Capabilities of the attenuated total reflection (ATR) at THz wavelengths for increased sub-surface depth characterisation of (bio-)materials are presented. The penetration depth of a THz evanescent wave in biological samples is dependent on the wavelength and temperature and can reach 0.1–0.5 mm depth, due to the strong refractive index change ∼0.4 of the ice-water transition; this is quite significant and important when studying biological samples. Technical challenges are discussed when using ATR for uneven, heterogeneous, high refractive index samples with the possibility of frustrated total internal reflection (a breakdown of the ATR reflection mode into transmission mode). Local field enhancements at the interface are discussed with numerical/analytical examples. Maxwell’s scaling is used to model the behaviour of absorber–scatterer inside the materials at the interface with the ATR prism for realistic complex refractive indices of bio-materials. The modality of ATR with a polarisation analysis is proposed, and its principle is illustrated, opening an invitation for its experimental validation. The sensitivity of the polarised ATR mode to the refractive index between the sample and ATR prism is numerically modelled and experimentally verified for background (air) spectra. The design principles of polarisation active optical elements and spectral filters are outlined. The results and proposed concepts are based on experimental conditions at the THz beamline of the Australian Synchrotron.

The penetration depth of an evanescent wave in Attenuated Total Reflection (ATR) is dependent on the wavelength of the radiation utilised. At THz frequencies, the penetration depth into biological tissues is in the order of 0.1 to 0.5 mm; rendered pig lard was used as a model sample in this study. A method for the direct measurement of the evanescent wave penetration depth is presented which allows for the estimation of the dispersion of the complex refractive index by using the reflection of the evanescent wave from varying sample depths. The method employs frustrated total internal reflection, and has been demonstrated by using the THz/Far-IR beamline at the Australian synchrotron, and modelled using finite difference time domain (FDTD) simulations.

A novel method of investigating the temperature dependent variation of aspects of the complex refractive index n* in samples in the THz range using continuous, non-polarised, synchrotron radiation is presented. The method relies on the use of ATR apparatus, and retains the advantage of minimal sample preparation, which is a feature of ATR techniques. The method demonstrates a “proof of concept” of monitoring temperature reflectance whilst continuously heating or cooling samples by using a temperature variable Thermal Sample Stage. The method remains useful when the refractive index of the sample precludes attenuated total reflection study. This is demonstrated with the water reflectance experiments. The temperature dependent ATR reflectance of tissue-representative fats (lard and Lurpak® butter) was investigated with the novel approach. Both are within the ATR range of the diamond crystal in a “true” ATR mode. Lard showed no clear temperature variation between −15 °C and 24 °C at 0.7 to 1.15 THz or 1.70 to 2.25 THz. Lard can be regarded as having invariable, constant, dielectric properties within mixtures when biological substances are being assessed for temperature dependent dielectric variation within the stated THz ranges. Lurpak® butter (water content 14.7%) displayed temperature dependent reflected signal intensity features with a steady decline in reflectivity with increasing temperature. This is in line with the temperature-dependent behaviour of liquid water. There is no rapid change in reflected signal intensity even at −20 °C, suggesting that emulsified water retains liquid-water-like THz properties at freezing temperatures.

There is increasing interest in applications which use the 30 to 90 GHz frequency range, including automotive radar, 5 G cellular networks and wireless local area links. This study investigated pulsed 30-90 GHz radiation penetration into the human ear canal and tympanic membrane using computational phantoms. Modelling involved 100 ps and 20 ps pulsed excitation at three angles: direct (orthogonal), 30° anterior, and 45° superior to the ear canal. The incident power flux density (PD) estimation was normalised to the International Commission on Non-Ionizing Radiation Protection (1998) standard for general population exposure of 10 Wm −2 and occupational exposure of 50 Wm −2. The PD, specific absorption rate (SAR) and temperature rise within the tympanic membrane was highly dependent on the incident angle of the radiation and frequency. Using a 30 GHz pulse directed orthogonally into the ear canal, the PD in the tympanic membrane was 0.2% of the original maximal signal intensity. The corresponding PD at 90 GHz was 13.8%. A temperature rise of 0.032° C (+20%, −50%) was noted within the tympanic membrane using the equivalent of an occupational standard exposure at 90 GHz. The central area of the tympanic membrane is exposed in a preferential way and local effects on small regions cannot be excluded. The authors strongly advocate further research into the effects of radiation above 60 GHz on the structures of the ear to assist the process of setting standards.

Terahertz radiation (THz) is highly absorbed by liquid water. This creates the possibility of medical imaging on the basis of the water content difference between normal and diseased tissue. The effective penetration of THz is limited, however, to a tissue depth of 0.2–0.3 mm at body temperature. A unique feature of the 0.1–2.0 THz frequency is that there is a high disparity between liquid water absorption and ice absorption, with ice being 100 times more permeable to the radiation than liquid water. This results in 90% of the radiation surviving to 1.0 mm in ice, permitting the imaging of frozen tissues to a depth of 5.0 mm. This method is practical as an in vivo procedure before or during surgical excision. Finite difference time domain (FDTD) computational modeling of frozen normal skin and frozen melanoma was undertaken using tissue phantoms. The study suggests that sufficient contrast exists to differentiate normal frozen skin and melanoma on the basis of the difference of water content alone. When the melanin pigment in melanomas is modeled as a significant absorber of THz, the contrast changes. Based on the modeling, further exploration of the “THz‐skin freeze” imaging technique is justified. In the modeling, the boundary between the frozen tissue and non‐frozen tissue is shown to be strongly reflective. If the reflective properties of the boundary are substantiated, the “THz‐skin freeze” technique will have applications in other areas of skin diagnostics and therapeutics. Bioelectromagnetics. 40:118–127, 2019. © 2019 Bioelectromagnetics Society

Membrane model systems capable of mimicking live cell membranes were used for the first time in studying the effects arising from electromagnetic fields (EMFs) of 18 GHz where membrane permeability was observed following exposure.

Mobile communication has achieved enormous technology innovations over many generations of progression. New cellular technology, including 5G cellular systems, is being deployed and making use of higher frequencies, including the Millimetre Wave (MMW) range (30-300 GHz) of the electromagnetic spectrum. Numerical computational techniques such as the Finite Difference Time Domain (FDTD) method have been used extensively as an effective approach for assessing electromagnetic fields’ biological impacts. This study demonstrates the variation of the accuracy of the FDTD computational simulation system when different meshing sizes are used, by using the interaction of the critically sensitive human cornea with EM in the 30 to 100 GHz range. Different approaches of base cell size specifications were compared. The accuracy of the computation is determined by applying planar sensors showing the detail of electric field distribution as well as the absolute values of electric field collected by point sensors. It was found that manually defining the base cell sizes reduces the model size as well as the computation time. However, the accuracy of the computation decreases in an unpredictable way. The results indicated that using a cloud computing capacity plays a crucial role in minimizing the computation time.

No abstract