WO2005088271A1

WO2005088271A1 - Imaging ellipsometer with a synchronized sample advance and ellipsometric measuring method

Info

Publication number: WO2005088271A1
Application number: PCT/EP2005/002380
Authority: WO
Inventors: Dirk HÖNIG
Original assignee: Nanofilm Technologie Gmbh
Priority date: 2004-03-12
Filing date: 2005-03-07
Publication date: 2005-09-22
Also published as: DE102004012161B3

Abstract

The invention relates to a method and device for the imaging ellipsometeric detection of an, in essence, flatly extending sample array (10). According to the invention, the sample array is, by means of an illuminating arrangement, illuminated at least in areas and at an illuminating angle (ϕ) diagonal to the sample normal with illuminating light having adjustable polarizing properties. Detection light, which is reflected by illuminated areas of the sample (10), in a polarization-sensitive detection arrangement, which is situated at an illuminating angle (ϕ) diagonal to the sample normal, is projected onto a detector (12), which can be read out in a controlled manner and which has a number of ordered photosensitive detector elements. Parameters of the projection of sample areas onto the detector (12) are varied. At least one read-out area (12a; 12b) of the detector (12) is read out synchronous to the variation of the imaging parameters in order to generate images representative of sample areas.

Description

Imaging egg lipometer with synchronized sample feed and ellipsometric measuring method

description

The invention relates to a method for imaging, ellipsometric detection of a sample arrangement that is mainly extended over a large area, in which the sample assembly is illuminated at least in regions by means of an illumination arrangement with polarization properties that can be set with illuminating light under an illumination angle inclined to the sample normal, detection light reflected from illuminated areas of the specimen in an under a suitable polarization-sensitive detection arrangement arranged at an inclination angle normal to the sample is imaged on a controlled readable detector with a plurality of ordered, photosensitive detector elements, parameters of imaging sample areas on the detector are varied and at least one readout area of the detector is synchronized with the variation of the imaging parameters is read out to generate sample areas. The invention further relates to an imaging egg lipometer for imaging, ellipsometric detection of a sample that is mainly extended over a large area, comprising an illumination arrangement for illuminating the sample at least in regions with illuminating light selectable polarization properties under an illumination angle inclined to the sample normal, one at a suitable detection angle inclined to the sample normal Arranged detection arrangement for polarization-sensitive, imaging detection of detection light which is reflected by illuminated areas of the sample, comprising a controlled readable detector with a plurality of photosensitive, ordered detector elements, variation means for varying parameters of imaging sample areas on the detector and control means for controlling the Reading out at least one reading area of the detector in synchronism with the variation of the imaging parameters.

Finally, the invention relates to a quality assurance module for the detection of manufacturing defects in a product manufacturing device.

Ellipsometry has established itself as a powerful measuring principle for imaging very thin and in particular transparent samples. The basic principle of known ellipsometric methods can be explained most simply using the schematic sketch in FIG. 1. A light source L with associated optics is generated a light beam that illuminates a sample arrangement at least in regions at an angle, φ. The

Angle φ is usually measured against the sample standard and can be selected with conventional devices. In addition to the light source and suitable light guide optics, e.g. a suitable lens and / or mirror system, the lighting arrangement comprises polarization-relevant elements with which the polarization properties of the

Illumination light can be modified. In the diagram shown, a polarizer P and a compensator C, which can be designed, for example, as a quarter-wave plate, are used. The sample arrangement is therefore illuminated at least in regions with illuminating light of known polarization properties. A detection arrangement for detecting detection light is also provided at an angle to the sample normal. Since light is generally used as the detection light from the illuminated sample areas, the angle of the detection arrangement to the sample normal, following the law of reflection of the beam optics, generally also corresponds to the angle φ. Detection light is directed to a photosensitive detector D via suitable light guide optics and further polarization-relevant elements (in the diagram shown an analyzer A). The intensity detected on the detector D depends, among other things, on the relative setting of the polarization-relevant elements to one another. The choice of the angle φ also influences the detected intensity.

Imaging and non-imaging applications of ellipsometry are known. While at the non-imaging If the ellipsometry optics and detector are aligned with the illumination of a single point or the detection of reflection light from this point, larger sample areas are simultaneously illuminated in the imaging ellipsometry and an imaging optic 0 is provided which assigns individual points of the illuminated area to corresponding photosensitive detector elements. from which the detector is constructed in an orderly manner.

Two basic principles are known in particular for carrying out an ellipsometric measurement. For example, several individual measurements with different relative settings of the polarization-relevant elements to one another and / or different settings of the

Angle φ performed and values of interest are derived according to the physical relationships known to the person skilled in the art from the resulting sequence of measurement value units (individual measurement values in the case of non-imaging or individual images in the case of imaging application). In contrast, in so-called zero-ellipsometry, which is used in particular in imaging applications, the polarization-relevant elements are set relative to one another in such a way that certain characteristic values of a result image, for example a contrast between two imaged sample areas, assume a special value. For example, when inspecting so-called micro-arrays or biochips in which the quality of the application of a pattern of biomolecules on a carrier substrate is to be checked, it is known to adjust the polarization-relevant elements relative to one another in such a way that the effectively usable contrast between molecular-occupied areas and free areas of the carrier substrate is maximized. Erroneous forms of molecule-occupied areas are then particularly easy to identify and in particular without carrying out a complex, multi-step process.

Since the sample arrangement is illuminated and observed at an angle deviating from the sample normal in all ellipsometric methods, the unavoidable difficulty arises in imaging applications that neighboring sample areas, which can be illuminated simultaneously and can be detected by the detector at the same time, have different distances from the detector. To enable a "sharp" imaging of all areas on the detector, imaging optics with extreme depth of field would have to be used in the detection arrangement. This cannot be generally reconciled with the other requirements for light intensity, high spatial resolution and flexible choice of the observation angle This problem is therefore usually avoided by successively reading out a plurality of adjacent readout areas of the detector in order to generate partial images of results, the imaging optics being refocused before each readout of a partial image, so that the sample area that is focused is “focused” on the detector readout area to be read out next is mapped. This has the disadvantage that the corresponding imaging optics have to be mechanically complex. This procedure is also difficult to automate for the acquisition of larger sample areas. A method and a device for imaging ellipsometry are known from WO 96/24034 A1. In particular, an oscillating movement of the sample is provided during the integration time of the imaging detector in order to achieve an averaging of a speckle pattern, which arises from the interference of partial beams of the illuminating light scattered on the sample with one another. The known device and the known method are not suitable for solving the above-explained problems of automatically recording large image areas with a consistently high image sharpness, since only a single sample area with blurred edges is recorded. On the contrary, the sample movement during the integration time of the detector leads to a strong "blurring" of the generated image and thus to a clear decrease in the image sharpness.

DE 694 15 641 T2 (EP 0 652 415 B1) discloses a scanning ellipsometry device in which an illuminating beam from beam deflectors is guided over the sample to be examined. Disadvantages here are the high mechanical outlay, which has to be done for an exact beam deflection, and the high optical outlay, for correct focusing of the illuminating beam moving over the sample and for "sharp" imaging of the likewise moving, reflected beam onto a two-dimensional one Detector is required.

US 6,052,188 discloses a non-imaging, ellipsometric spectral analysis device in which a two-dimensional detector for recording polarimetric and spectrometrically resolved result arrays are used for a single sample point.

It is an object of the present invention to develop a generic method in such a way that automation, in particular for industrial use, is possible simply and inexpensively.

It is a further object of the present invention to further develop a generic device in such a way that overall result images built up from partial images can be generated in an automated manner, in particular if the surface to be imaged exceeds the image field of the detector.

The first-mentioned object is achieved in connection with the features of the preamble of claim 1 in that the sample arrangement is moved in a motorized manner relative to the detection arrangement, different sample areas being mapped in succession onto the same detector readout area, the detector readout area synchronously with the movement of the sample is read out, and result partial images read out successively from the detector read-out area are combined to form a result image.

This principle of the method according to the invention is completely detached from the previously pursued idea of "focusing" different image areas. Instead, the sample itself is read out with the detector proceed synchronously. This has the advantage that the imaging optics can be designed in a correspondingly simple manner and can be optimized in terms of optically relevant boundary conditions.

The sample arrangement is advantageously moved linearly and perpendicularly to the sample normal and parallel to the plane of incidence of the illuminating light. The movement can take place continuously or step by step. Continuous movement of the sample arrangement lends itself to applications in which the time factor is critical. However, it can only be implemented if the lighting conditions allow a sufficiently short integration time of the detector, so that despite the continuous movement, it is possible to generate sufficiently sharp partial images. A step-by-step movement of the sample arrangement, on the other hand, enables longer integration times of the detector without the risk of “smearing” of partial images.

The polarization properties of the illuminating light and the polarization-sensitive detection arrangement are particularly preferably set relative to one another at the beginning of the method in such a way that a result image of a reference sample arrangement with respect to at least one image characteristic value shows an extreme value. This image characteristic value is preferably a contrast between two selected sample areas. This embodiment of the method according to the invention can be used in particular in so-called zero-ellipsometry. Here, the settings can be made, for example, in such a way that there is no or one between two adjacent areas of the reference sample minimal contrast is recognizable, so that the occurrence of a contrast between equivalent regions of a sample to be measured characterizes a deviation from a desired value. On the other hand, the setting can also be made in such a way that a maximum contrast arises between two sample areas of a reference sample, so that, for example, certain shapes of sample areas become particularly well recognizable during the actual measurement. The result image, which is preferably displayed to the user via a display device, advantageously forms a kind of continuous window. This means that a result image made up of result partial images corresponds in each case to the previously built result image without its first read partial image and supplemented by a newly read result partial image. If the resulting sequence of result images, which can also be stored digitally and / or analogously, is displayed, the result is the impression of an observation window through which the sample is passed.

The synchronization of motorized sample feed and readout of the detector readout area can be guaranteed in different ways. In the simplest case, known movement data, for example speed, duration of a movement step, feed distance during a movement step, etc., are used in order to suitably control the detector reading. It can be cheaper, albeit technically more complex, to generate trigger signals from the actual sample movement by means of suitable sensors for triggering the detector reading. The problem with both variants can be that the Readout control is based only on mechanical movement parameters. However, it may be that, for example, by unevenness of the substrate, undesirable optical deviation ^'s added. For example, a wave in the substrate would result in a change in the reflection angle, which would result in a shift in the imaging area of the sample on the detector due to the lack of corresponding adaptation of the detection beam path.

In a particularly favorable embodiment of the method according to the invention, it is therefore provided that an interest region (ROI: region of interest) is defined in a reference image on the basis of image information parameters and its position and / or shape is determined in a first partial image, in a second partial image after a shift the position and / or shape of the ROI is determined on the basis of the same image information parameters of the sample, the second partial image is corrected at least in part by translation, rotation, compression and / or stretching in such a way that the position and / or shape of the ROI in the second partial image also the position and or shape of the ROI in the first partial image matches, and position and / or shape-corrected partial images or sections thereof are used to construct the result image.

The result of this measure is that the individual partial images or sections thereof are corrected from themselves, ie from image information contained in them can, so that the optical errors mentioned above are compensated in a result image constructed from them. The synchronization between reading and sample movement can therefore be carried out in a simple manner on the basis of known or measured purely mechanical parameters. The image information parameters that are used for correction can include contrast, intensity, color and / or shape information. For example, certain markings on the substrate, whose setpoints (eg position, shape, size, etc.) are known, can be used as correction markers. Structures of the sample itself can also be used as correction markers. In particular, the structure of the illuminating light, for example an intensity maximum of a laser beam, can also be used for the correction. ROIs are particularly relevant as relevant sections of individual drawing files. The reference image used to define the ROI can be identical to the first partial image or can be a separately recorded image.

A particularly preferred application of the method according to the invention is in the field of quality assurance, in particular in the production of micro-arrays or biochips. It is preferably provided that sample areas to be recorded in the future are prepared with the detection of sample areas and result images are subjected to an automated or visual check and, in the event of deviations from predetermined target values, parameters of the sample preparation are varied accordingly. In the case of the production of micro-arrays or biochips, these are purely exemplary and in no way As a limitation, this can be implemented, for example, by feeding substrates, which are mechanically coated with patterns of biomolecules, comparatively shortly after their preparation to a device suitable for carrying out the method according to the invention. The resulting result images can then be inspected by a human quality controller or automatically by using suitable image processing algorithms (e.g.

Pattern recognition algorithms) can be compared with specified target values (e.g. shape, contrast, intensity, etc.). Deviations from the target values indicate errors in sample preparation. These can be corrected manually or automatically for those samples that are being prepared. The success of the correction can be checked shortly thereafter if the test method according to the invention explained above is then applied to these samples. The shorter the time between sample preparation and review, the smaller the amount of rejects that can be kept.

If a correction of defective samples is not possible or too complex due to the specific production process, the described ellipsometric analysis gives the possibility to determine a correction factor that can be taken into account in the intended use of the sample and thereby improves the quality of the result. In the case of the micro arrays, this could, for example, be the specification of a

Area occupancy factor, which is used, for example, as a weighting factor in a fluorometric analysis. On An element of a microarray that was incompletely occupied with biomolecules would therefore have a weighting factor between 0 and 1, for example, while an element that was occupied in excess would be assigned a weighting factor> 1.

The second above-mentioned object is achieved in connection with the features of the preamble of claim 8 in that the variation means comprise a movable sample holder with which the sample can be moved in such a way that different sample areas are mapped in succession onto the same detector read-out area, the control means in this way are set up in such a way that the detector read-out area can be read out synchronously with the movement of the sample, and image processing means are provided and set up in such a way that result partial images read out successively from the detector read-out area can be combined to form a result image.

To avoid repetition, with regard to the effects and advantages of these measures, reference is made to the effects and advantages of the corresponding procedural measures explained above.

A sensor is advantageously provided, via which the control means receive information regarding the current positioning of the sample. Such information can be used to generate trigger signals for the detector reading and thus for the synchronization of sample movement and detector reading. Come as sensors optical, mechanical, magnetic and / or electrical sensors in question.

The motor-movable sample carrier is preferably designed as a conveyor belt, which can move the sample linearly perpendicular to the sample normal and parallel to the plane of incidence of the illuminating light, continuously or stepwise. Such a conveyor belt, which can be designed as a real belt, but also as a grid, net, chain or the like, is particularly suitable for use in larger devices in which samples are prepared in one area and in another area, which they are conveyed with the aid of the conveyor belt can be supplied, their quality can be checked.

In an alternative embodiment, the sample carrier is designed as a rotary table, which can move the sample azimuthally to the sample normal, also continuously or step by step.

The lighting arrangement advantageously includes beam shaping means in order to match the shape of the illuminating light to the shape of the sample areas of interest. Adjacent sample areas from which disturbing reflection light could be imaged on the same detector area are not illuminated, so that no interference light can arise. Another advantage of this embodiment is that the radiation exposure to the sample can also be reduced and the image contrast that can be achieved can be improved. Sample areas that are not clearly focused on the detector at a time are also not illuminated, so that the illumination of the sample area of interest, for example with regard to the contrast that can be achieved, can be optimized, that is to say maximized, as a rule, without other radiation areas, possibly later to be measured, being affected by the radiation.

In the case of a desired strip-shaped illumination, this can be a cylindrical lens or a corresponding mirror optic, through which the illuminating light can be directed onto a strip-shaped sample area which is in particular perpendicular to the plane of incidence of the illuminating light. As is known to the person skilled in the field of optics, a cylindrical lens or a corresponding mirror optic has a focal line instead of a focal point. In connection with the device according to the invention, it is therefore suitable for reducing the illuminated sample area to a narrow strip perpendicular to the direction of movement. High intensities (and thus high image contrasts) can be achieved within this strip without unnecessarily burdening other sample areas by radiation. However, the use of cylindrical lenses or corresponding mirror optics for imaging processes is usually prohibited. In combination with the sample movement according to the invention and the synchronization with the detector readout, however, the advantages of precise illumination by the cylinder optics can also be used in the context of an imaging method.

It is particularly expedient if the detector is a line detector which is extended perpendicularly to the direction of movement of the sample and in which a plurality of adjacent detector elements can be defined as a strip-shaped detector reading area. Since only a narrow strip is illuminated due to the cylinder optics, the corresponding detector readout area can also be designed as a narrow strip. The alignment of the line detector is preferably adapted perpendicularly to the failure level of the detection light or, in the case of deflection optics in the detection arrangement, accordingly.

However, it is also possible to design the detector as an area detector, of which a strip-shaped region of adjustable width or adjustable length and width extending perpendicular to the direction of movement of the sample can be defined as the detector read-out region. For such a configuration, there are significant simplifications in terms of focusing the system and advantages in the flexibility of the selection of the readout area. In addition, when evaluating several lines at the same time, the sensitivity of the system increases due to a correspondingly possible longer exposure time.

The cylinder optics are advantageously arranged such that the illuminating light has a beam waist in the sample plane. This is particularly advantageous if a laser beam with a Gaussian beam profile is used as the source of the illuminating light, since an essentially flat wavefront is then realized in the sample plane.

It is preferably provided that the lighting arrangement has a light source and beam expansion means arranged in front of the cylinder optics for generating an expanded, parallel light beam. For example, a laser can be used as the light source, the narrow beam of which is initially widened in both dimensions perpendicular to the beam direction by means of radially symmetrical beam expansion optics and then focused by the cylinder optics in one of these dimensions, while the expansion is retained in the other dimension. However, it is also possible to use a further cylinder optic as beam expansion means, which expands the beam of the light source in only one dimension. This widening is maintained by the focusing cylinder optics, while the beam tapering according to the invention takes place perpendicular to it. In this case, the focusing cylinder optics preferably have a long focal length.

In an advantageous embodiment of the invention, the lighting arrangement has polarization-influencing light guide means, for example a polarizer and a compensator, such as a quarter-wave plate, which are arranged between the beam expansion means and the cylinder optics. In this way, the polarization properties of the illuminating light can be set particularly precisely, since subsequent disturbances in the polarization properties by the beam expansion optics are excluded. On the other hand, the polarization-influencing elements in this embodiment must have a sufficiently large diameter to allow the expanded beam to pass. However, large diameter polarizer crystals are relatively expensive. In an alternative embodiment, it is therefore provided that the lighting arrangement has polarization-influencing light guide means which are arranged between the light source and the beam expansion means. These only have to have a small diameter and are therefore comparatively cheap. The following

However, beam expansion optics should then be selected very carefully in order to minimize the above-mentioned disturbances in the polarization properties.

Further features and advantages of the device according to the invention and of the method according to the invention result from the following special description and the drawings, in which

FIG. 1 shows a schematic sketch of the structure of an imaging egg lipometer according to the prior art,

FIG. 2 shows a schematic sketch of the detection beam path of an egg lipometer according to the prior art,

FIG. 3 shows a schematic sketch of the detection beam path of an egg lipometer according to the invention,

FIG. 4 schematically shows the construction of a result image from partial result images according to the invention and Figure 5 shows the schematic structure of a particularly preferred embodiment of an ellipsometer according to the invention.

FIG. 6 shows a schematic illustration of the beam paths of an embodiment of an ellipsometer according to the invention with a cylindrical lens,

FIG. 7 shows a schematic top and side view of the illumination beam path of the ellipsometer from FIG. 6,

FIG. 8 shows a schematic representation of the superposition of desired and interference reflections from adjacent beams,

FIG. 9 shows a schematic illustration of a beam superimposition when a parallel beam is incident and

FIG. 10 shows a schematic illustration of a beam superimposition when a beam of rays is incident with a beam waist in the sample plane.

The basic structure of an ellipsometer has already been explained with reference to FIG. 1 in the general part of this description. In order to avoid repetitions, a new description of FIG. 1 is omitted. Figure 2 shows schematically the structure of the

Detection beam path of an ellipsometer according to the prior art. Note that to simplify the figure, the polarization-relevant optical elements, such as an analyzer, have been omitted. A sample 10 is shown, from which reflected light is imaged onto an imaging surface detector 12 by means of optics 11. Because of the

Angle φ, at which the optical axis of the detection beam path is inclined relative to the sample normal, different areas of the sample 10 have different distances from the detector 12. It is therefore not possible to adjust the imaging optics 11 while maintaining essential optical properties, such as the light intensity that all areas of the sample 10 are “sharply” imaged on the detector 12. The apparent flight arrangement known from photography is only suitable for small magnifications, since for the frequently required high resolution and magnification, an extreme tilt of the detector and thus distortion Usually, this is done by first adjusting the imaging optics 11 such that a first sample area 10a is focused on the area 12a of the detector 12 assigned to it. With this setting, the remaining sample areas become "unsharp" on the others detector areas shown. With these settings, a first image is read out of the detector. The optics 11 are then readjusted (shown in broken lines as imaging optics 11 ′), so that a second sample area 10b is sharply imaged on a second detector area 12b read an image from the detector again. This is done as often as necessary to sharply image all sample areas at least once on a corresponding detector area. Subsequently, the "sharp" sub-areas have to be cut out of the recorded sequence of individual images and put together to form a result image.

As shown schematically in FIG. 3 while retaining the reference numerals, according to the invention, after an initial adjustment of the imaging optics 11 in such a way that a sample area 10a, the sample 10 is “sharply” imaged on the corresponding detector area 12a, no further settings on the imaging optics 11 are provided This can be correspondingly simple or optimized with regard to optical parameters other than the adjustability or depth of field. After the detector has been read out once, the sample 10 is moved according to the invention, for example in the direction of the movement arrow 13, so that a sample area adjacent to the sample area 10a is “sharp”. is imaged on the same detector area 12a. If the detector area 12a onto which the “sharp imaging is carried out” is initially suitably defined, the “sharp” sub-areas of the resulting individual images can be cut out and put together to form a larger-area result image. It is even cheaper to use only the “sharp” image "To read out the imaging area of the detected detector area alone and to dispense with reading out the other detector areas. This can accelerate data acquisition and reduce the amount of data to be saved. 005/088271

4 schematically shows the composition of a result image 14 from a plurality of partial images 15a to 15f. The partial images 15a to 15f each correspond to the data read out from the “sharp” detector area 12a. In the exemplary embodiment shown, each partial image comprises only a single line of pixels. Of course, other definitions or drawing file dimensions are also possible.

In particular when the device according to the invention is used in a quality assurance device with visual monitoring, the result image 14, e.g. displayed on a monitor. As also indicated in FIG. 4, the result image structure can be carried out “continuously”. Thus, each result image 14 shown can each have the same dimensions, ie be constructed from the same number of partial images. The partial images contributing to the result image 14 can, as shown in FIG. 4, each correspond to those partial images which contributed to the construction of the previous result image, but without the oldest partial image 15x and supplemented by a recently recorded partial image .15y.This gives the viewer the impression of a continuous result image 14.

FIG. 5 shows a particularly favorable embodiment of an ellipsometer according to the invention. A so-called micro-array or a biochip is schematically indicated as sample 10. The detection beam path essentially corresponds to the beam path indicated in FIG. 3, with here too Simplification of the drawing, the representation of the polarization-relevant optical elements has been omitted. As a special feature of the embodiment shown in FIG. 5, the detector 12 is designed as a line detector. Only one illumination optics 16 of the illumination beam path of the ellipsometer is shown. The polarization-relevant elements, such as the polarizer and compensator, and the light source were not shown in the illumination beam path either. As a special feature of the embodiment shown in FIG. 5, the illumination optics 16 is shown as a cylindrical lens. A widened parallel laser beam with a Gaussian beam profile is preferably coupled into this cylindrical lens. Since the cylindrical lens has a focal line instead of a focal point, this arrangement leads to a strip-shaped illumination of the sample 10. The illuminated sample area is shown as the sample area 10a. The cylindrical lens 16 is advantageously selected and arranged such that, as indicated in FIG. 5, the illuminating beam has a beam waist in the sample plane. This leads to flat wavefronts in the sample plane and thus to a uniform distribution of the polarization properties of the illuminating light. According to the invention, the sample is moved by a motor, as indicated by the movement arrow 13, so that all sample areas are illuminated one after the other and imaged on the detector 12. As shown in FIG. 5, the sample 10 is advantageously moved perpendicular to the longitudinal extent of the strip-shaped illumination area. FIG. 6 shows the particularly favorable embodiment of an ellipsometer according to the invention with a cylindrical lens from FIG. 5 in more detail. The detection beam path, ie the elements between the light source 21 and the sample arrangement 45 are shown again in FIG. 7, FIG. 7 a showing a plan view of the beam path and FIG. 7b shows a side view as in FIG. 6. The light source 21, which is preferably designed as a laser, emits an illumination beam 40, which preferably has a Gaussian beam profile. The beam 40 is expanded perpendicular to its plane of incidence onto the substrate 50 (paper plane in FIG. 6) by means of a beam expansion optic, which in the exemplary embodiment shown is made up of two cylindrical lenses 23, 24 (see FIG. 7a). The polarization properties of the illumination beam 40 can be adjusted by means of an analyzer 25 and a compensator 26. Another cylindrical lens 27, which is aligned perpendicular to the expansion lenses 23, 24, focuses the illumination beam 40 on the sample, preferably in such a way that it forms a beam waist in the sample plane. This leads to a strip-shaped illumination of the sample, as indicated in the rectangular field in FIG. 6, which represents a schematic top view of the sample arrangement 50. The reflected light 40 ′ is imaged on an imaging detector 30 by means of imaging optics 28, which can have a usual radial symmetry. To carry out ellipsometric measurements, an analyzer 29 is also arranged in the detection beam path. The movement arrow 60 indicates the sample movement according to the invention which is synchronized with the reading out of the detector. FIG. 8 schematically shows the problem of superimposing adjacent beams, which results from large-area lighting. This problem can be avoided by the lighting described above using a cylindrical lens. The sample arrangement, of which only a transparent carrier substrate 50 is shown in FIG. 8, is illuminated with an illuminating beam 40 which, according to conventional technology, illuminates a large sample area in parallel. FIG. 8 shows parallel partial beams 41, 42, 43, 44, 45 adjacent to the illumination beam 40. A portion of each beam 41-45 is applied directly to an upper surface 51 of the substrate 50 (or to the actual sample, provided that this is applied to the surface 51, for example as a thin biomolecule layer) as a reflection beam 41'-45 ' reflects the detector. Another part of each beam 41-45 penetrates the upper surface 51 of the substrate 50 and is reflected on the lower surface 52 of the substrate 50 (or on the actual sample if this is applied to the surface 52) (41 '' -45 ''). After passing through the upper surface 51 of the substrate 50 again, this light also approaches the detector as indirectly reflected rays 41 '''-45'''. The indirectly reflected rays 41 '''-45''' overlap with the respectively directly reflected rays 41'-45 '. On the detector, the overlapping steels fall on the same detector elements, so that they contribute indiscriminately to the resulting signal. Depending on which of the surfaces 51, 52 of the substrate 50 the sample layer to be measured is applied, the directly reflected portions 41'-45 'or the indirectly reflected portions Shares 41 '''-45''' can be regarded as a desired signal or as a disturbance component.

FIG. 9 shows the case in which the transparent carrier substrate 50 is so thin that, despite the lighting according to the invention, an overlap region 46 results in a cylinder optic in which directly and indirectly reflected light falls on the same detector elements, although a clear separation is possible in other regions , The overlap area 46 is shown hatched in FIG. FIG. 9 is based on the case that the sample is illuminated with an essentially parallel light beam.

FIG. 10 shows the case of particularly advantageous illumination of the sample with a light beam with a pronounced beam waist in the sample plane. Such lighting can be implemented, for example, with the arrangement in FIG. 6. By widening a beam, in particular with a Gaussian beam profile, and then focusing it in the sample plane by means of a cylindrical lens, effective beam tapering can be achieved in one dimension in the sample plane. This has the effect that there is no overlap of directly and indirectly reflected light in the sample plane. The hatched overlap region 37, on the other hand, is harmless, since the beams overlapping one another have different directions, so that they can be separated by the imaging optics and directed to different areas of the detector. Of course, the embodiments of the invention shown in the figures and explained in the special part of the description are only illustrative examples. In particular with regard to the specific sequence of measurement and evaluation steps and the specific design of the individual components of the device according to the invention, the person skilled in the art has a wide range of possible variations.

Claims

claims

1. A method for imaging, ellipsometric detection of a sample arrangement (10) which is mainly extended over a large area, in which the sample arrangement is illuminated at least in regions by means of an illumination arrangement with polarization properties which can be adjusted with illuminating light and under an illumination angle (φ) inclined to the sample normal, from illuminated areas of the sample (10). reflected detection light in a polarization-sensitive detection arrangement arranged at a suitable illumination angle (φ) inclined to the sample normal is imaged on a controlled readable detector (12) with a plurality of ordered, photosensitive detector elements, parameters of imaging sample areas on the detector (12) are varied and at least one read-out area (12a; 12b) of the detector (12) is read out in a manner synchronous with the variation of the imaging parameters for generating sample areas characterized in that the sample arrangement (10) is moved in a motorized manner relative to the detection arrangement, different sample areas (10a; 10b) are mapped onto the same detector readout area (12a), the detector readout area (12a) is read out synchronously with the movement of the sample (10), and result partial images (15a-15f.) Read out successively from the detector readout area (12a) , 15x, 15y) to form a result image (14).

2. The method according to claim 1, characterized in that the sample (10) is moved linearly perpendicular to the sample normal and parallel to the plane of incidence of the illuminating light, continuously or stepwise.

3. The method according to any one of the preceding claims, characterized in that the polarization properties of the illuminating light and polarization-sensitive elements (A) of the detection arrangement are set relative to one another at the beginning of a method section such that a result image of a reference sample arrangement with respect to at least one image characteristic value shows an extreme value ,

4. The method according to claim 3, characterized in that the relevant image characteristic is a contrast between two selected sample areas.

5. The method according to any one of the preceding claims, characterized in that a result image made up of result partial images in each case corresponds to the previously created result image without its first read partial image and supplemented by a newly read result partial image.

6. The method according to any one of the preceding claims, characterized in that an interest region (ROI: region of interest) is defined in a reference image on the basis of image information parameters and its position and / or shape is determined in a first partial image, the position and in a second partial image / or shape of the ROI is determined on the basis of the same image information parameters, the second partial image is corrected at least in part by translation, rotation, compression and / or stretching such that the position and / or shape of the ROI in the second partial image with the position and / or The shape of the ROI in the first partial image matches, and position and / or shape-corrected partial images or sections thereof are used to build up the result image.

7. The method according to any one of the preceding claims, characterized in that sample areas to be recorded in the future are prepared with the detection of sample areas and result images are subjected to an automated or visual check and in the event of deviations compared to predetermined target values, parameters of the sample preparation can be varied accordingly.

8. Imaging egg lipometer for imaging, ellipsometric detection of a sample (10) which is mainly extended over a large area, comprising an illumination arrangement for illuminating the sample (10) at least in areas with illuminating light selectable polarization properties under an illumination angle (φ) inclined to the sample normal, one under a suitable one of the sample normal inclined detection angle (φ) arranged detection arrangement for polarization-sensitive, imaging detection of detection light which is reflected by illuminated areas of the sample (10), comprising a controlled readable detector (12) with a plurality of photosensitive, ordered detector elements, variation means for varying Parameters of the imaging of sample areas on the detector (12) and control means for controlling the reading out of at least one reading area (12a; 12b) of the detector (12) in a manner synchronous with the variation of the imaging parameters, since characterized in that the variation means comprise a movable sample carrier with which the sample can be moved in such a way that different sample areas are in succession the same detector readout area (12a) are mapped, the control means are set up in such a way that the detector readout area (12a) can be read out synchronously with the movement of the sample (10), and image processing means are provided and set up in such a way that the detector Read-out area (12a) read-out partial images (15a-15f, 15x, 15y) can be combined to form a result image (14).

9. Eilipsometer according to claim 8, characterized in that at least one sensor is provided, via which the control means receive information regarding the current positioning of the sample.

10. Eilipsometer according to one of claims 8 or 9, characterized in that the motor-movable sample carrier is designed as a conveyor belt, which can move the sample linearly perpendicular to the sample normal and parallel to the plane of incidence of the illuminating light, continuously or stepwise.

11. Eilipsometer according to one of claims 8 or 9, characterized in that the motor-movable sample carrier is formed on a rotary table which can move the sample azimuthally to the sample normal, either continuously or stepwise.

12. Eilipsometer according to one of claims 8 to 11, characterized in that the lighting arrangement comprises beam shaping means for illuminating sample areas of interest in the correct shape.

13. Eilipsometer according to claim 12, characterized in that the lighting arrangement comprises a cylindrical lens (16; 27) or a corresponding mirror optic through which the illuminating light can be guided onto a strip-shaped sample area (10a).

14. Eilipsometer according to one of claims 8 to 13, characterized in that the detector (12; 30) is designed as a line detector extended perpendicular to the direction of movement of the sample, in which a plurality of adjacent detector elements as a strip-shaped detector reading area (12a) are definable.

15. Eilipsometer according to one of claims 13 or 14, characterized in that the cylinder optics (27) is arranged such that the illuminating light (40) has a beam waist in the test plane.

16. Eilipsometer according to one of claims 13 or 15, characterized in that the lighting arrangement is a light source (21) and before of the cylinder optics arranged beam expansion means (23, 24) for generating an expanded, parallel light beam.

17. Eilipsometer according to one of claims 13 or 16, characterized in that the lighting arrangement has polarization-influencing light guide means (25, 26) which are arranged between the beam expansion means (23, 24) and the cylinder optics (27).

18. Eilipsometer according to one of claims 13 or 17, characterized in that the lighting arrangement has polarization-influencing light guide means (25, 26) which are arranged between the light source (21) and the beam expansion means (23, 24).

19. Eilipsometer according to one of claims 8 to 18, characterized in that the detector is designed as an area detector, from which a detector reading area corresponding to a sample area of interest can be defined.

20. Quality assurance module for the detection of manufacturing defects in a product manufacturing device comprising an egg lipometer according to one of claims 8 to 19 and suitable for carrying out a method according to one of claims 1 to 7.