Open Access
Add to BibSonomy
Abstract
Quantitative detection of neurotransmitters in aqueous environment is crucial for the early diagnosis of many neurological disorders. Terahertz waves, as a non-contact and non-labeling tool, have demonstrated large potentials in quantitative biosensing. Although the detection of trace-amount analyte has been achieved with terahertz metamaterials in the recent decades, most studies have been focused on dried samples. Here, a hexagonal asymmetric metamaterial sensor was designed and fabricated for aqueous solution sensing with terahertz waves in the reflection geometry. An absorption enhancement of 43 was determined from the simulation. Dilute adrenaline solutions ranging from 30 µM to 0.6 mM were measured on our sensor using a commercial terahertz time-domain spectroscopy system, and the effective absorption was found to be linearly correlated with the concentration (R2 = 0.81). Furthermore, we found that as the concentration becomes higher (>0.6 mM), a non-linear relationship starts to take place, which confirmed the previous theory on the extended solvation shell that can be probed on the picosecond scale. Our sensor, without the need of high-power and stable terahertz sources, has enabled the detection of subtle absorption changes induced by the solvation dynamics.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Neurotransmitters are a group of chemical messengers released between neurons or from neurons to muscles. They play a significant role in brain functioning, and serve as candidate biomarkers for the early detection, prognosis, and real-time follow-up of neurological diseases and disorders [1–5]. Therefore, the quantitative detection of neurotransmitters in biofluids is vital for the early diagnosis of certain diseases in order to prevent their further development. However, the neurotransmitters are usually released and cleared in the extracellular space very rapidly at extremely low concentrations, making their in vivo monitoring very challenging. As an essential member of the neurotransmitters, catecholamine, including adrenaline (AD), noradrenaline (NAD) and dopamine, is produced by the adrenal glands and has close connections with stress regulation, mental activity, mood, learning, sleep, and memory [6]. However, due to the special structure of adrenal glands (1–15 mm length) [7], it is quite difficult to make a point-of-care monitoring with the well-developed methods, such as microdialysis combined with high-performance liquid chromatography (HPLC), fluorescence spectroscopy, capillary electrophoresis, and fast-scan cyclic voltammetry (FSCV) [8–12]. Besides the tedious sample preparation process, these methods also face technical limitations for clinical uses. For instance, HPLC requires a few minutes to collect sufficient sample volume and the currents of neurotransmitters decay rapidly by hundreds of milliseconds due to diffusion, re-uptake, binding to receptors, or enzymatic breakdown. These bio-processes will increase the difficulty of sample collection and pretreatment. In consequence, real-time and ultrasensitive methods for the quantitative identification of neurotransmitters are in urgent demand.
Terahertz radiation generally refers to the electromagnetic waves with frequencies between 0.1∼10 THz (wavelength of 3000∼30 µm). It is located between the microwave and infrared light, which is in the transition zone from electronics to photonics. The collective vibrational and rotational modes of many biomolecules together with the weak intermolecular interactions are situated within the terahertz spectrum. This enables the changes of binding states and vibrational modes between biomolecules to be probed by terahertz waves [13,14].
Terahertz spectroscopy has a great potential for the early diagnosis of tumors as well as real-time monitoring of disease-related indicators [15,16]. These trace-amount indicators together with other components such as mineral salt, glucose, amino acid, urea, polypeptide, and water (most absorbing), are always dispersed in our body fluids, which makes the quantification of the target analytes very challenging in the terahertz regime [17]. With the advancement of nano-fabrication techniques, metamaterials have been developed as biosensors which could increase the interaction strength between the analytes and the incident terahertz waves significantly, hence allow for a much higher sensitivity compared with the traditional terahertz time-domain spectroscopy (THz-TDS) technique [18,19]. However, up to now, a heavy body of work has been focused on the identification of target biomolecules in the solid state, dried, or embedded in low-absorption medium, which is not in accordance with the actual situation and there is no practical value in medical treatment [20–24]. Aqueous-based biosensing has been demonstrated using microfluidic cells [25,26], which confine aqueous solutions in microscale level [27,28]. These structures have not only reduced the water absorption of terahertz waves, but also improved the detection efficiency of the biosensors [17]. The combination of microfluidic cells with metamaterials have also been explored recently [17,29], which have demonstrated a great progress in the detection accuracy and sensitivity. Nevertheless, the prone to bubble formation in the microcavity has posed issues in the practical in-situ analysis.
Here, we designed a hexagonal asymmetric split plasmonic ring biosensor and demonstrated its potential ability to measure trace-amount (0.03 to 0.6 mM) AD directly in aqueous solutions using a THz-TDS system in reflection geometry. Through a parametric study, the sensitivity to the refractive index of our sensor is determined to be 60 GHz/RIU, and the absorption enhancement factor is 43, which suggests that our structure is very sensitive to the subtle changes in the absorption coefficient of the analyte. Furthermore, by measuring a wider concentration range (up to 15.1 mM), we observed the linear to nonlinear solvation effect which has only been detected with a stable p-Germanium laser or a high-power synchrotron source before.
2. Experimental section
2.1 Sample preparation and THz-TDS measurements
AD (bitartrate salt, ≥98%) was purchased from Macklin (Shanghai, China). Deionized water was obtained from the flow water purification system (18 MΩ·cm−1, Qingdao, China). Stock standard solutions of AD (50000 µg/mL) was prepared by dissolving accurately weighed reagent in deionized water. Working solutions were prepared immediately before use by taking appropriate aliquots of solutions and diluting them with deionized water to the desired concentrations.
Terahertz spectra were collected in the 30° reflection mode utilizing a commercial THz-TDS system from Menlo Systems GmbH (Germany) with a frequency range from 0.1 to 3.0 THz [30], a schematic of which is shown in Fig. 1 (a). The incident terahertz wave is s-polarized with its focal plane aligned to the upper surface of the device. Air, pure water, as well as 200 µL of different concentrations of AD solutions were measured on our metamaterial device. A silicon wafer without the metamaterial structure was used as a reference to obtain the effective refractive index and absorption coefficient of the samples on the surface of the metamaterial. The details of the parameter extraction procedure can be found in the supplementary file. An air-compressed system was used to remove the water vapor, and the humidity was purged down to less than 1% in the measurement chamber.
Fig. 1. (a) A schematic of the THz-TDS setup in the 30°reflection geometry. (b) The designed pattern of our metamaterial with dimensions. (c) The microscope image of the fabricated metamaterial with a scale bar. (d) The calculated effective absorption coefficient and refractive index of the blank metamateral.
2.2 Design and fabrication of terahertz metamaterials
The terahertz metamaterial was fabricated with a standard photolithography technique. A metal layer of Cr/Au with a thickness of 20 nm/200 nm was fabricated on a 500 ± 10 µm high-resistivity silicon wafer (>6000 Ω·cm) polished on both sides (SiO2, 300 ± 2 nm). As illustrated in Fig. 1 (b), each unit cell of the arrays consists of two concentric hexagonal rings with complementary openings (outer edge length L1= 49.85 µm, inner edge length L2= 43.45 µm, distance between the inner and outer rings d1= 3.95 µm, distance between the two rings d2 = 8 µm, line width w = 1.6 µm, and the opening width d3 = 3.2 µm). Figure 1 (c) shows the microscope image of our fabricated metamaterial.
2.3 Simulations
The numerical simulations were performed with a finite-element method (FEM) supplied by COMSOL Mutliphysics. In the simulation, two perfectly matched layers (PML) were selected for the bottom and top layers of the established model. The high-resistivity silicon wafer and air were set as lossless dielectric materials with a refractive index of 3.42 and 1, respectively. The metamaterial layer was set as a transition boundary with a relative permittivity calculated with the Drude model as shown in Eq. (1):
In the model [31], ɛr is the relative permittivity of gold in the terahertz regime with a plasma frequency of ${\omega _p}$=1.37×1016 rad/s. ${\varepsilon _\infty }$ is the relative permittivity at high frequencies, which was set to 1. Γ represents the collision frequency and was set to 3.95×1013 rad/s. The incident terahertz wave was s-polarized which is consistent with the experiment. The incident angle at the silicon/metamaterial interface was set to 8.4°, which was calculated by Snell’s law with a 30° incident angle in the air. Another structure without the metamaterial layer was also simulated with the same settings and was used as a reference. The amplitude and phase of S11 of both models were extracted to calculate the effective absorption coefficient (${\alpha _{eff}}$) and the refractive index (${n_{eff}}$) of the metamaterial/air interface. As can be seen in Fig. 1 (d), the ${n_{eff}}$ shows two abnormally high resonant peaks at 535 GHz and 755 GHz, respectively, while ${\alpha _{eff}}$ went through a sign flip at the same frequencies which is caused by the abrupt change of phase at the resonant frequencies. At the first resonant frequency, the effective refractive index reached 46 as plotted in Fig. 1 (d).
2.4 Bruggeman's effective-medium theory
In order to simulate the response of the metamaterial loaded with sample solutions, the complex dielectric constants of the sample solutions were obtained using the Bruggeman's effective-medium theory (EMT) [32,33], in which the various components of a mixture are treated as spheres of radius R in a background medium. When the radius of the sphere is much smaller than the incident wavelength, the complex dielectric constant of the mixture can be approximated with Eq. (2):
3. Results and discussion
3.1 Sensitivity of the metamaterial
To investigate the sensitivity of the metamaterial, we performed a parametric study with COMSOL Multiphysics. The refractive index (n) and the extinction coefficient ($\kappa $) of the medium on the surface of the metamaterial were varied separately, and the effective refractive index and absorption coefficient of the medium were calculated with the simulated phase and amplitude of S11. From the results shown in Fig. 2, the resonant frequency (${f_R}$) is solely affected by the refractive index: a linear red-shift was observed when n was increased from 1 to 3. This can be explained by the equivalent LC circuit model where the openings in the hexagonal ring serve as capacitances in the circuit and the resonant frequency is proportional to $1/\sqrt {LC} \; $. When the refractive index of the medium in the openings is increased, the equivalent capacitance of the structure becomes larger, causing a red-shift of the resonant frequencies. Meanwhile, the absolute phase change at the metamaterial interface is also increasing with the refractive index, shifting the zero-crossing point of the absorption coefficient to lower frequencies as illustrated in Fig. S3 in the supplementary document. A refractive index sensitivity of 60 GHz/RIU was then determined from the slope of the linear fitting in Fig. 2 (c). The peak-to-peak value of the effective absorption coefficient, as labeled in Fig. 2 (d), showed a negative correlation with the extinction coefficient of the loaded sample. Without the metamaterial layer, the phase shift at the substrate/sample interface is induced by $\kappa $ of the sample and the absolute value of the phase shift is well below π. The metamaterial layer induces a huge frequency-dependent phase shift, which causes the phase difference between the sample and reference to cross $- \pi $, as shown in Fig. S2 (d) in the supplementary document. This phase shift results in a local maximum and minimum of the effective absorption coefficient as plotted in Fig. 2 (d). The extracted absorption coefficient is therefore increased compared to that measured without the metamaterial. In the meantime, the sensitivity to the change in the absolute value of the reflection coefficient is amplified. A more detailed analysis can be found in the supplementary document. The absorption coefficient of the medium can be calculated with ${\alpha _{fR}} = \frac{{4\pi {f_R}}}{c} \cdot \kappa $. Therefore, by plotting ${\alpha _{pp}}$ against ${\alpha _{fR}}$, as shown in Fig. 2 (f), a strong absorption enhancement of 43 was calculated.
Fig. 2. (a) and (b) The effective absorption coefficient and refractive index when ${\mathbf \kappa } = \mathbf 0$ and n was changed from 1 to 3; (c) The resonant frequency shift plotted against n; (d) and (e) The effective absorption coefficient and refractive index when $\mathbf {n} = \mathbf 2$ and ${\mathbf \kappa }$ was changed from 0.1 to 0.5. (f) The peak-to-peak value of ${{\mathbf \alpha }_{\mathbf {pp}}}$ plotted against the absorption coefficient at the resonant frequency.
3.2 Sensor response to AD solutions approximated with a binary EMT model
To theoretically characterize the spectral response of the sensors loaded with different concentrations of AD solutions, we calculated the complex dielectric constants $\tilde{\varepsilon }\; $ of the solutions with volume ratios ranging from 0.1% to 50% using the EMT model as described in the Methods section. By inputting these parameters into the simulation, the phase and amplitude of S11 were extracted and used to calculate ${n_{eff}}\; $ and ${\alpha _{eff}}\; $ for various sample loadings. When loaded with AD solutions, the resonant frequency of the sensor shifted to lower frequencies for all of the concentrations in the simulation compared with that of the blank device as can be seen in Fig. 3 (a) and (b). With the increasing of volume percentage, the resonant frequency ${f_R}$ underwent a linear blue-shift, and the peak-to-peak value of the effective absorption coefficient also increased linearly with the volume ratio as expected.
Fig. 3. (a) and (b) The effective absorption coefficient and refractive index simulated when the sensor is loaded with different volume ratios of AD. (b) The absorption coefficient with different volume ratios of adrenaline; (c) The correlation of ${{\mathbf \alpha }_{{\mathbf pp}}}$ and the volume ratio; (d) The resonant frequency shift predicted for different volume ratios of AD solutions.
3.3 Experimental detection of adrenaline in dilute aqueous solutions
The fabricated metamaterial without any sample solution was measured first, and the results are plotted in the inset of Fig. 4 (a) and (b). As predicted by the simulation, a double-resonant feature was observed with an abnormally high effective refractive index of 24.2, and two zero-crossing points of the absorption coefficient were observed at 585 GHz and 775 GHz, respectively. Compared to the simulation, the experimental result is blue-shifted for 50 GHz. This discrepancy is most likely to be caused by the fabrication process. A 5% (2.5 µm) error in the length of outer edge (L1) will induce a frequency shift of ∼100 GHz. Sequentially, 200 µL of AD solutions with concentrations ranging from 0.03 to 0.6 mM were dropped onto our sensor with a pipette. The calculated refractive index and absorption coefficient of the aqueous solutions are shown in Fig. 4 (a) and (b), respectively. Both ${f_R}$ and ${\alpha _{pp}}$ showed a linear relationship (R2≈0.81) with the concentration as has been predicted by the simulation. These results indicate that our metamaterial sensor is able to detect trace amount of AD directly from aqueous solutions due to enhanced light-matter interactions.
Fig. 4. (a) and (b) The effective absorption coefficient and refractive index for different concentrations of AD solutions extracted from the experiment; (c) and (d) The measured correlation diagrams of ${\alpha _{pp}}$ and $\mathrm{\Delta }{f_R}$ with concentration.
3.4 Solvation of AD molecules in aqueous environment
Interestingly, from Fig. 4, we found that the resonant frequency showed a red-shift with increasing concentration, which is opposite to the simulation results presented above. Meanwhile, ${\alpha _{pp}}$ also decreased with increasing concentration as opposed to the simulation. The experimental results suggest that, at the measured concentrations, the refractive index as well as the absorption coefficient has increased with the concentration. The deviation from the binary EMT model we used could be due to the solvation effect of AD molecules in aqueous environment [34,35]. The breaking and rearrangement of hydrogen bonds due to the presence of solutes has a unneglectable influence in the terahertz frequencies, as has been reported by Havenith’s group [34–36]. The perturbed water molecules form hydration shells around the solute which has an effectively higher absorption coefficient compared to that of bulk water. For very dilute solutions, the volume of solvation water increases linearly with the concentration so that the macroscopic absorption of the solution increases, as has been demonstrated for some proteins and carbohydrates [37–39]. The observation of the so called “THz excess” phenomenon [40] needs precise measurement setups which is difficult for conventional THz-TDS systems, and has only been realized with either a p-Germanium laser or a coherent synchrotron radiation source [41]. Here, by utilizing an absorption-enhanced metamaterial, we are able to detect the subtle changes induced by the solvation effect in a simple reflection setup without any intense terahertz sources.
To further verify our observation, we extended the concentration range of our sample solution in the hope to see the non-linear effect caused by the overlapping of hydration shells. AD solutions with concentrations up to 15.1 mM were measured on our metamaterial as well as on a piece of crystalline quartz window. As can be seen from Fig. 5, at very dilute concentrations (<0.6 mM), the variation of ${\alpha _{pp}}$ measured on our metamaterial linearly increased with the concentration, and a maximum absolute variation of 11% was determined compared to that of water. As the concentration becomes moderate (0.6 mM to 15mM), the variation of ${\alpha _{pp}}\; $ drops steeply at first (0.6 mM to 2 mM) and then declines at a slower rate. This nonlinearity effect starts to take place when the AD molecules are brought closer together as the concentration increases, and the dynamic hydration shells begin to overlap. When the distance between the solute molecules further decreases, the increasing of the solute (with neglectable absorption) will become the main factor to the mixture, causing the overall absorption to drop. The absorption coefficient measured directly on a quartz window is also shown in Fig. 5 for comparison. As can be seen, the solvation effect could not be clearly observed for measurements performed on the quartz window, as we have expected.
Fig. 5. The signal variation plotted against concentration. The red dots are the absolute variations of ${\alpha _{pp}}$, calculated with $\frac{{|{{\alpha_{pp}} - {\alpha_{pp0}}} |}}{{{\alpha _{pp0}}}} \cdot 100\%$, measured on our metamaterial, while the black squares illustrate the variation of absorption coefficient measured directly on a piece of quartz window. The dash line indicates the point where the hydration shell starts to overlap and the linear relationship begins to transform into a nonlinear one.
4. Conclusion
To summarize, we have designed and fabricated a hexagonal asymmetric split plasmonic ring biosensor. The sensor demonstrated a strong absorption enhancement factor of 43. With such enhancement, trace-amount AD sensing directly in aqueous solutions have been realized with THz-TDS in the reflection setup. The minimum concentration identified with our sensor is 30 µm. Furthermore, our sensor, without any high-power terahertz sources, have enabled the detection of solvation effect for AD molecules. Our measurement confirmed the previously reported “THz excess” and “Hydration shell overlapping” theory. Within a certain concentration range, the absorption of the solutions could increase linearly first and then drops in a non-linear manner due to the solvation overlapping effect, which would influence the linearity of terahertz biosensors. Therefore, our findings also have instructive meanings to the future development of terahertz sensors for dilute aqueous solutions.
Funding
National Natural Science Foundation of China (61805150, 61975135); Natural Science Foundation of Guangdong Province (2019A1515010869, 2021A1515012296); Key Technologies Research and Development Program (2019YFB2204500); Shenzhen Science and Technology Innovation Program (KQTD20180412181422399); Science, Technology and Innovation Commission of Shenzhen Municipality (JCYJ20180507181858539); Shenzhen International Cooperation Research Project (GJHZ20190822095407131).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper can be obtained upon request from Shuting Fan via email.
Supplemental document
See Supplement 1 for supporting content.
References
1. W. Lee, B. H. Kang, H. Yang, M. Park, J. H. Kwak, T. Chung, Y. Jeong, B. K. Kim, and K. H. Jeong, “Spread spectrum SERS allows label-free detection of attomolar neurotransmitters,” Nat. Commun. 12(1), 159 (2021). [CrossRef]
2. M. M. Karim, S. M. Alam, and S. H. Lee, “Spectrofluorimetric estimation of norepinephrine using ethylenediamine condensation method,” J. Fluoresc. 17(4), 427–436 (2007). [CrossRef]
3. K. E. Secor and T. E. Glass, “Selective Amine Recognition: Development of a Chemosensor for Dopamine and Norepinephrine,” Org. Lett. 6(21), 3727–3730 (2004). [CrossRef]
4. M. J. Whiting and M. P. Doogue, “Advances in biochemical screening for phaeochromocytoma using biogenic amines,” Clin Biochem Rev 30, 3–17 (2009).
5. A. M. Wehry, L. Ramsey, S. E. Dulemba, S. A. Mossman, and J. R. Strawn, “Pharmacogenomic testing in child and adolescent psychiatry: an evidence-based review,” Current Problems in Pediatric and Adolescent Health Care 48(2), 40–49 (2018). [CrossRef]
6. D. S. Goldstein and I. J. Kopin, “Linking Stress, Catecholamine Autotoxicity, and Allostatic Load with Neurodegenerative Diseases: A Focused Review in Memory of Richard Kvetnansky,” Cell. Mol. Neurobiol. 38(1), 13–24 (2018). [CrossRef]
7. D. Molinnus, G. Hardt, L. Käver, H. S. Willenberg, J.-C. Kröger, A. Poghossian, M. Keusgen, and M. J. Schöning, “Chip-based biosensor for the detection of low adrenaline concentrations to support adrenal venous sampling,” Sensors and Actuators B: Chemical 272, 21–27 (2018). [CrossRef]
8. A. Lund, “Simultaneous fluorimetric determinations of adrenaline and noradrenaline in blood,” Acta Pharmacol Toxicol (Copenh 6(2), 137–146 (2009). [CrossRef]
9. E. Nalewajko, A. Wiszowata, and A. Kojło, “Determination of catecholamines by flow-injection analysis and high-performance liquid chromatography with chemiluminescence detection,” J. Pharm. Biomed. Anal. 43(5), 1673–1681 (2007). [CrossRef]
10. S. G. a. Palop, A. Mellado Romero, and J. Mart’nez Calatayud, “Oxidation of adrenaline and noradrenaline by solved molecular oxygen in a FIA assembly,” J. Pharm. Biomed. Anal. 27(6), 1017–1025 (2002). [CrossRef]
11. S. Wei, G. Song, and J.-M. Lin, “Separation and determination of norepinephrine, epinephrine and isoprinaline enantiomers by capillary electrophoresis in pharmaceutical formulation and human serum,” J. Chromatogr. A 1098(1-2), 166–171 (2005). [CrossRef]
12. B. Si and E. Song, “Recent Advances in the Detection of Neurotransmitters,” Chemosensors 6(1), 1 (2018). [CrossRef]
13. H. Cheon, H. J. Yang, S. H. Lee, Y. A. Kim, and J. H. Son, “Terahertz molecular resonance of cancer DNA,” Sci. Rep. 6(1), 37103 (2016). [CrossRef]
14. M. Walther, P. Plochocka, B. Fischer, H. Helm, and P. Uhd Jepsen, “Collective vibrational modes in biological molecules investigated by terahertz time-domain spectroscopy,” Biopolymers 67(4-5), 310–313 (2002). [CrossRef]
15. V. P. Wallace, B. C. Cole, R. M. Woodward, R. J. Pye, and D. A. Arnone, “Biomedical applications of terahertz technology,” in The 15th Annual Meeting of the IEEE Lasers and Electro-Optics Society, (2002), pp. 308–309.
16. B. Ng, J. Wu, S. M. Hanham, A. I. Fernández-Domínguez, N. Klein, Y. F. Liew, M. B. H. Breese, M. Hong, and S. A. Maier, “Spoof Plasmon Surfaces: A Novel Platform for THz Sensing,” Adv. Opt. Mater. 1(8), 543–548 (2013). [CrossRef]
17. K. Shih, P. Pitchappa, L. Jin, C.-H. Chen, R. Singh, and C. Lee, “Nanofluidic terahertz metasensor for sensing in aqueous environment,” Appl. Phys. Lett. 113(7), 071105 (2018). [CrossRef]
18. K. V. Sreekanth, Y. Alapan, M. ElKabbash, E. Ilker, M. Hinczewski, U. A. Gurkan, A. De Luca, and G. Strangi, “Extreme sensitivity biosensing platform based on hyperbolic metamaterials,” Nat. Mater. 15(6), 621–627 (2016). [CrossRef]
19. W. Xu, L. Xie, and Y. Ying, “Mechanisms and applications of terahertz metamaterial sensing: a review,” Nanoscale 9(37), 13864–13878 (2017). [CrossRef]
20. K. Yang, J. Li, M. L. de la Chapelle, G. Huang, Y. Wang, J. Zhang, D. Xu, J. Yao, X. Yang, and W. Fu, “A terahertz metamaterial biosensor for sensitive detection of microRNAs based on gold-nanoparticles and strand displacement amplification,” Biosens. Bioelectron. 175, 112874 (2021). [CrossRef]
21. K. Liu, R. Zhang, Y. Liu, X. Chen, K. Li, and E. Pickwell-Macpherson, “Gold nanoparticle enhanced detection of EGFR with a terahertz metamaterial biosensor,” Biomed. Opt. Express 12(3), 1559–1567 (2021). [CrossRef]
22. W. Xu, L. Xie, J. Zhu, L. Tang, R. Singh, C. Wang, Y. Ma, H.-T. Chen, and Y. Ying, “Terahertz biosensing with a graphene-metamaterial heterostructure platform,” Carbon 141, 247–252 (2019). [CrossRef]
23. M. R. Konnikova, O. P. Cherkasova, M. M. Nazarov, D. A. Vrazhnov, Y. V. Kistenev, S. E. Titov, E. V. Kopeikina, S. P. Shevchenko, and A. P. Shkurinov, “Malignant and benign thyroid nodule differentiation through the analysis of blood plasma with terahertz spectroscopy,” Biomed. Opt. Express 12(2), 1020–1035 (2021). [CrossRef]
24. M. Hu, M. Tang, H. Wang, M. Zhang, S. Zhu, Z. Yang, S. Zhou, H. Zhang, J. Hu, Y. Guo, X. Wei, and Y. Liao, “Terahertz, infrared and Raman absorption spectra of tyrosine enantiomers and racemic compound,” Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 254, 119611 (2021). [CrossRef]
25. K. Shirai, K. Mawatari, and T. Kitamori, “Extended nanofluidic immunochemical reaction with femtoliter sample volumes,” Small 10(8), 1514–1522 (2014). [CrossRef]
26. Z. Geng, X. Zhang, Z. Fan, X. Lv, and H. Chen, “A Route to Terahertz Metamaterial Biosensor Integrated with Microfluidics for Liver Cancer Biomarker Testing in Early Stage,” Sci. Rep. 7(1), 16378 (2017). [CrossRef]
27. A. J. Baragwanath, G. P. Swift, D. Dai, A. J. Gallant, and J. M. Chamberlain, “Silicon based microfluidic cell for terahertz frequencies,” J. Appl. Phys. 108(1), 013102 (2010). [CrossRef]
28. B. Ng, S. M. Hanham, V. Giannini, Z. C. Chen, M. Tang, Y. F. Liew, N. Klein, M. H. Hong, and S. A. Maier, “Lattice resonances in antenna arrays for liquid sensing in the terahertz regime,” Opt. Express 19(15), 14653–14661 (2011). [CrossRef]
29. R. Zhou, C. Wang, Y. Huang, K. Huang, Y. Wang, W. Xu, L. Xie, and Y. Ying, “Label-free terahertz microfluidic biosensor for sensitive DNA detection using graphene-metasurface hybrid structures,” Biosens. Bioelectron. 188, 113336 (2021). [CrossRef]
30. Y. Sun, J. Huang, L. Shan, S. Fan, Z. Zhu, and X. Liu, “Quantitative analysis of bisphenol analogue mixtures by terahertz spectroscopy using machine learning method,” Food Chem. 352, 129313 (2021). [CrossRef]
31. Y. Zhang, T. Li, B. Zeng, H. Zhang, H. Lv, X. Huang, W. Zhang, and A. K. Azad, “A graphene based tunable terahertz sensor with double Fano resonances,” Nanoscale 7(29), 12682–12688 (2015). [CrossRef]
32. D. A. G. Bruggeman, “Berechnung verschiedener physikalischer Konstanten von heterogenen Substanzen. I. Dielektrizitätskonstanten und Leitfähigkeiten der Mischkörper aus isotropen Substanzen,” Ann. Phys. 416(8), 636–664 (1935). [CrossRef]
33. D. B. Bennett, W. Li, Z. D. Taylor, W. S. Grundfest, and E. R. Brown, “Stratified Media Model for Terahertz Reflectometry of the Skin,” IEEE Sens. J. 11(5), 1253–1262 (2011). [CrossRef]
34. M. Zhou, K. Cheng, and G.-z. Jia, “Molecular dynamics simulation studies of dopamine aqueous solution,” J. Mol. Liq. 230, 137–142 (2017). [CrossRef]
35. H. Wang, Z. Huang, T. Shen, and L. Guo, “Hydrogen-bonding interactions in adrenaline–water complexes: DFT and QTAIM studies of structures, properties, and topologies,” J Mol Model 18(7), 3113–3123 (2012). [CrossRef]
36. U. Heugen, G. Schwaab, E. Bründermann, M. Heyden, X. Yu, D. M. Leitner, and M. Havenith, “Solute-induced retardation of water dynamics probed directly by terahertz spectroscopy,” Proc. Natl. Acad. Sci. 103(33), 12301–12306 (2006). [CrossRef]
37. S. Ebbinghaus, S. J. Kim, M. Heyden, X. Yu, U. Heugen, M. Gruebele, D. M. Leitner, and M. Havenith, “An extended dynamical hydration shell around proteins,” Proc. Natl. Acad. Sci. 104(52), 20749–20752 (2007). [CrossRef]
38. M. Heyden, E. Bründermann, U. Heugen, G. Niehues, D. Leitner, and M. J. J. o. t. A. C. S. Havenith, “Long-Range Influence of Carbohydrates on the Solvation Dynamics of Water– Answers from Terahertz Absorption Measurements and Molecular Modeling Simulations,” J. Am. Chem. Soc. 130(17), 5773–5779 (2008). [CrossRef]
39. J. Xu, K. W. Plaxco, and S. J. Allen, “Probing the collective vibrational dynamics of a protein in liquid water by terahertz absorption spectroscopy,” Protein Sci. 15(5), 1175–1181 (2006). [CrossRef]
40. D. A. Schmidt, Ö Birer, S. Funkner, B. P. Born, R. Gnanasekaran, G. W. Schwaab, D. M. Leitner, and M. Havenith, “Rattling in the Cage: Ions as Probes of Sub-picosecond Water Network Dynamics,” J. Am. Chem. Soc. 131(51), 18512–18517 (2009). [CrossRef]
41. J. W. Bye, S. Meliga, D. Ferachou, G. Cinque, J. A. Zeitler, and R. J. Falconer, “Analysis of the Hydration Water around Bovine Serum Albumin Using Terahertz Coherent Synchrotron Radiation,” J. Phys. Chem. A 118(1), 83–88 (2014). [CrossRef]
