.

[0195] The illumination can be suitably achieved with a monochromatic light source which radiates continuously. For this purpose, lasers are suitable as well as other broadband beam sources which can be combined with suitable interference filters. The only condition is that the coherence length of the light used is sufficiently large for the working range to be achieved.

[0196] It is sufficient for the method according to the invention to generate a phase shift between the two light bundles, i.e. between the first mirror beam 490 and the second mirror beam 500 of the "reference arm", which phase shift differs from an integer multiple of the wavelength for all wavelengths used. As shown in the embodiment, this can be achieved by spatially offsetting the mirrors 60 and 70. However, it is also possible to provide mirrors with one or two different coatings for generating the phase shift. For example, on an applied transparent layer, the phase shift is formed by the interference of the beams reflected on the front side and the rear side of the layer.

[0197] Likewise, a bonded prism with a square cross section can be used instead of beam splitter 90, which has a partially mirrored layer on the bonded 45° surface. A semi-transparent mirror is thus formed. When using such a prism, the "reference arm" can be produced by vapor deposition of the mirrored portion directly on the outer surface of the prism. By applying two different layer thicknesses or different materials, the desired phase shift can be produced again. Due to the significantly different dispersion of the "reference arm" and the "measurement arm" of the interferometer, this arrangement leads to a reduction in the interference contrast; however, due to its robustness, it is advantageous in any case for simple requirements. Figure 5 and Figure 6 An embodiment with bonded prisms is shown.

Claims

1. An optical device for sensing the surface profile of an object surface (30) of an object by interferometric distance measurement, comprising: A beam splitter (3) for splitting a beam (84) of a light source (7) into a first sub-beam (410) and a second sub-beam (420); A beam splitter (4) for splitting each sub-beam (410, 420) of the light source (7) into a reference beam (470, 480) and a measuring beam (430, 440); a mirror (5) for reflecting the two reference beams (470, 480), the mirror being designed to produce a phase difference between the reflected beams (490, 500); a detector unit (8) for detecting the evaluation light beam (520, 510); and a signal evaluation unit (9) for evaluating the detected evaluation light beam (520, 510) and determining the surface contour; in Each measuring beam (430, 440) is guided to be reflected on a measuring area of ​​the object surface (30) and guided to the beam splitter (4) as an object beam (450, 460) after reflection; Each reference beam (470, 480) is reflected by a mirror and directed to a beam splitter (4) as a reflected beam (490, 500); The object beam (450, 460) and the reflected beam (490, 500) interfere with each other after hitting the beam splitter (4) and are each sent to the detector unit (8) as an evaluation beam (520, 510) for evaluation; The signal evaluation unit (9) is designed to determine the intensity of the two evaluation light beams (520, 510); and The signal evaluation unit (9) is designed to determine the height profile of the surface (30) by using the intensities of the two evaluation beams (520, 510), wherein the influence of the reflectivity of the surface (30) on the measured intensity is compensated by calculation, so that a correct profile is determined for object surfaces (30) with a reflection that is not 100% and for object surfaces (30) with different reflection coefficients in the area to be measured, and The signal evaluation unit is further designed to, from the determined reflection coefficient of the object surface (30), Each wavelength creates a monochromatic inspection image and the reflectivity is calculated from a previously determined height profile and the measured signal intensity of at least one of the two evaluation beams (520, 510), so that only one measurement of the intensity of the two evaluation beams (520, 510) is sufficient for the determination of the height profile of the observed measurement area and for eliminating the influence of the reflectivity.

2. The device according to claim 1, characterized in that The signal evaluation unit (9) is designed to determine the phase difference of the two evaluation beams (520, 510) and to determine the height profile of the surface (30) from the intensities and the phase difference of the two evaluation beams (520, 510).

3. The device according to claim 1, characterized in that The mirror (5) has two parallel mirror segments (6), the mirror surfaces of which are offset in the normal direction.

4. The device according to claim 1, characterized in that The beam splitter (4) and the mirror (5) are integrated in the optical element.

5. The device according to claim 4, characterized in that The optical element is a prism (600) formed from two glass blocks brought close together and two mirrors or mirrors (60, 70) vapor deposited on one side of the glass blocks.

6. The device according to claim 1, characterized in that The beam splitter (4) comprises a semi-transparent mirror (90) which allows a part of the impinging light beam (84, 410, 420) to pass as a reference beam (470, 480) and reflects another part of the impinging light beam as a measuring beam (430, 440) towards the surface of the object.

7. The device according to claim 1, characterized in that The detector unit (8) comprises a time-delay integration camera.

8. The device according to claim 7, characterized in that The detector unit (8) comprises a time-delay-integration color camera or a time-delay-integration multi-channel line scan camera.

9. The device according to claim 6, characterized in that It comprises an optical element for splitting the bundled light beam, the optical element being arranged in such a way that the light beam is split before it reaches the detector unit (8).

10. The device according to claim 1, characterized in that A lens (50) is arranged between the optical device (2) and the object surface (30) of the object for guiding a measuring light beam (430, 440) to a measuring region of the object surface (30).

11. The device according to claim 10, characterized in that The lens (50) is a fixed lens or an interchangeable lens with different magnifications.

12. A system for sensing the surface profile of an object surface (30) of an object by interferometric distance measurement, comprising: A light source (7) for generating a monochromatic light beam (84); as well as An optical device (2) according to any one of claims 1 to 11.

13. The system according to claim 12, characterized in that A drive unit is provided for generating a relative movement between the optical device (2) and the object.

14. The system according to claim 12, characterized in that The light source (7) is a laser.

15. The system according to claim 12, characterized in that A plurality of light sources (7) for generating monochromatic light beams (84) are provided, wherein the light beams (84) of the light sources have different wavelengths, wherein the wavelength of no light beam (84) is an integer multiple of the wavelength of another light beam (84), and wherein the light beams (84) of the light sources are bundled in a beam coupler.

16. A method for sensing the surface profile of an object surface (30) of an object by interferometric distance measurement, comprising the following steps: A light beam (84) consisting of one or more monochromatic light fractions is emitted in the direction of the optical device (2) by means of a light source (7); Splitting a light beam (84) from a light source (7) into a first sub-beam (410) and a second sub-beam (420) by a beam splitter (3); Splitting each sub-beam (410, 420) of the light source (7) into a reference beam (470, 480) and a measuring beam (430, 440) by a beam splitter (4); directing the reference beam (470, 480) to a mirror (5); reflecting the reference beam (470, 480) on a mirror (5) and forming a reflected beam (490, 500) having a phase shift; directing the reflected light beam (490, 500) to a beam splitter (4); reflecting the reflected light beam (490, 500) on the beam splitter (4); directing a measuring beam (430, 440) to a measuring area on a surface (30) of an object; Reflecting the measuring beam (430, 440) on the object surface (30) and forming an object beam (450, 460); directing the object beam (450, 460) to a beam splitter (4); The reflected light beam (490, 500) interferes with the object light beam (450, 460) to form an evaluation light beam (510, 520), wherein the evaluation light beam (510, 520) has a phase difference; detecting the evaluation light beam (510, 520) by means of a detector unit (8); evaluating the detected evaluation beam (510, 520) and determining a surface contour of the object surface (30) by means of a signal evaluation unit (9); Two measured intensity values ​​of an evaluation beam (510, 520) reflected from the same position on the object surface (30) are used to compensate for the influence of the reflectivity of the beam on the object surface (30) when determining the surface profile, thereby For object surfaces (30) whose reflection is not 100% and for object surfaces with different reflection coefficients in the area to be measured, the correct profile is found; A monochromatic inspection image is created for each wavelength from the determined reflection coefficient of the object surface (30), wherein the reflection coefficient is calculated from the previously determined height profile and the measured signal intensity of at least one of the evaluation beams (510, 520), so that the correct profile is determined by only one measurement of the intensity of the evaluation beams (510, 520) without changing the distance to the object.

17. The method according to claim 16, characterized in that There are the following steps: evaluating the detected evaluation beams (510, 520) and determining phase differences; and When determining the surface profile, the influence of the reflectivity of the light beam on the object surface (30) is compensated by the determined phase difference.

18. Method for optically inspecting the surface of an object (30) for defects, wherein the defect assessment is based on a test image ascertained by the method according to claim 16 or 17 and on a subsequent back calculation of the reflection of light on the surface of the object (30), wherein the reflection r xy Obtained through: where i xy = the measured value of wavelength x at pixel y of the sensor; d xy = dark signal value; q xy = measured irradiance intensity; = the maximum value of the transmission function for wavelength x at pixel y of the sensor; and Δz y =z mirror，z -z sample where z mirror，z = the distance from the beam splitter (4) to the mirror (5), z sample = the distance from the beam splitter (4) to the object surface.

Images