Dielectric probe, method and apparatus including its use

DESCRIPTION

This invention relates to a dielectric probe according to the generic content of the claims.

The interest in the use of electrical measurements to characterize materials has existed for more than fifty years. There are many books that deal with this subject, for example: Dielectric Spectrosco of Pol mers, by P. Hedvig, John Wiley, New York, 1977; Dielectric properties and Molecular Behavior, by N. Hill, W. Vaughan, A. Price and M. Davies, Van Nostrand Reinhold, New York, 1969; Anelastic and Dielectric Effects in Polymeric Solids, by N. McCrum, B. Read and G. Williams, John Wiley, London, 1967; J. of Polymer Science Symposium 50, 345-358, 1978, by A. North; Dielectric and Related Molecular Processes 3, 143175, 1977, by Y. Wada; Dielectric Properties of Polymers, by F. Karasz, Plenum Press, New York, 1972; The Determination of Dinamic Properties of Polymers and Composites, by BE Read and GD Dean, John Wiley, New York, 1978, and Chemorheology of Thermosetting Polymers, by CA May, American Chemical Society Symposium Ser., 227, 1983. The widespread use of electrical measurements, mainly using dielectric techniques, to characterize materials has caused widespread interest in the use of electrical measurements to monitor resin curing procedures. The reason for this interest is that there are very few techniques for convenient and continuous monitoring of the curing procedure over a wide range of resin viscosity, that is, between less than 10² more than 10⁸ Pa.s.

Until a few years ago, research on the use of electrical measurements has not enjoyed sufficient interest. Although, in most cases, the reasons for this lack of expectation are complex, usually related to improper instrumentation, that is, the use of instrumentation, eg sensors, which have not been specifically designed for that purpose, the failure to be able to make measurements over a wide range of frequencies, not understanding the molecular basis for the signal in terms of dipolar and ioonic phenomena, and the lack of a integrated approach to interpret the electrical measurements in a way that correlates with other measurements of chemical characterizations.

The effective and satisfactory use of dynamic electric measurements to monitor the healing cycle requires a great effort that translates into the chemical study of the resin chemistry of each resin or other medium using various characterization techniques, which are then correlated with dynamic electric measurements. made over a wide range of frequencies. Only with this background information is it possible to use the electrical measurements with maximum efficiency to monitor the healing procedure or other reaction and provide the basis for a complete mathematical treatment model to monitor the cure cycle with quality assurance as well as for the control "Ingenious" closed circuit healing cycle.

As indicated above, convenient and accurate measurements of the electrical characteristics of chemical compounds that undergo chemical reactions to monitor and / or correlate the chemical compounds and the fossil characteristics of the products have not been easily used in the past. of reaction with them. Now, it has been observed in particular that the complex permittivity of a polymerizable resin was closely related to the progress of the polymerization reaction, but so far the technique has not developed the apparatus and the techniques necessary to exploit that relationship along the broad frequency range and wide variation in the magnitude of the complex permittivity required for a useful correlation.

For example, in US-A-4,448,943, a fairly elaborate system for controlling the polymerization process variables is described, but uses a "slotted die within which a parallel plate capacitance cell is incorporated" for the measuring device . Although the parallel plate capacitance cell gave an approximate measure of capacitance, it has certain disadvantages, compared with the planar interdigital device developed by the authors of the present invention, including the inability to maintain a controlled spatial relationship between electrodes at small desirable distances. And what is more important, the method of the present invention employs the capacitance and conductance of the medium to first calculate the most useful complex permittivity. Golba and Hansen, in US-A-4-448943, do not use the complex permittivity of the compounds that react, as is done here.

The authors of the present invention have already described, in a general way, the measurements over a wide frequency range, see the article "Dynamic Dielectric Characterization of the Cure Process: LARC-160" (Dynamic dielectric characterization of the process of cure: LARC-160), SAMPE Journal, July / August 1983, page 18 and the chapter “Electrical Methods of Characterization of Cure processes in Resins”, which will be published in Developments in Reinforced Plastics -5 by Elsevier publishers. However, many of the dielectric or capacitance probes of the prior art use parallel flat electrodes or, if they are in the same plane as with those of the applicants of the present invention, they were either covered with insulators or the authors were not They recognize the importance of the dimension of geometry and construction materials, which have been mentioned by the applicants of this invention. In this regard, the reader may be interested in US-A-3,515,987 (Zurbrick et al.) Which describes a dielectric probe used illustratively as a humidity detector. "Humid conditions" detectors are also described in US-A-3,873,927 (Overall), US-A-3,841,610 (Hanzawa et al.), US-A3,777,257 (Geisselmann) and US-A-4,057. 823 (Burkhardt et al.).

020 243

US-A-4,296,630 (Jung et al.) Shows a particularly clear illustration of an interdigital pectiniform configuration; However, the detector, which has been constructed to detect the presence or absence of a liquid, that is, to detect the level to which the capacitor scale is covered, is indifferent in terms of dimensions and characteristics of the material, which are necessary for the purposes of the present invention.

In most of the configurations of the prior art, the probe is, in general, a rigid and relatively solid structure. The configuration of the electrodes is relatively or improvisedly constant. The particular materials used, the distance between the electrodes, the width of the electrodes and the accuracy / reproducibility in the electrode model are not critical or considered, since the purpose of these probes is, in general, to determine the presence or absence of a material or its proximity to the probe.

Many workers in the technique have used field effect transistors and / or load flow transistors in probes or sensors to observe the characteristics of various media. See, for example, patents USA-4,437,969 (Covington et al.), 4,247,903 (Grudkowski et al.), USA-4,322,680 (Janata et al.) And various patents by Senturia et al., Eg, 4,317,084, 4,316,140, ​​4,209,796, 4,158,807, 4,236,121 and 4,352,059, which again describe a configuration particularly for moisture measurements.

Patent DE-A-3,403,844 describes another dielectric probe in the form of a box that carries an electrode arrangement on its outside for the medium being monitored and another electrode arrangement inside the box, for the reference medium. Both electrode arrangements comprise parallel interdigital electrically conductive and chemically resistant electrodes. The spacing between the digits of the interdigital electrodes is not greater than 0.028 mm. The wall of the box that carries the electrode arrangements outside and inside respectively functions as an insulating substrate of electricity having a thermal conductivity. This substance may have the form of a thin plate.

US-A-4,399,100 describes another dielectric probe comprising two electrode arrangements in the form of aluminum sheets, located on both sides of the material being tested. Between each electrode arrangement and the material being tested, there is a layer of electricity insulating substrate.

The object of the invention is to provide a probe, an apparatus and a method for monitoring or controlling chemical reactions.

This object is solved by the characteristics of claims 1, 6 and 10 respectively.

Preferred embodiments are described in the dependent claims.

Applicants for this patent have invented a probe specially adapted to monitor and control chemical reactions and, more specifically, polymerizations when using the apparatus and the methods described below, or to control the complex permittivity of the reaction media.

The probe of this invention is a capacitor probe, flat and thin, comprising an arrangement of electrode lines, straight or curved, normally parallel, preferably in an interdigital pectiniform configuration. The electrode lines are conductive, preferably metallic, for example of titanium, tungsten, gold, copper, platinum, palladium, chromium or combinations thereof, are less than 0.25 mm apart and have a width less than 0.5 mm. The two line arrangements form the two terminals of a capacitor. The probe of the present invention, instead of using the usual configuration of a capacitor consisting of plates in parallel planes between which the material being measured is found, what it does is measure the capacitance between a disposition of lines, preferably in The same plane. The material that is measured is put in contact with the arrangement of lines. When a voltage is applied through the two isolated arrangements of electricity, the electric field between the lines rises and passes through the material that is in contact with the probe. The probe uses the edge effects of the electric field to measure the dielectric properties of the material facing the side of the probe, as well as the electric field that passes through the small amount of material that lies directly between the lines. The layered arrangement is stationary on the surface of a substrate comprising a low-loss polymeric film (such as Kapton), glass, ceramic, Al2O3 or other dielectric material with a low loss tangent, that is, whose conductivity remains below about 10^-7 ohm^-1.cm^-1 along the range of use (which is 0 ° to 400 ° C) - that is, over a frequency range of 1Hz to about 10 megaHz. Preferably, the lines have been photographed on the substrate, on whose surface a thin metal film has been adhered or deposited. The total thickness of the probe is less than 2.5 mm preferably.

The geometry and design of the probe offer several advantages. A fundamental advantage is that the material that is monitored or measured can be easily placed in front of the capacitor probe instead of having to place it between the capacitor plates. Therefore, it is ideal for monitoring the properties of thin films, coatings and adhesives, since it can measure its dielectric properties by considering them from the side only. The probe can be used to measure very fine samples - samples whose thickness is comparable to the spacing between the lines, that is, as thin as a millimeter or even less. The capacitor probe is ideal for monitoring the dielectric properties of laminates and adhesives, since it can be inserted between the layers of the laminate or on its surface. Dielectric measurements are not altered by changes in the thickness of the laminate, as is usually the case when the laminate is inserted between the electrodes of the capacitor, since the probe lines are attached to the electrical substrate.

Geomatric reproducibility, dimensions3

020 243 nes and the stability of the electrodes, as well as the use of a reliable, low-loss substrate, allow the determination of the complex permittivity e * = e - ie, an intensive property, not as C and tangent of δ. The configuration of the dielectric probe also allows the measurement of a material during the curing of a resin or a laminate over a wide range of temperatures (0^or C at 400 ° C) while the material undergoes changes in its physical state, that is, from liquid to gel to rubber to glass to solid (SAMPE 28 608 (1983)). The probe is less sensitive to the problem of contraction and dilation of the material than capacitance monitors that use plates or electrodes in parallel planes. The probe is also ideal for measuring the dielectric properties of liquids, thin films, uncured tissue-resin compositions ("prepreg"), rubbers, and gels with high viscosities, since it is enough to wet the probe surface. .

In addition to a well-designed probe, specially adapted for the purpose, there are three other factors that must be considered to effectively use electrical measurements to control a resin curing procedure. The first is instrumentation. Great progress has been made in this regard. During the last years, many fast and low-noise automated bridges have been introduced that go up to six tens of frequencies. Suitable examples are impedance analyzers sold by Hewlett-Packard, conversion spectrometers (time domain) of General Radio (Gen. Rad.), Tetrahedron and Fourier that can be constructed in a conventional manner (see Mopsik Rev. Sci. Instrum 55 (1), January 1984).

The electrical properties vary greatly from one resin to another and may change in factors of 10⁶ you omit during the healing procedure. ⁴⁰ therefore, the choice of instrumentation and its components should be designed and adjusted to the properties of a particular resin, the objectives of use, the operating conditions, the limitations of treatment of the plant, the system ⁴⁵operational, etc.

A second very important factor, which often constitutes more than half of the problem, is the interpretation and understanding of the molecular basis of the electrical signal. The aspec-⁵⁰Critics of this problem are to make rapid measurements over a wide range of frequencies and then correlate the molecular basis of the frequency dependence of the signal with the chemistry and physics of the system. ⁵⁵resin. Frequency-dependent electrical measurements may be correlated with carbon-13 nuclear magnetic resonance, gel infiltration chromatography, infrared spectroscopy, viscosity, ultrasoin and⁶⁰differential analytical calorimetry data.

The third aspect of the problem is to integrate the chemical and flow information contained in the magnitude of the complex electrical signal and the frequency dependence in the treatment model. ⁶⁵used for quality assurance and / or "fast" closed circuit control of the cure cycle. The development of treatment models is an extensive subject that is also currently being developed. See CA May, Chemorheoloqy of Thermosettinq Polymers, American Chemical Society Symposium Ser., 227, 1983.

Measurements of the electrical properties of a material are preferably made with an impedance analyzer. An impedance analyzer measures the opposition presented by the material to an alternating current in terms of the complex ratio of voltage to current.

_Z* = V (ω) ^{τ (ω)} (1) where Z^* It is the complex impedance. The analyzer output is indicative of the magnitude and variation of the voltage over time with respect to the current. Often the properties of the material and its impedance are conveniently represented by an equivalent circuit of a capacitor (Cs) and a resistor (Rs) in series. In this case:

^Rs ⁺ τω (2)

Similarly, the properties of the material can be represented by an equivalent circuit of a resistor (Rp) and a capacitor (Cp) in parallel, in which case:

= ^{Ι (ω)} Y*

Z * V (ω) ¹ (1)

Y * = -jtó + iωC_p = G_p * iωC_p where Y * is the admittance and the inverse of the resistance in parallel is the conductance Gp.

To describe the properties of a capacitor charged with a low conductive material, a third notation can be used. In this notation, the equivalent circuit is simply visualized as a capacitor with capacitance Co without the material, and capacitance C = eC_or With the material.

The magnitude and variation of the current over time with respect to the voltage is represented by the complex value of the permittivity of the material e * = e - ie. Again, according to the theory of the elementary circuit:

- = V © ίω e * Co = e ωϋ „Co + iwe 'Co (4)

All these representations are simply representations of equivalent circuits of two parameters that describe the propagation of an electromagnetic field of a material. Given a pair of parameters, it is possible to restate them in terms of another pair of an equivalent circuit.

The electrical conductance of the resins is usually low and the capacitive properties are moderate to high. Therefore, the electrical properties of the resin would be expressed in terms of the third notation, the complex permittivity e *. In addition, since polymeric materials are poorly conductive, the term "dielectric properties" can be used instead of electrical properties.

020 243

Referring now to the drawings:

Figure 1 is a more or less schematic, enlarged front view of a preferred embodiment of the probe of this invention.

Figure 2 is a conceptual scheme of a preferred system for monitoring the complex permittivity of a reaction mass.

Figure 3 is a simplified block diagram of a preferred system for controlling the time-temperature cycle of a polymerization reaction.

Referring to Figure 1, a preferred probe comprises a pair of electrodes 1, each of which has fingers 2 in a pectiniform configuration and which are interdigitated. The drawing is only intended to illustrate geomometric relationships and not true dimensions. Preferably, they are located between about 100 and 200 fingers or lines 2 of less than about 0.5 mm wide, with a separation distance between them 3 not greater than about 0.25 mm. The number of fingers may vary depending on the frequency used, but in general, it will be at least 13 per cm. Each electrode 1 can have two terminals 4 to connect them to an appropriate bridge or analyzer if it uses a four-terminal measurement. Electrodes 1 may be made of copper or, preferably, for many of the applications that probe I may have, are made of a conductor more resistant to chemical agents, such as gold, platinum, chromium, titanium, tungsten, palladium, or combinations of them. These electrodes are fixed to the substrate 5 in a known manner, such as by vapor deposition and / or with a suitable adhesive, and can be attacked using a continuous layer, also in a known manner. Chemical resistance is, of course, a relative and subjective matter; The choice of materials to build the electrodes depended on the medium to be measured. Most of the applications of the probe of this invention are of one use only and the probe can be thrown after it is a use.

The substrate 5 is important and plays a significant role in this invention; It will be a material that has a low dielectric loss tangent, as explained above, over the range of temperatures and frequencies used, and should have a reliable capacitance curve (eg, stable) presenting little or no hysteresis along the range from 1 hertz to 10 megahertz at operating temperatures. Its dielectric permittivity should be less than 13, with reproducible change along the frequency of use. The dielectric characteristics of the substrate will be known by the user, as is evident from the formulas and calculi. For example, when an additional cross-linking system (or reaction) is to be controlled, a probe is normally needed having a loss tangent of less than 0.001 and a replaceable capacitance of the equivalent parallel plates of maos of 10 pF at over the interval of frequencies and temperatures of use.

In Figure 2, Ztotal represents the global impedance measurement that can be made with a bridge or dielectric analyzer when a probe of Figure 1 is contacted with a material to be measured that has a Z2 impedance. The impedance of the probe substrate is represented by Z1; Due to its high value, it did not mask the effect of Z2 on Ztotal.

With reference to Figure 3, the probe 20 similar to that of Figure 1 lies or is contained in a medium of a polymer mold 21 contained in a heater (22), which is controlled by a temperature controller (23). An iron constantane thermocouple 27 was directly attached to the mold and the temperature is measured by a Keithley 179 TRMS 28 Digital Multimeter. The probe inputs to the dielectric bridge 24 give an indicative output of complex permittivity according to the formulations discussed above; The computer (25) monitors, records and analyzes the frequency dependence of complex permittivity. Based on this information and the predetermined quality assurance criteria of other samples and tests, the computer passes a signal to the digital-to-analog converter 26, which is the input to the temperature controller and thus controls the temperature profile in line -time of the medium 21.

Typically, the operation of the apparatus is as follows: The temperature-time profile of the mold is controlled by a computer program that is based on predetermined correlations thereof with the desired properties of the product. The upper layer can be formed by several layers of Kapton to isolate the probe from the metabolic mold, a probe that can be inserted directly between the layers of the Latin American resin and, in turn, the layers of Kapton between it and the upper part of the mold. Capacitance (Cp) and conductance (Gp) afrequency measurements are made from 5 to 5 x 10⁶ Hz in the resin in contact with the probe, at regular intervals, during the curing cycle and stored on a computer disk. Complex permittivity is calculated for each of these measurements. The current temperature and time for each measurement are also calculated. Graphical representations of the results can be prepared from the stored data and printed at 29 using a Hewlett-Packard 7475A six-pen instrument.

The impedance of a resin, measured as its response to an applied oscillatory electric field, was derived, in general, from two types of molecular processes, namely ionic and dipolar.

Using a bridge or time domain technique, the real and imaginary components of the macroscopic impedance of the material, Z *, are determined as a function of frequency. In the following formulas, these components are expressed as Cp and Gp. Knowledge of the replaceable capacitance with equivalent air C_or of the measurement circuit allows to calculate the complex permittivity e *, an intensive property that is independent of the size and shape of the material.

^and 3C0 (5)

Both real and imaginary components of e * have a dipole component and a component

020 Ionic 243

e = e_zz d + e _zie = ed + ei (6) 5

The dipole component is due to the rotational diffusion of molecular dipole moments.

In its simplest form, the frequency dependence of the polar component is represented ₁₀for a single time of relaxation τ and + e_or

1 + ίωτ ¹ ed = ') (7) where e_or ye _x are the high and low frequency values ​​limiting e *. The speed of the rotational diffusion or reorientation of polar moments is characterized by the relaxation time τ. In most systems, a distribution of relaxation times is observed. The distribution is due to a combination of molecular phenomena, including the presence of masses of a polar species, the asymmetric shape of the polar group and intra- and intermolecular forces. Intermolecular forces determine the degree of correlation that exists between the movement of the dipoles. In this case, e * can be represented as _ 1 (1 + ΐωτ) (8) where β is a measure of the distribution of relaxation times.

The presence of ions often has a dominant effect on the e * properties of the resins and their changing electrical properties during the curing procedure. For example, the presence of mobile ions generates localized layers of charge. Since these load layers are separated by very small distances, of the order of 0.1 mm, the corresponding effective capacitance can be extremely large. It is still common to observe values ​​of e, for example, of the order of 10⁶ in the curing or elaboration of resins, in comparison with values ​​of e less than 10, which are common for more typical liquid and saolids. Similarly, the translational diffusion of ions significantly increases conductance, resulting in values ​​of e that are equally large, particularly when the frequency decreases.

The magnitude of the electrical impedance of the resin usually depends heavily on the presence of ions. Many previous studies on the electrical characterization of the healing procedure despise the role of ions.

This may be due to the huge bibliography that exists on the state of heat that deals with widely used polymers (polyethylene, polyesters, polyamides, etc.) that focuses on the dipole properties of these polymers. However, it has been observed that ioan processes play a fundamental role in determining impedance during the curation procedure, in particular at low frequencies.

The literature that discusses attempts to develop molecular models and equivalent circuits to represent the ionic phenomena is long and complex. In a nutshell, the ionic systems involve irreversible, non-equilibrium processes, which depend on the types, concentration and mobility of the ions, as well as the electrode dimensions and the discharge conditions. Ionic properties are determined in part by the diffusion of charge, the impaired translational mobility of ions in the mass and their interaction with the neighboring charged species. They are also determined by the spatial charge, the formation of ionic regions charged around the electrodes as well as any other interface in the material, and the discharge properties at the interface of the material. No single or unified representation of all these effects has been accepted. Although limited and approximate, a convenient representation of these effects is to consider them together as an effective electrode impedance. In this case:

ei = C_orZ_orsen ηπ ω ^(η+¹⁾ (-) _σ_2

8.85x10^-14 (9) where C_or is the replaceable capacitance in Faradios, Z = Ζ © ω)^-η is the impedance of the electrode induced by the ions and η will vary between 0 and 1. The imaginary part of the ioanic component has the form:

_σ

8.85x10^-14

C_orZ_orwith ηπ ω^{- (η + 1)} (—) (10)² _σ 2

8.85x10^-14 σ is the conductivity (ohm^-1 cm^-1), an intensive variable, unlike the conductance G (ohm^-1) which depends on the size of the cell and the sample. The first term of equation 10 is due to the DC conductance of the ions that move in the middle. The term Zo is due to the effects of electrode impedance. This term is more significant as the frequency of measurement decreases.

It is important to remember that any procedure that is developed and used for the polymer curation process is, in general, uranic for a resin system and / or a particular application. The essential requirement is to understand the molecular basis for the frequency dependence of the dielectric signal. Using this information, then the ianic contribution e *, the dipole contribution e * and the magnitude of e are used to monitor and control the healing cycle.

020 243