Bio-chemical Process Monitoring with Terahertz Sensor

  • Y. Zhang
  • V. Matvejev
  • S. Declerck
  • S. Stroobants
  • G. Pandey
  • G. He
  • D. Mangelings
  • D. Maes
  • S. Muyldermans
  • J. Stiens
Conference paper
Part of the NATO Science for Peace and Security Series B: Physics and Biophysics book series (NAPSB)


An in-house developed Label-free Immobilization-free Terahertz (THz) sensor has shown its potential in various applications [1]. In this paper, we present the experimental results in the field of High Performance Liquid Chromatography (HPLC) and protein crystallization. Ultra-Violet (UV) detectors are very commonly used as a detection technique in HPLC systems. However, it relies on labelling when target substances do not absorb UV light. Therefore, it has the drawback of increased cost and time consumption. Our experimental result shows that THz sensor has the potential to replace UV sensor integrated on the HPLC machine as it is able to detect both UV and non-UV absorbing substances without labelling. Meanwhile, the THz sensor is also deployed for the first time to follow a protein crystallization process.


Millimeter wave HPLC Protein crystallization THz Dielectric permittivity 

4.1 Introduction

In biological solutions, the presence of bio-molecules alters the hydrogen bond network dynamics of water molecules in its vicinity. These molecules, also known as hydration shell, exhibit different dielectric permittivity than normal bulk water molecules in the THz frequency range (30 GHz–1 THz) [2]. The amount of hydration shell water molecules in liquid is closely related to the amount of bio-molecules, their shape, conformation flexibility and the distribution of hydrophobic and hydrophilic surface. In other words, the change in concentration or conformation of these bio-molecules will translate to dielectric permittivity change in the THz frequency range. In previous work [3], a THz sensor operating @ 270 GHz has been made to probe permittivity changes in aqueous solutions. It is a label-free, immobilization-free real-time sensor that features very high sensitivity. In this study, another sensor topology operating @ 60 GHz is developed for the same purpose. High performance liquid chromatography (HPLC) and protein crystallization measurements are chosen as proof-of-concept studies to validate the sensor design.

4.2 THz Sensor & Setup

The THz sensor developed in house consists of a metal waveguide section, and a capillary tube as liquid fixture (shown in Fig. 4.1a). THz wave is launched from one side of the waveguide. The interaction between incident wave and liquid results in reflection and transmission. If the impedances of the sensor and the rest of the THz set-up are properly matched, very low reflection (S11 Parameter) can be achieved at a certain frequency. When the dielectric permittivity of the liquid changes as a result of bio-molecular interactions, the wave impedance of the sensor will also deviate from its original value. The mismatch of impedance between the THz sensor and the other parts creates a larger reflection signal. In real measurements, a reference liquid is first injected into the waveguide via capillary tube. Then impedance tuning is applied to achieve desired S11 signal level (blue curve in Fig. 4.1b). When bio-molecular interactions occur (increased concentration of certain substance, protein interaction etc.), the change of the liquid’s dielectric permittivity leads to an increased S11 Parameter (red curve in Fig. 4.1b). This difference will be registered for further analysis.
Fig. 4.1

(a) THz sensor internal structure (b) different responses of reference liquid and sample liquid

The measurement set-up consists of (Fig. 4.2a):
  • Vector network analyzer that produces and detects 60GHz signal

  • Directional coupler that separates reflected wave from the forward propagating wave

  • THz sensor that is very sensitive to dielectric change of liquid under test.

Fig. 4.2

(a) THz sensor measurement set-up. (b) Bench marking performance comparison of different THz sensors. A1 used in this paper. A2, B1, B2 are designs made in the past. The other yellow dots are theoretical designs. Blue diamonds are sensors presented in literature (J[5], C[6], M[7], L[8], D[9], I[10], H[11], E[12], F[13], G[14])

Such measurement set-up has shown its capability in the previous work [3]. When compared to other THz sensor designs by bench marking water-ethanol contrast measurements (Fig. 4.2b), it shows superior sensitivity. Note that the THz sensor used in this paper is denoted as A1 in the graph which is 30 dB more sensitive than the closest K [4].

4.2.1 High Performance Liquid Chromatography System Setup & Measurements

HPLC system is often used to separate individual ingredients in a mixture, and measure the concentration separately. Such system typically consists of (Fig. 4.3):
  • HPLC pump(s) to keep mobile phase flowing through the entire system

  • An injector to inject samples onto the mobile phase

  • A stationary phase to separate individual substances from the sample injected

  • Detector to register chromatograph in order to measure the concentration of each component in the sample. There exists various techniques [15]. In our case, UV detector is integrated on the available machine.

Fig. 4.3

High performance liquid chromatography system (Red, green and blue colors depict different substances in a mixture)

To avoid extra complexity, the stationary phase is left out since the target of this study is detection of substances instead of separation. As a result, liquid with only one target substance can be injected each time. The THz sensor coupled HPLC system is shown in Fig. 4.4.
Fig. 4.4

THz sensor coupled HPLC system

Three substances as candidates have been chosen for this proof-of-concept study. Two UV-absorbing substances are: Trans-Stilbene-Oxide (TSO) and Praziquantel (PRA). One non-UV-absorbing substance is: Sorbitol. Since TSO and PRA only dissolve poorly in water, acetonitrile (ACN) is chosen as mobile phase in such case. Water is chosen to be the mobile phase for the measurement with sorbitol for the same reason.

During the measurements, impedance tuning is applied when only mobile phase (ACN or water) is in the THz sensor. A continuous-wave measurement on this frequency is started at the moment of injection. Since the distance between the THz sensor and the UV detector is fixed, we expect that the time delay between the two sensor responses is also constant. A series of measurements has been conducted and the results are shown in Fig. 4.5.
Fig. 4.5

(a) Blank injection: ACN injection on mobile phase ACN. (b) 10 mg/ml PRA injection on mobile phase ACN. (c) 10 mg/ml TSO injection on mobile phase ACN (d) 40 mg/ml Sorbitol injection on mobile phase water

As shown in Fig. 4.5a, very little signal variation can be observed from both sensors with a blank injection. For THz sensor, since the impedance of the entire system is matched with ACN, the injected ACN does not introduce any permittivity change. The UV detector has a flat line response since ACN is not UV absorbing. As PRA and TSO are injected (Fig. 4.5b, c), both detectors respond to the substances. The time delay also stays constant around 1.3 min. Since sorbitol is non-UV absorbing, the UV detector was not able to detect it (Fig. 4.5d). However, the THz sensor gave a very obvious response.

4.2.2 Protein Crystallization Measurements

Protein crystallization plays a vital role in understanding the structure of proteins [16]. Though it has been extensively studied, the mechanism of this process is still not clear. UV-VIS and Dynamic light scattering are very commonly used to follow this process among various other techniques. In this study, the capability of THz sensor in this field will be tested.

The THz set-up is slightly modified to measure protein crystallization. The liquid to be measured will be extracted from an eppendorf tube into the sensor structure (Fig. 4.6a). The driving force will be provided by a 25ul syringe.
Fig. 4.6

(a) Crystallization THz sensor setup (b) Illustration of reference liquid and measured sample (c) Crystallization measurement results with different lysozyme concentrations

When the concentrations of lysozyme (40, 50 &60 mg/ml) and NaCl (40 mg/ml) are properly chosen in a mixture(0.1 M Sodium Acetate as buffer), crystallization process can be established at an predictable speed. Increasing the concentration of either lysozyme or NaCl will leads to reduction of time needed for such process. In each measurement, the freshly made crystalizing mixture is chosen as reference liquid. The sensor system is tuned so that very low S11 parameter is achieved. This reference is then compared to the same process at different times (Fig. 4.6b).

As shown in Fig. 4.6c, with increasing lysozyme concentration, the time needed to observe a rise in the THz signal shortens. This corresponds to the accelerated crystallization process.

4.3 Conclusion

Although further studies are still required to validate the THz sensor for the two applications demonstrated above, through our proof-of-concept study, it is already clear that this sensor can track protein crystallization processes. The label-free feature also gives it an advantage over the UV detector in the field of High-performance Liquid Chromatography. In the near future, the focus will be shifted to quantitative measurements of HPLC so that the concentration of injected substances can be determined from measurement curves. For protein crystallization, efforts need to be made to increase the time resolution of the measurements.



The authors acknowledge the Vrije Universiteit Brussel (VUB) through the SRP-project M3D2, the FWO-Vlaanderen through FWOAL682 “Building blocks of lab-on-chip system for label-free monitoring of bio-molecular interactions” and the FWOAL611 “Millimeter wave sensor solutions for chromatographic analyzer systems of today and tomorrow”, the, the COST-action MP1204, TERAMIR the NATO support for the Workshop on THz Diagnostics of CBRN effects and Detection of Explosives & CBRN.


  1. 1.
    V. Matvejev, et al. Integrated waveguide structure for highly sensitive THz spectroscopy of nano-liter liquids in capillary tubes, Progress in Electromagnetic Research, 121, 89–101 (2011)Google Scholar
  2. 2.
    Leitner, D.M., et al. Solvation dynamics of biomolecules: modeling and terahertz experiments, HFSP J., 2008, 2, (6), pp. 314–323Google Scholar
  3. 3.
    V. Matvejev, Y. Zhang, J. Stiens, High performance integrated THz sensor for detection of biomolecular processes in solution, IET Microwaves, Antennas & propagation, 8(6), 394–400 (2014)Google Scholar
  4. 4.
    Weina Liu, A Novel Technology for Measurements of Dielectric Properties of Extremely Small Volumes of Liquids, International Journal of Antennas and Propagation, Volume 2016, Article ID 1436798, 5 pagesGoogle Scholar
  5. 5.
    Cui Y., Y. He, P. Wang. “A Quadrature-Based Tunable Radio-Frequency Sensor for the Detection and Analysis of Aqueous Solutions” IEEE MICROWAVE AND WIRELESS COMPONENTS LETTERS, vol. 24, pp. 490–492, 2014.Google Scholar
  6. 6.
    R. Mendis, V. Astley, J. Liu, D.M. Mittleman, Appl. Phys. Lett 95, 171113 (2009)Google Scholar
  7. 7.
    S. M. Hanham et al. APPLIED PHYSICS LETTERS 107, 032903 (2015)Google Scholar
  8. 8.
    Vaishali Rawat et al., Journal of Applied Physics 116, 164106 (2014)Google Scholar
  9. 9.
    M. Neshat et al., Elect Lett 44(17), 1020 (2008)Google Scholar
  10. 10.
    L. Wendling, et al. Real Time Monitoring and Detection of Alcohol Using Microwave Sensor Technology, IEEE Proc. Conf. Developm eSystems, 2009Google Scholar
  11. 11.
    N. Wiwatcharagoses, et al, Michigan State Uni report 2009Google Scholar
  12. 12.
    A. Sklavounos, N. Barker, IRMMWTHz-IEEE Proc, p2 (2011)Google Scholar
  13. 13.
    Laurette, S, et al. Subterahertz characterization of ethanol hydration layers by microfluidic system, Applied Physics Letters, 97, 111904 (2010), DOI:
  14. 14.
    T. Chretiennot et al. A Microwave and microfluidic planar resonator for efficient and accurate complex permittivity characterization of aqueous solutions. IEEE Trans Microw Theory Techn. 2013 Feb;61(2):972–978Google Scholar
  15. 15.
    W.W. Christie. Detectors for HPLC of lipids with special reference to evaporative lightscattering detection. In Advances in Lipid Methodology – One. Edited by W.W. Christie, Oily Press, Ayr. pp. 1992. 239–271Google Scholar
  16. 16.
    N. E. Chayen, “Turning protein crystallisation from an art into a science,” Current Opinion in Structural Biology, vol. 14, pp. 577–583, Oct. 2004Google Scholar

Copyright information

© Springer Science+Business Media B.V. 2017

Authors and Affiliations

  • Y. Zhang
    • 1
  • V. Matvejev
    • 1
  • S. Declerck
    • 2
  • S. Stroobants
    • 3
  • G. Pandey
    • 1
  • G. He
    • 1
  • D. Mangelings
    • 2
  • D. Maes
    • 3
  • S. Muyldermans
    • 4
  • J. Stiens
    • 1
  1. 1.Laboratory of Micro- and Photoelectronics, LAMI-ETROVrije Universiteit Brussel (VUB)BrusselsBelgium
  2. 2.Department of Analytical Chemistry and Pharmaceutical Technology, Center for Pharmaceutical Research (CePhaR)Vrije Universiteit Brussel-VUBBrusselsBelgium
  3. 3.Structural Biology BrusselsVrije Universiteit BrusselBrusselsBelgium
  4. 4.Cellular and Molecular ImmunologyVrije Universiteit BrusselBrusselBelgium

Personalised recommendations