Free-Space Dielectric Property Measurement of Biological Tissues at W-Band Frequencies
Article information
Abstract
This paper presents a free-space measurement method for the dielectric properties of small biological tissues (in this study, micropig skins that are 60 mm × 60 mm in size and 1.4 mm in thickness are used) that do not meet the size requirement of conventional free-space material measurement at W-band (75–110 GHz) frequencies. The first step involved enhancing the material under test (MUT) holder of an existing free-space material measurement system to ensure the reliable measurement of small biological tissues. This enhancement was confirmed by comparing the measured dielectric properties of an air gap with its theoretical properties. The measured dielectric properties of the micropig skins were then validated through a comprehensive comparison using data based on the double Cole-Cole parametric model, which is a widely referenced model in many numerical dosimetry studies. The good agreement between the two results demonstrates the effectiveness of the proposed free-space material measurement method, as well as the reliability of the measured dielectric property of micropig skins.
I. Introduction
The usage and application of electromagnetic (EM) waves have rapidly expanded to various engineering and science fields, such as communication, military, security, aerospace, transportation, sensing, detection, medicine, radio astronomy, and geophysics. In addition, their operating frequencies have recently ventured into high-frequency ranges, while bandwidths have widened and power levels have increased [1]. Under the circumstances, more attention must be paid to the interaction between EM waves and the human body to ensure safety and healthcare [2–4].
The dielectric properties of biological tissues are a prerequisite for safety assessments [5–7] pertaining to exposure of the human body to EM waves and various biomedical applications [8]. For example, these properties contribute to the calculation of the specific absorption rate (SAR) [9, 10], which is widely used as a measure of the rate at which the human body absorbs radio frequency (RF) EM energy. Dielectric properties are also crucial in the design of biomedical devices and systems for disease detection and treatment, such as non-invasive diagnosis [11], hyperthermia [12], and biomedical implants [13].
In this context, coaxial probes and free-space material measurements determine the dielectric properties of biological tissues non-destructively [14]. The coaxial probe method [15] is based on reflection measurement by a semi-infinite material under test (MUT) or a planar MUT without (or with) a ground plane. However, a large MUT is required to ignore the scattering effect caused by the finite boundary of the MUT, especially for low-loss materials in a low-frequency range [16]. In contrast, the free-space method is based on reflection and transmission measurements by a planar MUT situated between two antennas in free space [17, 18], as shown in Fig. 1. It is especially suitable for a high-frequency range with an antenna, used as an EM transmit/receive (Tx/Rx) device, of a manageable size. While the former offers wideband material properties because the coaxial line can support the propagation of the transverse electromagnetic (TEM) mode, the latter yields frequency-banded material properties due to the limitations of the operating frequency range of the antenna.
Conventional free-space material measurement scheme for a planar MUT of M > 3S, where M is the size of the MUT and S is the waist radius of a Gaussian beam incident upon the MUT.
Free-space material measurement usually involves the following three steps:
1. The free-space material measurement system is calibrated to set the measurement reference plane (Reference planes #1 and #2, as shown in Fig. 1) at both sides of a planar MUT in free space.
T2. he scattering parameters and thickness of the MUT are measured using the calibrated material measurement system and thickness-measuring device, respectively. Conventionally, the size (M) of a planar MUT must be three times larger than the waist radius (S) of a Gaussian beam incident upon the MUT (i.e., M > 3S, as shown in Fig. 1) so as to ignore the diffraction effect at the edges of the MUT [17].
3. The material properties of the MUT are extracted from the measured scattering parameters and thickness using explicit expression and iterative techniques based on the Nicolson–Ross–Weir (NRW) method [19–21].
Several methods for calibrating free-space material measurement systems have been proposed in the literature. One such method is the thru-reflect-line (TRL) method, which employs a precision positioning system to adjust the separation distance between the antenna and the calibration standard (or MUT) in free space [17, 18]. As a variant of the TRL method, the thru-reflect-match (TRM) method requires a wideband low-reflectivity absorber [22], which replaces the line standard used in the TRL method. Furthermore, the gated-reflect-line (GRL) method necessitates a time-domain gating measurement and a planar metal plate used as a reflecting device [23], which is recommended to be as thick as an MUT. This method can afford to ignore the measurement uncertainty caused by movements of the RF cable, since all parts of this material measurement system remain stationary throughout the calibration and measurement procedures. Recently, unknown thru and two-tier one-port calibration methods [24, 25], usually used for characterizing coaxial and waveguide devices, have demonstrated their adaptability to application in free-space material measurement [26, 27] by using a planar offset short [28] as a free-space reflect standard. Notably, these methods demand at least three reflect standards for free-space one-port calibration, such as planar offset shorts equipped with different offsets.
Porcine tissues are widely used in human safety assessments and biomedical applications due to their similarity to human skin. The dielectric properties of these tissues can be obtained through TRL-calibrated free-space material measurements operating at the 50–110 GHz frequency range [14, 29]. For this measurement, biological tissues that meet the conventional size requirement of M > 3S must be used. The tissues were placed between two glass plates in free space. It also involves a deembedding process to account for the influence of the glass plates on the measurement and a precision positioning system for TRL calibration.
Recently, a free-space material measurement method has been developed for determining the dielectric properties of micropig skins, which are often used as reliable porcine tissue, without using glass plates at W-band (75–110 GHz) frequencies [30]. This measurement was conducted on small biological tissues of M < 3S, such as micropig skins of size 60 mm × 60 mm with a thickness of 1.4 mm [31], since large ones that met the size requirements were not readily available. The measurement system was calibrated using the GRL method, and the dielectric properties of the micropig skins were extracted through an iterative process based on the NRW method. Notably, this approach eliminates the need for a de-embedding process and precision positioning system.
This paper extends the above-mentioned small biological tissue measurement method by providing comprehensive details about the material measurement system, investing its validity, explaining thickness measurement process for the biological tissues, and examining the repeatability and reproducibility of the measured dielectric properties of small micropig skins. Section II describes the process of enhancing the MUT holder of an existing free-space material measurement system to ensure reliable measurement of small biological tissues. Section III describes the micropig skins used in this study, including their thickness measurement and the measurement condition. The feasibility of the enhanced free-space small material measurement is then confirmed by comparing the measured dielectric properties of an air gap with its theoretical properties. Furthermore, the measured dielectric properties of micropig skins are validated through a comparison based on the double Cole-Cole parametric model [14], a widely referenced model in many numerical dosimetry studies. Finally, Section IV summarizes this work.
II. Free-Space Biological Tissue Measurement System
1. Free-Space Material Measurement System in the W-Band
Existing free-space material measurement systems operating in the W-band usually comprise a W-band scattering parameter-measuring instrument and a quasi-optic free-space system [32], as shown in Fig. 2(a). Furthermore, the W-band scattering parameter-measuring instrument consists of a 0.01–67 GHz vector network analyzer (VNA), which is used as the main frame of the measuring instrument, and two 67–110 GHz frequency extenders. Meanwhile, the quasi-optic free-space system is composed of two Gaussian-beam Tx/Rx modules and an MUT holder mounted on a holder base located in the middle of the two Tx/Rx modules, as shown in Fig. 2(b). Each Tx/Rx module contains a corrugated horn antenna to generate a Gaussian beam, and an ellipsoidal mirror for 90° bending and for collimating the propagating Gaussian beam. The Tx/Rx module is set on a bench that can adjust its distance from the MUT holder using a positioning system.
Free-space material measurement system operating in the W-band: (a) photograph and (b) quasi-optic free-space system.
In Fig. 2(b), it is assumed that a Gaussian beam (as a localized plane wave) generated by a corrugated horn antenna connected to a frequency extender via a 1-mm coaxial cable is normally incident upon a planar MUT fixed at the MUT holder after 90° bending and collimating by an ellipsoidal mirror. The incident Gaussian beam is reflected, absorbed, and transmitted by the MUT in different ratios based on the material properties of the MUT.
In conventional free-space material measurement, a planar MUT needs to meet the size requirement of M > 3S. To meet this size requirement, existing free-space material measurement systems, which use a Gaussian beam with a 25-mm waist radius, can measure an MUT that is larger than 75 mm. However, the biological MUT of interest in this study—micropig skins that are 60 mm × 60 mm in size and 1.4 mm in thickness [31]—do not meet this size requirement. This underscores the crucial need to enhance the existing measurement system to reliably measure small biological tissues.
2. Enhancement of MUT Holder
Building upon the MUT holder of existing free-space material measurement systems, a combination of an MUT fixture and a fixture holder was devised for small biological tissue measurement, as shown in Fig. 3. First, the MUT fixture fixes a biological tissue between two metal plates (#1-1 and #1-2 in Fig. 4). This MUT fixture, along with the biological tissue, is then held between the two metal plates of the fixture holder (#2-1 and #2-2 in Fig. 5). To prevent scattering by the fixture holder, the outer surfaces of the fixture holder, which are illuminated by the Gaussian beam, are shielded by a flat absorber of 11 mm thickness, as shown in Fig. 6.
Enhanced free-space material measurement scheme for a small planar MUT of M < 3S, where M is the size of the MUT and S is the waist radius of a Gaussian beam incident upon the MUT.
MUT fixture consisting of two metal plates (#1-1, #1-2) for fixing a small MUT.
The fixture holder bears a central circular hole with a diameter (H) of 110 mm for holding the MUT fixture, as shown in Fig. 5. In contrast, the MUT fixture and the flat absorber have a central circular hole with a diameter of N (N = 54 mm and 51 mm for the MUT fixture and the flat absorber, respectively), as shown in Figs. 4 and 6. This hole, which is slightly smaller than the size of the biological tissue (in this study, M = 60 mm), is crucial for propagating a Gaussian beam.
To accommodate a biological tissue thicker than 1 mm between the two metal plates (#1-1 and #1-2 in Fig. 4) of the MUT fixture, the central area of the inner surfaces of the metal plate, which face each other, was removed to achieve a depth of 0.5 mm and a diameter of 107 mm. This cut-away side serves as the measurement reference plane (Reference planes #1 and #2 in Fig. 3) for scattering parameter measurement of the biological tissue. The distance between the two metal plates was precisely adjusted using two micrometer-heads at the short side of the MUT fixture and four springs with spring guide at the corner, as shown in Fig. 4. These components ensure that the biological tissue is held in place without the application of excessive pressure.
To measure the scattering parameters of the biological tissues, the enhanced free-space material measurement system was first calibrated using only the fixture holder. Subsequently, both the MUT fixture and the biological tissue were held between the two metal plates of the fixture holder for the scattering parameter measurement.
III. Dielectric Property Measurement Of Biological Tissues
1. Biological Tissue under Test
Dielectric property measurements were conducted using micropig skins that were 60 mm × 60 mm in size and 1.4 mm in thickness [31]. The biological tissues were delivered in a dry-ice frozen state. After thawing them at room temperature (23°C±1°C), they were used for conducting the measurements without delay to avoid tissue necrosis and drying. The measurements were repeated twice on two different days to ensure the repeatability and reproducibility of the results. The first measurement used four tissues evenly collected from two micropigs. The second measurement employed four tissues collected from one micropig, as shown in Fig. 7, since it aimed to reveal better homogeneity in the biological tissue under test compared to the first one. Notably, each measurement was first performed considering the thickness of the micropig skins and then the scattering parameters.
The four micropig skins used in the second measurement (60 mm × 60 mm in size and 1.4 mm in thickness).
2. Thickness Measurement of Biological Tissues
Typically, the thickness of a planar MUT is measured using a micrometer of rotating spindle calibrated by a gauge block. However, this method is unsuitable for measuring the thickness of biological tissue because spindle rotation may damage soft tissue. To avoid this, a 20 mm-diameter disc micrometer of a non-rotating spindle was employed, ensuring no damage or application of excessive pressure on the micropig skins. The average± standard deviation of the thickness of the four micropig skins used in the second measurement were 1.432±0.008, 1.434±0.028, 1.412±0.029, and 1.403±0.020 in millimeter, respectively. Notably, these measurements were calculated from the thickness measured at four different positions of each skin.
3. Measurement Condition
The scattering parameters of the micropig skins were measured using the enhanced free-space material measurement system operating at 100 Hz intermediate frequency (IF) bandwidth and 801 stepped frequency sweep points in the W-band at room temperature (23°C±1°C).
The enhanced system was calibrated using the GRL method [23, 32]. As a two-tier two-port calibration method, the GRL method first involved calibrating the material measurement system at the feeder of the two corrugated horn antennas using a waveguide TRL method, followed by the measurements below:
1. Gated, using the time-domain gated reflection coefficient seen looking into an antenna without an MUT in free space.
2. Reflect, using a metal plate (in this study, 3.924 mm in thickness) secured at the fixture holder.
3. Line, using an air delay line separated by the thickness of the metal plate.
The dielectric properties of the micropig skins were extracted from the measured thickness and scattering parameters, which were de-embedded to compensate for the effect of the difference in thickness of the skins and metal plate used in the GRL calibration. The extraction process involved an iterative technique [21, 33] that began by estimating the dielectric property. This initial estimate was then updated by minimizing the difference between the measured scattering parameters and those calculated from the updated dielectric property. This iterative process continued until the difference was less than the expected criterion. Notably, this technique relied on transmission coefficients (S21, S12), assuming that the skins were nonmagnetic.
4. Measurement Results
The feasibility of the enhanced free-space material measurement system, featuring an MUT holder specifically designed for testing small biological tissues that do not meet conventional size requirements, was first checked using air gap measurement. The measured complex relative permittivity (
Measured complex relative permittivity of an air gap of 3.924 mm thickness: (a) real and (b) imaginary parts.
The dielectric properties of the micropig skins were measured twice on two different days to check their repeatability and reproducibility. The results of the first measurement, which used four skins evenly collected from two micropigs, are shown in Fig. 9, and those of the second measurement, which employed four skins collected from one micropig, are shown in Fig. 10. The real and imaginary parts of the measured complex relative permittivity of the eight skins collected from three micropigs roughly showed a similar magnitude, decreasing with an increase in the frequency. Figs. 9 and 10 also show that the standard deviation of the measured complex relative permittivity obtained through the second measurement is better than that of the first, owing to better homogeneity of the biological tissue under test. Furthermore, Table 1 indicates that the maximum and minimum standard deviations of the measured complex relative permittivity of the second measurement were smaller than those of the first measurement.
Complex relative permittivity of the four micropig skins used in the first measurement, showing data based on the double Cole-Cole parametric model: (a) real and (b) imaginary parts.
Complex relative permittivity of the four micropig skins used in the second measurement, showing data based on the double Cole-Cole parametric model: (a) real and (b) imaginary parts.
Maximum and minimum standard deviations (SDs) of the measured complex relative permittivity of micropig skins
Since other biological tissue measurements, such as coaxial probe measurement, could not be performed in the laboratory, the measurement results obtained using the micropig skins were validated by comparing them with data based on the double Cole-Cole parametric model [14] (This model was determined from the permittivity measurement results of porcine skin with thicknesses of approximately 1.1 mm and 1.5 mm. These measurements were conducted using a coaxial probe method at a frequency range of 0.5–50 GHz, and a free-space method at a frequency range of 50–110 GHz, at temperatures ranging from 34°C to 37°C. This model exhibited an average deviation of approximately 2% for real and imaginary parts compared to the measured data), which is a widely referenced model in many numerical dosimetry studies. Figs. 9 and 10 show that the two measurement results roughly fall within the range of the epidermis and dermis of the parametric model, which is considered reasonable, since micropig skins are typically composed of these layers.
Fig. 11 shows that each average of the two measurement results agrees well. It also shows that the average of the real part approaches the value of the dermis of the parametric model with increasing frequency. In contrast, that of the imaginary part is similar to the average values of the epidermis and dermis.
Average of the complex relative permittivity of the micropig skins used in the first and second measurements, showing data based on the double Cole-Cole parametric model: (a) real and (b) imaginary parts.
IV. Conclusion
This paper presents a free-space measurement method for the dielectric properties of small biological tissues, thus challenging the size requirement for conventional free-space material measurement in the W-band (75–110 GHz) at room temperature (23°C±1°C). This proposed method utilized micropig skins, 60 mm × 60 mm in size and 1.4 mm in thickness, as the biological tissue under test. The measurement system was calibrated using the GRL method, and the dielectric properties of the micropig skins were extracted through an iterative process based on the NRW method.
The MUT holder of an existing free-space measurement system was enhanced for the reliable measurement of small biological tissues. This enhancement was confirmed by comparing the measured dielectric properties of an air gap with its theoretical properties. The dielectric properties of the micropig skins (this study used eight skins collected from three micropigs) were measured twice on two different days to ensure the repeatability and reproducibility of the measurement results. The measurement results were then comprehensively validated by comparison with data based on the double Cole-Cole parametric model. The good agreement between the two results provides evidence of the effectiveness of the proposed enhanced free-space material measurement method, as well as the reliability of the measured dielectric properties of micropig skins.
The enhancement of a conventional free-space material measurement system and the accurate measurement of the dielectric properties of micropig skins have significant implications for safety assessments pertaining to exposure of the human body to EM waves and various biomedical applications. These also provide a reliable basis for future research and practical applications.
Notes
This work was supported by Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) (No. 2021-0-00103, Research and development of technologies for utilization of THz frequency band and evaluation of electromagnetic safety) and the Korea Research Institute of Standards and Science (No. KRISS-2025-GP2025-0009, Developing digital safety measurement to enhance the availability of smart structural monitoring of facilities).
References
Biography
Jin-Seob Kang, https://orcid.org/0000-0002-0370-5810 received his B.S. degree in electronic engineering from Hanyang University, Seoul, Korea, in 1987, and his M.S. and Ph.D. degrees in electrical and electronic engineering from Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, in 1989 and 1994, respectively. In 1995, he was a postdoctoral research associate in the Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign. From 1996 to 1997, he was an assistant professor at the School of Electrical and Electronic Engineering in Chungbuk National University, Korea. Since 1998, he has been a principal research scientist at the Korea Research Institute of Standards and Science (KRISS). His research interests include electromagnetic (EM) measurement standards, material measurements, material parameter reference standards, (sub-)mm-wave measurements, and non-destructive testing.
Young Seung Lee received his B.S. degree in radio communications engineering from Korea University, Seoul, Rep. of Korea, in 2006 and M.S. and Ph.D. degrees in electrical engineering from the Korea Advanced Institute of Science and Technology, Daejeon, Rep. of Korea, in 2008 and 2012, respectively. Since 2012, he has been with the Electronics and Telecommunications Research Institute, Daejeon, Rep. of Korea, where he is presently a principal researcher. His current research interests are in the fields of exposure system design, compliance assessment, and numerical dosimetry with an emphasis on 5G frequencies. He is also concerned with electromagnetic theory, antenna propagation, and measurement.
Youngcheol Park, https://orcid.org/0000-0001-6275-4957 received his B.S. degree in electrical engineering from Yonsei University, Seoul, Korea, in 1992, and his Ph.D. degree in electrical engineering from the Georgia Institute of Technology, Atlanta, in 2004. He is currently a professor at Hankuk University of Foreign Studies, Korea. From 1994 to 2007, he worked with Samsung Electronics, Korea, where he designed mobile handsets. His research interests include high-efficiency power amplifiers, precise calibration of test equipment, and backscattering communication.