DEVICE FOR GENERATING A SYNTHETIC WAVEL LENGTH AND MEASURING DEVICE

Description

[0001] The invention relates to a device for generating a synthetic wavelength, in particular a miniaturized coherent multi-wavelength light source for generating light which emits spectrally stable radiation and thus enables the setting of synthetic wavelengths for measurement and observation systems. The invention further relates to a measuring device.

[0002] Interferometric methods have long been established in metrology. They are characterized by high measurement accuracy in the nanometer range over a relatively small measuring range.

[0003] To increase the measurement range, synthetic wavelengths can be used instead of the wavelength of light when lower resolution requirements are needed. These are created by the coherent superposition of two light waves with (slightly) different fundamental wavelengths λ₁, λ₂. The synthetic wavelength Λₙ is given by Λ s = λ 1 ⋅ λ 2 λ 1 − λ 2

[0004] In order for the synthetic wavelength to be used for interferometric measurements and other quantitative phase measurements, it must be very stable over time. This requires, in particular, that the difference between the output wavelengths be very constant.

[0005] A disadvantage of the known devices is that significant technical effort is required to ensure stable operation. For example, laser systems are used, which represent a considerable cost factor. Another possibility is active temperature stabilization, but this requires the use of high-quality temperature sensors, which are also comparatively expensive.

[0006] The generation of a synthetic wavelength from two similar fundamental wavelengths has already been described in the literature (see, e.g., Wyant JC 1971 "Testing aspherics using two-wavelength holography" Appl. Opt. 10 2113-8). The use of synthetic wavelengths has now become established. For example, EP 2 589 923 A1 describes a device for measuring the distance between a reference and a test surface.

[0007] US 6 002 705 A describes a device with four surface emitters, which are positioned on a GaAs support element.

[0008] The object of the present invention was to overcome the disadvantages of the prior art and to provide a device by means of which a user is able to generate a synthetic wavelength in a simple and cost-effective manner. In particular, an object of the invention was to provide a (miniaturized) coherent multi-wavelength light source for generating time-stabilized emitted wavelengths, which produces a stable, and preferably adjustable, synthetic wavelength for measurement and observation systems. A further object was to provide an improved measurement device.

[0009] This task is solved by a device according to the claims.

[0010] The device according to the invention for generating a synthetic wavelength comprises at least two surface emitters and a support element, wherein the support element is a heat conductor and both surface emitters are fixedly arranged in, on or at the support element.

[0011] A surface emitter (also known as a "VCSEL," short for "vertical-cavity surface-emitting laser") is a laser diode in which the light is emitted perpendicular to the plane of the semiconductor chip, typically in the milliwatt range. This distinguishes a surface emitter element from edge-emitting laser diodes, in which the light exits at one or two sides of the semiconductor chip. A surface emitter, as defined in the invention, comprises at least one surface-emitting semiconductor chip (which can also be considered the actual surface emitter). The semiconductor chip can be arranged in a surrounding housing, or it can be directly mounted on, attached to, or integrated within the substrate. Examples of surface emitters include single-mode vertical-emitting laser diodes.

[0012] Such single-mode VCSELs (e.g., longitudinal monomode) are available on the market as standard products with, for example, intended fundamental wavelengths of 795 nm, 852 nm, 895 nm, and 940 nm. VCSELs have the properties advantageous for the invention that their bandwidth (Δλ < 2 pm) is sufficiently small and their coherence length (approximately 50 cm) is comparatively large and particularly suitable for interferometry.

[0013] A support element is understood to be a component that serves as a thermal bridge between the surface emitters, thus bringing them into thermal contact with each other via the support element. For this to function, the support element must be a thermal conductor. While virtually any material is thermally conductive, only materials with a comparatively high thermal conductivity > 2 W / (m·K) are considered thermal conductors within the meaning of this invention. Thermal insulators or materials with a low thermal conductivity coefficient are not thermal conductors within the scope of this invention. Preferably, the support element is made essentially of materials whose thermal conductivity is at least greater than 40 W / (m·K), in particular greater than 200 W / (m·K) or 270 W / (m·K), and most preferably greater than 300 W / (m·K). Preferred materials are metals, especially aluminum or copper, or carbon compounds."Essentially" here means that other materials (including non-thermal conductors) may be included, e.g., insulating materials for separating conductor tracks from other conductive areas, provided that these materials do not significantly impede the heat flow between the surface emitters, e.g., by being thin. Preferably, "essentially" means that the support element comprises at least 90% of the aforementioned thermally conductive material.

[0014] A preferred material for the support element is aluminum nitride. This typically has a thermal conductivity of 285 W / (m·K) and is particularly advantageous because its coefficient of thermal expansion of 4.4 to 5.3 × 10⁻⁶ < 1 / K is similar to that of the substrate material GaAs (5.7 × 10⁻⁶ < 1 / K), which is typically used for VCSEL. This is advantageous because it avoids the occurrence of mechanical stresses at different temperatures. Another preferred material is aluminum oxide.

[0015] Preferably, the surface of the support element comprises a metal. The surface emitters are preferably applied directly to the support element (the metal) with an electrically conductive connection (e.g., the cathode), e.g., soldered on.

[0016] The surface emitters are arranged very close together. Preferably, their distance *a* from each other is no greater than 1 cm, and particularly no greater than 5 mm. Since the temperature difference between the outer surfaces of the surface emitters should be very small during operation, and the overall temperature should preferably remain within a narrow operating range, their thermal expansion is negligible. Therefore, the surface emitters can even touch each other. Better thermal conductivity of the support element allows for a comparatively large distance between the surface emitters; with poorer thermal conductivity of the support element, a smaller distance between the surface emitters should be chosen.The ratio V of the distance a between two surface emitters and the effective thermal conductivity w of the substrate material, V = w / a, is, according to the invention, greater than 5000 W / (m²< √K), in particular greater than 20000 W / (m²< √K), and most preferably greater than 50000 W / (m²< √K). Depending on the thermal conductivity of the substrate material and the arrangement of the surface emitters used, it is also preferred that V be greater than 100000 W / (m²< √K), in particular greater than 150000 W / (m²< √K), and most preferably greater than 200000 W / (m²< √K). The distance a is considered to be at least the distance between the active areas of the surface emitters, since these are the critical areas with respect to thermal coupling. For surface emitters that are relatively small or whose housing walls are very thin, the distance *a* can also be the distance between the outer surfaces. Since it is advantageous for the surface emitters to be small (e.g.,For VCSL, it is preferred that the distance a is the distance between the outer surfaces of the respective facing sides of the surface emitters. This is also much easier to implement in a practical design.

[0017] Surface emitters, such as VCSELs, should be positioned as close together as technically possible to ensure a direct relationship between their temperatures. However, direct temperature coupling of the active zone in the surface emitter (the region where the laser radiation is generated) should be avoided. For VCSELs, this means, for example, that they can be positioned very close to each other, but without their substrates touching in the active zone.

[0018] Depending on the application within the context of VCSL, it is preferred to use adjacent VCSELs from a wafer without separating them. It is therefore possible for at least two surface emitters to be located on the same substrate. However, according to the invention, the distance between the two surface emitters (VCSELs) should then be at least 1 mm to prevent temperature coupling. With such a distance, the active zones of the surface emitters generally no longer touch.

[0019] Preferably, the support element serves as a heat sink. Components that allow rapid heat conduction and heat exchange with the environment are considered heat sinks. The dimensions of the heat sink, particularly its thickness, are designed to ensure sufficient heat flow between the surface emitters and the environment. Preferably, the support element is thicker than 0.1 mm, and particularly thicker than 1 mm. The temperature of the support element can be controlled, as explained in more detail below.

[0020] A surface emitter has a defined wavelength (fundamental wavelength), which is specified in its datasheet. At constant temperature and current, the emitted wavelength of a VCSEL is constant. The wavelength depends on the current at which the surface emitter is operated. In this example, it can be assumed that the wavelength λ and the current I are related by the equation λ = I · 0.6 nm / mA. Considering the surface temperature T of the VCSEL, it can be observed that the wavelength λ is related to this surface temperature by λ = T · 0.06 nm / K. It is preferred that at least two surface emitters have the same defined wavelength and are operated at different currents. This allows the generation of synthetic wavelengths down to the millimeter range.

[0021] The relationship between wavelength (fundamental wavelength) and current intensity results from the local heating of the semiconductor chip and the associated expansion of the resonator. However, the total amount of heat transferred to the semiconductor crystal per unit time is comparatively small, and the heated area is relatively small. Due to the thinness of the laser-active layer, with a resonator length typically of 2 to 5 µm, the thermal relaxation time is short, and the wavelength can be quickly adjusted by varying the current intensity. The aforementioned dependence on current intensity thus relates to a very localized heating within the surface emitter, while the dependence on temperature relates to the heating of the entire surface emitter.

[0022] It is particularly advantageous for the surface emitters to have the same intended wavelength. This may seem contradictory at first glance, but this can be easily resolved. Currently available VCSELs have wavelength differences greater than 10 nm. For example, an extreme combination of commercially available VCSELs (so) with wavelengths of 795 nm and 940 nm would emit a synthetic wavelength ΛS = 5.2 µm according to the formula (1) mentioned above. The closest fundamental wavelengths, 852 nm and 895 nm, result in ΛS = 17.8 µm. Since light with wavelengths in the µm range can also be emitted directly by lasers, the relevance of generating comparable synthetic wavelengths may be of minor importance in practice. Significantly larger synthetic wavelengths, however, are achieved when identical VCSELs (same intended wavelength) are operated at different current intensities.Depending on the type, the current can be meaningfully varied over a few mA, which, with the aforementioned factor of 0.6 nm / mA, allows for a wavelength difference in the single-digit nm range or even in the pm range. For a VCSEL pair with a fundamental wavelength of 852 nm, a wavelength difference of 1 nm results in a synthetic wavelength in the mm range.

[0023] It is preferred that at least two of the surface emitters have fundamental wavelengths that differ from each other by less than 10 nm, preferably less than 5 nm. Since surface emitters with the same intended wavelength can differ by several nm with respect to their fundamental wavelength due to manufacturing tolerances, it is also possible to use two surface emitters (e.g., VCSELs) in addition to or alongside current control, which at the same current intensity have a wavelength difference corresponding to that of the required synthetic wavelength. Preferably, surface emitters with the same nominal fundamental wavelength (intended wavelength) are used, whose actual wavelengths differ by more than 0.01 nm due to manufacturing tolerances, preferably by 0.1 nm or more. However, the wavelengths preferably differ by less than 10 nm, preferably by less than 5 nm, and most preferably by less than 1 nm.

[0024] It should also be noted that when mapping rough surfaces, the fundamental wavelength pairs should preferably be chosen to avoid speckle decorrelation (see S. Montresor and P. Picart, "Quantitative appraisal for noise reduction in digital holographic phase imaging," Opt. Expr. 24(13), 14322-14343 (2016)). Such deconstruction would result in increased noise in the resulting topographic map.

[0025] According to a preferred embodiment, the device comprises a temperature sensor and, more preferably, a temperature control element and a temperature control unit for regulating the temperature of a surface emitter and / or the support element. Preferred temperature sensors are thermistors or semiconductor sensors; preferred temperature control elements are heating elements, such as heating wires, cooling elements, or elements suitable for both heating and cooling, preferably thermoelectric elements, such as Peltier elements. In particular, a temperature control element comprises a combination of the aforementioned elements, such as a Peltier element combined with a heating element for fine-tuning. The temperature control unit can be a circuit that allows direct feedback between the temperature sensor and the temperature control element or an actively controlling circuit.

[0026] When using a thermoelectric element (e.g., a thermoelectric cooler, TEC), the surface emitters can theoretically be located anywhere on the element, since the temperature on its surface is essentially the same everywhere. However, they should preferably not touch, at least not with respect to the active zones. It is particularly preferred that the thermoelectric element has a similar coefficient of thermal expansion to the surface emitter to avoid mechanical stresses. A preferred material in this regard is aluminum oxide (Al₂O₃) with a coefficient of thermal expansion in the range of 4.5 to 10.9 × 10⁻⁶ < 1 / K), particularly preferably with a gold finish. Aluminum nitride, which has already been mentioned above, is also preferred.

[0027] The stability requirements for synthetic wavelengths depend on their intended use. For topographic measurement methods, such as iterative phase reconstruction, direct phase reconstruction (e.g., transport of intensity), interferometric, or holographic methods, measurement deviations should ideally be less than 0.1% of the measurement range, i.e., the synthetic wavelength. For synthetic wavelengths in the millimeter range, the wavelength difference between the fundamental wavelengths used must then be constant in the pm range (see formula (1)). The bandwidth of a surface emitter is sufficiently small for this purpose. However, with a temperature drift of 60 pm / K, the temperatures of the two surface emitters must be stabilized to approximately 0.01 K.

[0028] The support element achieves relative stabilization. Absolute stabilization, which does not require such precise results, can be achieved with the temperature control described above. The following example calculation illustrates absolute stabilization: With fundamental wavelengths λ₁ and λ₂ of 853.00 nm and 852.00 nm, respectively, a synthetic wavelength Λₙ of 726.76 µm is obtained. Increasing the combined temperature of the two surface emitters by 1 K results in fundamental wavelengths of 853.06 nm and 852.06 nm for λ₁ and λ₂, respectively, and a synthetic wavelength Λₙ' of 726.86 µm, which is acceptable. If, however, only a single surface emitter were to heat up by 1K (e.g. λ 1 and λ 2 would be 853.06 nm and 852.00 nm), the resulting synthetic wavelength would be Λ S' = 770.4µm.

[0029] Consequently, even with a simultaneous temperature change of 1 K at both surface emitters, the synthetic wavelength would only change by a fraction of a percent. Therefore, the temperature control of the device should be designed to maintain the temperature of the surface emitters or the support element at a constant 1 K or better.

[0030] This device has the advantage of providing a compact light source capable of generating a temporally stable synthetic wavelength from two fundamental wavelengths, which can also be separated if required. Compared to the prior art, the device according to the invention has very few components, is very robust against mechanical and thermal stresses, and can be implemented more cost-effectively.

[0031] To observe wavelength changes in spectrally unstabilized light sources, spectrometers or wavemeters (which provide higher spectral resolution) are necessary, since otherwise the determined phase image cannot be converted into a true topographic pattern, resulting in an extreme increase in measurement uncertainty. Because the device according to the invention represents a spectrally stabilized light source, measurements can also be carried out with it, in particular, without a spectrometer or wavemeter.

[0032] Operating two VCSELs with different current ratings can be achieved using two independent power supply units, preferably adjustable, for example, by means of two voltage regulators. However, it is also possible to power both VCSELs with a single power supply unit. For this purpose, a VCSEL is preferably connected in series with a parallel circuit consisting of another VCSEL and a resistor. If additional VCSELs are to be used, these can also be connected in parallel with a resistor, and this parallel circuit can then be included in the series circuit. The resistors preferably all have different resistance values. Of course, when using more than two VCSELs, a hybrid configuration is also possible, in which one group of VCSELs is connected in the aforementioned series circuit, and another group is powered by a separate power supply unit.

[0033] Preferential control or regulation of a fundamental wavelength can be achieved when using multiple power supply units by controlling or regulating one of them. In such a series circuit, the resistors can be controlled or regulated, for example, by a potentiometer or by adding different resistors.

[0034] Since there is a direct relationship between voltage and current via the resistance of the VCSEL, it can also be said that each VCSEL must have its own individual current source. This could be achieved with two voltage regulators or the aforementioned series connection, as the parallel connection with the resistor acts like an individual current source for each VCSEL.

[0035] Considering conventional VCSELs, the accuracy of the current control should be in the range of a few µA. Preferably, the current control is designed such that current fluctuations are less than 10 µA. Preferably, the device should have at least one control element with which the current can be changed separately for at least one surface emitter, in particular for a subset of the surface emitters or each one, and regulated to a stable value.

[0036] Preferably, the device is designed such that the light emission intensities of at least two of the surface emitters can be changed independently of one another, and in particular, at least two surface emitters or coupling units can be switched so that they emit light alternately. This can be achieved by directly controlling the energy supply to the surface emitters, or by controlling the light beams, for example, by means of an aperture or an element whose transmission can be controlled, such as an LCD element, an acousto-optic modulator, or a Pockels cell in conjunction with polarizing optics. Controlling the light beams independently of the energy supply to the surface emitters has the advantage that the wavelength of the emitted radiation can be kept very constant.This embodiment, in particular the alternating switching of light beams, has the advantage that different fundamental wavelengths can be recorded sequentially during a single measurement. It should be noted that the intensity of the light emission does not necessarily have to be switched directly at the surface emitters, but can essentially be switched anywhere in the beam path, e.g., at an output coupling unit.

[0037] According to a preferred embodiment, the device comprises a light guidance system designed to guide the light beams of the VCSEL. This can involve separate guidance of the beams and / or merging. For example, the beams of each VCSEL can be split, with two bundles from different VCSELs then being merged and two other beams being guided separately. The light guidance system preferably comprises mirrors (e.g., a roof-edge mirror), beam splitters, and / or light guides (e.g., glass or plastic fibers).

[0038] It is particularly preferred to apply a light guide directly to the VCSEL, for example by gluing or clamping. Light guides with a circular cross-section are especially advantageous because VCSELs have a rotationally symmetrical emission profile and can therefore be well coupled into these circular light guides. If polarized light is to be used (a single-mode VCSEL often emits linearly polarized light), the light guide system preferably also includes polarization elements. An output coupling unit can also be part of the light guide system. In a preferred application of light guides, a light guide is translucently bonded directly to the output of a surface emitter, for example by means of an epoxy resin bond.

[0039] When two beams merge, this part of the optical system can also be called a "unification unit". Preferred unification units include a merging of two optical fibers, a beam splitter (as a cube or plate), or other more complex optical elements.

[0040] Preferred configurations include the aforementioned merging of two light beams into a single output unit, or the separate guiding and extraction of two light beams using two separate output units, as well as the initial separate guiding and subsequent merging of two light beams in a beam splitter. It is particularly preferred that the extraction of a reference beam and the extraction of a target beam occur separately, especially in the form of different extraction methods. For example, the target beam can be extracted by merging two light beams, and the reference beam by combining two polarized light beams using a beam splitter.

[0041] The extraction of light from a light-generating device, and thus the aforementioned extraction units, are well known. Since the device according to the invention is intended to emit light, it automatically includes a number of extraction units. A simple extraction unit can be an opening, a window, or the end of a light guide. Preferably, an extraction unit includes lenses or lens systems for manipulating (e.g., widening or focusing) the emitted light.

[0042] For some applications, it is preferable for the device to have more than two VCSELs. Although only two fundamental wavelengths are necessary to generate a synthetic wavelength, having more than two fundamental wavelengths offers the advantage of being able to generate multiple synthetic wavelengths simultaneously. This can be useful, for example, for building a topography sensor that, at larger distances, solves ambiguity problems by combining different wavelengths, thus achieving a very large measurement range with high accuracy.

[0043] The laser emission from surface emitters can exhibit sidebands with slightly shorter or longer wavelengths in addition to the main wavelength. Even though these sidebands typically have a significantly lower intensity than the main wavelength, it is preferred that the device includes a filter element in the beam path of a (especially any) VCSEL. The filters are, of course, dimensioned to remove the sidebands, but not the main wavelengths. Particularly when transmitting the radiation via optical fibers, it is preferred to place the filters on the input and / or output coupling side. If two filter elements are used, one can be placed on the input coupling side and the other on the output coupling side relative to an element of an optical fiber system. Preferably, such an output coupling unit comprises a filter element.

[0044] The filters are preferably placed individually in the beam path for each VCSEL. A bandpass filter is preferably used for this purpose, in which the main wavelength of the VCSEL falls within the transmission range and the sidebands within the absorption range. If available filters are not narrow enough, the transmission range can also be reduced to the required narrow spectral width by combining two filters. For this, it is preferably possible to use two or more bandpass filters with slightly different central wavelengths, which can be achieved, for example, by selecting filters from a batch sample. Alternatively, the spectral range can also be adjusted by tilting a filter, since the spectral transmission is often dependent on the angle of incidence.

[0045] The following describes preferred devices for topographic measurement methods or measuring apparatus, such as for iterative phase reconstruction, interferometric measurements (which also includes shearing interferometry or measurements within the framework of ptychography), or holographic measurements (which also includes Gabor inline holography), which can be advantageously implemented with the device according to the invention and yield good measurement results. Within the scope of the invention, interferometric and holographic measurement methods are understood to be all quantitative phase reconstruction methods based on the superposition of light waves. Measurements can be taken with or without a reference beam, for example, in the iterative determination of a complex amplitude distribution of an object (see, e.g., R. Gerchberg and W.Saxton, "Phase determination for image and diffraction plane pictures in the electron microscope", Optik 34, 275-284 (1971). J. Fienup, "Reconstruction of an object from the modulus of its Fourier transform", Optics Letters 327-29 (1978) or on ptychography, H. Faulkner and J. Rodenburg, "Movable aperture lensless transmission microscopy: a novel phase retrieval algorithm", Phy. Rev. Lett., 93(2) 023903 / 1-4 (2004)), or on the non-iterative determination of the complex amplitude distribution (e.g. the principle of Transport of Intensity (TIE)).

[0046] A measuring device according to the invention always comprises a device according to the invention and an optical sensor (for detecting the fundamental wavelengths and the synthetic wavelength). Furthermore, a suitable measuring range should be defined within which an object can be measured. Since an object for measurement should preferably be illuminated with the synthetic wavelength, it is preferred that the beams from at least two surface emitters of the device for illuminating the measuring range are always combined (before they exit the respective output coupling unit of the device).

[0047] The optical sensor can be a (linear or area) image sensor, i.e., a sensor that captures a linear or area image with multiple pixels, or a point sensor, i.e., an optical sensor with a single pixel, e.g., a photoelement.

[0048] In addition, the measuring device preferably includes one or more of the following components: An optical element, preferably an element from the group consisting of beam splitters, mirrors (especially dichroic mirrors), prisms, and lenses. Preferably, the optical element comprises a deflecting element for deflecting the beam. Alternatively, a fiber end of an optical fiber can be positioned to directly illuminate the object. It is particularly preferred that a deflecting element is combined with a lens or lens system with which the light exiting the device is preferably collimated and directed perpendicularly onto the object under investigation. The object fiber AasVCSEL emitted light of at least two wavelengths BCCA from the light source. An evaluation module for determining the phase signal recorded by the optical sensor corresponding to a synthetic wavelength, in particular by calculating the difference between the at least two quantitative phase signals corresponding to the fundamental wavelength.The synthetic wavelength is determined from the fundamental wavelengths, in particular according to formula (1). A motion unit for moving an object to be measured within a defined measuring range. Such a motion unit is particularly advantageous when point measurement or ptychography is to be performed. It is not strictly necessary in a conventional interferometric or holographic arrangement, as area measurements are often performed there; however, it could be advantageous for adjusting or otherwise repositioning the object. An arrangement of an output coupling unit of the device, which directs a reference beam with a separately guided fundamental wavelength directly or deflected via an optical element (e.g., mirror) onto the optical sensor, so that a holographic image is achieved.The light beam can strike the optical sensor at a slight angle to generate a carrier frequency that enables the simultaneous acquisition of multiple wavelength-specific holgrams (a carrier frequency with different directions, allowing separation in Fourier space). Preferably, one light beam strikes the optical sensor obliquely from above or below, and a second beam with a different wavelength strikes obliquely from the side. An output coupling unit emits two different fundamental wavelengths of different polarizations. The measuring unit preferably comprises two optical sensors, particularly point sensors, and an optical unit, e.g., a polarization beam splitter (cube), configured to direct light with different polarizations in different directions (toward the optical sensors).An imaging optic that produces an image of the object (especially through an aperture), which is positioned so that it is captured by the optical sensor (e.g., a point sensor).

[0049] A preferred measuring device is designed for measurement with a reference beam, wherein the object beam (the beam directed at and reflected by the object) interferes with the reference beam on an image sensor. Here, the optical sensor can be an image sensor for capturing an area image or, depending on the acquisition mode, a point sensor. For other applications, the measuring device can also measure without a reference beam, i.e., it does not rely on the interference of the object beam with a reference beam. In this case, sequential recording using a point sensor as the optical sensor is preferred. Temporally parallel recording using polarization optics (with a point sensor or an area image sensor) is also preferred.According to a further preferred embodiment, the measuring device is designed for a measurement method based on self-interference, such as shearing interferometry or Gabor inline holography. For a snapshot measurement, the measuring device is preferably designed such that the associated reference beams are arranged in such a way that the respective wavelength-specific information can be separated by means of Fourier transformation of the recorded hologram.

[0050] Examples of preferred embodiments of the device according to the invention are shown schematically in the figures. Figure 1 shows an exemplary embodiment of a device according to the invention. Figure 2 shows an embodiment of a device according to the invention with an alternative coupling. Figure 3 shows a basic structure of a preferred measuring device. Figure 4shows a basic structure of another preferred measuring device. Figure 5 shows a basic structure of another preferred measuring device. Figure 6 shows a basic structure of another preferred measuring device. Figure 7 shows a basic structure of another preferred measuring device. Figure 8 shows a basic structure of another preferred measuring device.

[0051] Figure 1 Figure 1 shows an embodiment of a device 1 according to the invention for generating a synthetic wavelength. This device 1 comprises a thermally conductive support element 2 onto which two surface emitters 3 are applied. For example, the support element 2 can be assumed to be a copper block and both surface emitters 3 are soldered onto the support element 2. VCSELs 3 are small, typically with a diameter of approximately 5 mm. This allows them to be positioned close together (< 6 mm).

[0052] The support element 2 is cooled or heated as required by a temperature control element 9. The device may also include a temperature sensor (not shown) and a temperature control unit (also not shown) for controlling the temperature control element 9.

[0053] The two surface emitters are each supplied with current at a predetermined voltage by a power supply unit 6. For proper power supply, a power supply unit 6 can include everything a surface emitter 3 needs for its intended operation, e.g., a series resistor.

[0054] At least one of the power supply units 6 is adjustable so that the two surface emitters 3, which emit the same intended wavelength, can be operated with different current intensities. This allows the fundamental wavelengths to differ slightly, thereby generating the synthetic wavelength according to the invention. It is important to note that no "light beam" with a synthetic wavelength is generated. The wavelength range detected during a measurement always corresponds to the emitted fundamental wavelengths of the light sources. The synthetic wavelength is generated later during data processing.

[0055] The light generated by the surface emitters 3 is guided by optical fibers 5, which allow the light to exit at their ends 4. The figure shows two options, each of which can be used individually or together. First, the two beams from the surface emitters 3 can be combined, as shown in the center (solid lines of the optical fibers 5), and exit together as a light beam to illuminate the object with the fundamental wavelengths (for measurement with the synthetic wavelength). Second, the two beams from the surface emitters 3 can be guided separately, as shown at the sides (dashed lines of the optical fibers 5). These can later be combined to generate the synthetic wavelength or used separately.For example, the combined rays (solid lines) can be used to illuminate the object (as the object ray) and the separately guided rays (dashed lines) as separate reference rays of the two wavelengths, so that, for example, by local separation in two phase outputs, the different directions or solid angles (e.g., vertical and horizontal angle of incidence) necessary for a snapshot application can be implemented on an optical sensor.

[0056] The light can exit directly from an optical fiber end 4 (see center), be widened or focused by means of an exit lens 8, or be filtered by a filter element 7, whereby a combination of exit lens 8 and filter element 7 is also possible. All these light exit options can be considered as an output coupling unit K.

[0057] The core of the device 1 is the combination of the support element 2 with the surface emitters 3. The remaining components are optional and could theoretically also be external components.

[0058] Figure 2 Figure 1 shows an embodiment of a device according to the invention with an alternative output coupling. Here, the light from the surface emitters 3 is first guided separately through two optical fibers, enters a beam splitter 10 from different sides through the ends of the optical fibers, is combined in the beam splitter 10, and exits the beam splitter 10 as a light beam with the two fundamental wavelengths that can generate the synthetic wavelength. The beam splitter 10 or an output aperture (or an output window) can be considered here as the output coupling unit K.

[0059] If measurements are to be taken with polarized light beams, polarizers 15 can be arranged at suitable locations. The figure shows three possible positions in which polarizers 15 (dashed lines) can be arranged, with the positions between the lenses 8 and the beam splitter 10 being preferred.

[0060] Since half the light is lost with a standard 50%-50% beam splitter, it is advantageous to use an element with lower loss instead. A dichroic element, for example, is advantageous here because, under favorable wavelength conditions, it reflects one wavelength and transmits the other, thus allowing lossless merging of both beams. However, for clarity, beam splitters will be shown below.

[0061] Figure 3Figure 1 shows a basic structure of a preferred measuring device. For clarity, only the output coupling units K of a device according to the invention are shown. As already explained, these can be, for example, light guide ends 4 (optionally equipped with exit lenses and / or filters), the exit surface of beam splitters 10, or apertures or exit windows. The only important point here is that the object O shown on the left is irradiated with at least two different fundamental wavelengths, which strike the object at the same angle. This means that the beams from the surface emitters 3 of the device 1 are always combined before exiting the lower (left) output coupling unit K.

[0062] The measuring device shown here allows for lensless recording of the diffraction patterns of different wavelengths reflected back from object O. The light beam exiting the lower (left) output coupling unit K is deflected by the beam splitter 10 onto object O, reflected back by object O through the beam splitter 10, and strikes the optical sensor 11 (see dashed beam paths). The surface structure of object O can be determined using known evaluation methods based on the recorded information.

[0063] The measuring device in this figure can perform measurements without a reference beam. However, for interferometric or holographic measurements, a reference beam can also be introduced, e.g., by means of an additional output coupling unit K, appropriate beam guidance through further optical elements, or by rearranging the elements shown, so that the beam splitter 10 uses the light exiting the output coupling unit K as both an object beam and a reference beam. A preferred interferometric or holographic embodiment, in which the object beam can interfere with obliquely incident reference beams, is described below.

[0064] In Figure 3The figure also shows how the measuring unit can be modified for interferometric or holographic recording (components shown in dashed lines). Using two additional coupling units K, a reference beam, or, as shown here, two reference beams with the two separately guided fundamental frequencies, are directed either directly or via optical elements (e.g., mirrors) onto the optical sensor 11. Here, these light beams strike the optical sensor at a slight angle to avoid shadowing of the object beam and to generate the carrier frequency necessary for the snapshot recording. One light beam can strike the sensor obliquely from above or below, another obliquely from the side (off-axis). This results in a hologram, which can be used to measure the object using known evaluation methods.

[0065] Figure 4Figure 1 shows a basic setup of another preferred measuring device in the form of a point detector. The measuring device operates with a light beam that is focused onto the object at a single point by means of a lens 12. The object is moved during the measurement by means of a motion unit (symbolized by arrows) and scanned sequentially. Here, too, the light beam illuminating the object (again with the synthetic wavelength) exits the lower (left) output unit K, is deflected by the beam splitter 10 onto the object O, reflected back from the object O through the beam splitter 10, and, focused by a lens 12, strikes the optical sensor 11. For a topographic (e.g., interferometric) measurement, a reference beam exits the upper output unit K and is deflected by the beam splitter 10 onto the optical sensor 11.

[0066] In a simple embodiment, both the object beam and the reference beam can have the synthetic wavelength (or the fundamental wavelengths that result in the synthetic wavelength). Alternatively, it is also possible to measure with polarized light. For this purpose, an arrangement such as that described in the following could be positioned in front of each of the output coupling points K. Figure 2 can be seen, whereby the light guides supply light from a surface emitter 3 each and combine it in a beam splitter 10 (see. Figure 2 ), where the two beams have different polarizations (the beam from one surface emitter 3 is p-polarized and the other is s-polarized). It is very advantageous if the two surface emitters 3 are switched alternately, so that the optical sensor 11 alternately receives radiation with different polarizations.

[0067] In Figure 5 is a variation of Figure 4shown, which is particularly advantageous for simultaneous measurement with polarized light. For this purpose, a light beam should emerge from the upper left output coupling unit K, in which a first fundamental wavelength emerges with p-polarization and a second fundamental wavelength with s-polarization. This can be achieved, for example, by means of a device according to Figure 2 This can be achieved. Preferably, corresponding polarization filters are arranged after the two ends of the optical fiber. However, this is not strictly necessary, for example, if the light exiting the fiber is already polarized. This can be achieved, for instance, by guiding linearly polarized light from a single-mode VCSEL into a polarization-maintaining fiber. The resulting beam is directed towards the object O and passes through a quarter-wave plate 14, which rotates the polarization by 45°.

[0068] Polarized light from the corresponding reference beams is coupled out of the output unit K, positioned on the right in the image, and directed by a combination polarization beam splitter 10a, depending on its polarization, onto one of the two optical sensors 11 shown. The beam reflected from object O also passes through the combination polarization beam splitter 10a and is directed onto one or the other optical sensor 11 according to its polarization.

[0069] Due to the double transmission of the object beam through the quarter-wave plate 14, which is rotated by 45°, the polarization of the object beam at the combination polarization beam splitter 10a is rotated by 90°. Thus, the p-polarized object beam transmitted through the combination polarization beam splitter 10a, together with the corresponding s-polarized reference beam, can be focused at a common point on the right optical detector 11. To enable interference between the reference and object beams, a polarizer 15, rotated by 45°, is placed in the still collimated beam path. A corresponding arrangement is also used for the lower optical detector 11 to enable interference between the s-polarized object beam and the p-polarized reference beam. The optical sensors 11 are preferably point sensors.

[0070] The measuring device may also include a holographic surveying arrangement, as described in the modification to Figure 3 is shown.

[0071] In Figure 6 is a variation of Figure 5This is shown, which is particularly advantageous for simultaneous measurement with polarized light on rough surfaces. For this purpose, a light beam emerges from the lower output unit K, in which the first and second fundamental wavelengths of the object beam are not necessarily polarized. The object beam is directed towards the object O by a beam splitter 10 and passes through a polarizer 15 rotated by 45°. Polarized light from the corresponding reference beams is coupled out of the output unit K, which is positioned at the top of the image, and directed by the beam splitter 10 and a subsequent combination polarization beam splitter 10a, depending on its polarization, onto one of the two optical sensors 11 shown. The beam reflected from the object O also passes through the beam splitter 10 and the combination polarization beam splitter 10a and is directed onto one or the other optical sensor 11 according to its polarization.Thus, the p-polarized object beam transmitted through the combination polarization beam splitter 10a can be focused together with the associated s-polarized reference beam onto the right optical detector 11 at a common point.

[0072] Determining the phase of the wavelength is advantageous for reducing measurement uncertainty. This can be achieved by positioning a preferably 4-quadrant detector or four photodiodes with a preceding combination of a lambda-quarter plate rotated by 45° and differently rotated polarizers 15, so that the phase can be directly determined for the respective wavelength (see, e.g., S. Suja Helen, MP Kothiyal, and RS Sirohi, "Achromatic phase-shifting by a rotating polarizer", Opt. Comm. 154, 249 (1998)).

[0073] A corresponding arrangement also applies to the lower optical detector 11 to enable the interference of the s-polarized object beam and the p-polarized reference beam and to determine the phase. The optical sensors 11 are preferably point sensors.

[0074] The measuring device may also include a holographic surveying arrangement, as described in the modification to Figure 3 is shown.

[0075] In Figure 7 is another variation of Figure 5The diagram shows a circulator 16 with three coupling points A, B, and C. A synthetic wavelength from an output coupling unit K is coupled into coupling point B, exits at coupling point C, strikes an object O, and is reflected back by it. The reflected light re-enters coupling point C and exits at coupling point A. Polarized light can again be coupled out as reference beams from the output coupling unit K, positioned above in the diagram, and directed by the combination polarization beam splitter 10a, depending on the polarization, onto one of the two optical sensors 11 shown. This arrangement has the advantage that the intensity of the beam reflected by object O is comparatively high.

[0076] Figure 8Figure 1 shows a basic setup of another preferred measuring device in which the object is first imaged through an aperture 13 by means of a lens 12, and this image is then captured by an optical sensor 11, either in focus (distance between aperture 13 and optical sensor 11 within the depth of field of the optics 12) or out of focus (distance greater than the depth of field of the optics 12). The output coupling unit again emits light with two slightly different fundamental wavelengths to generate a synthetic wavelength. The modifications described above are also possible here.

[0077] Finally, it should be noted once again that the use of the indefinite articles "ein" or "eine" does not preclude the possibility that the characteristics in question may be present multiple times. Reference symbol list

[0078] 1 Device 2 Carrier element 3 Surface emitter 4 Light guide end 5 Light guide 6 Power supply unit 7 Filter element 8 Exit lens 9 Temperature control element 10 Beam splitter 11 Optical sensor 12 Lens 13 Aperture 14 Lambda quarter plate 15 Polarizer 16 Circulator A, B, C Coupling point K Output coupling unit O Object

Claims

1. Device (1) for generating a synthetic wavelength comprising at least two surface-emitters (3) and a carrier-element (2), wherein the carrier-element (2) is a heat conductor and the surface-emitters (3) are fixedly arranged in, at or on the carrier-element (2), characterized in that the ratio V of the distance a between two surface-emitters (3) and the effective thermal conductivity w of the carrier material V = w / a is greater than 5000 W / (m2·K), wherein at least two surface-emitters (3) are located on the same substrate and the distance between the two surface-emitters (3) is at least 1 mm to prevent cross-coupling of the temperature.

2. Device according to claim 1, characterized in that it comprises a temperature-sensor, a temperature-control-element (9) and a temperature-control-unit for controlling the temperature of a surface-emitter (3) and / or the carrier-element (2), preferably a thermal resistor or semiconductor sensor as a temperature sensor and / or a heating element or cooling elements or elements that are suitable for both heating and cooling as a temperature control element (9), wherein the device (1) particularly preferably comprises a Peltier-element.

3. Device according to one of the preceding claims, characterized in that the carrier-element (2) is essentially made of materials whose thermal conductivity is at least greater than 40 W / (m·K), in particular greater than 200 W / (m·K) and preferably contains aluminum or copper, wherein the surface-emitters (3) are preferably applied directly on the carrier-element (3) in an electrically conductive manner.

4. Device according to one of the preceding claims, characterized in that the ratio V of the distance a between two surface-emitters (3) and the effective thermal conductivity w of the carrier material V = w / a is greater than 20,000 W / (m2·K), in particular greater than 50,000 W / (m2·K), wherein the distance between the surface-emitters (2) is preferably not greater than 1 cm, in particular not greater than 5 mm.

5. Device according to one of the preceding claims, characterized in that it has at least one controllable energy supply unit (6) which enables an accuracy of the current control in the range of a few µA or below 19 µA, wherein the energy supply unit (6) is preferably designed such that its current fluctuations are less than 10 µA.

6. Device according to one of the preceding claims, characterized in that it has more than two surface-emitters (3) which are electronically and optically connected in such a way that more than two fundamental wavelengths are emitted and two or more synthetic wavelengths are generated simultaneously.

7. Device according to one of the preceding claims, characterized in that it comprises a light-guide-system which is designed to guide the light beams of a number of surface-emitters (3) and preferably to separately guide or combine the beams of two surface-emitters (3), wherein the light guide system preferably comprises an element from the group of mirrors, beam splitters (10), lenses (8), filter elements (7), polarization elements and light guides (5).

8. Device according to one of the preceding claims, characterized in that it comprises a number of filter elements (7) in the beam path of a number of surface-emitters (3), wherein the number of filter elements (7) is dimensioned such that sidebands of the emissions of the surface-emitters (3) are removed and the main wavelengths are allowed to pass, wherein a filter element (7) is preferably mounted on the input and / or output side relative to a light-guide-system.

9. Device according to one of the preceding claims, characterized in that at least two surface-emitters have the same intended wavelength, preferably wherein at least two of the surface-emitters have fundamental wavelengths that differ from one another by less than 10 nm, preferably less than 5 nm.

10. Device according to claim 9, characterized in that the fundamental wavelengths of at least two surface-emitters differ due to manufacturing tolerances.

11. Device according to one of the preceding claims, characterized in that the surface-emitters are positioned in such a way that cross-coupling of the temperature of active zones of the surface-emitters is avoided, wherein the surface-emitters preferably comprise substrates and are positioned in such a way that the substrates do not touch in the area of the active zones.

12. Device according to one of the preceding claims, characterized in that at least two surface-emitters are located on a substrate, in particular on a wafer.

13. Device according to one of the preceding claims, characterized in that the carrier material comprises or consists of aluminum nitride and / or aluminum oxide.

14. Measuring device for topographical measurement, characterized in that it comprises a device (1) according to one of the preceding claims, and an optical sensor (11).

15. Measuring device according to claim 14, characterized in that it comprises one or more of the following components: - an optical element (10, 12), preferably an element from the group of beam splitters (10), mirrors, prisms and lenses (12), with which the light emerging from the device is directed onto the object (O) to be examined, - an evaluation module for determining the phase signal recorded by the optical sensor (11) corresponding to a synthetic wavelength, in particular by forming the difference between the at least two quantitative phase signals corresponding to the fundamental wavelength, - a movement unit for moving an object (O) to be measured in a designated measuring range, - an arrangement of a coupling-out unit (K) of the device (1), which directs a reference beam with a separately guided fundamental wavelength directly or deflected via an optical element onto the optical sensor (11), so that a holographic recording is achieved, - a coupling-out unit (K) which emits two different fundamental wavelengths of different polarization, wherein the Measuring unit preferably has two optical sensors (11) and an optical unit which is designed to direct light with different polarizations in different directions, - an imaging-optics (12) which generates an image of the object (O) which is positioned such that it is recorded by the optical sensor (11).