Skip to main content

Photonic Terahertz Signal Analyzers

Periodic Reporting for period 3 - Pho-T-Lyze (Photonic Terahertz Signal Analyzers)

Reporting period: 2020-06-01 to 2021-11-30

For a long time, the Terahertz range (100 GHz-10 THz), situated at the junction between microwave electronics and infrared optics, has been considered as the “Terahertz gap” as there were only inefficient ways of signal generation and detection. Within the past two decades, the THz range has developed into a vivid research field with many -so far mostly academic- applications. Many ways of generating and detecting Terahertz radiation have become available, however, this part of the electromagnetic spectrum still suffers from the lack of adequate and affordable component- and spectral analysis techniques.
So far, spectrum analysis and vector network analysis are enabling technologies for component development throughout the microwave and millimeter wave band. Spectrum analyzers measure the output power of a source or generator vs frequency, allowing for metrological characterization. Vector network analyzers launch an electromagnetic wave with tunable frequency into a device under test and measure the transmitted and the reflected fraction. With an appropriate device model, individual component contributions can be analysed and the measured transmission and reflection data be compared to simulations. Both information obtained by spectrum analysis or by vector network analysis help component designers to optimize the respective structure, finally making them viable for commercial applications.
Due to the lack of affordable electronics for frequencies above ~100 GHz, vector network analyzers (VNAs) have to use frequency-extenders to reach into the THz frequency band. The bandwidth of frequency extended electronic systems is restricted to about 50% of the center frequency. Therefore, about seven different extender setups are required to cover the range of 100 GHz- 1 THz. Each extender setup has to be individually realigned and tediously calibrated. Further, extenders become increasingly expensive the higher the THz frequency and are, therefore, only used by a handful of groups per country. Electronic spectrum analyzers face similar problems as VNAs. Concerning optical solutions, only a few examples of highly expensive photonic, pulsed frequency comb-based systems have been demonstrated. Affordable, large bandwidth commercial THz metrology tools are missing so far.
This lack impedes the development of components and systems in this very attractive frequency range with large societal impact. Key applications of Terahertz components lie, e.g. in future communications. Already 5G targets for carrier frequencies up to 80 GHz. Future communication systems for 6 G and beyond will use frequencies within the Terahertz range in order to keep pace with the world’s data hunger: In the terahertz band, there are many so far unused bands which can be exploited for communication. Another important application are imaging and spectroscopy systems in security. Terahertz radiation penetrates through many optically opaque materials, such as clothes or paper, allowing to analyse and image the content below. Combined with spectroscopic techniques potential threats or may be uniquely identified. Also medical imaging applications have been demonstrated. Last but not least, non-destructive testing is a versatile field of applications for terahertz technology. Many objects are difficult to analyse with conventional techniques such as ultrasound or X-rays. Examples are plastic objects containing air or distribution of the active pharmaceutic ingredient within pills.
So far, none of these applications became commercial. The systems developed within this proposal will boost component development in the Terahertz range and foster Terahertz technology. The objectives are the development of photonic THz characterization tools based on telecom-wavelength compatible photomixing technology as an alternative to conventional electronic concepts, namely:
1.) Photonic vector network analysers with extreme frequency coverage (at least two octaves in one system), and
2.) Photonic Terahertz spectrum analyzers (PSAs)
Vector network analyzers (VNAs) measure the transmitted and reflected wave launched into a device or circuit under test. In most cases, VNAs are two-port systems requiring one source and one receiver in each port. In a first step, we tailored the photonic sources and receivers to the requirements of the project. In terms of sources, a photomixer mixes a pair of lasers with a difference frequency of, say, 1 THz to a THz current by absorbing the beat note. An attached antenna converts the current into THz light. The detection process works inversely, i.e. a THz signal is mixed with a beat note of two lasers, resulting in a down-conversion to DC.
In a first step, the absorption within the source photomixer was increased by a factor of 2.7 by integrating sources with an optical waveguide. Photoconductive receivers are now realized in packages with enhanced screening from external noise. For on-chip designs, we have developed an endfire-antenna-coupled receiver for directly launching a waveguide-coupled mode to the receiver without any transition to free space. Photoconductors are further implemented in a pulsed version of the photonic VNA (PVNA).

With powerful photoconductors at hand, we have completed the setup of a pulsed two port PVNA. The bandwidth so far covers frequencies from 100 GHz to 1.5 THz in a single photonic system, i.e. it covers about 5-7 separate waveguide bands of an electronic VNA. The peak dynamic range is currently 50 dB at 100 µs integration time per point. Higher dynamic range is possible by longer integration. We have developed data analysis techniques for parallel evaluation of transmission and reflection in order to improve the data quality of extracted material and device parameters of a device under test. We have proven the effectiveness at the example of a THz isolator and a lossy dielectric plate. While conventional THz systems require a mechanical determination of the thickness for accurate evaluation of the refractive index and absorption coefficient, we demonstrated the determination of the refractive index, the absorption coefficient (alternatively, the complex dielectric permittivity) and the physical thickness only by a THz measurement. The thickness error was of the order of that of a caliper (~10 µm).

For on-chip solutions, we developed first versions of on-chip dielectric waveguides with various components included, such as splitters/combiners, bends, and coupling structures. To this point, we are still optimizing the in- and outcoupling. The waveguide transmission losses are small enough for on–chip dielectric circuits.

Photonic free space spectrum analyzers are yet fully functional. We have developed a concept based on a photoconductor where a laser beat note acts as local oscillator. In electronic systems, in contrast, a THz local oscillator is required, typically with severe power and tunability limitations. The photonic concept offers several advantages: Firstly, laser power and components are abundant at telecom wavelengths. Secondly, tuning a laser by, say, 1 THz is comparatively simple. At 1550 nm center wavelength, 1 THz tuning corresponds roughly to 0.5% of the laser frequency. The bandwidth of a single photonic PSA therefore exceeds that of electronic PSAs by several octaves. The noise floor is currently at 520 nW/sqrt(Hz) at 100 GHz and increases to 8.3 µW/sqrt(Hz) at 1 THz. This is sufficient for analyzing many source types. The spectral resolution is currently around 10 MHz.
In summary, all targeted free space systems are yet functional and in operation. The main results achieved so far are:
-Waveguide-coupled ballistic pin diodes with 2.7 times improved optical to photocurrent conversion efficiency
-Operational free space photonic PSA
-Operational free space pulsed photonic VNA
-Data evaluation routines for materials parameter extraction from 1.5 port VNA measurements and first routines for two-port measurements
-First broadband on-chip dielectric waveguide components
-First applications of the PSA and pulsed PVNA carried out and published
Progress beyond the state of the art achieved so far:
-Operational two port pulsed vector network analyzer
-Operational photonic spectrum analyzer
-Developed data evaluation routines show superior performance of the systems beyond the state of the art. Refractive index, absorption coefficient and sample thickness can be determined accurately without any mechanical measurement for both lossy and transparent materials.

Within the second half of the project the focus will lie on on-chip solutions.
Progress beyond the state of the art expected till the end of the project:
- Increase the dynamic range of the pulsed photonic VNA by at least 20 dB
- Advanced on-chip dielectric circuitry
- On-Chip continuous-wave photonic VNA
- On-Chip photonic PSA
- Applications of the newly developed systems
Dielectric THz waveguides
1550 nm waveguide-coupled fully ballistic p-i-n photomixer with antenna
Logo of the ERC Starting Grant "Pho-T-Lyze", grant No. 713780
Image of the pulsed PVNA setup