Detection of nanoparticles gets more and more relevant in various fields of application, e.g., semiconductor technology, medicine or biology. A fast and universal detection method of nanoparticles is essential for these leading-edge technologies. Classical imaging methods struggle with the resolution of nanoparticles smaller than the diffraction limit. Established optical near-field methods are complex and time consuming. We demonstrate that using Coherent Fourier Scatterometry can overcome most of these challenges. For this purpose, we study the optical response of gold nanosphere arrangements on silicon wafers. In particular, we examine arrays of particles arranged in geometrical structures such as lines, squares, triangles, and L-shapes. The dimensions of these structures are in the order or even smaller than the Airy diameter. Correlations between the geometry of the particle arrangement and the intensity distributions in the Fourier plane are characterized by applying different analyzing methods. Thus, it is possible to distinguish between different structures and, furthermore, to extract their geometrical orientations. In addition, gold nanostructures of defined shape, but different sample thicknesses top-down fabricated on silicon are also investigated. These nanostructure shapes studied comprise the same geometries as mentioned above as well as spirals with a diameter comparable to the spot size of the used microscope objective.
Over the last two decades solid immersion objectives have been developed for various applications, offering the opportunity to achieve a higher resolution than is possible for conventional air objective lenses. For semiconductor applications hemispherical solid immersion lenses (SILs), incorporating tips fabricated in semiconductor materials with a high refractive index (up to n = 3:5 for silicon, for example) are commonly used. With such high refractive index materials it is possible to design an objective with a numerical aperture of NA = 3:2. An apochromatic color correction is mandatory if a broad spectral range from λ = 1200 nm to λ = 2000 nm is required. It is well known that glasses with anomalous partial dispersion must be used to realize apochromatic color correction. It will be shown that the anomalous partial dispersions of some glasses in the IR range differ from the known behavior in the visible region. Therefore, glass selection plays a significant role for the design of a high NA broadband IR objective and will be discussed. Monte Carlo tolerance analysis shows that even with state of the art manufacturing capabilities the tolerance induced aberrations of an objective with NA = 3:2 will lead to a dramatic loss of image performance, indicated by a significant drop in the Strehl-ratio. This becomes even worse when the interface of the exchangeable hemispherical SIL tip with the main objective is considered. With the aim to manufacture such objectives within a stable production process it makes no sense to overload the requirements regarding the NA. Therefore, due to manufacturing issues the numerical aperture of this objective has been restricted to NA = 2:9. Wave front measurements of the manufactured objectives shows Strehl-ratios of SR > 97 % which guarantees a diffraction limited resolution.
Increasing challenges of the industry to improve camera performance with control and test of the alignment process will be discussed in this paper. The major difficulties, such as special CFAs that have white/clear pixels instead of a Bayer pattern and non-homogeneous back light illumination of the targets, used for such tests, will be outlined and strategies on how to handle them will be presented. The proposed algorithms are applied to synthetically generated edges, as well as to experimental images taken from ADAS cameras in standard illumination conditions, to validate the approach. In addition, to consider the influence of the chromatic aberration of the lens and the CFA’s influence on the total system MTF, the on-axis focus behavior of the camera module will be presented for each pixel class separately. It will be shown that the repeatability of the measurement results of the system MTF is improved, as a result of a more accurate and robust edge angle detection, elimination of systematic errors, using an improved lateral shift of the pixels and analytical modeling of the edge transition. Results also show the necessity to have separated measurements of contrast in the different pixel classes to ensure a precise focus position.
This article describes the progress in the area of modern centration technology by using digital image processing. This work is motivated by the continuously increasing demand for high-end optics. During the last years the surface lens quality has been continuously improved. Today the image quality is more determined by the manufacturing tolerances for the mechanical interface which is responsible for decenter and tilt of the lenses respectively the subgroups. Some of the aberrations are directly linked to the decenter of the lenses, Coma for example. Hence it is necessary to realize the subgroups with tolerances below lpm. To determine the decenter of a lens an auto collimation telescope is used to image the reflex of the lens surfaces onto a detector, commonly a half covert photodiode. Rotating the lens generates a sinusoidal signal, which is evaluated by a lock-in amplifier to drive two actuators to adjust the alignment chuck. Typical internal reflections caused by stray light for example disturb the current procedure in such a way that it is impossible to get a stable alignment process. Digital image processing allows us to fix these problems with image recognition. We will demonstrate how a modified auto collimation telescope in combination with the developed software algorithms made the manufacturing process more accurate, faster and useable for a broad spectrum of lenses. It has been proofed by some thousand diverse lenses that with these new technique subgroups can be centered within 0.25μm.
White-light interferometry is an absolute 3D-measurement technique, used for the inspection of structured silicon and
other materials with high quality surfaces. In this technique, each pixel of the camera detects a separate interference
signal, which correlates with the height of the corresponding object point. Different signal processing algorithms are
used, which extract the height from the interference signal by using the coherence or the phase information of the signal.
However, measurement errors can occur if there are chromatic aberrations in the interferometer system. Then the phase
information correlates with the height information in an unexpected manner and there are often disturbing 2&pgr; phase
jumps in the numerical evaluation process, although the topography of the object is continuous and a light source with a
short coherence length is used. We examined a Mirau type white-light interferometer with chromatic aberrations and
explain how mirrorlike, tilted objects cause a correlation of the phase and the height information in each interference
signal. We also show that this measurement error depends on both the slope of the object point and its field position. A
comparison of measurements and a simulation, which shows the described correlation effect, is given.
During the last years, new microscope applications require an increased resolution which enforces the development of new state of the art high NA immersion objectives. With the introduction of the 4Pi confocal fluorescence microscope, the increase of the numerical aperture from NA=1.4 to NA=1.46 makes sense, although the gain of lateral resolution is quite small. On the other hand, for inspection and metrology in the semiconductor industry the continuously decreasing structures need the highest possible resolution, which can be achieved with high NA water immersion objectives working in the DUV wavelength range. Building this kind of objectives requires special measuring and testing technologies and a manufacturing precision which has never been realized before in series production.
Cemented doublets and triplets can not be used for objectives working at wavelengths of 248 nm and shorter, because the optical cement can not withstand the high photon energies. It will be shown that high NA deep UV objectives can be designed and built successfully with the help of air spaced doublets. Assuring Strehl ratios above 95% enforces very tight tolerances. For example the distance error of the lens vertex to its mount has to be less than 1 μm. This calls for a new manufacturing precision never realized before in series production. We show how a white light Mirau interferometer can be used to measure lens vertex positions with an accuracy of 200 nm. We also demonstrate how the fine-tuning process can be optimized by using a "simulated star test", where the point-spread function is calculated in real time with a FFT-algorithm from the optical path difference data, acquired by a Twyman-Green interferometer.
To realize the required precision, today various measurement techniques and production processes are used. Picking up the subgroups on different machining tools and measurement systems will loosen the accuracy. Here, we present the concept and the layout of a new manufacturing tool where we implemented the different measurement techniques needed in one CNC machining center. This tool is able to 1) adjust automatically the optical axis of the subgroups related to the machining axis better than 0.5 μm with the help of the stick-slip effect where a mechanical impulse is transferred by an electromagnetically driven hammer, 2) measure the lens vertex relative to the shoulder of the mount with an accuracy of 250 nm and 3) do all steps which are necessary to process the lens mount within the accuracies described above.
To realize the image quality of high end objectives, e. g. high NA microscope objectives working in the DUV spectral region the subgroups have to be manufactured with a mechanical precision which is difficult to achieve cost effectively. For high end microscope objectives the accuracy of the diameter of the lens mount must be within 1 µm, the run-out must be met within 1 µm and the distance of the lens vertex relative to the shoulder of the mount must fit within 1 µm. To realize the required precision, today various measurement techniques and production processes are used. Picking up the subgroups on different machining tools and measurement systems will loosen the accuracy. Here, we present the concept and the layout of a new manufacturing tool where we implemented the different measurement techniques in one CNC machining center.
Thin-film coatings in modern optical systems as wideband AR-coatings may have >10 layers and an optical thickness of several λ. Such complex thin-films may introduce pronounced changes in transmission phases with varying angles of incidence, polarization and/or wavelength. "Polarization ray tracing" as utilized by current optical design programs models a "ray" as a "localized plane wave" hitting the air/thin-film/glass system and the transmission properties in phase and amplitude for the p- and s-components are taken into account. However, this only approximates the thin film as a pure phase object of vanishing thickness on a flat surface. Any "ray" crossing a layer of finite thickness will undergo lateral displacement and on a surface of notable curvature, this displacement will further change the direction of the refracted "ray". Both effects might become important in high NA, deep UV microscope objectives based on an air-spaced design that involves a large number of highly curved air/glass interfaces, large angles of incidence and tight tolerances. This paper shows how the equivalent lateral ray displacement and bending can be calculated from the film/glass properties and the surface curvature and how it can be incorporated into a polarization ray-tracing program. It also addresses other problems encountered in polarization ray tracing of thin films, as proper conversion from phase shifts to optical path length and how to easily "unwrap" the thin-film induced phase.
The precise positioning of the individual optical elements is essential for attaining diffraction limited performance in high-numerical-aperture (high-NA) microscope objectives. Tolerances are in the micron range or lower for high-end objectives, e.g. for broad-band scanning confocal applications, metrology objectives in general, and especially for deep ultraviolet (DUV) applications. The ever increasing demands on imaging performance ask for the continuous development and improvement of specialized measurement equipment for the production line. Our award-winning 150x/0.90-DUV-AT-infinity/0 objective for wafer inspection and metrology at 248nm employs air
spacings in its doublets because of the instability of optical cements against DUV radiation. This comes however at the cost of a higher number of surfaces and even higher precision demands on their geometry, orientation and positioning. We present several tools enabling us to meet these requirements. A Fourier transform fringe
analysis scheme is adapted to high-NA Fizeau interferometry for surface characterization. A white light Mirau interferometer for dimensional measurements on lens groups with sub-μm resolution enables us to keep surface distance errors lower than 2 μm. Residual aberrations of the objective are compensated for by translating special correction elements under observation of the wave-front using a DUV-Twyman-Green interferometer, which also
incorporates a 903nm branch for the parfocal adjustment of the infrared (IR) autofocus feature of the objective. To adjust the shifting element for the elimination of on-axis coma, we compute an artificial (real-time) star test from the interferogram, allowing interactive manipulations of the element while monitoring their influence on the point spread function (PSF).
Cemented doublets and triplets, which are the principle parts in high quality, high numerical aperture (NA) objectives, can not be used for objectives working at wavelengths of 248 nm and shorter, because the optical cement can not withstand the high photon energies. We will show that high NA deep UV objectives can be designed and built successfully with the help of air spaced doublets. Assuring Strehl ratios above 95% enforces very tight tolerances. For example the distance error of the lens vertex to its mount has to be <1 μm. This calls for a new manufacturing precision never realized before in series production. We show how a white light Mirau interferometer can be used to measure lens vertex positions with an accuracy of ±200 nm. We also demonstrate how the fine-tuning process can be optimized by using a "simulated star test," where the point-spread function is calculated in real time with a FFT-algorithm from the optical path difference data, acquired by a Twyman-Green interferometer.
The quality of high performance microscope objectives is usually verified by interferometric measurement of the wave front. However, a non-interferometric method might be preferable, when appropriate light sources of sufficient coherence length or an interferometric setup are difficult to realize. Under these circumstances, the complex pupil function of an optical system can also be determined from its intensity point spread function (PSF), i.e. from the image of a sub-resolution point object. We present a system that can determine the pupil function of high-NA microscope objectives from defocused images of an artificial point source. An extended version of the "Misell" algorithm is used, which utilizes 4 or more PSF images to overcome the Fourier phase ambiguity and which generally converges rapidly to the correct pupil function both in phase and amplitude. The algorithm can also compensate small errors in the x-, y- and z-positions of the images, which might be caused by vibrations, thermal drift or the limited accuracy of the z-drive.
We discuss the requirements on design and production regarding geometric and chromatic aberrations for objectives used in 4Pi confocal microscopy. We show that even the selection of a category 1/1 glass will not automatically assure that these requirements are met, due to residual variations in the Abbe number v within the manufacturer's tolerances. Consequently, the optical design has to take into consideration the possibility of balancing chromatic aberrations by varying selected air spacings in the final assembly of each individual objective. We also demonstrate, that for analyzing the influence of aberrations on the intensity distribution along the optical axis, a scalar diffraction theory is still applicable and very useful.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.