The remarkable explosion of wireless devices and bandwidth-consuming Internet applications have boosted the demand for ultra-high data rate wireless communication. To meet the exponentially increasing traffic demand, new regions in the radio spectrum are explored. The terahertz band (0.1 THz-10 THz), sandwiched between microwave frequencies and optical frequencies, is considered as the next breakthrough-point to revolutionize communication technology due to its rich spectrum resources. It is recognized as a promising candidate for the future 6G communication.
Bandpass filters are indispensable components for Terahertz communication. They are used to reduce and filter out signal interference and distortion. Terahertz bandpass filters have wide applications in imaging, spectroscopy, security systems and detection of materials. Although they have promising advantages, they also have some challenges and limitations. For instance, electrical conductivity, surface roughness and losses can have a major effect on the performance of the filter.
HFWorks Modeling of Terahertz Filters
To contribute to this research effort, we used our electromagnetic virtual prototyping software, HFWorks, to study the impact of electrical conductivity, surface roughness and losses on filter performance. We examined two types of filters: low pass and bandpass. The design of the filters was obtained from the two journal papers [1] and [2].
The Effect of Electrical Conductivity and Surface Roughness
Electrical conductivity is an important parameter, especially at terahertz. The choice of conductors may not be crucial at low frequencies. However, it is an important factor to consider at high frequencies. Surface roughness can have a significant influence on terahertz filters as it becomes larger than the skin depth.
Spoof surface plasmon polaritons (SSPP) filter [1]
The above filter is simulated, and the following results are obtained
Electric field intensity (303.5 GHz)
Insertion loss of SSPP filter made of different materials
The effect of surface roughness
From the above plots, we can notice that decreasing the electrical conductivity results in a decreased bandwidth. The filter made of copper has a cutoff frequency of 730 GHz, and this drastically drops to 100 GHz for the gold filter. Increasing surface roughness leads to the rise of losses. As a result, the bandwidth is reduced. For instance, the filter with a flat surface has a cutoff frequency of 1200 GHz, while the filter with a surface roughness of 5 um has a cutoff frequency of 730 GHz. This is a significant change.
The Effect of Changing the Physical Dimensions
Variations in the physical dimension can impact the operation of the filter. The terahertz wavelength is very short. Hence, any slight change in the geometry can degrade the filter performance. In the below example, we studied the effect of changing the length L.
Spoof surface plasmon polaritons (SSPP) filter
The effect of changing dimension L
We can observe from the above graph that a minor change, even in the order of microns, in the physical dimensions can lead to a major change in the insertion loss. For example, at the frequency of 1200 GHz, there is a difference of about 2 dB in the insertion loss for a change of 2.5 um in the length L.
Losses and Heat
Losses can be significant at terahertz. During the operation of the filter, the conductor and dielectric losses can be converted into heat, and this can lead to the filter breakdown.
Ridged waveguide filter [2]
Electric field animation (240 GHz)
Insertion and return loss of the filter
Temperature distribution of the filter (240 GHz)
The filter has a wide bandwidth of 120 GHz and a passband from 180 GHz to 300 GHz. The filter reaches a maximum temperature of 43.6 °C at 240 GHz. This is an acceptable value, and it is within the standard limits.
In the exploration of terahertz filters, crucial factors like electrical conductivity, surface roughness, and physical dimensions significantly influence their performance. Through virtual prototyping with HFWorks, we demonstrated how material choice and geometry adjustments affect filter bandwidth and losses. This study underscores the precision required in designing terahertz filters to mitigate heat generation and ensure optimal operation, highlighting HFWorks' capability in navigating these intricate design challenges for enhanced filter efficiency at THz frequencies.
References
[1] K. Xu, F. Zhang, Y. Guo, L. Ye, and Y. Liu, 2020, “Spoof Surface Plasmon Polaritons Based on Balanced Coplanar Stripline Waveguides”
[2] J. Ding, J. Hu, D. Liu, D. Wang, S. Shi and W. Wu, 2016, “A 240-GHz Wideband Ridged Waveguide Filter Based on MEMS Process “