Veeco Instruments Inc.

Bruker Nano Gmbh

Other

Veeco Corporation

Joanna Schmit

Erik Novak

Artur Olszak

Florin Munteanu

Paul R Unruh

Matthew Jarrod Novak

Son Hoang Bui

Dong Chen Chen

Colin T Farrell

Mark A Schmucker

Jaime Duran

Akiko Harasaki

Shawn David McDermed

Parameswaran Hariharan

John Christopher Boney

Anthony L Martinez

Tay-Chang Wu

Michael B Krell

Side illumination is combined with scanning interferometry to provide a means for measuring with a single data-acquisition scan surfaces that contain sections suitable for interferometric processing as well as sections that are not suitable because of lack of fringes produced by the measurement. In the sections where no fringes are produced, the irradiance detected during the scan is processed using a depth-from-focus mapping method to yield a corresponding measurement. The result is a complete profilometric measurement of the sample surface with a single scan. In addition, by increasing sample irradiance through side illumination, the structural features of the sample become markedly more visible than when illuminated only by the object beam of the interferometer, which greatly facilitates finding focus and identifying regions of interest for the measurement.

Side illumination in interferometry

The rough bottom surface of a recessed feature partially obscured by an overlying structure is profiled interferometrically with acceptable precision using an objective with sufficiently large numerical aperture to illuminate the bottom under the obscuring structure. The light scattering produced by the roughness of the surface causes diffused light to return to the objective and yield reliable data fringes. Under such appropriate numerical-aperture and surface roughness conditions, the bottom surface of such recessed features can be profiled correctly simply by segmenting the correlograms produced by the scan and processing all fringes that correspond to the bottom surface elevation.

Signal sectioning for profiling printed-circuit-bord vias with vertical scanning interferometry

A coordinate measuring machine is calibrated by taking optical measurements of an optical flat that includes a grating placed on its surface. The arm of the CMM capable of multi-directional translation in relation to an object is fitted with a white-light interferometric objective and optical measurements are taken of the flat while translating the objective (or the flat) in the coordinate direction subject to calibration. The objective may serve also as the probe of the CMM.

Optical plate for calibration of coordinate measuring machines

A true-color image of a sample is obtained from interference data captured with a color camera. The irradiance of each color on the respective photo-sensor represents the sum of DC components received from the object and the reference surface and a modulated interference component. The color is determined at each pixel by removing the interference component and the reference-surface component from the irradiance data. The color map so derived is then combined with the height map produced with the same data to yield a true-color 3D map of the sample.

Full-color images produced by white-light interferometry

The bottom surface of a via drilled in a fiber-reinforced PCB is profiled interferometrically with acceptable precision using an objective with sufficiently large numerical aperture to illuminate the bottom under the fibers. The light scattering produced by the inherent roughness of the surface of the via bottom causes diffused light to return to the objective and yield reliable data fringes. Under such appropriate numerical-aperture and surface roughness conditions, the bottom surface of vias can be profiled correctly simply by segmenting the correlograms produced by the scan and processing all fringes that correspond to the bottom surface elevation.

Signal sectioning for profiling printed-circuit-board vias with vertical scanning interferometry

An adaptive algorithm is tailored to fit the local fringe frequency of single-frame spatial-carrier data under analysis. Each set of data points used sequentially by the algorithm is first processed with a Fourier Transform to find the local frequency of the fringes being analyzed. That information is then used to adapt the algorithm to the correct phase step thus calculated, thereby optimizing the efficiency and precision with which the algorithm profiles the local surface area. As a result, defects are identified and measured with precision even when the slope of the surface varies locally to the point where the algorithm without adaptive modification would not be effective to measure them. Once so identified, the defects may be measured again locally with greater accuracy by conventional temporal PSI.

Large-surface defect detection by single-frame spatial-carrier interferometry

The conventional two plates with beamsplitter of the Mirau interferometer are replaced with two achromatic &#x3bb;/4 retarders. The upper surface of the second retarder is coated with a 50 percent reflecting film, so that it also functions as a beamsplitter. The objective is illuminated with a linearly polarized beam. As a result of this arrangement, the test and reference beam emerging from the Mirau interferometer are orthogonally polarized beams suitable for achromatic phase shifting, thereby facilitating the use of the Mirau interferometer for monochromatic to broadband phase-shifting interference microscopy. Alternatively, it can be used for equalization of beams intensity by placing a rotatable polarizer at the exit of the objective.

Polarization mirau interference microscope

An interferometric intensity equation includes parameters that depend on bandwidth and numerical aperture. An error function based on the difference between actual intensities produced by interferometry and the intensities predicted by the equation is minimized iteratively with respect to the parameters. The scan positions (i.e., the step sizes between frames) that minimized the error function are then used to calculate the phase for each pixel, from which the height can also be calculated in conventional manner. As a result, the phase map generated by the procedure is corrected to a degree of precision significantly better than previously possible.

Interferometric iterative technique with bandwidth and numerical-aperture dependency

An interferometric profiler is used to measure object motion by modifying the motion of the scanner so that the phase variation at each scanning step is kept within the acceptable limits of the algorithm used to calculate phase changes. The scanner motion is so manipulated on the basis of prior knowledge about the nature of the object motion, or knowledge obtained by pre-calibration, or by real-time feedback based on current measurements. The object motion is recovered from the scanning information by subtracting the scanner position from the object position as it evolves during the scan.

Measurement of object deformation with optical profiler

Two threshold parameters are used to identify the intensity modulation peaks corresponding to the interfaces of the two sides of a thin film with the adjacent media. The first parameter is used to distinguish modulation data from noise and is set on the basis of actual background noise data measured during the interferometric scan. The second parameter is used to separate actual contrast data from signals of relatively high modulation that satisfy the first parameter but do not in fact result from interference fringes. Data that satisfy both parameters are considered valid modulation data and the peak of each modulation envelope is then calculated using conventional means. The thickness of the film at each pixel is obtained by dividing the scanning distance corresponding to the two peaks by the group index of refraction of the film material.

Film thickness and boundary characterization by interferometric profilometry

Composite height profiles are produced by taking successive interferometric measurements of different sections of a sample surface by sequentially placing them within the field of view of the instrument. A reference signal is used to provide a full history of scanner motion during each measurement scan. The reference signal is independent of the fringes collected for profile-measurement purposes and is used to produce a z-position history of the scan that is independent of scanner nonlinearities and other error sources. As a result, errors caused by scanner nonlinearity and lack of repeatability are removed from the process and it is possible to combine profiles of sections that are spatially disconnected without loss of precision attributable to scanner imperfections.

Reference signal for stitching of interferometric profiles

An electronic template delineating distinct selected patterns corresponding to predetermined regions of interest in a sample part is used to limit analysis to those regions. The surface of the sample is first measured using conventional techniques. The data so acquired are used to identify boundaries between distinct regions, which are then compared to a predetermined pattern boundary in the template to find a best-fit match. The position of the pattern is then shifted to overlay the match, thereby automatically aligning the template's selected patterns with the regions of interest in the sample surface. As a result, profilometric analysis can be limited to the regions of interest. Correction factors are also assigned to each selected pattern in the template to account for physical differences in the corresponding regions of interest of the sample part that affect the profilometric measurement.

Alignment and correction template for optical profilometric measurement

Interferometric measurements are carried out in conventional manner to produce a correlogram corresponding to successive scanner steps. An approximation of the actual scan-step size between frames is calculated from multiple-frame intensity data collected around the frame of interest using common irradiance algorithms. The scan-step size so measured is then used in standard PSI, VSI or PSIOTF algorithms, instead of the scanner's nominal phase step. According to one embodiment, the invention utilizes a five-frame PSI algorithm to produce an average scan-step size of four scan steps. According to another embodiment, the phase step between frames is calculated directly utilizing a novel five-frame algorithm that produces an approximation of actual phase step for a given frame, rather than an average value of four steps around the frame. The method requires reduced data processing and can advantageously be applied &ldquo;on-the fly&rdquo; as intensity data are acquired during scanning.

Correction of scanning errors in interferometric profiling

A reference signal is used to track the actual behavior of the scanner in an interferometer to produce scanner-position data that can be used to correct errors introduced by scanner nonlinearities and other error sources. A narrow-band light source is advantageously utilized to cover the entire range of operation of the scanner. Because of the independent reference channel, the invention is suitable for implementation with all types of conventional interferometric techniques. The concept is preferably implemented by utilizing an additional light source and the same scanner used for the measurement, so that the OPD varies in synchronization of both the reference-signal and data-collection procedures. Alternatively, a high temporal-coherence filter may be used with the same light source and optical path used for the interferometric measurement.

Scanning interferometry with reference signal

Bat wings are removed at step discontinuities within the coherence length of the light source of a vertical-scanning interferometer. A first height profile is obtained from a correlogram using a coherence-sensing technique. A second height profile is obtained from phase measurements at the best-focus frame position of the scanner. The two profiles are compared, and phase ambiguities are removed in conventional manner. In addition, during unwrapping the differences in height between two adjacent pixels obtained both by coherence sensing and by phase measurements are compared to f&lgr;/4. Where the inter-pixel height difference calculated by coherence sensing is smaller and the inter-pixel height difference calculated by phase is larger than f&lgr;/4, the phase measurement is corrected by 2&pgr; increments until both coherence and phase inter-pixel height differences are within f&lgr;/4. This additional step removes bat-wing effects from profiles obtained by phase measurement.

Growing community of inventors

Tucson, AZ, United States of America

Joanna Schmit