Title:

Kind
Code:

A1

Abstract:

A method is disclosed for processing, optimization, calibration, and display of measured dielectrometry signals. A property estimator is coupled by way of instrumentation to an electrode structure and translates sensed electromagnetic responses into estimates of one or more preselected properties or dimensions of the material, such as dielectric permittivity and ohmic conductivity, layer thickness, or other physical properties that affect dielectric properties, or presence of other lossy dielectric or metallic objects. A dielectrometry sensor is disclosed which can be connected in various ways to have different effective penetration depths of electric fields but with all configurations having the same air-gap, fluid gap, or shim lift-off height, thereby greatly improving the performance of the property estimators by decreasing the number of unknowns. The sensor geometry consist of a periodic structure with, at any one time, a single sensing element that provides for multiple wavelength within the same sensor footprint.

Inventors:

Goldfine, Neil J. (Newton, MA, US)

Zahn, Markus (Lexington, MA, US)

Mamishev, Alexander V. (Cambridge, MA, US)

Schlicker, Darrell E. (Watertown, MA, US)

Washabaugh, Andrew P. (Menlo Park, CA, US)

Zahn, Markus (Lexington, MA, US)

Mamishev, Alexander V. (Cambridge, MA, US)

Schlicker, Darrell E. (Watertown, MA, US)

Washabaugh, Andrew P. (Menlo Park, CA, US)

Application Number:

10/040797

Publication Date:

06/20/2002

Filing Date:

01/07/2002

Export Citation:

Assignee:

JENTEK Sensors, Inc. (Waltham, MA)

Primary Class:

International Classes:

View Patent Images:

Related US Applications:

20090222224 | Multi-circuit direct current monitor with Modbus serial output | September, 2009 | Lewis et al. |

20020005730 | Comprehensive application power tester | January, 2002 | Heiman et al. |

20080246493 | Semiconductor Processing System With Integrated Showerhead Distance Measuring Device | October, 2008 | Gardner |

20090306839 | Aircraft monitoring equipment | December, 2009 | Youngquist et al. |

20100066371 | INTRABODY MRI STACKED FLAT LOOP ANTENNAS AND RELATED SYSTEMS | March, 2010 | Vij |

20080265913 | Sealed Capacitive Rain Sensor | October, 2008 | Netzer |

20070013408 | Inspection device for display panel and interface used therein | January, 2007 | Hamamoto |

20100060259 | Determining Electric Grid Endpoint Phase Connectivity | March, 2010 | Vaswani et al. |

20080164870 | Arrangement for Measuring the Position of a Magnet Relative to a Magnetic Core | July, 2008 | Beichler et al. |

20090027057 | Ethernet Electrometer | January, 2009 | Anderson et al. |

20080157782 | Analog boundary scanning based on stray capacitance | July, 2008 | Krah |

Primary Examiner:

WEST, JEFFREY R

Attorney, Agent or Firm:

HAMILTON, BROOK, SMITH & REYNOLDS, P.C. (CONCORD, MA, US)

Claims:

1. A method for generating, and evaluating, property estimation grids for use with a dielectrometer for measuring preselected properties of a material comprising the steps of: a) defining electrical, physical, and geometric properties for a material including preselected properties of the material; b) defining operating point parameters and an electrode geometry, electrode configuration, substrate material and dimensions, and electrical source excitation for the dielectrometer; c) inputting the material properties, the operating point parameters, and the dielectrometer electrode substrate geometry, configuration and source excitation into a model to compute and input/output terminal relation value; d) recording in a database the terminal relation value as a property estimation grid point; e) adjusting the preselected properties of the material and repeating steps c) and d).

2. A method as claimed in claim 1 where the additional step is added of f) analyzing the resultant grid(s) to determine their numerical properties and properties as mappings between measurement and property spaces to allow their comparison, determination of fitness for various measurements and the implications upon the whole measurement strategy, and to allow selection among grid and measurement alternatives.

3. A method as claimed in claim 1 wherein the terminal relation values of part c) are transcapacitance and transconductance values; transadmittance values; transimpedance values; self-admittance values; self-impedance values; complex gain; or any electrical equivalent circuit or network representation.

4. A method as claimed in claim 1 wherein one or more of the operating point parameters in parts b) and c) are single or multiple shims of known property and geometry.

5. A method as claimed in claim 1 wherein one of the operating point materials in parts b) and c) is a variable liquid mixture of unknown properties.

6. A method as claimed in claim 1 further comprising the step of plotting the terminal relation values on a single or multidimensional grid.

7. A method as claimed in claim 1 where the grid points are magnitude and phase for a single wavelength dielectric sensor.

8. A method as claimed in claim 1 where the grid points are magnitude at one wavelength and magnitude at a second wavelength.

9. A method as claimed in claim 8 where the magnitude-magnitude grids are for measurements on substantially nonconducting media.

10. A method as claimed in claim 1 where the grid points are a magnitude or phase measured with a dielectric sensor and a parameter measured with a non-dielectric sensor.

11. A method as claimed in claim 1 wherein one or more of the operating point parameters in b) and c) are temperature dependent and variations in the temperature are used to alter the operating point.

12. A method as claimed in claim 2 wherein step b) comprises: defining initial dielectrometer operating point parameters and an electrode geometry, electrode configuration, substrate material and geometry, and electrical source excitation for the dielectrometer; inputting the material properties, the dielectrometer operating point parameters, and the dielectrometer electrode geometry, configuration, substrate material and geometry, and source excitation into a model to compute an input/output terminal relation value; adjusting the preselected properties of the material to compute another terminal relation value; computing the Jacobian elements which are measures of the variation in said terminal relation values due to the variation in the preselected material properties; computing a singular value decomposition for the Jacobian elements to obtain singular values, singular vectors and condition number of the Jacobian elements; evaluating sensitivity, selectivity, and dynamic range of the dielectrometer electrode and substrate structures and operating point using the singular values, singular vectors, and condition numbers for material property estimate requirements; adjusting the dielectrometer operating point parameters and electrode geometry, configuration, substrate material and geometry, and source excitation and repeating Steps b-f until the material property estimate requirements are achieved.

13. A method as claimed in claim 12 wherein the singular values, singular vectors, and condition number are stored with grid points to support a grid interpolation algorithm to obtain property estimates.

14. A property estimator as claimed in claim 12 wherein the property analyzer converts each sensed electromagnetic response into a transadmittance or transimpedance magnitude and phase or equivalently into real and imaginary parts; or into equivalent electrical circuit or network representation.

15. A method as claimed in claim 1 where the material under test is composed at least in part of a viscous material.

16. A method as claimed in claim 15 where the viscous material is curable such as an epoxy.

17. A method as claimed in claim 15 where the material is monitored in an on-line configuration as part of a quality control process.

18. A method for selection of a dielectrometer electrode and substrate structures and operating point for measuring one or more preselected properties of an material comprising the steps of: a) defining electrical, physical, and geometric properties for a material including preselected properties of the material; b) defining dielectrometer operating point parameters and an electrode geometry, electrode configuration, substrate material and geometry, and electrical source excitation for the dielectrometer; c) inputting the material properties, the dielectrometer operating point parameters, and the dielectrometer electrode geometry, configuration, substrate material and source excitation into a model to compute an input/output terminal relation value; d) adjusting the preselected properties of the material to compute another terminal relation value; e) computing Jacobian elements which are measures of the variation in said terminal relation values due to the variation in the preselected material properties; f) computing a singular value decomposition for the Jacobian elements to obtain singular values, singular vectors and condition number of the Jacobian elements; g) evaluating sensitivity, selectivity, and dynamic range of the dielectrometer electrode structure and operating point using the singular values, singular vectors, and condition numbers for material property estimate requirements; h) repeating Steps c-g with adjusted dielectrometer operating point parameters and electrode geometry, configuration, substrate material and geometry, and source excitation until the material property estimate requirements are achieved.

19. A method using a property estimator which accesses a property estimation grid for translating the sensed electromagnetic responses of a dielectrometer into estimates of the preselected properties of the material, the property estimator generating the property estimation grid by successively implementing a model which provides for each implementation prediction of a response for the preselected properties based on a set of properties characterizing the electrode and substrate structures and the material.

20. A property estimator which accesses a property estimation grid for translating each sensed electromagnetic response into a proximity (or lift-off) estimate, the property estimator generating the property estimation grid by successively implementing a model which provides for each implementation a prediction of a response for a particular proximity based on a set of properties characterizing the dielectrometer electrode and substrate structures and the material under test.

21. A method for generating property estimates of one or more preselected properties of a material comprising: a) providing an electromagnetic structure capable of imposing an electric field in the material when driven by an electrical signal and sensing an electromagnetic response, an analyzer for applying an electric signal to the electromagnetic structure and sensing the response, and a property estimator for translating sensed responses into estimates of one or more preselected properties of the material; b) defining a dynamic range and property estimate tolerance requirement for the preselected properties of the material; c) selecting an electrode geometry, configuration, substrate material and geometry, and source excitation for the electromagnetic structure d) generating property estimation grids for the preselected material properties and operating point response curves for operating point properties and analyzing the grids and curves to define a measurement strategy; e) optimizing operating point properties and electrode geometry, configuration, substrate material and geometry, and source excitation, said optimizing including generating property estimation grids and operating point response curves at each operating point; f) sensing electromagnetic response at each operating point; g) converting electromagnetic responses into estimates of the preselected properties; and estimating property estimate technology as a function of values of the estimated preselected properties over the defined dynamic range using the property estimation grids and operating point response curves.

22. A sensor comprising: a first and a second interdigital conductors; and a meandering conductor which has elements which parallel the first interdigital conductor.

23. A sensor of claim 22 wherein the elements of the meandering conductor are equally spaced on either side of each of the digits of the first interdigital conductor.

24. A sensor of claim 23 wherein the ratio of the distance between the digits of the first interdigital conductor and the elements of the meandering conductor and the distance between the digits of the first interdigital conductor and the digits of the second interdigital conductor is approximately 1.6

25. A sensor of claim 23 further comprising a switching device for selecting one of the first interdigital conductor, the second interdigital conductor and the meandering conductor as a driven electrode, selecting another of the first interdigital conductor, the second interdigital conductor and the meandering conductor as a sensing electrode and selecting the last as a guard electrode.

26. A method for translating the sensed electromagnetic responses of a dielectrometer into estimates of the preselected properties of the material, comprising the steps of: accessing a property estimation grid using a property estimator; generating a property estimation grid for the preselected properties based on a set of properties with the property estimator; and incrementing a model which provides for each implementation prediction of a response for the preselected properties based on a set of properties characterizing the electrode and substrate structures and the material.

27. A property estimator comprising: an input device for receiving a sensed electromagnetic response by at least one sensor; a property estimation grid for translating each sensed electromagnetic response into a proximity estimate; and a property analyzer for generating an improved property estimation grid by successively implementing a model which provides for each implementation a prediction of a response for a particular proximity based on a set of properties characterizing the dielectrometer electrode and substrate structures and the material under test.

28. A method of determining properties of material under test comprising the steps of: providing a pair of substantially identical sensors; immersing the material under test in a first liquid dielectric; pressing one of the sensors against the material under test; immersing the other sensor in the first liquid dielectric and spaced from the material under test; measuring the capacitance of each of the two sensors; adding a second miscible liquid with a higher dielectric permittivity to the first liquid; and comparing the capacitance of the sensors as the second liquid is added.

29. A method of determining properties of material under test of claim 28 wherein when the two capacitances of the two sensors become identical, the liquid mixture dielectric permittivity equals the dielectric permittivity of the material under test.

Description:

[0001] This application is a divisional of U.S. application Ser. No. 09/310,507, filed May 12, 1999, which claims the benefit of Provisional Application No. 60/085,201, filed May 12, 1998, the entire teachings of which are incorporated herein by reference.

[0002] The technical field of this invention is dielectrometry and, in particular, the electromagnetic interrogation of materials of interest to deduce their physical, chemical, geometric, or kinematic properties. The disclosed invention applies to semiconducting, both lossy and lossless dielectric media, very thin metalizations, and shape/proximity measurements for conducting and dielectric objects and surfaces.

[0003] Dielectric sensors are commonly used for material property characterization and defect detection in a material under test (MUT). The sensors respond to the absolute properties of the MUT, such as the electrical permittivity, electrical conductivity, thickness, and proximity, and changes in those properties. Factors that affect the dielectric properties include composition, chemistry and the state of cure, density, porosity, and contamination with other substances such as moisture. The property variations may be a normal part of the manufacturing process or a result of the presence of defects or damage. These defects can be created during the manufacturing process, such as improper curing or incorrect layer thickness for stratified media, or when the material is placed into service by use- and/or age-related degradation processes, such as fatigue. In manufacturing, the continuing drive toward defect-free products, yield improvement and operation near the capability limits of the production system require sensing technologies for monitoring as many critical process variables as possible. In operations, service maintenance, and repair and replacement activities, the continuing push toward a retirement-for-cause philosophy from the retire-for-time approach requires reliable measurements on all fatigue-critical components in the system, even at difficult-to-access locations.

[0004] Dielectric measurements can be performed with a wide variety of devices. The simplest devices involve parallel plate capacitors where the MUT is placed between a pair of electrodes. Often guard electrodes are used to minimize the effects of fringing electric fields at the electrode edges so that MUT is exposed to an essentially uniform electric field. The electrical terminal admittance or impedance of the device is then related to the material properties through geometric factors associated with the sensor geometry.

[0005] In many applications both sides of the MUT are not easily accessible and single-sided sensor configurations are required. A common implementation of a single-sided sensor is the interdigitated electrode structure used for chemical and moisture sensing applications. U.S. Pat. No. 4,814,690 further discloses the use of multiple sets of interdigitated electrodes as part of the imposed frequency-wavenumber dielectrometry approach for spatial profiling of stratified dielectric media. These devices have been effective in determining the dielectric properties of fluids. However, the determination of solid dielectric properties is more complicated due to the presence of microcavities or unintentional air gaps between the solid dielectric and the sensor.

[0006] While one can attempt to compensate for the air gap, the thickness is usually unknown and variable across the surface of the sensor. Therefore, effective compensation may be difficult to achieve even with multiple sensors placed onto a single substrate. One of the difficulties is due to the fact that a sensor having a number of sensor elements, each with different electrode spacings, those sensor elements are not co-located and therefore are not located at exactly the same places relative to the MUT.

[0007] Generally dielectrometry measurements require solving of an inverse problem relating the sensor response to the physical variables of interest. Such inverse parameter estimation problems generally require numerical iterations of the forward problem, which can be very time consuming often preventing material identification in real-time. Real-time parameter estimations often need to be provided for such applications as manufacturing quality control. In some cases, simple calibration procedures can be applied, but these suffer from requiring and assuming independent knowledge about the properties. More advanced model-based techniques utilize multivariable parameter estimation algorithms to estimate the properties of interest, but these are generally slow, precluding real-time measurement capabilities, and may not converge on the desired solution.

[0008] The present invention comprises of a method for generating property estimates of one or more preselected properties or dimensions of a material. Specific embodiments of the methods are disclosed for generating, calibrating, measuring properties with, and selecting among two-dimensional response databases, called measurement grids, for both single wavelength dielectrometry applications and multiple wavelength dielectrometry applications.

[0009] One step in a preferred method requires defining or estimating the range and property estimate tolerance requirements for the preselected properties or dimensions of the material under test. The next step is the selecting at least one of each of an electrode geometry, configuration, excitation source, and measurement instrument operating point. A continuum model, either analytical or numerical or an experimental approach using calibration test pieces of known properties and dimensions or both are then used to generate measurement grids as well as operating point response curves for preselected operating point parameters.

[0010] The measurement grids and operating point response curves are subsequently analyzed to define a measurement strategy. Operating point parameters and an electrode geometry, configuration, and excitation source are then determined to meet the dynamic range and tolerance requirements. To accomplish this, property estimation grids and operating point response curves are generated and analyzed for various operating points. The sensitivity and selectivity is calculated for grids representing varying electrode designs and operating conditions. Then the best of the lot of prechosen design parameters and operating conditions are selected. If inadequate to requirements, this evaluation process can be reiterated with improved selections based on what was learned in prior rounds.

[0011] A property estimator implements a model for generating a property estimation grid, which translates sensed responses into preselected material property or dimension estimates. Accordingly, the present invention includes a method for generating a property estimation grid for use with a dielectrometer for estimating preselected properties or dimensions of a material under test. The first step in generating a grid is defining physical and geometrical properties of the MUT and the electrode geometry, configuration, and source excitation for the dielectrometer are defined.

[0012] The material properties, the operating point parameters, and the dielectrometer electrode geometry, configuration, and source excitation are input into a model to compute an input/output terminal relation value. In a preferred embodiment, the input/output terminal relation is a value of transadmittance magnitude and phase. The terminal relation value is then recorded and the process is repeated after incrementing the preselected properties of the material under test. After a number of iterations, the terminal relation values are saved in a database of material responses and plotted to form a property estimation grid.

[0013] A preferred embodiment of the method according to the invention includes the incorporation of geometric properties into the grid databases for the representation of multi-layered media and the use of analytic properties of the measurement grid to map from measurement space to property space, such as singular value decomposition, condition numbers (and visualizations of these), to improve selection amongst alternative sensor designs and operating conditions. In addition, the preferred embodiment uses methods for measuring with and calibrating dielectrometers, using single and multiple grids for multiple wavenumber dielectrometry Exploiting the characterization and understanding of other properties of such mappings to aid in choosing and selecting among measurement grid alternatives is also feasible.

[0014] A preferred method of the invention includes enhancing sensitivity and selectivity by using fluids or solids of known properties and dimensions to intentionally move the sensor response within the grid or to alter the grid itself. Movement within grids can also be achieved by varying other parameters such as temperature that also affect permittivity. The method supports both measurement and calibration. One method uses “shims” of known dielectric constant, conductivity, and thickness between the sensor and the material under test.

[0015] The need is also recognized for a sensor device configuration that reduce the sensor sensitivity to undesired inhomogenities across the face of the sensor. In a preferred embodiment, a sensor has multiple electric field penetration depths but each with the same air-gap, fluid gap, or shim lift-off height, thereby greatly reducing the number of unknowns in parameter estimation algorithms.

[0016] The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular description of preferred embodiments of the invention, as illustrated in the accompanying drawings in which like reference characters refer to the same parts throughout the different views. The drawings are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention.

[0017]

[0018]

[0019]

[0020]

[0021]

[0022] _{0}

[0023]

[0024]

[0025]

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

[0035]

[0036]

[0037]

[0038]

[0039]

[0040] Methods, techniques, and devices are disclosed for measurements of electrical, physical, and dimensional properties of a material under test as well as geometric and kinematic properties of the measurement arrangement. These measurements are made with both contact and non-contact of the material under test by a sensor. For contact measurements, the sensor may be embedded in material under test (MUT).

[0041] A measurement apparatus

[0042] The driven electrode

[0043] The measurement apparatus _{S }_{S}

[0044] Through proper design and selection of the electromagnetic elements

[0045] This increased capability through the selection of the elements, points, and properties as explained below results in measurement of properties and or dimensions of interest for the material under test which are not measurable with traditional electrode structures.

[0046] The grid measurement methods provide a real-time capability for solving the inverse problem relating the material properties to the sensor response. These grid measurement methods use a database of sensor responses to map the measured signals into the desired properties for the material. The database is derived, prior to the data acquisition, using a “forward model” of the sensor response using either a continuum model or a finite element or other numerical method for the sensor and the specific problem of interest and/or a preselected set of calibration experiments. The measurement databases can be visualized as grids, as seen in

[0047] These measurement methods are applied to capacitive sensing dielectrometry, where the dielectric properties of a material can be described by two parameters, the permittivity and conductivity. The permittivity is a constitutive parameter that relates the displacement current density in the material to the applied electric field, whereas the conductivity applies to the conduction current density. The dielectric properties of materials vary significantly and can provide a means for characterization of the materials and their geometric properties such as size or layer thickness.

[0048] It is convenient to represent the complex permittivity of a material as ∈*=∈′−j∈″, where ∈′ is the real part and ∈″ is the imaginary part of the complex permittivity. The real part is the dielectric constant, or permittivity, of the material (∈′=∈); whereas, the imaginary part (∈″=σ/ω where σ is the conductivity and ω is the angular frequency of the electric field) describes the power dissipation or loss of the material. The dielectric spectrum of a material is a representation of its complex permittivity, expressed as a function of frequency. The dielectric spectrum provides a signature of a material in a particular state.

[0049] Classical dielectrometry extracts information about the state of a material construct from its dielectric spectrum. The application of a sinusoidally varying potential of complex magnitude v and angular frequency ω=2πf results in the flow of a terminal current with complex amplitude I, whose magnitude and phase are dependent on the complex permittivity of the material.

[0050] Referring to

[0051] The depth of sensitivity of the sensor is determined by the electrode spacing. The electric scalar potential in the materials above and below the sensor obeys Laplace's equation. In Cartesian coordinates with linear lossy dielectrics the potential can be written as an infinite series of sinusoidal Fourier modes of fundamental spatial wavelength

[0052] where k_{n}_{n }

[0053] _{1 }_{2 }

[0054] A three wavelength sensor

[0055] Measurement of the gain and phase, the real and imaginary parts of the transadmittance between the driven and the sensing electrodes, or the transconductance and transcapacitance, provides two parameters which can be related to conductivity and dielectric permittivity of a material. Liquid and gaseous dielectrics are most suitable for this type of measurement because the fluid conforms to the sensor surface. This eliminates the uncertainty in geometry that can exist for measurements with solid dielectrics due to surface roughness and deformation of the solid material and sensor electrodes. Alternatively, the comb-serpentine-comb structure disclosed herein allows for improved accuracy in the determination of the properties of solid dielectrics by allowing different depths of penetration to be achieved within the same sensor footprint. An alternative method to achieve multiple sensing wavelengths is disclosed in U.S. patent application Ser. No. 09/003,390, filed on Jan. 6, 1998 titled “Magnetometer and Dielectrometer Detection of Subsurface Objects,” the entire contents of which are incorporated herein by reference.

[0056]

[0057] While property grids for the real and imaginary parts of the complex permittivity for semi-infinite materials have been generated, measurement grids have not been incorporated into the design or operation of dielectrometry measurement systems. The property grids for semi-infinite material have been used to illustrate the mapping between the model response and the measured response, but the inverse problem of estimating the properties of the material under test use iterative procedures which minimize the error (for example, the least square error) between the measure response and the response for estimate parameter. In contrast, the method of the invention uses measurement grids to estimate properties of the material under test. Different forms of measurement grids are required to solve specific problems. The measurement grids express two properties relative to each other. For example,

[0058] The incorporation of geometric properties into the grids for the representation of multi-layered media, methods for measuring with and calibrating dielectrometers, the use of single and multiple grids for multiple wavenumber dielectrometry, and the use of singular value decomposition and condition numbers to improve selection amongst alternative grids and grid representation candidates, as described below. This invention also includes new methods for enhancing sensitivity and selectivity by using fluids or solids of known properties and dimensions to intentionally move the sensor response within the grid. Movement within grids, i.e., shifting of operating points, can also be achieved by varying other parameters such as temperature that also affect permittivity.

[0059] For general dielectrometry measurements of homogeneous solids, there are usually at least three unknowns that need to be determined; the material dielectric constant and conductivity and an unknown air-gap thickness between sample and sensor. This air gap can be intentional in the case of noncontact measurements or it can be the unintentional result of voids between the sample and the sensor due to sample roughness and deformation. Although these voids are usually quite small, on the order of a few micron spacing for smooth samples and greater for rough samples, the voids are located in the region of strong electric field and consequently have a significant effect on the sensor response. With a device containing two wavelength sensors, each wavelength provides two independent measurements of gain and phase, so that in most cases the four measured values give more than enough information necessary to evaluate the three unknowns. With more wavelengths, the additional redundant information can be used to further improve parameter estimations via mathematical fits, such as using a least squares fit between theory and measurements, or using single or multiple measurement grids at each frequency and averaging the results. For nonhomogeneous dielectrics, other physical parameters of interest may be the layer thickness, porosity, moisture content, or anisotropic property variations.

[0060] In some situations it is possible to simplify the measurements of the solid MUT's dielectric properties by either eliminating one of the unknown parameters or operating the sensor in a regime that is independent of one of the material properties. In the simplest case, the air gap may be negligible for a contact measurement if, for example, the MUT is fluid or soft enough to conform the sensor geometry. Then a single wavelength measurement of the transcapacitance and transconductance yields the effective permittivity and conductivity of the MUT, as in

[0061]

[0062] In another implementation, a single wavelength sensor can be used in combination with information from other sources to provide estimates of the solid MUT's dielectric properties. In this generalization of the measurement grid approach, variations in a dielectric or geometric property of the MUT are mapped into two measurable parameters, only one of which is from a conventional dielectrometry measurement of capacitance or conductance. The other measurable parameter, such as a layer thickness, temperature, or pressure, is an input from some other sensing device. As an example, consider a nonhomogeneous MUT consisting of four layers of insulating polymers, with the only unknown parameters being the permittivity and the thickness of the second layer away from the sensor. This represents, for example, a curing polymer layer sandwiched between two protective films and placed onto an insulating substrate. A representative measurement permittivity-thickness grid, with simulated data, is shown in

[0063] One example application that illustrates the grid measurement approach is the curing of epoxies as shown in

[0064]

[0065] In cases where the materials are insulating, the magnitude values for two separate sensors are used in a two-dimensional grid to estimate permittivity, thickness, lift-off, or other geometric parameters with two unknowns. Examples of such magnitude-magnitude grids are shown in

[0066] In a related operation,

[0067]

[0068] In the case of a single wavelength measurement, only two measurement values (gain and phase) are determined, which are insufficient to uniquely determine three or more unknowns. A well-calibrated shim of known permittivity, conductivity and thickness can be inserted between sensor and sample; placed on the other side of the sample; or multiple precisely positioned shims including air gaps can be used. The gain/phase measurement can be repeated for any combination of these shim variations. The shim can be either solid or a fluid (liquid or gas) located between the solid MUT and the sensor. One embodiment of a liquid shim is described below with respect to

[0069] A variation of the calibrated shim measurement method is to use liquid dielectrics of well known permittivity, conductivity and thickness with two identical sensors. One sensor has an unknown dielectric while the second sensor uses well-calibrated known dielectrics. Well characterized miscible liquid dielectrics of precisely controlled volume are added to the calibrated sensor until the gain/phase results match those from the unknown dielectric measurement, so that geometric and physical properties of the unknown dielectric can be determined from comparison to measurements of known dielectrics.

[0070]

[0071] Generalized Material Under Test Property Estimation Framework

[0072] The method and techniques of the disclosed invention comprise a general property estimation framework. This approach is related to the one developed by Goldfine et al. in U.S. Pat. No. 5,629,621, “Apparatus and Methods for Obtaining Increased Sensitivity, Selectivity and Dynamic Range in Property Measurements using Magnetometers,” the entire contents of which are incorporated herein by reference. The application to dielectometry as opposed to magneometry is complicated by differences in both the nature of the sensing techniques and differences in the responses of materials. In magnetometry, the decay of the sensing field into the material is governed by the (vector) magnetic diffusion equation, which has partial derivatives with respect to both time and space. In contrast, dielectometry is governed by the (scalar) Laplace's equation, which has only partial derivatives with respect to space. Thus, achieving multiple spatial decay rates with magnetometry requires changing only the temporal frequency of excitation. Achieving this same capability for dielectometry requires specific designs in the electrode structures, as described both above and further below.

[0073] Similarly, there are differences in responses of materials; not all materials have strong magnetic or conducting responses required to interact with magnetometers, but all materials have some dielectric response. Therefore, air-gap lift-off layers, which are mere separation layers in magnetometry have more direct influence in dielectometry, which complicates the application to measurements of solids, due to the inavoidable, and typically non-uniform, sensor lift-off. Many other, but less troublesome differences exist, which are well known to those versed in both arts, magnetometry and dielectometry, which preclude the mechanical transferrance of methods from one domain to the other. The unavoidable lift-off layer is one of if not the most troublesome in applying this methodology to measurements of properties of solids, so several means of overcoming this obstacle are disclosed herein.

[0074] A typical measurement procedure flow would include the following steps as shown in the procedure flow diagram in

[0075] Step 1 (

[0076] Step 2 (

[0077] Step 3 (

[0078] Step 4 (

[0079] Step 5 (

[0080] Step 6 (

[0081] Step 7 (

[0082] For any application, calibration experiments can be used to tune the model parameters and improve MUT property estimation accuracy. Such calibration, although not always required, should always be used when available.

[0083] Property Estimation Grid Database and Operating Point Response Curve Generation

[0084] Each parameter estimation application will require a set of property estimation grids, i.e., databases/measurement grids

[0085] 1) Develop a measurement strategy and select the measurement operating points by evaluating the MUT property estimation grids and operating points response curves, at a variety of different operating points over the required dynamic range for the MUT properties of interest (Step 3 (

[0086] 2) Graphical estimation of the MUT properties of interest (Step 6 (

[0087] 3) Determination of the estimate tolerances, as a function of the estimated values for the MUT properties of interest (Step 7 (

[0088] 4) Provide comparison and evaluation of measurement strategy options. All too often measurements are performed with inadequate understanding of whether the measurement strategy (a.k.a., protocol, methodolgy) is adequate to the task, what properties the strategy has, whether such properties of the strategy could be improved or are already optimal. In situations where measurements of certain kinds have been needed for years, or decades, ad-hoc standards and rules of thumb have often accumulated. Often they are sanctified by various standards bodies. In the absence of means of evaluating and comparing measurement strategies, this is the best that can be expected. But in measurement domains where means of evaluating and comparing measurement strategies have been developed, ad-hocracy has been supplanted with objective evaluation. This has occurred long since for many simple measurements, e.g., circuit measurement of voltage or impedance. The methods disclosed herein, and their obvious extensions, now enable the choice of dielectometry measurement strategies to be made on objective scientific bases, instead of rules of thumb and ad-hoc techniques. It is the analysis of the properties of measurement grids and operating response curves, such as condition numbers and singular value decompositions, when applied to comparison and selection of measurement strategies and the details that comprise such strategies, that enables objective comparisons and rational choices.

[0089]

[0090] Fringing Field Multi-Penetration Depth Dielectrometry Sensor with Common Lift-Off Height

[0091] The property estimators with multi-penetration depth dielectrometry sensors can be most accurate if the lift-off height for each electrode configuration is the same. This can be achieved with a three electrode fringing field dielectrometry sensor ^{31 }^{1/2 }

[0092] The configuration of

[0093] The sensor topology of

[0094]

[0095] While this invention has been particularly shown and described with references to preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.