Abstract: A system and method of inspecting semiconductor wafers that is capable of determining a scattering power associated with a wafer surface defect whether or not the scattering power associated with the defect exceeds the dynamic range of the system. The scattering power of the detected defect is obtained by determining the height of a Gaussian shape representing data collected by the system. The height of the Gaussian shape may be determined by defining a cross-sectional area of the Gaussian shape at an intermediate height, determining a cross-sectional area value and combining the area value, intermediate height and a slope value m that is representative of a relationship between the area of a cross-section in a Gaussian pulse and the height of the pulse at the cross-section. The technique increases the dynamic range of the equipment with a uniform scan process that can determine all defect sizes in a single scan pass.

Abstract: A system and method of inspecting a semiconductor wafer that may be employed to detect and to characterize defects occurring on an edge of the wafer. The wafer inspection system includes an optical module for providing a light source to scan the wafer edge, a light channel detector for detecting light reflected from the wafer edge, and a processor and memory for converting detected signals to digital form, and for filtering and processing the digital data. The module includes a wafer edge scanning mechanism for projecting a collimated laser beam toward the wafer edge at a predetermined angle of incidence to scan the wafer edge for defects. The light channel detector detects light reflected from the wafer edge to obtain wafer edge data, which are applied to thresholds to determine the location of defects in the wafer edge.

Abstract: A system for inspecting semiconductor wafers capable of determining a scattering power associated with a wafer surface detect whether or not the scattering power exceeds the dynamic range of the system. The scattering power is obtained by determining the height of a Gaussian shape representing data collected by the system. The height is determined by defining a plurality of cross-sectional areas of the Gaussian shape, determining a value of each area, determining a value of the natural logarithm of intermediate heights of the Gaussian shape corresponding to the cross-sectional areas, plotting the area values as function of the natural logarithm of the intermediate height values to form a linear plot, determining a natural logarithm of the height value corresponding to a zero area value based on the linear plot, and determining the inverse natural logarithm of the value to obtain the height of the Gaussian shape.

Abstract: An apparatus for measuring semiconductor wafer shape that minimizes wafer distortion. The apparatus includes a plurality of wafer gripping fingers for holding a wafer in a predetermined position during wafer measurement. Each finger includes a groove that contacts the edge of the wafer. The groove and the wafer edge have respective radii of curvature, in which the radius of curvature of the groove is greater than that of the wafer edge. Each finger includes a rigid member having a recess formed in a central location at one end thereof, and a compliant material such as PEEK disposed in the recess in which the groove is formed. The compliant material extends a first distance beyond the rigid member at the central groove location and a second shorter distance beyond the rigid member on each side of the central location.

Abstract: This method removes high frequency noise from shape data, significantly improves metrology system (10) performance and provides very compact representation of the shape. This model-based method for wafer shape reconstruction from data measured by a dimensional metrology system (10) is best accomplished using the set of Zernike polynomials (matrix L). The method is based on decomposition of the wafer shape over the complete set of the spatial function. A weighted least squares fit is used to provide the best linear estimates of the decomposition coefficients (Bnk). The method is operable with data that is not taken at regular data points and generates a reduced data field of Zernike coefficients compared to the large size of the original data field.

Abstract: An improved non-contact capacitive displacement sensor that may be employed for accurately measuring small distances between the sensor and shaped targets. The non-contact capacitive displacement sensor includes a probe having a sensor element and a guard element. The guard element substantially surrounds the sensor element. At least the sensor element has a shape that substantially matches the shape of a target element.

Abstract: A ring chuck that holds a wafer with a vacuum uses a vacuum trough that contacts the entire outer edge of the wafer. The chuck has a base having a top surface equal to or slightly smaller than a water to be tested with vacuum channels in the base. The base provides the mechanism to connect the chuck to a measurement instrument and a vacuum source. An annulus of non-contaminant material that has concentric rings extending upward from its outer edge is fixed to the base top surface with the trough between the concentric rings connected to the vacuum channels. The vacuum trough holds the wafer securely to the chuck and minimizes vibrations when the wafer is rotated.

Abstract: A population of data points each having three or more parameters associated therewith, such as multi-channel defect data from an optical scanner, are plotted in three dimensions, and groupings of data points are identified. Boundary surfaces are defined in the three-dimensional space for delineating groupings of data points. The different groupings correspond to different data classifications or types. Classification algorithms based on the boundary surfaces are defined. When applied to defect classification, the algorithms can be exported to an optical scanner for runtime classification of defects. An algorithm for identifying a particular grouping of data points can be defined as a Boolean combination of grouping rules from two or more different n-dimensional representations, where n can be either 2 or 3 for each representation.

Abstract: A population of data points each having three or more parameters associated therewith, such as multi-channel defect data from an optical scanner, are plotted in three dimensions, and groupings of data points are identified. Boundary surfaces are defined in the three-dimensional space for delineating groupings of data points. The different groupings correspond to different data classifications or types. Classification algorithms based on the boundary surfaces are defined. When applied to defect classification, the algorithms can be exported to an optical scanner for runtime classification of defects. An algorithm for identifying a particular grouping of data points can be defined as a Boolean combination of grouping rules from two or more different n-dimensional representations, where n can be either 2 or 3 for each representation.

Abstract: An improved non-contact capacitive displacement sensor that may be employed for accurately measuring small distances between the sensor and shaped targets. The non-contact capacitive displacement sensor includes a probe having a sensor element and a guard element. The guard element substantially surrounds the sensor element. At least the sensor element has a shape that substantially matches the shape of a target element.

Abstract: In an optical inspection system, defects such as particles, pits, subsurface voids, mounds, or other defects occurring at or near the smooth surface of a substrate are classified by type and size based on the magnitude S of a signal produced by collected light for each of a plurality N of different test configurations, yielding a plurality of signal magnitudes S1 through SN. A database is consulted, comprising a relationship of S versus defect size d for each test configuration and for each of a plurality of idealized defect types, so as to determine a defect size d corresponding to each measured signal magnitude S, and an average defect size is determined for each defect type. Signal magnitudes <S1> through <SN> that would be produced by a defect of the average size are determined for each defect type, and defect type is determined based on a smallest deviation between the measured magnitudes and the determined magnitudes.

Abstract: Full surface maps of slope and height are determined for the surface of a highly smooth surface such as a silicon wafer, by an apparatus which includes a light source for creating a light beam and scanning and wafer transport systems which cause the incident beam to be scanned over the full surface of the wafer. A quad cell light detector is positioned to receive the beam specularly reflected from the wafer surface, the quad cell detector having four cells arranged in quadrants with each cell providing an electrical signal indicative of the amount of light received by the cell. A processor is programmed to calculate changes in spot location on the detector based on the signals from the cells, and to calculate changes in surface slopes based on the changes in spot location. Full maps of X- and Y-slope are produced, and a line integration algorithm is used for calculating full surface height maps. Regions of surface height gradient exceeding a threshold are flagged as defects by the processor.

Abstract: A non-contact capacitive displacement measurement gage that provides high accuracy displacement measurements of well-connected targets and poorly-connected targets. The capacitive displacement measurement gage includes a capacitive probe, first and second amplifiers, and a signal generator. The capacitive probe includes a sensor electrode, a guard electrode, and a compensating electrode. The signal generator provides a predetermined voltage signal directly to the sensor electrode, to the guard electrode through the first amplifier having unity gain, and to the compensating electrode through the second amplifier having a predetermined transfer function. The second amplifier assures that substantially zero current is driven into the target element by the probe during gage operation, thereby allowing highly accurate displacement measurements of target elements having unknown or poorly controlled impedance.

Abstract: A method to determine the systematic error of an instrument that measures features of a semiconductor wafer includes the following sequential steps. Collecting sensor data from measurement runs on front and back surfaces of a wafer while the wafer is oriented at different angles to the instrument for each run, yielding a front data set and a back data set for each angle. Then organizing the data in each set into a wafer-fixed coordinate frame. Reflecting all back surface data about a diameter of the wafer creates a reflected back data set. Subtracting the reflected back data from the front data for each wafer angle, and dividing the result by two, yields an averaged wafer shape for each load angle. Adding the reflected back data to the front data and dividing the result by two, yields an instrument signature for each load angle. The symmetric corrector is calculated by taking the average over all instrument signatures at each load angle.

Abstract: A method for facilitating field replacement of sensors is presented. The replacement sensor's transfer function need only be measured, not adjusted, and the signal-processing unit's transfer function need only be adjusted, but not measured, to achieve simple field-replacement of sensors. With this method simple calculations and adjustments, typically expedited by means of a computer program, can be made in the field when sensor replacement is required. By way of the presently disclosed method sensor cost and size are not increased since no normalization techniques or components are required.

Abstract: A process for obtaining the periodic component in a data vector of data taken of a sensed property of an element driven by one or more rotary components over several rotations. The procedure comprises first forming the data vector asynchronously with respect to rotation of said one or more rotary components. A series of marker signals representing the rotation of the one or more rotary components is also established. From these the periodic component of data in said data vector is determined as a function of the data in the data vector and the marker signals.

Abstract: A ring chuck that holds a wafer with a vacuum uses a vacuum trough that contacts the entire outer edge of the wafer. The chuck has a base having a top surface equal to or slightly smaller than a wafer to be tested with vacuum channels in the base. The base provides the mechanism to connect the chuck to a measurement instrument and a vacuum source. An annulus of non-contaminant material that has a plurality of concentric rings extending upward from its outer edge is fixed to the base top surface with the trough between the concentric rings connected to the vacuum channels. The vacuum trough holds the wafer securely to the chuck and minimizes vibrations when the wafer is rotated. When the plurality of concentric rings are contained within the wafer exclusion band, the print through onto the tested area is minimized. The tops of the concentric rings are minimally elevated from the top surface of the base and form a sealed volume that supports the interior portion of the wafer.

Abstract: A system and method for realizing vector Kerr magnetometry are disclosed. The system enables simultaneous longitudinal and transverse Kerr effect measurements at each point on a sample surface. An optional component includes a sample platform for achieving precise linear and rotational relocation. The repositionable platform enables complete, 360 degree characterization about a single point. Additionally, the platform control mechanism may be utilized in obtaining longitudinal and transverse Kerr effect measurements at succeeding points on the surface of a sample. Rapid sample characterization is thus achieved.

Abstract: An edge-gripping wafer handling apparatus for holding and transporting wafers in a clean room environment is implemented by mounting gripping fingers on a paddle arm adapted to be controlled by a robot. At the free end of the paddle arm, a plurality of fixed fingers is mounted. Each of these fixed fingers includes a notch for gripping the edge of a wafer. A moving finger is mounted close to the fixed end of the paddle arm. The moving finger can be smoothly moved to engage the edge of the wafer and cause the opposite edge of the wafer to engage the fixed fingers. The movement of the moving finger is caused by a bellows flexing a hinge connected to the finger. The flexing hinge is formed as a parallelogram of two stable ends and two flexing arms. The movement of the hinge is smooth and causes no particulate matter to be generated due to rubbing or other friction generating mechanism.

Abstract: P-polarized light or having a strong P-polarized component is directed onto a filmed substrate at two (or more) different incidence angles, one angle being relatively large and the other angle being relatively small as measured from a surface normal. Light that is scattered into a back region of the hemispherical space above the substrate surface is collected and the intensity of the collected light is measured for each of the two incident angles. A defect can be classified as either a hole in the film or a particle on the film based on the relative intensities of the collected light.