Electrical transmission systems and methods for semiconductor, nanoparticle, cell analysis and identification

Abstract

Embodiments described herein provide for high frequency transmission spectroscopy (HFTS) system and methods. The HFTS system and methods provided herein allow for identification or classification of materials of interest in a wide range of industries. The HFTS system and methods are capable of identifying unknown particles, characterizing particle size / charge, or monitoring solutions for specific materials in semiconductor, biomedical, and environmental applications. The HFTS system includes an ENA operable to generate an AC as an excitation signal and sweep a frequency of the AC from an initial frequency to a final frequency. A test cell operable to retain a MOI is positioned between an input antenna and an output antenna. The ENA is configured to calculate parameters at each frequency step to plot magnitude response and / or phase response for identification or classification.

Description

ELECTRICAL TRANSMISSION SYSTEMS AND METHODS FORSEMICONDUCTOR, NANOPARTICLE, CELL ANALYSIS AND IDENTIFICATIONBACKGROUNDField

[0001] Embodiments described herein generally relate to transmission spectroscopy. More specifically, embodiments described herein relate to a high frequency transmission spectroscopy (HFTS) system and HFTS methods.Description of the Related Art

[0002] Spectroscopic methods are common to scientific inquiry in part due to their ability to make precise and meaningful observations across a wide range of domains. While the majority of spectroscopic measurements are performed using electromagnetic signals with frequencies at or near visible light, recent advances in instrumentation have enabled spectroscopy in the radio frequency spectrum. One of the most common of these techniques is impedance spectroscopy, which measures the change in complex impedance within a material over a range of frequencies. Impedance spectroscopy has been applied to a wide range of problems from semiconductors to biomedical analysis. Here, impedance spectroscopy measurements are typically dominated by the impedances associated with electrochemical reactions at the interface of the material under test and the working electrode. This makes it difficult for direct measurement of material properties to be conducted. When a direct measurement of the material properties must be made, the impedance spectroscopy experimental setup must be augmented. Furthermore, reflectance spectroscopy to measure the complex reflection coefficients (Sn parameters) require significant cycle time and calculations to classify materials.

[0003] Accordingly, what is needed in the art is a system and methods for high frequency transmission spectroscopy (HFTS) for in-situ material classification.SUMMARY

[0004] In one embodiment, a high frequency transmission spectroscopy (HFTS) system is provided. The system includes an electric network analyzer (ENA). The ENA has an input port and an output port, where the ENA is operable to generate anAC (Alternating Current) as excitation signal and sweep a frequency of the AC from an initial frequency to a final frequency. An input antenna is electrically connected to the input port and the output antenna is electrically connected to the output port, and a test cell positioned between the input antenna and the output antenna. The test cell is operable to retain a material of interest (MOI). Conducting surfaces of the input antenna and the output antenna are electrically isolated from the MOI. In operation of the HFTS system: the ENA is configured to generate an excitation signal having an initial voltage that is transmitted by the input antenna to the MOI as an electromagnetic (EM) wave such that the EM wave propagates through the MOI until is received by the output antenna as a transmitted signal having an output voltage, and the ENA is configured to calculate a parameter of the output voltage to input voltage from the initial frequency to the final frequency to plot at least one of a magnitude response or a phase response of the EM wave at each frequency step.

[0005] In one embodiment, an HFTS system is provided. The system includes an ENA having an input port and an output port, where the ENA is operable to generate an AC as excitation signal and sweep a frequency of the AC from an initial frequency to a final frequency. The input antenna is electrically connected to the input port and an output antenna is electrically connected to the output port. A circuit board having a test socket positioned between the input antenna and the output antenna, the test socket is operable to retain a respective integrated circuit (IC), the input antenna and the output antenna are connected to the circuit board with the test socket therebetween. A relay matrix of a plurality of relays are connected to the test socket with traces, and a digital controller is connected to the relay matrix to provide a DIO signal to the relay matrix to activate or deactivate relays to correspond to each pin permutation of the respective IC.

[0006] In yet another embodiment, a high frequency transmission spectroscopy (HFTS) method is provided. The method includes retaining a material of interest (MOI) with a test cell, where the test cell is positioned between an input antenna and an output antenna electrically isolated from the MOI. The input antenna is connected to an input port and the output antenna is connected to an output port of an electric network analyzer (ENA). The method further includes generating an AC (Alternating Current) with the ENA as excitation signal and sweeping a frequency of the AC froman initial frequency to a final frequency. An initial voltage of the excitation signal is transmitted by the input antenna to the MOI as an electromagnetic (EM) wave that propagates through the MOI until is received by the output antenna as a transmitted signal having an output voltage, and the ENA is configured to calculate a parameter of the output voltage to input voltage wave at each frequency step from the initial frequency to the final frequency to plot at least one of a magnitude response or a phase response of the EM wave.BRIEF DESCRIPTION OF THE DRAWINGS

[0007] So that the manner in which the above recited features of the present disclosure can be understood in detail, a more particular description of the disclosure, briefly summarized above, may be had by reference to embodiments, some of which are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only exemplary embodiments of the disclosure and are therefore not to be considered limiting of its scope, as the disclosure may admit to other equally effective embodiments.

[0008] Figure 1A is a schematic view of a high-frequency transmission spectroscopy (HFTS) system according to embodiments.

[0009] Figure 1 B is a system architecture of an HFTS system according to embodiments.

[0010] Figure 2 is a schematic view of an integrated circuit (IC) configuration of an HFTS system according to embodiments.

[0011] Figure 3 is a flow diagram of a method of IC testing according to embodiments.

[0012] Figure 4A is a confusion matrix of the classification performance for lot IDs according to embodiments.

[0013] Figure 4B is a chart of the classification accuracy scores at each relay permutation for pin permutations of a lot ID according to embodiments.

[0014] Figure 5 is a schematic view of an analyte configuration of an HFTS system according to embodiments.

[0015] Figure 6 is a flow diagram of a method of analyte testing according to embodiments.

[0016] Figure 7 is a magnitude response plot of NaCI solutions generated by an ENA.

[0017] Figure 8A is a magnitude response plot of solutions generated by an ENA.

[0018] Figure 8B is a phase response plot of solutions generated by an ENA.

[0019] Figure 9A is a magnitude response plot of nanoparticle solutions.

[0020] Figure 9B is a magnitude response plot of microparticle solutions.

[0021] Figures 10A and 10B are magnitude response plots of microparticle solutions.

[0022] Figure 11 is a chart of classification accuracy of identification of plastic microparticles.

[0023] Figure 12A is a magnitude response plot of bodily tissue tested via an HFTS system according to embodiments.

[0024] Figure 12B is a phase response plot of bodily tissue tested via an HFTS system according to embodiments.

[0025] To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures. It is contemplated that elements disclosed in one embodiment may be beneficially utilized on other embodiments without specific recitation.DETAILED DESCRIPTION

[0026] Embodiments described herein generally relate to transmission spectroscopy. More specifically, embodiments described herein relate to a high frequency transmission spectroscopy (HFTS) system and methods. The HFTS system and methods provided herein provide for identification or classification of materials of interest in a wide range of industries. The HFTS system and methods provided herein are capable of identifying unknown particles, characterizing particlesize / charge, or monitoring solutions for specific materials in semiconductor, biomedical, and environmental applications.

[0027] An integrated circuit configuration of the HFTS system and methods provided herein have the capability of production testing and counterfeit detection in the semiconductor industry. An analyte configuration of the HFTS system and methods provided herein have the capability to screen for diseases and infections, and test for particles, such as environmental particles. The FTS system and methods provided herein have the capability to screen in-situ. The HFTS system and methods may be utilized for identifying unknown particles, characterizing particle size / charge, or monitoring solutions for specific components within a complicated system with applications in blood borne illness diagnosis, water pollution monitoring, and nanoparticle characterization. An additional application of the HFTS system and methods include using the HFTS system as a nano-scale analyzer for nanoparticle tracking analysis including the detection of particle size and / or concentration of particles. For microbial identification, the HFTS system is operable to be used as a microbial analyzer identifying the presence of specific microbes through markers or by conducting comparisons against a standardized library. The HFTS system and methods may characterize solid and soft matter such as muscle mass and bones.

[0028] The HFTS methods are able to determine qualities of dispersed components in solution to obtain concentration, size, and / or charge changes of the components. The components include, but are not limited to, salts, nanoparticles, microparticles, cells, and other microscale particles within the sample solution. The HFTS system described herein may be applied to cell counting, cell identification, blood cell profiling, nanoparticle detection, nanoparticle sizing, determining nanoparticle concentration, microbe detection, microbe counting, determination of tissue composition, and microbe identification. The microbial detection, counting, and identification is applicable to virus microbes including bacteria, fungi, and viruses. In some embodiments, the HFTS methods via the HFTS system are operable to determine changes in concentration with a detection limit of about 15 pM. In other embodiments, the HFTS methods via the HFTS system are operable to determine changes in concentration with a detection limit of about 0.15 pM. In additionalembodiments, the HFTS methods via the HFTS system are operable to determine changes in concentration with a detection limit of about 15 nM.

[0029] Figure 1A is a schematic view of a high-frequency transmission spectroscopy (HFTS) system 100. Figure 1 B is a system architecture of the system 100. The system 100 and system architecture may be or include an experimental or prototype system, and other HFTS systems including a test cell 102, an input antenna 104, an output antenna 106, and an electric network analyzer (ENA) 108 may be used to perform the methods described herein.

[0030] The system 100 further includes a host computer 114. In some embodiments, the system 100 further includes a digital controller 112. Each of the ENA 108, the digital controller 112, and the host computer 114 are connectable to each other via communication links 113. The communication links 113 send and receive files, data, and / or instructions. The communication links 113 can temporarily or permanently store files, data, or instructions in the cloud, before transferring or copying the files, data, or instructions between the ENA 108, the digital controller 112, and / or the host computer 114.

[0031] The host computer 114 includes a central processing unit (CPU), support circuits, and memory. The CPU can be one of any form of computer processor that can be used in an industrial setting for controlling the system 100. The memory is coupled to the CPU. The memory can be one or more of readily available memory, such as random access memory (RAM), read only memory (ROM), floppy disk, hard disk, or any other form of digital storage, local or remote. The support circuits are coupled to the CPU for supporting the processor in a conventional manner. These circuits include cache, power supplies, clock circuits, input / output circuitry, subsystems, and the like. The host computer 114 can include the CPU that is coupled to input / output (I / O) devices found in the support circuits and the memory or the digital controller 112. The digital controller 112 is connected to an interface 120, such as a relay matrix 204 (as shown in Figure 2). The interface 120 is connected to the test cell 102. The test cell 102 includes a material of interest (MOI) 101 . The test cell 102 includes the MOI 101 such that the interface 120 connects to the MOI 101. For example, the relay matrix 204 is connected to the test cell 102 via traces. The digital controller 112 is operable to manage states of the interface 120 by providing digitalinput / out (DIO) signal. The DIO signal controls the state of the interface 120. For example, the digital controller 112 is connected to the relay matrix 204 which is connected to the MOI 101 , i.e., an integrated circuit (IC) 201. The digital controller 112 provides a DIO signal to the relay matrix 204 to turn activate / deactivate relays 206 of the relay matrix 204 for each pin permutation of the IC 201 .

[0032] The test cell 102 is operable to retain a MOI 101. The MOI 101 includes, but is not limited to, integrated circuits (ICs), bodily fluids, bodily tissues, hard matter, soft matter, micropartical solutions, nanoparticle solutions, or electrolyte solutions. An example of hard matter is bone. An example of soft matter is flesh. The MOI 101 is any material having the capability for an electromagnetic (EM) wave to be transmitted therethrough. Thus, the system 100 and the methods provided herein identify and classify components of MOI 101 with the capability for an EM wave to be transmitted therethrough. The system 100 and the methods provided herein are applicable to semiconductor, biomedical, and environmental applications.

[0033] The input antenna 104 and the output antenna 106 are aligned opposite to each other. The test cell 102 with the MOI 101 is positioned between input antenna 104 and the output antenna 106 such that the input electrode antenna 104 and the output electrode antenna 106 oppose each other. The conducting surface of the antennas 104, 106 are electrically isolated from the MOI 101. The input antenna 104 and the output antenna 106 are electrodes. In some embodiments, the input antenna 104 and the output antenna 106 are patch antennas. The patch antennas are printed circuit board (PCB) patch antennas. The input antenna 104 and the output antenna 106 are electrically isolated from the MOI 101 to prevent direct electrical current from flowing between them while still allowing signal transmission. The electrical isolation is achieved by physical separation or insulation as further described herein. In some embodiments, the input antenna 104 and the output antenna 106 are not placed in physical contact with the MOI 101. In other embodiments, the conducting surface of the antennas 104, 106 includes a layer of non-conductive material disposed thereon. The non-conductive material can be an epoxy material, plastic or glass. The non- conductive material may be commonly known as a silkscreen. This electrical isolation prevents electrochemical effects, common to impedance spectroscopy, from dominating frequency dependent measurements.

[0034] The ENA 108 is a two-port instrument operable to obtain the frequency response of a system via calculation of the system’s scattering parameters. The ENA 108 is operable to generate an alternating current (AC) excitation potential on Port 1 , i.e. , input port 116, and sweep the signal frequency from an initial frequency to a final frequency. The signal frequency is swept at a frequency step (fstep). In one example, the fstep is 250 kilohertz (kHz). The ENA 108 is operable to generate and record frequencies of 1 kHz to 3 gigahertz (GHz). In some embodiments, the initial frequency is 50 kHz and the final frequency is 500 MHz. In other embodiments, the final frequency is 500 gigahertz (GHz). The ENA 108 is connected to the input antenna 104 and the output antenna 106 with coaxial cables connected to RF connectors 122 of the input antenna 104 and the output antenna 106. In one example, the RF connectors 122 are SubMiniature version A (SMA) connectors. The input port 116 is connected to the input electrode antenna 104 and the output electrode antenna 106 is connected to the output port 118 of the ENA 108 via RF connectors 122.

[0035] The ENA 108 is operable to receive an AC signal on Port 2, i.e., output port 118, and calculate the ratio of the received signal to the excitation signal. The ENA 108 is operable to generate the AC as the excitation signal and sweep the frequency of the AC from an initial frequency to a final frequency. In operation, the input antenna 104 transmits the excitation potential from the input port 116 of the ENA 108 to the MOI 101. The output antenna 106 receives the transmission which has been attenuated and phase-shifted as it propagated through the MOI 101. The signal is received by Port 2. The ENA 108 is operable to calculate the frequency response. The MOI 101 is then classified. The host computer 114 is operable to process the frequency response and perform an analysis to classify the MOI 101 .

[0036] The Port 1 (input port 116) excitation potential results in an electromagnetic (EM) wave, i.e., excitation signal, to be transmitted from the TX antenna (input antenna 104). The EM wave propagates through the MOI 101 until it is received by the RX antenna (output antenna 106). Port 2, i.e., output port 118, receives the signal, i.e., received signal. The ENA 108 calculates a parameter. In some embodiments, the parameter is a S21 parameter, i.e., forward transmission coefficient, at each frequency step (fstep) from the initial frequency to the final frequency. The S21 parameter is a ratioof the power of the received signal at Port 2 (output port 118) to the power of the excitation signal at Port 1 (input port 116). The S21 parameter is denoted as S21= Power at Port 2 (output) . voita e (t) -| QQ plots the magnitude responsePower at Port 1 (input)VoltageTX(f)ra r(representing the signal strength) and the phase response (representing the time delay) at each fstep. The host computer 114 receives data 103 from the ENA 108 including the plot of the magnitude response and the plot of the phase response at each fstep. The system 100 models the MOI 101 as a linear, time-invariant system such that the EM wave experiences only modulation in magnitude and phase as the EM wave propagates through the MOI 101.

[0037] The host computer 114 includes a suite 105 of at least one of one or more machine learning-based (ML-based) models 107 or one or more deep learning-based (DL-based) models 109 that infer / predict information related to MOI 101 based on the data 103. The data 103 includes at least one of the magnitude response or the phase response at each frequency step from the initial frequency to the final frequency. The ML-based models 107 include, but are not limed to, classification ML-based models, clustering ML-based models, or combinations thereof. Example ML-based models 107 include, a k-Nearest Neighbor (KNN) model, a Principal Component Analysis (PCA) model, Logistic Regression model. The DL-based models 109 include, but are not limed to, Convolutional neural networks (CNNs), Recurrent neural networks (RNNs), Adversarial Neural Networks, Generative Neural Networks, or combinations thereof. The one or more ML-based models 107 and / or DL-based models 109 are applied to the data 103 including at least one of the magnitude response or the phase response at each frequency step to provide at least one output 111. The output 111 includes, but is not limited to, at least one of identification or classification of particle formulations or concentrations, electrolyte formulations or concentrations, ICs, bodily tissue compositions, bodily fluid compositions, hard matter compositions, soft matter compositions, particle sizes, or particle charges.

[0038] Figure 2 is a schematic view of an integrated circuit (IC) configuration 200 of the system 100. The IC configuration 200 includes the test cell 102, the input antenna 104, the output antenna 106, the digital controller 112, and the host computer 114. The test cell 102 corresponds to a test socket 202 to retain a respective IC 201 . The test socket 202 may retain various types of ICs to classify ICs by at least one oflot ID or device family. The test socket 202 may be a burn-in test socket. Each IC 201 to be tested corresponds to the MOI 101. Each IC 201 may be a small outline integrated circuit (SOIC). The SOIC may be a SOIC package. The IC configuration200 includes a relay matrix 204 having relays 206. The relays 206 are connected to the test socket 202 via traces 210. The digital controller 112 is connected to the relay matrix 204 (as shown in Figure 2). The IC configuration 200 includes a circuit board 203 having the test socket 202, relay matrix 204 having the relays 206, and traces 210. The input antenna 104 and the output antenna 106 are connected to the circuit board 203. The input antenna 104 and the output antenna 106 are electrically isolated from the test socket 202 to prevent direct electrical current from flowing between them while still allowing signal transmission. The electrical isolation is achieved by physical separation of the test socket 202 from the circuit board 203.

[0039] The digital controller 112 is connected to the relay matrix 204 which is connected to the test socket 202. The test socket 202 connects each pin of the IC201 to be tested to a respective relay 206. The digital controller 112 provides a DIO signal to the relay matrix 204 to turn activate / deactivate, i.e. , turn on or off, relays 206 of the relay matrix 204 for each pin permutation of the IC 201 to be tested. The relay matrix 204 includes a matrix. For example, the relay matrix 204 includes the relay 206 arranged in a 2x8 matrix such that all 256 permutations of an 8-pin IC may be tested. In some embodiments, the relay matrix 204 is connected to a power source 208. The power source 208 is operable to provide VDC power to the relay driver of the relay matrix 204.

[0040] Figure 3 is a flow diagram of a method 300 of IC testing. At operation 301 , a first IC is retained by the test socket 202. A device under test (DUT) ID of the first IC is provided to the host computer 114. At operation 302, the relay matrix 204 is set to an initial relay permutation and the ENA 108 generates an input voltage (VoltageRX) and sweeps the signal frequency from an initial frequency to a final frequency. The relay 206 is set to the initial relay permutation by the digital controller 112 providing a DIO signal to the relay matrix 204. At the initial permutation, the ENA 108 generates an AC excitation potential on the input port 116. The excitation potential results in an EM wave, i.e., excitation signal, to be transmitted from the input antenna 104 to the circuit board 203. The EM wave propagates through a respective IC 201 until it isreceived by the output antenna 106. The output port 118, receives the signal, i.e., output voltage (VoltageTX). The ENA 108 calculates a parameter at each frequency step (fstep) from the initial frequency to the final frequency. In some embodiments, the recorded parameters are S21 parameters. The recorded parameters at each fstep at the initial relay permutation of the relay matrix 204 corresponding to the initial pin permutation of the first IC are provided to the host computer 114 under the corresponding DUT ID.

[0041] At operation 303, the relay matrix 204 is set to a subsequent permutation and the ENA 108 generates VoltageRXand sweeps the signal frequency. The recorded S21 parameters at the subsequent permutation of the relay matrix 204 for the first IC are provided to the host computer 114 under the corresponding DUT ID. Operation 303, is repeated for each pin permutation of the first IC to be tested. The S21 parameters are recorded for each relay permutation of the corresponding pin permutation of the first IC. The host computer 114 is provided with the recorded S21 parameters for each relay permutation of the corresponding pin permutation under each respective DUT ID.

[0042] Operations 301 -303 are sequentially repeated for additional ICs 201 such that a plurality of ICs 201 may be tested. At operation 304, the ICs 201 are classified. The host computer 114 includes the suite 105 of at least one of one or more machine learning-based (ML-based) models 107 or one or more deep learning-based (DL- based) models 109 for analyzing the data 103 of the recorded S21 parameters for each relay permutation of the corresponding pin permutation under each respective DUT ID. The data 103 includes at least one of the magnitude response or the phase response at each frequency step from the initial frequency to the final frequency. The one or more ML-based models 107 and / or DL-based models 109 are applied to the data 103 including at least one of the magnitude response or the phase response at each frequency step to provide at least one output 111. The output 111 includes classification of ICs 201 by at least one of lot ID or device family.

[0043] Figure 4A is a confusion matrix of the classification performance for lot IDs. Figure 4B is a chart of the classification accuracy scores at each relay permutation for pin permutations of a lot ID. For example, as shown in Figure 4A, a batch of 31 ICs 201 were tested. The batch of 31 ICs 201 included were from three different lots, eachhaving a lot ID. A lot ID is a unique identification code for a specific batch (lot) of chips manufactured together. The lot IDs were from the same device family. A device family is a grouping of chips or devices based on shared features, architectures, or functionalities. In other embodiments, the method 300 may test ICs 201 of different device families with the same or different lot IDs. As shown in Figure 4A, ICs 201 of three lot IDs were tested. Only one IC was misclassified. The mean (p) accuracy scores are represented by black squares, while the error bars (p-o) illustrate the standard deviation below the mean.

[0044] Figure 5 is a schematic view of an analyte configuration 500 of the system 100. The analyte configuration 500 includes the test cell 102, the input antenna 104, the output antenna 106, the ENA 108, and the host computer 114. The test cell 102 is a vessel 502 with a volume to retain a sample solution. The MOI 101 corresponds to the sample solution. The vessel 502 may be 3D printed. The vessel 502 may include an input antenna slot 504 and an output antenna slot 506. The input antenna 104 is disposed in the input antenna slot 504 and the output antenna 106 is disposed in the output antenna slot 506. The input antenna 104 and the output antenna 106 are electrically isolated from the sample solution to be disposed in the volume of the vessel 502 to prevent direct electrical current from flowing between them while still allowing signal transmission. The conducting surface of the antennas 104, 106 includes a layer of non-conductive material 508 disposed thereon when disposed in the slots 504, 506.

[0045] Figure 6 is a flow diagram of a method 600 of analyte testing. In some embodiments, prior to operation 601 , the vessel 502 is cleaned. In some embodiments, the vessel 502 is rinsed with ultra-pure water and wiped using a sterile paper towel. The vessel 502 is wiped for a minimum of 10 seconds. In other embodiments, the vessel 502 is washed for 15 minutes in isopropyl alcohol and is sterilized under UV irradiation. This reduces the likelihood of external factors (e.g., the resonant frequency shifted by parasitic inductance caused by the orientation of the coaxial cables) from contributing misleading clustering results during data analysis.

[0046] At operation 601 , a first sample solution is dispensed in the test cell 102. The sample solution is dispensed at a specified volume. For example, the specific volume of sample solution is 2 mL. At operation 602, an ENA measurement processis performed. The ENA 108 generates an AC excitation potential on the input port 116. The excitation potential results in an EM wave, i.e., excitation signal, to be transmitted from the input antenna 104 to the sample solution. The EM wave propagates through the sample solution until it is received by the output antenna 106. The output port 118, receives the signal, i.e., output voltage (VoltageTX). The ENA 108 calculates the S21 parameter from at each frequency step (fstep) from the initial frequency to the final frequency. The ENA 108 plots at least one of the magnitude response (representing the signal strength) or the phase response (representing the time delay) at each fstep. At operation 603, the host computer 114 receives data 103 from the ENA 108 including the plot of the magnitude response and the plot of the phase response at each fstep. The system 100 models the MOI 101 as a linear, timeinvariant system such that the EM wave experiences only modulation in magnitude and phase as the EM wave propagates through the MOI 101 .

[0047] In some embodiments, the first sample solution is compared to previous measurements or data 103 to identify one or more of the material, formulation, concentration, size, cation radius, cation charge, anion radius, or anion charge of one or more analytes in the first sample solution. The analytes include, but are not limited to, salts, nanoparticles, microparticles, cells, or other microscale particles within the sample solution. In other embodiments, operations 601 -603 are repeated for at least a second sample solution (i.e., operations 601 -603 are conducted for additional sample solutions). The sample solutions are compared to each other to identify one or more of the material, formulation, concentration, size, cation radius, cation charge, anion radius, or anion charge of one or more analytes in each sample solution. The host computer 114 receives data 103 for each sample solution from the ENA 108 including the plot of the magnitude response and the plot of the phase response at each fstep.

[0048] The host computer 114 includes the suite 105 of at least one of one or more machine learning-based (ML-based) models 107 or one or more deep learning-based (DL-based) models 109 to analyze the data 103 at least one of the magnitude response plots or the phase response plots for each sample solution. The one or more ML-based models 107 and / or DL-based models 109 are applied to the data 103 including at least one of the magnitude response or the phase response at eachfrequency step to provide at least one output 111. The output 111 includes at least one of classification or identification of the one or more sample solutions. The output 111 includes, but is not limited to, one or more of the material, formulation, concentration, size, cation radius, cation charge, anion radius, or anion charge of one or more analytes in each sample solution.

[0049] In one example, the at least one output 111 is provided by selecting features for clustering selected from the magnitude response and phase plots. Dimensionality is reduced to features based on the measured resonance peaks. In some embodiments, visual analysis of the magnitude response and phase plots resonance peaks may be performed to determine the resonance peaks. The features selected for clustering include the frequency, magnitude response, and phase response of the resonance peaks. These features may be normalized using min / max normalization to map each feature to a continuous range. One or more ML-based models 107 and / or DL-based models 109 may be applied to reduce the dimensionality. For example, a PCA model is used. To classify the data 103 one or more ML-based models 107 and / or DL-based models 109 may be applied. The ML-based models 107 and / or DL- based models 109 are trained models. For example, a KNN model is used. The ML- based models 107 and / or DL-based models 109 are trained such that an individual sample solution is identifiable in a single test to determine particle formulation, size, charge, or concentration, electrolyte formulation, size, charge, or concentration, hard matter composition, soft matter composition, or bodily fluid composition (i.e., only operations 601 and 602 are performed to generate the at least one output 111 ).

[0050] In some embodiments, the HFTS methods via the HFTS system are operable to determine changes in concentration with a detection limit of about 15 pM. In other embodiments, the HFTS methods via the HFTS system are operable to determine changes in concentration with a detection limit of about 0.15 pM. In additional embodiments, the HFTS methods via the HFTS system are operable to determine changes in concentration with a detection limit of about 15 nM. The HFTS system 100 provides the sensitivity to differentiate anions or cations for identical solute concentrations with the same charge. HFTS is developed as a new approach to aqueous chemical and physical sensing in bulk solutions that is applied to analyze a range of analytes in aqueous solution. HFTS provides the sensitivity to detect changesin concentration, system chemistry, surface chemistry, and analyte size. Thus, HFTS has potential to perform both qualitative and quantitative analysis of mixtures.Concentration Analysis

[0051] One implementation of the method 600 is determining concentrations of at least one component in the sample solution. Sample solutions having the same analyte with different concentrations may be tested to obtain data 103 corresponding to at least one of the magnitude response or phase response. For example, the analyte is NaCI. Concentrations of different molarities are analyzed. In some embodiments, the method 600 using the system 100 provides the ability to distinguish concentration changes on the order of 250 pM.

[0052] In this example, NaCI solutions include 0 M, 15 pM, 150 pM, 1.5 mM, and 150 mM concentration. Figure 7 is a magnitude response plot of the NaCI solutions generated by the ENA 108. The host computer 114 identifies the magnitude and the peak frequency locations. In this example, at concentrations below 150 mM there is a magnitude peak centered at approximately 60 MHz. The magnitude and the peak frequency locations decrease as the concentration increases. The peak response is shown in Table 1. The NaCI solutions from 15 pM to 15 mM show an inverse correlation with peak response. The response of the 150 mM NaCI solution shows the emergence of another peak at a lower frequency (14.5 MHz). The rate of decrease in magnitude is the greatest from 1.5 mM to 15 mM.Concentration (M) Peak Response0 M 0.8715 M 0.851500.821.5 mM 0.6015 mM 0.29Table 1

[0053] The one or more ML-based models 107 and / or DL-based models 109 may be applied to the data 103 to conduct a least-squares linear regression. The leastsquares linear regression for the above table show a strong inverse correlation between changes in concentration and changes in magnitude response. When alinear concentration is between the concentration and the peak value of the magnitude response for data 103, the slope may be found to identify the concentration of the sample solutions. For example, in Table 1 , a slope of 0.0249±0.005 M-1is found with an intercept of 0.0204±0.004 M and an R2value of 0.88. In this example, the method 600 using the system 100 provides a linear response to changes in salt concentration. The calculated slope and the sensitivity of the ENA 108 provide an ability to distinguish concentration changes of salts on the order of 250 pM.Electrolyte Concentration

[0054] One embodiment of the method 600, includes testing one or more sample solutions to classify electrolytes by at least one of cation radius, cation, charge, and anion radius. Sample solutions having the different analytes are tested to obtain data 103 corresponding to at least one of the magnitude response or phase response. For example, the sample solutions include sodium chloride (NaCI), potassium chloride (KCI), and magnesium chloride (MgCl2) at a concentration of 0.15pM. The sample solutions at 1 ,5pM and 15pM may also be tested. Figure 8A is a magnitude response plot of the solutions generated by the ENA 108. Figure 8B is a phase response plot of the solutions generated by the ENA 108. The frequency, magnitude response, and phase response of each resonant peak are chosen as features for clustering as shown as a statistical summary in Table 2.Selected Features (p, a)Concentration Analyte Frequency (MHz) Magnitude Phase (rad)058, 0.862 0.875, 0.005 -2.80, 1.123KCI 271 , 0.625 0.535, 0.013 -18.1 , 1.126458, 4.390 0.536, 0.031 -26.2, 1.143061 , 1.326 0.864, 0.005 -2.70, 0.0680.15pM MgCI2249, 15.83 0.595, 0.022 -17.0, 0.654456, 12.93 0.479, 0.036 -25.9, 0.482059, 1.138 0.854, 0.013 -2.60, 0.063NaCI 262, 9.311 0.420, 0.057 -17.0, 1.821466, 11.12 0.322, 0.107 -25.6, 1.963059, 0.654 0.865, 0.014 -2.60, 0.0371.5pM KCI 271 , 0.656 0.538, 0.018 -17.8, 0.027456, 4.518 0.533, 0.021 -25.9, 0.159062, 0.770 0.872, 0.005 -2.80, 0.046MgCI2238, 1.245 0.594, 0.011 -16.6, 0.050463, 3.330 0.449, 0.017 -26.1 , 0.130060, 0.629 0.868, 0.005 -2.70, 0.053NaCI 270, 5.116 0.527, 0.027 -17.6, 1.372455, 7.598 0.454, 0.073 -25.5, 1.416059, 0.873 0.868, 0.006 -2.80, 1.129KCI 270, 4.848 0.541 , 0.027 -17.8, 1.789453, 6.684 0.510, 0.074 -25.7, 1.695061 , 1.326 0.864, 0.005 -2.70, 0.06815pM MgCI2249, 15.83 0.595, 0.022 -17.0, 0.654456, 12.93 0.479, 0.036 -25.9, -0.482059, 1.138 0.854, 0.013 -2.60, 0.063NaCI 262, 9.311 0.420, 0.057 -17.0, 1.821466, 11.12 0.322, 0.107 -25.6, 1.963Table 2

[0055] The one or more ML-based models 107 and / or DL-based models 109 may be applied to the data 103 to conduct one or more of KNN clustering, PCA modeling, or training to successfully identify or classify electrolytes such that an electrolyte in a sample solution is identifiable in a single test.Nanoparticle or Microparticle Concentration Analysis

[0056] Another implementation of the method 600 using the system 100 includes determining the concentrations of nanoparticles or microparticles in sample solutions. Example, nanoparticles include carbon dot nanoparticles. Example, microparticles include plastic microparticles. The method 600 is performed with sample solutions having nanoparticles or microparticles of different concentrations. In one example, sample solutions with concentrations of carbon dot nanoparticles of 0%, 20%, 60%, and 100% are tested. In another example, sample solutions with concentrations of carbon latex microparticles of 0%, 20%, 60%, and 100% are tested. Figure 9A is a magnitude response plot of nanoparticle solutions. Figure 9B is a magnitude response plot of microparticle solutions. There is the peak response at each concentration and correlation between concentration and magnitude response at a frequency. A strong linear relationship exists between nanoparticle concentration and magnitude responsewith an expected sensitivity to concentration changes of approximately 7%. The system 100 can be used to determine the concentration of nanoparticles and microparticles in aqueous solution with a relatively high sensitivity.Cell Analysis

[0057] Another implementation of the method 600 using the system 100 includes characterization of bulk living cells. By passing the high-frequency signals through the cells and measuring the frequency response, metrics such as type, size, and health may be determined. Sulfate latex microparticles are analogue for living cells. In another example, microparticle solutions having particles of 4pm and 10pm in water are tested via the method 600. Figures 10A and 10B are magnitude response plots of microparticle solutions. As shown in Figure 10A, the larger 10 pm particles have lower frequency response. This is consistent with the analysis of electrolytes at higher concentrations where the larger molecule also yielded a lower response. As shown in Figure 10B, there is a secondary peak in magnitude response around 260 MHz where a larger discrimination between particle size is visible. The one or more ML-based models 107 and / or DL-based models 109 may be applied to the data 103 to conduct one or more of KNN clustering, PCA modeling, or training to successfully identify or classify living cells in a bodily fluid or sample solution in water.Environmental Analysis

[0058] Another implementation of the method 600 using the system 100 includes identification of plastic microparticles in water. For example, there is a measurable difference between the frequency responses of the 10 pm and the frequency of the 4 pm particles. There is a secondary peak in magnitude response around 260 MHz where a larger discrimination between particle size is visible. In contrast to the lower frequency results, the larger 10 pm particles show a higher peak magnitude response than the smaller 4 pm particles. Figure 11 is a chart of classification accuracy of identification of plastic microparticles.Biomedical

[0059] In one or more embodiments, the system 100 is used for biomedical testing. Another implementation of the system 100 includes characterization of bodily tissueand the like. Figure 12B is a phase response plot of bodily tissue tested via the system 100. By passing the high-frequency signals through the bodily tissue and measuring the frequency response, metrics such as tissue, flesh, and / or bone composition may be determined. As shown in Figures 12A and 12B, bodily tissue is classified by muscle, fat, and a combination of muscle and fat. Another implementation of the system 100 includes characterization of hard matter and / or soft matter. For example, as shown in Figures 13A and 13B, system 100 can characterize flesh where flesh is classified by bone and air. Figure 13A is a magnitude response plot of flesh tested via the system 100. Figure 13B is a phase response plot of flesh tested via the system 100. By passing the high-frequency signals through flesh, and measuring the frequency response, metrics such as bone composition of the flesh may be determined. As shown in Figures 13A and 13B, flesh is classified by bone and air.

[0060] Embodiments described herein generally relate to transmission spectroscopy. More specifically, embodiments described herein relate to a high frequency transmission spectroscopy (HFTS) system and methods. The HFTS system and methods provided herein provide for identification or classification of materials of interest in a wide range of industries. The HFTS system and methods provided herein are capable of identifying unknown particles, characterizing particle size / charge, or monitoring solutions for specific materials in semiconductor, biomedical, and environmental applications. The material of interest be analyzed via the HFTS system and methods provided herein includes, but is not limited to, integrated circuits (ICs), bodily fluids, bodily tissues, hard matter, soft matter, micropartical solutions, nanoparticle solutions, or electrolyte solutions. An integrated circuit configuration of the HFTS system and methods provided herein have the capability of production testing and counterfeit detection in the semiconductor industry. An analyte configuration of the HFTS system and methods provided herein has the capability to screen for diseases and infections and test for particles, such as environmental particles. The HFTS system and methods provided herein have the capability to screen in-situ. The HFTS system and methods may be utilized for identifying unknown particles, characterizing particle size / charge, or monitoring solutions for specific components within a complicated system with applications in blood borne illness diagnosis, water pollution monitoring, and nanoparticle characterization. An additional application of the HFTS system and methods includeusing the HFTS system as a nano-scale analyzer for nanoparticle tracking analysis including the detection of particle size and / or concentration of particles. For microbial identification, the HFTS system is operable to be used as a microbial analyzer for identifying the presence of specific microbes through markers or by conducting comparisons against a standardized library. The HFTS system and methods may characterize solid and soft matter such as muscle mass and bones.

[0061] The descriptions of the various embodiments of the present disclosure have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the embodiments, the practical application or technical improvement over technologies found in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed here.

[0062] Furthermore, although embodiments disclosed herein may achieve advantages over other possible solutions or over the prior art, whether or not a particular advantage is achieved by a given embodiment is not limiting of the scope of the present disclosure. Thus, the aspects, features, embodiments, and advantages described are merely illustrative and are not considered elements or limitations of the appended claims except where explicitly recited in a claim(s). Likewise, reference to “the disclosure” shall not be construed as a generalization of any inventive subject matter disclosed herein and shall not be considered to be an element or limitation of the appended claims except where explicitly recited in a claim(s).

[0063] While the foregoing is directed to embodiments of the present disclosure, other and further embodiments of the disclosure may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow. While the foregoing is directed to embodiments of the present disclosure, other and further embodiments of the disclosure may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow.

Claims

What is claimed is:1 . A high frequency transmission spectroscopy (HFTS) system comprising: an electric network analyzer (ENA), the ENA having an input port and an output port, wherein the ENA is operable to generate an AC (Alternating Current) as excitation signal and sweep a frequency of the AC from an initial frequency to a final frequency; an input antenna electrically connected to the input port; an output antenna electrically connected to the output port; and a test cell positioned between the input antenna and the output antenna, the test cell is operable to retain a material of interest (MOI), conducting surfaces of the input antenna and the output antenna are electrically isolated from the MOI, wherein in operation of the HFTS system: the ENA is configured to generate an excitation signal having an initial voltage that is transmitted by the input antenna to the MOI as an electromagnetic (EM) wave such that the EM wave propagates through the MOI until is received by the output antenna as a transmitted signal having an output voltage; and the ENA is configured to calculate a parameter of the output voltage to input voltage from the initial frequency to the final frequency to plot at least one of a magnitude response or a phase response of the EM wave at each frequency step.

2. The HFTS system of claim 1 , further comprising: a circuit board having the test cell corresponding to a test socket to retain a respective integrated circuit (IC) corresponding to the MOI and a relay matrix of a plurality of relays connected to the test socket with traces; the input antenna and the output antenna connected to the circuit board with the test socket therebetween, the input antenna and the output antenna are physically separated from the test socket; and a digital controller connected to the relay matrix to provide a DIO signal to the relay matrix to activate or deactivate relays to correspond to each pin permutation of the respective IC.

3. The HFTS system of claim 2, wherein in operation of the HFTS system:the relay matrix is settable at an initial relay permutation and the ENA is configured to generate the excitation signal that is transmitted through the circuit board to the input antenna such that the EM wave propagates through the respective IC until received by the output antenna as the transmitted signal; the ENA is configured to calculate a parameter at each frequency step to plot at least one of the magnitude response or phase response at the initial relay permutation; and the relay matrix is settable at subsequent permutations and the ENA is operable to recorded parameters at the subsequent permutations for the respective IC.

4. The HFTS system of claim 3, further comprising a host computer connected to the ENA operable to receive the recorded parameters at each pin permutation for a plurality of ICs and classify the plurality of ICs by at least one of lot ID or device family.

5. The HFTS system of claim 2, wherein the test socket may retain various types of ICs.

6. The HFTS system of claim 1 , wherein the test cell is a vessel with a volume to retain a sample solution corresponding to the MOI.

7. The HFTS system of claim 6, wherein the vessel includes slots to retain the input antenna and the output antenna, the conducting surfaces of the input antenna and the output antenna have a layer of non-conductive material disposed thereon to prevent direct electrical current from flowing between them.

8. The HFTS system of claim 7, wherein the ENA is configured to calculate parameters at each frequency step to plot at least one of the magnitude response or phase response for each sample solution disposed in the vessel.

9. The HFTS system of claim 8, further comprising a host computer connected to the ENA operable to receive the parameters of each sample solution and classify sample solutions by particle formulation or concentration, electrolyte formulation or concentration, particle size, or particle charge.

10. The HFTS system of claim 1 , further comprising a host computer connected to the ENA, the host computer is operable to receive data from the ENA including at least one of the magnitude response or the phase response to identify particle formulations or concentrations, electrolyte formulations or concentrations, ICs, bodily tissue compositions, bodily fluid compositions, hard matter compositions, soft matter compositions, particle sizes, or particle charges of the MOI.11 . The HFTS system of claim 10, wherein the host computer includes a suite of at least one of one or more machine learning-based (ML-based) models or one or more deep learning-based (DL-based) models to classify the MOI.

12. The HFTS system of claim 1 , wherein the ENA is operable to generate and record frequencies of 1 kilohertz (kHz) to 3 gigahertz (GHz).

13. The HFTS system of claim 1 , wherein the parameter is an S21 parameter.

14. A high frequency transmission spectroscopy (HFTS) system comprising: an electric network analyzer (ENA), the ENA having an input port and an output port, wherein the ENA is operable to generate an AC (Alternating Current) as excitation signal and sweep a frequency of the AC from an initial frequency to a final frequency; an input antenna electrically connected to the input port; an output antenna electrically connected to the output port; and a circuit board having a test socket positioned between the input antenna and the output antenna, the test socket is operable to retain a respective integrated circuit (IC), the input antenna and the output antenna are connected to the circuit board with the test socket therebetween; a relay matrix of a plurality of relays connected to the test socket with traces; and a digital controller connected to the relay matrix to provide a DIO signal to the relay matrix to activate or deactivate relays to correspond to each pin permutation of the respective IC.

15. A high frequency transmission spectroscopy (HFTS) method comprising: retaining a material of interest (MOI) with a test cell, wherein:the test cell is positioned between an input antenna and an output antenna electrically isolated from the MOI; and the input antenna is connected to an input port and the output antenna is connected to an output port of an electric network analyzer (ENA); and generating an AC (Alternating Current) with the ENA as excitation signal and sweeping a frequency of the AC from an initial frequency to a final frequency, wherein: an initial voltage of the excitation signal is transmitted by the input antenna to the MOI as an electromagnetic (EM) wave that propagates through the MOI until is received by the output antenna as a transmitted signal having an output voltage; and the ENA is configured to calculate a parameter of the output voltage to input voltage wave at each frequency step from the initial frequency to the final frequency to plot at least one of a magnitude response or a phase response of the EM wave.

16. The HFTS method of claim 15, wherein: the test cell is a test socket of a circuit board that retains a respective integrated circuit (IC), the circuit board having a relay matrix of a plurality of relays connected to the test socket with traces; the input antenna and the output antenna are connected to the circuit board with the test socket therebetween, the input antenna and the output antenna are physically separated from the test socket; a digital controller connected to the relay matrix provides a DIO signal to set relay matrix at relay permutations corresponding to pin permutations of the respective IC; and the ENA calculates parameters at each frequency step to plot at least one of the magnitude response or phase response at the relay permutations.

17. The HFTS method of claim 16, wherein a plurality of ICs are tested and a host computer connected to the ENA receives recorded parameters at the relay permutations corresponding to pin permutations for each IC and classify the plurality of ICs by at least one of lot ID or device family.

18. The HFTS method of claim 15, wherein the parameter is an S21 parameter.

19. The HFTS method of claim 15, wherein the test cell is a vessel with a volume retaining a sample solution corresponding to the MOI.

20. The HFTS method of claim 15, a host computer connected to the ENA is operable to receive plots of at least one of the magnitude response or the phase response to classify sample solutions by particle formulation or concentration, electrolyte formulation or concentration, hard matter composition, soft matter composition, particle size, or particle charge.

Images