Background
In modern sensing systems, dielectric spectroscopy is considered a promising tool with respect to microwaves and is widely used in the fields of agriculture, food, automotive industry and biomedicine. The method relies primarily on the measurement of the complex permittivity and the variation of the complex permittivity of the medium with frequency. For example, in agriculture, changes in complex dielectric constants of vegetables and fruits can be estimated as changes in water and inorganic content; in the automobile industry, the measurement of dielectric constant is the first choice for oil quality detection; in biomedicine, it is widely used in blood sugar detection and in vivo cancer detection and evaluation. Although dielectric constant measurement systems have shown great potential in modern life, conventional measurement methods often rely on large experimental devices such as vector network analyzers. This limits the application of the method to outdoor scenarios, remote positioning measurements, etc., and is costly.
On the other hand, to truly exploit the potential of microwave permittivity sensors in real-world applications, miniaturized designs of sensors and measurement systems are essential. In addition, miniaturized sensors may facilitate new applications, such as two-dimensional sensor array designs for real-time dielectric constant measurement and visualization in the microwave frequency band. In order to meet the application requirements of such imaging systems, great emphasis must be placed on reducing the size of the sensor and its signal conditioning circuit, and at the same time, the characteristics of fast data readout and high resolution are required.
Furthermore, in nature, the dielectric constant is often a real number, i.e.r *=r′-jr", wherein the real partr' represents the storage of energy, while the imaginary partr"then characterizes the loss of material. Most of the existing measurement modes usually focus on the measurement of the real part, and the imaginary part information of the dielectric constant of the medium cannot be acquired. For example, the most commonly used resonance cavity perturbation method in industry, the imaginary part value cannot be measured by measuring the frequency change before and after the medium is loaded on the resonance cavity to obtain the dielectric constant.
Therefore, the above two aspects become important factors that restrict further applications of the dielectric sensor. How to simply and efficiently extract and convert the output signal of the sensor, study and develop a miniaturized sensing system for measuring the dielectric constant, and how to simultaneously extract the real part and the imaginary part of the dielectric constant has become a research hotspot in academia and industry.
Disclosure of Invention
In view of the above, the present invention provides a small portable complex permittivity sensor system to overcome the above-mentioned difficulties.
The sensor system mainly comprises a near-field patch sensor, a balanced Wheatstone bridge and a down-conversion circuit. The patch sensor is used for placing a medium to be measured, and the Wheatstone bridge is used for measuring the impedance of the patch sensor.
The wheatstone bridge and down-conversion circuit are well known in the art.
The near-field patch sensor comprises a silicon substrate, a metal stratum, an oxide layer, a metal patch and a passivation layer which are sequentially arranged from bottom to top; a groove is formed in the upper surface of the oxide layer, a metal patch is placed in the groove, and a passivation layer opening is reserved on the metal patch;
during detection, a medium to be detected (MUT) is placed on the upper surface of the sensor and completely covers the metal patch, and a passivation layer opening, namely an air layer, is left between the medium to be detected and the metal patch at the moment and is equivalent to a fixed capacitor;
the working principle is as follows: the near field patch sensor is equivalent to a fixed capacitor, and when a medium to be Measured (MUT) is placed on the near field patch sensor, the total admittance of the near field patch sensor changes, and the conductance and the capacitance of the admittance are related to the dielectric constant.
The near-field patch sensor replaces a capacitor on one branch of a Wheatstone bridge, the bridge is driven by radio frequency signals, real part and imaginary part information of the impedance of the patch sensor can be acquired through the amplitude and the phase of a differential output signal of the bridge, finally, an intermediate frequency signal is output through a down-conversion circuit, and the size of a dielectric constant is extracted through ADC and digital signal processing.
The invention has the beneficial effects that:
the invention avoids the use of a vector network analyzer, thereby greatly reducing the test cost; both the real and imaginary parts of the dielectric constant can be measured.
Detailed Description
The following are specific embodiments of the present invention and are further described with reference to the drawings, but the present invention is not limited to these embodiments.
Aiming at the defects in the prior art, the applicant finds that the dependence on large experimental equipment such as a vector network analyzer can be avoided by avoiding the measurement of the frequency response of the medium sensor. The applicant therefore focused on the detection of the admittance of the sensor. Fig. 1 is a schematic 3D diagram of a patch sensor, which can be used as a good sensing element because a single metal patch has deeper electric field lines and larger near field strength. The near field patch sensor is designed by adopting a typical CMOS process, and the cross section of the process is shown in FIG. 2. The lowermost layer is a silicon substrate (5), a metal stratum (4) is connected to the upper surface of the silicon substrate, an oxide layer (3) is connected to the upper surface of the metal stratum, a passivation layer (2) is connected to the upper surface of the oxide layer, a groove is dug in the upper surface of the oxide layer, a metal patch (7) is embedded in the groove, and the passivation layer on the metal patch is removed to form a passivation layer opening (6). And a medium to be tested (MUT) (1) is placed on the passivation layer and completely covers the metal patch. Wherein the metal layer is made of copper, the oxide layer is made of silicon dioxide, and the silicon substrate is made of silicon.
The equivalent circuit of the near-field patch sensor is shown in fig. 2, and the patch sensor is equivalent to a fixed capacitor C0,C0One end of which is grounded and the other end is marked as point P, and the admittance Y of the medium to be measuredMUTEquivalent to a capacitor CMUTAnd a conductance GMUTAfter parallel connection, one end of the parallel connection is grounded, and the other end is connected with a point P.
When no medium to be measured is placed, the admittance of the node P is the fixed capacitance C formed by the metal stratum and the metal patch layer0When a MUT is put in, the admittance of the point P will change with the change of the loaded medium, and since the real part and imaginary part of the complex permittivity represent the energy storage and loss of the medium, respectively, as shown in fig. 2, the dielectric layer can be equivalent to a parallel connection of a capacitor and a resistor, the admittance of which is YMUT≈GMUT(″)+jωCMUT(') then the total admittance of the node P is YP=jωC0+YMUT。
To quantify the relationship between the admittance and the dielectric constant of the P-spot, a 100X 100 μm process based on 40nm was used2And performing EM electromagnetic simulation on the large and small patch sensors. Fig. 3(1) and 3(2) show the variation of capacitance and conductance with real and imaginary parts of dielectric constant at 1 GHz. As can be seen from the figure, the 'sum capacitance', and the conductance are approximately in a linear relationship, i.e. have
YP(′,″,ω)≈αi·ω·″+jω·(C0+αr′) (1)
Wherein alpha isiAnd alpharAre real parameters. Table I summarizes the values of the parameters after the simulation and fitting by the least square method
Parameter(s)
|
Numerical value
|
C0 |
82.56fF
|
αr |
2.754fF-1 |
αi |
17.5μS-1·GHz-1 |
The capacitance and conductance due to point P are also affected by "and", respectively. Applicants have therefore fitted a rational function model from the EM simulation for calibration. The model formula is shown below
Where a is a scaling parameter equal to the patch sensor size, αnpAnd betamqAre the actual model parameters of the NxP and MxQ order matrices. In order to conform the model parameters to the admittance variation with 'and' in EM simulations, applicants found that the error rate calculated for dielectric constants in the frequency range of 0.1-10GHz is within 1% when N-P-Q-M-1.
The invention converts the measurement of the complex permittivity into an admittance measurement by means of a patch sensor, in which the admittance of the patch sensor is measured using a wheatstone bridge, since this bridge provides a very good quantitative admittance measurement with respect to a reference admittance, in which patch sensor C0I.e. the reference admittance. The basic schematic diagram of the bridge is shown in fig. 4, and the bridge mainly comprises a drive amplifier and four groups of branches formed by parallel capacitance conductance, wherein the input signal of the bridge is viOutput v after connecting with the drive amplifierin,vinDivided into two branches connected to one end of the first branch and one end of the third branch, the other end of the first branch is connected to one end of the second branch, the connection point is marked as node A, and the voltage is vb,O+The other end of the second branch is connected to ground, the other end of the third branch is connected to one end of the fourth branch, the connection point is marked as node B, and the voltage is vb,O-. The other end of the fourth branch is connected to ground. Wherein the admittance of the three branches including the first branch, the second branch and the third branch is a reference admittance Y1,Y1To fix the conductance G1And a fixed capacitor C1Are connected in parallel. Initiation of the fourth branchAdmittance being Y1On the basis, the patch sensor loaded with the medium is connected into a fourth branch circuit, and the admittance of the fourth branch circuit is changed into the reference admittance Y1And loading admittance YLAnd (4) summing. The bridge is excited by a signal with frequency omega through a driving amplifier, and differential output signals of nodes A and B and an input signal vinThe relationship of (A) is as follows
Wherein Y isL=GL+jBL,Y1=G1+jB1G is the conductance of the real part of the admittance, B is the susceptance of the imaginary part of the admittance, assuming YLNot equal to 0, taking the reciprocal of formula (3), having
By GL,BL,G1And B1To replace YLAnd Y1Then, there are:
wherein G isLω=GL/|YL|2,BLω=BL/|YL|2Respectively, weighted load conductance and susceptance values. As can be seen from equations (5) and (6), regardless of the load admittance YLOff-reference admittance Y1The real and imaginary parts of the bridge differential output are both linear combinations of the weighted load conductance and susceptance values. Based on the linear relation of the input and the output of the electric bridge, the load admittance values (the real part and the imaginary part) can be conveniently obtained by avoiding the fitting of a high-order polynomial, and further the dielectric coefficient can be obtainedConstant information. However, the circuit structure shown in fig. 4 has a large common-mode signal component at the output terminal, which imposes a strict requirement on the common-mode rejection ratio (CMRR) of the subsequent readout chain circuit. In view of this, we adopt a double-balanced configuration (double-balanced) differential bridge design scheme to solve the above problems. The double balance configuration differential bridge is composed of two Wheatstone bridges, wherein one Wheatstone bridge is not loaded with admittance as a reference bridge, the other Wheatstone bridge is loaded with admittance as a measuring bridge, and the output is the differential output of the two bridges.
Meanwhile, in order to facilitate subsequent digital signal processing, the output of the double-balanced differential bridge is down-converted to an intermediate frequency and then connected with an analog-to-digital converter (ADC) for processing. As shown in FIG. 5, a double-balanced mixer (double-balanced mixer) converts the RF output signal of the bridge into an intermediate frequency signal v by connecting the differential output of the bridge and the local oscillator signalIF+,vIF-. Because the intermediate frequency signal frequency is lower, the low-cost ADC can be used for conveniently processing, and the implementation cost of the system is greatly reduced. In addition, the local oscillator LO signal is a square wave signal, and the mixer works in a linear region, so that the intermediate frequency differential output delta vIFCompletely retains the delta v of the differential output of the bridgeb,oCan therefore be represented by Δ vIFAnd obtaining load admittance information so as to obtain a complex dielectric constant value.
Example 1
The present invention uses the balanced differential bridge shown in fig. 6 to measure the dielectric constant information of the material. The patch sensor passes through a switchcConnected in parallel to a third branch of the measuring bridge, which circuit is composed of a pair of differential bridges, the reference admittance of which branch is Cb. In order to adapt to wide capacitive load change and carry out experimental research on bridge behaviors in various unbalanced states, C is adopted in the implementation processbIn particular by 8 switchable capacitors C3Are connected in parallel to form a series of capacitors C3Free turn-on and turn-off is achieved by connecting a 10- μm/40-nm CMOS switch in series, thus providing 8 reference admittance values, where C3Is 100 fF. In addition, the bridge middle nodes C, D, C 'and D' are respectively connected with a resistor in series with the middle of the ground so as to ensure that a proper direct current path exists when the switch is biased. As can be seen from equations (5) and (6), in order to measure the capacitance and conductance values of the load admittance, the amplitude and phase of the bridge output need to be obtained. It should be noted that, in order to ensure the consistency of the phase measurement, a reference phase is also measured. This is to ensure that the relative phase change at the bridge output is caused only by changes in the patch load. For measuring relative amplitude and phase change of output quantity before and after loading of dielectric material, the invention adopts a reference capacitor C connected in parallel with a bridge load admittance branch, namely a node CfAnd said patch sensor, the capacitor and the sensor passing through a switchcSwitching gating is implemented. CfThe output phase at the time of gating is the reference phase. When the sensor is gated, the introduction of the medium to be measured causes an additional phase shift and amplitude shift. Thus, it is possible to obtain a continuous measurement track of the two outputs of the containing bridge in the case of two load connections by switching the switches a number of times in a continuous time domain. By down-converting and analog-to-digital converting the acquired signals and synchronizing with the control switch lc, two independent outputs can be isolated in the digital domain. Then, Fast Fourier Transform (FFT) of the two outputs is calculated and divided, so that consistent relative phase difference and amplitude ratio can be obtained, and further, the real part and the imaginary part of the load admittance are obtained.
The above description of the embodiments is only intended to facilitate the understanding of the method of the invention and its core idea. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.