DE102016006107A1 - Device and method for confocal measurement of a surface topography - Google Patents

Abstract

The present invention discloses an extension of the existing confocal measurement technique such that simultaneous detection of position and local slope of a surface is possible. The method presented here uses a coding of the incident light in the aperture of the optical system. Subsequent decoding makes it possible to analyze the portions of the reflected or scattered light as a function of the angle and to deduce the local inclination of the surface. Compared to existing methods, this offers new possibilities of reconstruction in terms of a better resolution or greater dynamics of the measuring principle.

Description

The present invention relates to a device and a method for confocal measurement of a surface topography, wherein at the same time the three-dimensional coordinates of individual measuring points on the surface and their inclination are determined.

The confocal measurement technique or confocal microscopy is a widely used, powerful, non-contact measurement technique z. B. for the measurement of surface topography. The basic measuring method is designed to measure the three-dimensional position of a focused light beam at the interface (i.e., the surface). The measurement volume is scanned three-dimensionally by lateral and axial positioning (that is to say focusing) of the focal point or parallel to several focal points. The reconstruction of the surface is via a mostly linear interpolation between the scanned points, the sampling rate limiting the minimum resolvable details of the surface topography. According to the Nyquist theorem, the desired reconstructable spatial frequency requires twice the sampling frequency, resulting in a large number of measurement points and thus long measurement times.

For the identification of small surface defects and the measurement of fine topographies, the direct measurement of the inclination instead of the three-dimensional position is purely superior in resolution (cf. US 8,224,066 B2 or DE 10 2008 023 599 A1 ). A method based on phase-measuring deflectometry is presented, with which the inclinations of a surface can be determined. The advantage of this method is that at the same time a complete gradient field can be recorded. The method fails, however, to measure the shape of the object directly. The latter is realized by integrating the measured gradients, which, however, represents a poorly-placed inverse problem and therefore leads to ambiguity problems.

In order to realize a simultaneous identification of errors and a direct shape measurement, a simultaneous measurement of the local inclination and the position of the measuring points on the surface is required. However, only a few methods are known from the prior art, with which this can be realized. This was z. B. a classic confocal microscope equipped with an additional measuring unit (s. DE 42 11 875 A1 . US 2010/0225926 A1 or US 8,928,891 B2 ), with which the displacement of the back-reflected beam can be analyzed with a position-measuring detector and thereby closed to the direction of the back-reflected beam and the inclination of the surface. The received intensity values can then be used either as a measurement or to optimize the angles of incidence of the illumination.

In addition, in the Article U. Neuschaefer-Rube and W. Holzapfel, "Simultaneous measurement of surface geometry and material distribution by focusing ellipsotopometry," Appl. Opt., Vol. 41, no. 22, pp. 4526-4535, 2002 , presented a combined optical measuring method capable of simultaneously measuring the position and inclination. The system determines the position of the points on the surface by an astigmatic autofocus system, which focuses the light on the surface of an object. The incident light is linearly polarized and is analyzed after reflection by a Polarizer Sample Analyzer ellipsometry. From the recorded polarization parameters of the reflected light, both the inclination of the surface and the material properties can be determined. The disadvantage of this presented method is that due to depolarization effects, the z. B. caused by roughness on the surface, not allowed to measure the inclination of rough surfaces.

In addition, there are other measurement systems based on different principles of structured illumination or aperture coding. Such solutions can be found, for example, for ultrafast multiphotonic processes ( US 7,698,000 B2 ).

Concerning the confocal measurement technique, in the prior art principles of spectral coding along the depth ( DE 197 13 362 A1 ) and the lateral position ( WO 1999/044089 A1 ) known.

The simultaneous determination of the inclination and position is not possible with classical confocal systems so far. However, it brings great benefits, as the additional information allows the introduction of the slope-based interpolation method for the reconstruction of the surface. This leads to a significantly better lateral resolution or to a much smaller number of necessary measurement points and is thus particularly well suited for moderately curved surface profiles, such. B. optical freeform surfaces.

The simultaneous measurement of the position and inclination of the surface at the measuring points can also be used to predict the position of the adjacent measuring points. This allows z. B. an on-the-fly optimization of the sampling strategy in the ongoing measurement process such that the sampling rate (ie, the density of the measuring points) is increased in more inclined areas to a constant resolution of the surface profile in tangential direction. Furthermore, the prediction of the position of the next measuring point can be used to reduce the axial scanning range necessary for precise detection of the position. Thus, the confocal measurement process can be significantly accelerated or an improvement in resolution can be achieved with increased dynamic range.

In addition, the confocal measuring technique also offers the possibility of non-contact analysis of the vibrations of micromechanical components (such as MEMS, for example) DE 10 2006 013 773 A1 ). There is a specific challenge. Since the MEMS are often operated in the resonance range, so-called standing waves often arise when vibration excitation. Standing surface waves produce the so-called nodal points, whose position remains constant, while the local surface tilt changes periodically. Since the existing confocal measuring principles can only measure the position of the points and not characterize the local inclination at these points, these methods are not optimally suitable for vibration analysis. To do this, a system and method is required for simultaneously detecting the position of the points and the inclination at these points.

The present invention discloses an extension of the existing confocal metrology such that simultaneous detection of position and local slope of the surface is possible. The method presented here uses a coding of the incident light in the aperture of the optical system. Subsequent decoding makes it possible to analyze the portions of the reflected or scattered light as a function of the angle and to deduce the local inclination of the surface.

Compared to existing methods, this offers new possibilities of reconstruction in terms of a better resolution or greater dynamics of the measuring principle.

• • Vermessung der Oberflächentopographie von zumindest teil-reflektierenden Objekten, z. B. Spiegel, Linse, Wafer-level Mikrotechnische Objekte,
• • Vermessung der Rauigkeit von Oberflächen
• • Identifizierung der Oberflächenfehler.
• • Schwingungsanalysen und Vibrationsmessung
• • Überwachung industrieller Prozesse
• • Medizintechnische Anwendungen, Automotive

The modification of the confocal measuring principle proposed here as an invention can be used wherever confocal measurements are already used today. Specific advantages arise, inter alia, in the following fields of application:

• Surveying the surface topography of at least partially reflecting objects, e.g. Mirror, lens, wafer-level micro-technical objects,
• • Measuring the roughness of surfaces
• • Identification of surface defects.
• • vibration analysis and vibration measurement
• • Monitoring of industrial processes
• • Medical applications, automotive

The invention will be described in more detail below with reference to drawings. It shows:

1a - schematic representation of the confocal measuring technique (prior art)

1b - Schematic representation of the confocal measuring device according to the invention

2 - Schematic representation of the confocal measuring method according to the invention

3 Various embodiments for the aperture coding

4 - Schematic representation of two embodiments of the method according to the invention with optomechanical coding

5 - Schematic representation of an embodiment of the method according to the invention with polarization coding

6 - Schematic representation of an embodiment of the method according to the invention with spectral coding

The present invention proposes a confocal measuring method with spatial coding of the light incident in the aperture of the optical system. Either one or more lasers, LEDs, or other lighting fixtures can be used as the light source. Their arrangement or further optoelectronic transfer function ensure that the optical aperture of the objective is coded in such a way that at a certain position within the aperture a light differentiable from the other positions is incident. The distinctness of the light, hereinafter referred to as coding, can be achieved either by temporal multiplexing, mechanical or optical scanning or by exploiting the other degrees of freedom which affect the electromagnetic properties of the light, such as light. B. polarization and wavelength, realized. The thus encoded light enters the aperture of the lens, which focuses the light to a point (the measuring point). In this case, the wavefront of each incident beam can be optimized so that the optical aberrations in the focus are as small as possible.

In 1a is shown schematically the measurement of a topography by means of classical confocal measurement technique (prior art). The light of a light source ( 1 ) (also Called illumination beam) via a collimating optics ( 2 ) and a beam splitter ( 3 ) on a lens ( 4 ) distracted. The objective ( 4 ) focuses the light on the surface to be examined ( 5 ). From the surface to be examined ( 5 ) the light is thrown back (reflected or scattered) (reflection beam) and over the lens ( 4 ) and the beam splitter ( 3 ) to a detection system consisting of a lens (or focusing optics) ( 6 ), in whose focal plane a pinhole (Pinhole) ( 7 ), and a photosensitive detector ( 8th ), focused and depending on the size of the hole in the pinhole ( 7 ) shadowed.

Is there a lit point on the surface to be examined exactly in the focal plane of the lens ( 4 ), a maximum intensity from the detector ( 8th ) detected. If a lighted spot is located on the surface to be examined in front of or behind the focal point of the system, the light is reflected by the lens ( 4 ) no longer ideal collimated (parallel beam path) but divergent or convergent. Moving the focus along the optical axis, the maximum intensity is detected exactly where the object in the focal plane of the lens ( 4 ) is located. From the travel path of the surface to be examined ( 5 ) or the lens ( 4 ) and the measured intensity profile at the detector ( 8th ), the position of the surface can thus be determined.

To avoid a focus adjustment by the mechanical or optical Z-scanning chromatic confocal methods can be applied. In this case, an optical system with a strong longitudinal chromatic aberration is used to focus different spectral components of the light at different axial positions. From the spectrum of the detected light, the z. B. can be recorded by a spectrometer at the detector output, the depth profile of the surface structure is determined without scanning.

1b schematically shows the aperture-coded confocal measuring device according to the invention. The collimation optics ( 2 ) by an optical system ( 9 ), in which the illumination beam is split into individual beams. The individual beams are separated from the optical system ( 9 ) and depending on their position in the aperture after focusing at different angles on the surface ( 5 ) steered. After reflection, some beams no longer reach the aperture, depending on the slope of the surface. If the incident light is coded, it is possible by decoding to deduce from the back-reflected light which ray bundles or which associated angle of incidence from the surface to be examined (FIG. 5 ) through the aperture of the optics ( 4 ) are reflected (reflected or scattered). If the intensity components of the reflected light of different positions within the aperture are compared, then the inclination of the surface to be measured (FIG. 5 ) shut down.

As a detector according to the invention a decoding system ( 10 ) which decodes the coded and reflected beams. From the determined intensities of the individual ray bundles it is possible to deduce the inclination of the surface. With the help of an integral measurement of the intensity over all light bundles, as with the classical confocal measuring technique, the three-dimensional coordinates of the points on the surface to be measured can be determined.

In 2 the proposed aperture coded confocal measuring method is shown schematically, wherein 2a the measurement of the three - dimensional coordinates of measuring points on the surface to be measured and in 2 B the measurement of the inclination of the surface is shown unfolded. If the individual beams are focused on the measuring points on the surface to be measured, and the measuring points are located either in front of or behind the focal plane of the objective ( 4 ), the reflected or scattered light becomes divergent or convergent and thereby at the pinhole ( 8th ) shaded ( 2a ). Are the measuring points on the surface in the focus of the lens ( 4 ), you get the maximum intensity of all light bundles. From the intensity can thus determine the three-dimensional coordinates of the measuring points on the surface.

If one tilts the object, intensity differences result between the individual light bundles ( 2 B ).

The present invention proposes to disassemble the illumination beam into individual beam bundles and then to code the individual beam bundles. The coded beams hit the aperture of the confocal optic ( 4 ) and are controlled by the decoding system ( 10 ) decoded separately. For this purpose, there are a variety of coding and decoding possibilities exemplified below with reference to the 3 . 4 . 5 and 6 to be introduced.

In a first exemplary embodiment, the coding takes place by temporal multiplexing with at least two or more light sources which are switched on and off in a time-shifted manner and alternately ( 3a ). In this case, the control of the light can be modulated either electronically or indirectly z. B. by digital micromirror (DMD) or liquid crystal modulators (LCD). The position of the light sources, their intensity as well as the Expansion and divergence can be arbitrary. At any time, the focus position is detuned, during which time the intensity distribution is recorded as a function of the focus position. After the sequential recording, the data evaluation for the detection of position and local surface tilt is performed from the recorded intensity distributions.

In a second embodiment, the coding is carried out via time-frequency multiplexing ( 3b ). The individual light bundles illuminating the aperture are modulated with different clock frequencies. By evaluation with a corresponding electronically clocked detector (eg a lock-in amplifier), the signals of the individual light bundles coding the aperture can be evaluated separately and / or integrally at the output. Thus, in turn, all data necessary for measuring the coordinates of a measuring point and the local inclination at the position of this measuring point are present.

Another way to implement coding of the aperture is to scan the aperture of the optical system (scanning). 9 ). The movement of the light beam can be realized either by mechanical or optical scanning (s. 3c ), whereby a mechanical scanning either via a translation or rotation of the light source ( 1 ), whereby the position of the light beam in the aperture varies, can be realized, or via the movement of a pinhole within the aperture. Optically, a scan system z. For example, a Risley prism system, rotating tilted glass plates, deformable optics, acousto-optic deflectors or liquid crystal modulators can be implemented.

These are exemplified in 4 two opto-mechanical coding variants shown. In 4a is the scanning with a glass plate ( 11 ), which causes the spatial offset of the incident into the aperture beam. By rotation of this glass plate ( 11 ), the illumination beam in the aperture of the lens ( 4 ) are moved. In 4b is the movement of the illumination beam within the aperture of the objective ( 4 ) with the aid of a tilting mirror ( 12 ) and a lens system ( 13 ), which again collimates the beam after reflection.

The detection of the position and the inclination of the surface is also done for this form of coding as in the coding by means of time division multiplexing after the aperture scanning was performed.

For the aperture coding in the confocal measurement technique, the polarization of the light can also be exploited ( 3d ). Polarization coding can z. B. passive polarization filters, the splitting of individual polarizations with beam splitters or Wollaston prisms, diffractive, refractive or hybrid optics or similar polarization groups are generated. In 5 an embodiment for polarization coding is shown. Using a first polarization beam splitter ( 14a ), both polarizations are split, with the P polarization passing through a tilted glass plate ( 11 ), and the S-polarization through a mirror periscope ( 15 ) is deflected so that an encoding of the aperture after the merger of both polarizations by means of a second polarization beam splitter ( 14b ) is achieved. After reflection, the decoding takes place via a splitting of the individual polarizations with a third polarization beam splitter ( 14c ) and the measurement of the intensities with two separate detection systems ( 6a . 7a . 10a ) and ( 6b . 7b . 10b ). The detection of the inclination and position of the surface takes place here via polarization-sensitive decoding systems ( 10a . 10b ), wherein the intensity distribution of individual polarizations are detected simultaneously and compared with each other.

Another embodiment of the aperture coding is a spectral coding ( 3e ). The aperture of the lens ( 4 ) is encoded such that certain positions within the aperture have a unique spectral profile consisting of at least two or more wavelengths. The spectral coding can be realized in different ways. As sources monochromatic and white light sources of any kind can be combined arbitrarily. This includes a combination of monochromatic light sources, heterodyne lasers, supercontinuum lasers, single color and multicolor LEDs, and other lighting fixtures. The arrangement of the light sources and the further optical assemblies is designed so that thereby a spectral coding in the aperture is achieved. Optical assemblies are any passive spectral filters and any combination of dichroic mirrors, special diffractive, refractive or hybrid diffractive-refractive optical elements and optical systems with prisms and dichroic beam splitters for generating the spectral encoding to understand.

Exemplary is in 6 the execution of a spectral coding shown. The three-wavelength broadband light is transmitted by means of a first dichroic beam splitter ( 16a ) and mirror periscopes ( 15a ) and ( 15b ) to a second dichroic beam splitter ( 16b ), in such a way that a spatial offset of individual wavelengths takes place and thus the coding of the aperture is realized. The decoding system ( 10 ) consists in this embodiment either of a spectrometer or a spectrally sensitive photodetector.

With the present invention, all arbitrary combinations of the described decompositions and codings of the illumination beam and decoding and evaluations of the reflection beam are also detected.

With the solution according to the invention, it is possible to simultaneously measure the axial position of a measuring point over the maximum intensity of all the beam and the local inclination of the surface at the measuring point on the intensity differences of individual beams. The axial resolution of the position measurement is dependent on the depth of field of the optical image as usual in confocal measurement technology. This can be described by the Full-Width-at-Half-Maximum Metric (hereinafter referred to as FWHM). Since individual beams within the aperture illuminate only a portion of the aperture, the corresponding numerical aperture of such beams is smaller than the numerical aperture of the lens employed. Therefore, the FWHM is larger and the axial resolution is worse than the maximum achievable resolution of the lens. To avoid this loss of resolution, the signals of the various beams can be coherently integrated with targeted evaluation algorithms, which again the entire aperture of the system is exploited, so that the FWHM of the system and, accordingly, the axial resolution is limited solely by the numerical aperture of the lens , The system is thus suitable for precise confocal measurement of the three-dimensional position (by the integral measurement over all beams) and the local inclination at the measurement point (by the separate analysis of the separated light bundles). As usual in confocal metrology, the concept can be extended by scanning or spatial parallelization so that the entire surface can be scanned point by point.

LIST OF REFERENCE NUMBERS

1

light source

2

collimating optics

3

beamsplitter

4

lens

5

surface to be measured

6, 6a, 6b

Focusing optics

7, 7a, 7b

Pinhole

8th

light-sensitive detector

9

Optical system for coding the illumination beam

10, 10a, 10b

decoding system

11

glass plate

12

tilting mirror

13

lens system

14a, 14b, 14c

Polarization beam splitter

15, 15a, 15b

Mirror Periscope

16a, 16b

dichroic beam splitters

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 8224066 B2 [0003]
• DE 102008023599 A1 [0003]
• DE 4211875 A1 [0004]
• US 2010/0225926 A1 [0004]
• US 8928891 B2 [0004]
• US 7698000 B2 [0006]
• DE 19713362 A1 [0007]
• WO 1999/044089 A1 [0007]
• DE 102006013773 A1 [0010]

Cited non-patent literature

• Article U. Neuschaefer-Rube and W. Holzapfel, "Simultaneous measurement of surface geometry and material distribution by focusing ellipsotopometry," Appl. Opt., Vol. 41, no. 22, pp. 4526-4535, 2002 [0005]

Claims (12)

Device for confocal measurement of a surface topography with a light source ( 1 ), which generates an illuminating beam which is incident on the surface to be measured ( 5 ) via an optical system ( 9 ), a beam splitter ( 3 ) and a lens ( 4 ) and from there to a decoding system ( 10 ) via the lens ( 4 ), the beam splitter ( 3 ), a focusing optic ( 6 ) and a pinhole ( 7 ) is reflected or scattered, the optical system ( 9 ) is adapted to divide the illumination beam into individual position-dependent coded radiation beams and the decoding system ( 10 ) is designed to decode and evaluate the reflected or scattered radiation beam angle-dependent. Apparatus according to claim 1, wherein the optical system ( 9 ) is adapted to the time, frequency, optical, mechanical, spectral and / or polarization dependent to encode the individual beam of the illumination beam. Apparatus according to claim 1, wherein the decoding system ( 10 ) is designed to decode and evaluate the reflected or scattered radiation beam time, frequency, optical, mechanical, spectral and / or polarization dependent. Apparatus according to claim 3, wherein the decoding system ( 10 ) is designed to evaluate integrally and / or separately the reflected or scattered and decoded beams. Method for confocal measurement of a surface topography with a device according to one of Claims 1 to 4, the illumination beam being detected by means of the optical system ( 9 ) is divided into individual position-dependent coded beams and the reflected or scattered beams by means of the decoding system ( 10 ) are decoded dependent on angle and evaluated. A method according to claim 5, wherein the beams of the illumination beam are time, frequency, optical, mechanical, spectral and / or polarization dependent coded. The method of claim 5, wherein the reflected or scattered beams are time, frequency, optical, mechanical, spectral and / or polarization dependent decoded. The method of claim 5, wherein the reflected or scattered beams are evaluated integrally and / or separately. The method of claim 8, wherein the three-dimensional coordinates of a point on the surface to be measured are determined from the integral evaluation of the reflected or scattered beam. The method of claim 8, wherein the inclination of the surface to be measured is determined from the separate evaluation of the reflected or scattered beam. The method of claim 10, wherein from the determined inclination of the surface to be measured, an optimized scanning strategy is realized. Method according to one of claims 10 or 11, wherein the three-dimensional coordinates of a plurality of points and the inclination of the surface to be measured are determined simultaneously by means of lateral scanning or parallelized.

Images