Optimized multilayer structure for sensitive THz characterization of thin-film glucose solutions

Terahertz time-domain spectroscopy (THz-TDS) is a highly advanced characterization technique with potential application in a wide range of fields. In particular, it has been widely used in biomedical research owing to its non-ionizing photon energies and high-water sensitivity. In most cases, biomedical samples, such as paraffin-embedded tissues and biological solutions, are often small in volume and thickness, resulting in high demand for high characterization sensitivity. The sample is usually considered a thin-film (TF) sample for THz frequencies due to the long THz wavelength.

Generally, the sample thickness in TF samples is smaller than or, in some cases, comparable to the wavelength, resulting in short interaction length between the electromagnetic wave and the sample. During sample measurements, however, the overlap between time-domain pulses is often associated with a reduction in the characterization sensitivity. This challenge can be addressed by either improving the signal contrast or reducing the system noise. Although a number of strategies have been proposed to measure different types of TF samples, most of them suffer from different limitations, including low characterization sensitivity, that hinders their widespread applications in different circumstances.

Using THz-TDS to study glucose solutions in the human body has drawn significant research attention. With the increasing number of diabetes cases, this could be important for monitoring blood glucose levels and concentration. However, this approach suffers from various limitations, including requiring a large amount of liquid sample and the inability to accurately detect any changes in the concentration of the TF solutions. Additionally, the relatively low characterization sensitivity of THz-TDS when used for TF glucose solutions is another limitation that has been the focus of recent research in this direction.

To overcome these problems, a team of researchers from the University of Warwick: PhD candidate Xuefei Ding, Dr. Arturo Hernandez-Serrano, Dr. Hannah Lindley-Hatcher, Dr. Rayko Stantchev, Dr. Jun Zhou and led by Professor Emma Pickwell-Macpherson developed an optimized multilayer structure to enhance the sensitivity of THz characterization of TF aqueous glucose solutions. This characterization was based on reflection measurement geometry. In their approach, the TF liquid sample was embedded between the imaging window and the top substrate with fixed slot thickness. Theoretical simulations were carried out for structural optimization to determine the most sensitive multilayer structure. Their work is currently published in the journal, Optics Express.

The researchers measured the TF glucose solutions with various concentrations and demonstrated the proposed structure’s ability to improve the signal contrast between various glucose concentrations than using ordinary transmission geometry. 0 – 20% TF glucose solutions were measured, and a notable spectral peak induced by the structure was observed in both the frequency- and time-domain signals, resulting in high sensitivity. Compared with traditional THz measurement geometries and other sandwich structures, the proposed 75 µm window-sample-mirror structures achieved the best characterization sensitivity. The refractive index and absorption coefficient exhibited linear correlation with the glucose concentration.

In summary, University of Warwick scientists successfully developed an improved sensitive THz characterization of TF glucose solutions based on the optimized sandwich structure. Theoretical simulations agreed well with the experimental results. The results indicated that the high sensitivity emanated from the overlapping effects of multiple reflected THz signals. This study provided a new strategy for customizing multilayer structures for THz TF characterization, even for sensitive measurements of certain TF samples. In a statement to Advances in Engineering, the authors explained that their findings will advance future research on THz sensing of blood glucose for effective diabetes diagnosis and treatment.