Open Access
Add to BibSonomy
Abstract
The suppression of the crosstalk in a CMOS THz detector is essential for enhancing the performance of detector arrays; however, it presents several technical challenges at the chip level. In this paper, a novel structure featuring a mushroom-like artificial magnetic conductor (M-AMC) is developed to suppress the crosstalk between CMOS THz detectors with on-chip antennas. Three-dimensional simulation results show that the M-AMC structure, which is designed by metal Al and doped-Si materials in the CMOS process, not only reduces the transmission coefficient of the electromagnetic wave between adjacent pixels but also enhances the electric field of the target pixels. A 0.65 THz detector array with a M-AMC structure based on the on-chip antenna was fabricated. Experimental results present that after implanting the M-AMC structure, the noise equivalent power (NEP) at the central frequency of pixels significantly decreases by 315.5%. Moreover, the distribution of NEP becomes more uniform, as evidenced by a reduction in the standard deviation coefficient of 26.3%. This demonstrates the effectiveness of the method in suppressing crosstalk and improving the responsivity of CMOS THz detectors, which can be used for high-performance THz detector arrays.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
CMOS Terahertz (THz) detectors have attracted great interest in room temperature THz imaging [1,2] due to their advantages such as fast on-state response time, low noise equivalent power (NEP), low cost, and high integration level. By integrating MOSFETs or diodes with on-chip antennas and readout electronics [3–6], the THz signal can be converted into electric signals for detection. In many applications, achieving high-density THz sensor integration is highly desirable which can achieve higher spatial resolution and higher sensitivity. However, as the pixel size reduces, the optical efficiency of the CMOS THz detectors deteriorates and the crosstalk between adjacent detector pixels becomes more pronounced. This effect mainly stems from a portion of the incoming THz radiation energy penetrates into the substrate [1], resulting in several detrimental effects on the detectors. The effects include the decrease of response amplitude and shift of response frequency of the detectors.
Several approaches have been proposed to mitigate the crosstalk effects in CMOS THz detectors. One solution involves modifying the structure of the lossy substrate through micromachining techniques to reduce the thickness of the Si substrate [6–9], or creating an air cavity at the backside of the substrate [10]. Alternatively, proton implantation in the Si substrate has been employed to increase its resistivity, thereby decreasing the power transmission in the Si substrate [11,12]. Additionally, off-chip microwave lenses [13] or superstrates [14] have been utilized to mitigate the crosstalk. In our previous studies, we explored the impact of spatial crosstalk on the CMOS THz detectors with on-chip surface plasmon resonance (SPR) antennas, and proposed thinning of the detector substrate increasing the isolation layer thickness of the pixel [15]. However, most existing methods are incompatible with CMOS processes and necessitate complex post-CMOS procedures. These additional steps not only increase overall costs but also introduce alignment uncertainties and mechanical stability issues in the technology.
AMC structure has been extensively applied in microstrip phased antennas, patch antennas, dipole antennas, cavity antennas, and chip antennas [16–21]. AMC structure is constructed by a periodic high-impedance electromagnetic surface, which can produce constructive reflections with the incident wave for a certain frequency range so that the electromagnetic wave can be localized more thoroughly. This technical breakthrough offers a method for reducing electromagnetic wave transmission in the substrate [22] and sensors. Recently, ultra-thin Artificial Magnetic Conductors (AMC) have been proposed to mitigate the crosstalk of the on-chip antennas [23], this advancement demonstrates improved the gain and radiation efficiency of antenna.
In this paper, the mushroom-like AMC (M-AMC) structure is proposed and introduced for the 0.65 THz CMOS detector under the framework of the 55 nm CMOS process. The engineering capability of M-AMC in reducing crosstalk, improving gain enhancement, and the uniformity of the responsivity in the detector array is fully presented theoretically and experimentally. To explore the performance of the M-AMC structure, the electric field profile of the silicon-based antenna with and without the M-AMC structure are simulated and compared using High-Frequency Structure Simulator (HFSS) method. Results reveal that the mutual coupling can be significantly reduced when the M-AMC structure is inserted between detector elements. Besides, the local electric field of the optical antenna is also enhanced after using the M-AMC structure. Furthermore, a 5 × 5 THz detector array is fabricated based on the standard CMOS process. Experimental results also demonstrate the positive effect of the M-AMC on the signal responsivity of the THz array, indicating that the proposed approach is potentially useful for optical antenna and detector arrays.
2. Device structure design and concept validation
The schematic diagram of the 0.65 THz detector array with M-AMC structure in the CMOS process is shown in Fig. 1(a). The detector pixel of the array comprises a MOSFET mixer (brown pattern) and a fan-rod shaped 0.65 THz SPR antenna (blue pattern). The SPR antenna is realized with gate polysilicon material, while the source and drain terminals of MOSFETs are implemented using heavily doped Si [24,25]. To ensure efficient THz responsivity, the non-symmetric gate-source structure is selected as a MOSFET mixer, as depicted in Fig. 1(b), which is placed in the center of the antenna. It has lower gate-source parasitic capacitance, which is favorable for reducing input THz signal leakage and increasing the THz responsivity [5]. According to the well-known self-mixing theory, when the THz signal radiates to the detector, the local electric field enhancement in the MOSFETs channel is realized due to the surface plasmon resonance effect of antenna, then the non-symmetric MOSFET mixer outputs the electrical signal by drain-source voltage [3,26].
Fig. 1. (a) The schematic illustration of the configuration of the M-AMC array used in CMOS THz detectors. (b) Parameters of MOSFET with the non-symmetric source-drain structure. (c) Side view and (d) top view of the single pixel with M-AMC structure.
To suppress the crosstalk effect in the detector arrays, the M-AMC structure is designed in the sides of the detector pixel. AMC structure normally includes three parts: metallic patches, a ground plane, and connecting vias [14]. Here, as illustrated in Fig. 1(c) and (d), the metal patch is a square periodic frequency selective structure (yellow pattern) made by the first Al layer in the CMOS process while reflective material is made by the heavily doped-Si (light green pattern) [27]. The vias (orange pattern) between the Al layer and Si substrate are used for connecting the medium. SiO2 (light blue pattern) is the dielectric material between the layers. All components of the M-AMC structure are designed in 55 nm CMOS process.
To study the influence of M-AMC structure on the crosstalk properties of the detector array, a 3 × 3 THz array including the central antenna (Tx) and adjacent antennas (Rx) is modeled in HFSS software, as shown in Fig. 2, wherein the central antenna is isolated by M-AMC. In the simulation, the material of metal patch, and the heavily doped-Si ground plane of M-AMC are modeled by aluminum with conductivity of 3.8 × 107 S/m. The vias material of M-AMC is copper with a conductivity of 5.8 × 107 S/m. The material of the SPR antenna is heavily doped polysilicon, which relative permittivity is calculated by the Drude model [28,29]. Its doping concentration and free carrier mobility are 3 × 1019 cm−3 and 800 cm2/V·s, respectively. The effective carrier mass m* is chosen as 0.98m0, which corresponds to the longitudinal effective mass of the electron. The relative dielectric constant of the Si substrate is 11.9. The rest of the free space is filled with SiO2 material with a relative dielectric constant of 4. The fixed dimensions of the THz detector with M-AMC structure are shown in Table 1.
Table 1. The fixed dimensions of the THz detector with the M-AMC structure
The transmission coefficient S parameter between Tx and Rx is selected to evaluate the level of antenna isolation. As illustrated in Fig. 2, Tx is set as port 1, while Rx along the y-axis and x-axis are set as port 2 and port 3, respectively. Since S11 is normally used interchangeably with return loss and reflection coefficient, S12 and S13 represent the transmission coefficients of Tx to Rx along the y-axis and x-axis, respectively. Smaller values of S12 and S13 indicate lower transmission coefficients and crosstalk optimization.
The influence of the parameters of M-AMC on the electric field gain of the antenna is further investigated. The plane-wave excitation polarized along the direction of the antenna rod length with its electric field of 1 V/m. By placing a field probe at the center of the antenna arm to monitor the localized field E concentrated by the antenna, the field enhancement (E/E0) of the antenna can be calculated as the ratio of the probed field power to the incident wave power [22,30].
At the center frequency of the antenna (650 GHz), a comprehensive study of the geometrical parameters (W0, W1, P) and the number (L/P) of M-AMCs on the transmission coefficient S12 and S13 are conducted. Figure 3(a) and (b) reveal the variation of S12 and S13 as a function of W0/P and W1, respectively. Here, W0/P varies from 0.6 to 1.0, W1 varies from 0.3 to 9.9 µm, while L/P is defined as 5. It is found that both S12 and S13 decrease gradually with W0/P while they almost keep constant values with W1. Both S12 and S13 reach minimum values when W0/P is set around 0.95. Figure 3(c) and (d) exhibit S12 and S13 as a function of W0/P and L/P with W1 = 4.8 µm, it is found that the minimum value of S12 and S13 occurs with L/P ≤ 6 and W0/P is around 0.95. It is considered when W0/P increases, the additional inductance in the M-AMC metal patch increases significantly, while the capacitance between the patch and the top surface of the vias decreases a little. Thus, the rising inductance leads to decreasing S parameters [23]. Figure 3(e) and (f) show S12 and S13 of the antenna with/without the M-AMC structure. It is found that after the introduction of the optimized M-AMC structure, the transverse and longitudinal transmission coefficients between antennas are significantly reduced under the optimal parameters of L/P = 5, W1 = 4.8 µm, and W0/P = 0.95. The minimum values of S12 and S13 are −59.77 dB and −48.31 dB, respectively, while in the case without M-AMC, S12 and S13 are −32.97 dB and −30.56 dB, respectively.
Fig. 3. Contour plot of (a) S12 and (b) S13 change with W0/P and W1. When W1 = 4.8 µm, (c) S12 and (d) S13 vary with W0/P and L/P. Comparison of S12 and S13 with/without M-AMC structure with respect to the variation in (e) W0/P and (f) L/P.
The electric field distributions in the xoy and yoz planes for arrays with/without optimized M-AMC are shown in Fig. 4. It is seen that for the pixels without M-AMC structure, shown in Fig. 4(a) and (b), a strong electric field leakage not only in Rx and Ry antennas but also in the Si substrate, which is consistent with the literature [31]. However, for the pixels coupled with the optimized M-AMC, as depicted in Fig. 4(c) and (d), the electric field is strongly confined in the M-AMC square with an extraordinarily low transmission signal. The significant reduction of the leakage is due to the fact that a periodic grid consisting of a large number of M-AMC cells has a high-impedance surface [32]. Under the theory of transmission lines and periodic circuits, M-AMC can be simply configured as RLC resonant circuit, here the resistor R is from the low resistivity of the silicon substrate, the capacitor C represents the coupling between two adjacent M-AMC cells, while the inductor L is from the traces of vias. Based on the optimized structure size, the M-AMC plane can produce constructively in-phase reflections when the incident wave is at a specified operating frequency band [33], therefore enhancing the radiation performance of the Tx antenna and preventing inter-pixel crosstalk.
Fig. 4. Electric field distribution of THz array without/with AMC structure in the xoy and yoz planes. Without M-AMC structure in (a) xoy and (b) yoz planes. With M-AMC structure in (c) xoy and (d) yoz planes.
The radiation pattern of the central antenna in the 3 × 3 array under two conditions (the central antenna with or without M-AMC) has been simulated. It can be seen from Fig. 5(a), for the antenna without M-AMC, the radiation pattern of the H_plane (xoy plane) is distorted due to the crosstalk of the surrounding antennas. For the antenna with M-AMC, as shown in Fig. 5(b), the overall width of the radiation pattern in the H_plane is narrowed. This informs that the M-AMC structure reduces the radiation and energy loss of the central antenna to the neighboring antennas. Moreover, the antenna gain of the E_plane (yoz plane) at ɵ = 90° significantly increased by approximately 6.87 dB after incorporating the M-AMC structure. This demonstrates that the M-AMC structure enhances the ability to receive electromagnetic waves in the z direction and effectively improves the local electric field capability of the antenna.
Fig. 5. Simulated 2-D radiation pattern (E-plane and H-plane) of the central antenna in 3 × 3 array under two conditions: (c) without M-AMC structure, and (d) with M-AMC structure.
Figure 6(a) and (b) exhibit the variation of electric field gain (E/E0) with the frequency of 3 × 3 THz antenna array with/without the optimized M-AMC structure, respectively. The maximum E/E0 of the antenna is at about 0.650 THz. In the array without M-AMC structure, the maximum values of E/E0 exhibit significant fluctuations, ranging from 100 to 165, while the values remain stable at about 225 after the introduction of M-AMC. Figure 6(c) presents the maximum values of E/E0 for each antenna with/without M-AMC. It demonstrates that E/E0 is significantly increased about 2 times after the integration of the M-AMC structure.
Fig. 6. Electric field enhancement of the 3 × 3 THz array (a) without and (b) with M-AMC structure as a function of frequency. (c) Comparison of peak E/E0 values of each antenna in the array.
3. Experimental demonstration
A 0.65 THz array with/without M-AMC structures based on the on-chip antenna is then fabricated using 55 nm standard CMOS process. The process flow is illustrated in Fig. 7(a-f). Firstly, the P well is formed by a boron implant. The shallow trench isolation (STI) and active area (AA) are shaped. Secondly, the gate of the MOSFETs and the silicon plasmonic antenna are constructed based on the reactive ion etching (RIE) technique. Thirdly, heavy N-type ions implantation is performed at the MOSFET source and drain. At the same time, heavy P-type ion implantation is carried out in P-well to form the ground plane of the M-AMC structure. Fourthly, the terminals of MOSFETs and heavy P-type regions are solicited for Ohmic contact. Subsequently, SiO2 is deposited on top of the active region. Fifthly, copper (Cu) is deposited between the first metal layer and heavily p-type well by Chemical Vapor Deposition (CVD) as vias for M-AMC. Finally, the first layer of Al metal is deposited to form the periodic metal patch of M-AMC. In the overall process flow, there is no additional process in the standard CMOS technology. Figure 8 depicts the micrograph of the fabricated 5 × 5 THz array with the M-AMC structure, in which the shape of the M-AMC is in fair agreement with the proposed structure. The three pixels incorporating with M-AMC structure in the array are labeled as Sample #1, Sample #2, and Sample #3, for comparison, these pixels without M-AMC are denoted as Sample #4, Sample #5, and Sample #6.
Fig. 7. Fabrication process flow of the silicon-based plasmonic THz detector with AMC structure in standard CMOS technology. (a) Formation of shallow trench isolation (STI), deep P well, gate oxide, and polysilicon; (b) To shape the gate of MOSFET and polysilicon antenna; (c) Formation of N and P heavily doped region; (d) The silicide layers and dielectric layer are deposited; (e) The vias are built; (f) Metal patch of M-AMC is deposited using first metal layer.
Fig. 8. (a) The fabricated THz detector array in which samples #1 - #6 present the pixels while the solid columns surrounding the sample #1 - #3 are M-AMC structures. (b) The enlarged pixel and M-AMC structures.
The M-AMC on the performance of the THz detector has been experimentally characterized. The schematic diagram of the experimental test is shown in Fig. 9. The THz sources are generated by a VDI-Tx-S140 source. Two off-axis parabolic mirrors are used to focus the THz beam onto the detector, which was biased with an applied voltage between the source and the gate. The induced voltage response of the detector was measured between the drain and the source of MOSFET using a lock-in amplifier. All measurements were performed at room temperature. The voltage responsivity (Rv) and noise equivalent power (NEP) are calculated by the following formula:
where ΔV is the response voltage amplitude, Ptotal is the total power of the THz source, St is the radiation beam spot area and Santanna is the area of the antenna. In our experiments, the beam area is about 3 mm × 3 mm, on the other hand, the size of the silicon antenna used in the responsivity calculation is 0.162 mm × 0.162 mm.Figure 10(a) and (b) exhibit the typical responsivities of detectors #1 and #4 with and without M-AMC structure at different gate voltages. It is seen that the Rv increases first and then decreases with the increase of gate voltage (Vg), showing a maximum value near the threshold voltage VTH of the MOSFET transistor. This characteristic is in good agreement with the plasma wave detection theory for MOSFETs THz detectors [26,34]. The maximum response appears at 0.655 THz. We compared the Rv of detectors with and without M-AMC as a function of the frequency when the gate voltage (Vg) is 0.4 V on the same chip, as depicted in Fig. 10(c). It is found that the available bandwidth of the antenna with/without the M-AMC structure is all about 30 GHz. The average responsivity of the detector increases by approximately 2.94 times with the incorporation of the M-AMC structure, which is almost consistent with the simulation results. In addition, we performed a statistical comparison of the performance of multiple detectors in several chips at the center frequency f = 0.655 THz, as shown in Fig. 10(d). It could be seen that the Rv of the sample with M-AMC is superior to the case without M-AMC, exhibiting an average value of about 0.094 kV/W and 0.032 kV/W, respectively. The standard deviation coefficients, which are typically calculated by dividing the standard deviation of the data by its mean, are 6.18% and 23.07% for the samples with and without M-AMC, respectively.
Fig. 10. The measured Rv of the sample (a) with and (b) without M-AMC at different gate voltages (Vg). (c) Comparison of measured Rv of the samples with and without M-AMC structure as a function of frequency. (d) A statistical comparison of the Rv of multiple samples with and without M-AMC structure in several chips at f = 0.655 THz.
Figure 11(a) and (b) illustrate the Vg dependence of NEP with the frequency ranging from 635 GHz to 660 GHz. With the increase of gate voltage, NEP exhibits a significant decrease followed by a gradual stabilization. At the operating gate voltage of 0.4 V, NEP consistently reaches its minimum value. Figure 11(c) presents a statistical comparison of NEP for pixels across multiple chip pixels at the center frequency f = 0.655 THz. Due to the implementation of the M-AMC structure, the average NEP value of samples is reduced by 315.48%, from 2.65 nW/Hz0.5 to 0.84 nW/Hz0.5. The standard deviation coefficients of the samples with and without M-AMC are 5.97% and 32.23%, respectively. It is proved that the adoption of M-AMC has greatly improved the uniformity of the NEP of the samples. The experimental results match the simulated ones, informing that the introduction of M-AMC enhances the responsivity and uniformity of the detector, and effectively reduces the crosstalk of the device. These findings validate the reliability of the antenna isolation predictions by the proposed M-AMC structure.
Fig. 11. The measured NEP of the sample (a) with and (b) without M-AMC at different gate voltages (Vg). (c) A statistical comparison of the NEP of multiple samples with and without M-AMC structure in several chips at f = 0.655 THz.
4. Conclusion
In this work, a mushroom-like AMC structure is implemented in the design of the CMOS THz detector array with silicon plasmonic antennas. This integration aims to reduce the strong mutual coupling between the detector elements without sacrificing their compact size. The M-AMC structure is constructed using metal layer in 55 nm standard CMOS process without additional fabrication steps. Experimental results demonstrate that the introduction of M-AMC leads to a significant reduction in the average NEP of the detector, with a decrease of approximately 315.5%. Moreover, at a frequency of 0.655 THz, the non-uniformity of the NEP of the detector is reduced by approximately 26.3%. The compatibility of M-AMC with CMOS manufacturing process offers a promising opportunity to realize high-performance and compact THz array detectors in the large-scale fabrication of detector arrays.
Funding
National Natural Science Foundation of China (61627804).
Disclosures
The authors declare no conflicts of interest.
Data availability
The data underlying results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. D. Coquillat, J. Marczewski, P. Kopyt, N. Dyakonova, B. Giffard, and W. Knap, “Improvement of terahertz field effect transistor detectors by substrate thinning and radiation losses reduction,” Opt. Express 24(1), 272 (2016). [CrossRef]
2. R. Al Hadi, H. Sherry, J. Grzyb, Y. Zhao, W. Forster, H. M. Keller, A. Cathelin, A. Kaiser, and U. R. Pfeiffer, “A 1 k-Pixel Video Camera for 0.7–1.1 Terahertz Imaging Applications in 65-nm CMOS,” IEEE J. Solid-State Circuits 47(12), 2999–3012 (2012). [CrossRef]
3. X. Zhang, X. Ji, Y. Liao, J. Peng, C. Zhu, and F. Yan, “Performance enhancement of CMOS terahertz detector by drain current∗,” Chinese Phys. B 26(9), 098401 (2017). [CrossRef]
4. Z. Ahmad and K. O. Kenneth, “THz Detection Using p+/n-Well Diodes Fabricated in 45-nm CMOS,” IEEE Electron Device Lett. 37(7), 823–826 (2016). [CrossRef]
5. Q. Yang, X. Ji, Y. Xu, and F. Yan, “Improved performance of CMOS terahertz detectors by reducing MOSFET parasitic capacitance,” IEEE Access 7, 9783–9789 (2019). [CrossRef]
6. I. Shcherback and O. Yadid-Pecht, “CMOS APS photoresponse and crosstalk optimization analysis for scalable CMOS technologies,” 11th IEEE Int. Conf. Electron. Circuits Syst. ICECS2004153–155 (2004).
7. C. Veerappan and E. Charbon, “A Substrate Isolated CMOS SPAD Enabling Wide Spectral Response and Low Electrical Crosstalk,” IEEE J. Sel. Top. Quantum Electron. 20(6), 299–305 (2014). [CrossRef]
8. H. Chu, Y. X. Guo, T. G. Lim, Y. M. Khoo, and X. Shi, “135-Ghz Micromachined on-Chip Antenna and Antenna Array,” IEEE Trans. Antennas Propag. 60(10), 4582–4588 (2012). [CrossRef]
9. L. Papapolymerou, R. F. Drayton, and L. P. B. Katehi, “Micromachined patch antennas,” IEEE Trans. Antennas Propag. 46(2), 275–283 (1998). [CrossRef]
10. W. T. Khan, A. Cagri Ulusoy, G. Dufour, M. Kaynak, B. Tillack, J. D. Cressler, and J. Papapolymerou, “A D-band micromachined end-fire antenna in 130-nm SiGe BiCMOS Technology,” IEEE Trans. Antennas Propag. 63(6), 2449–2459 (2015). [CrossRef]
11. K. T. Chan, A. Chin, Y. D. Lin, C. Y. Chang, C. X. Zhu, M. F. Li, D. L. Kwong, S. McAlister, D. S. Duh, and W. J. Lin, “Integrated Antennas on Si With over 100 GHz Performance, Fabricated Using an Optimized Proton Implantation Process,” IEEE Microw. Wireless Compon. Lett. 13(11), 487–489 (2003). [CrossRef]
12. R. Wu, W. Deng, S. Sato, T. Hirano, N. Li, T. Inoue, H. Sakane, K. Okada, and A. Matsuzawa, “A 60-GHz efficiency-enhanced on-chip dipole antenna using helium-3 ion implantation process,” Eur. Microw. Week 2014 Connect. Futur. EuMW 2014 - Conf. Proceedings; EuMC 2014 44th Eur. Microw. Conf. 108–111 (2014).
13. A. Babakhani, X. Guan, A. Komijani, A. Natarajan, and A. Hajimiri, “A 77-GHz phased-array transceiver with on-chip antennas in silicon: Receiver and antennas,” IEEE J. Solid-State Circuits 41(12), 2795–2806 (2006). [CrossRef]
14. H. Zhang and A. Shamim, “gain enhancement of millimeter-wave on-chip antenna through an additively manufactured functional package,” IEEE Trans. Antennas Propag. 68(6), 4344–4353 (2020). [CrossRef]
15. Y. Liao, K. Wang, H. Zhu, and X. Ji, “Crosstalk in CMOS terahertz detector array with on-chip SPR antenna,” IEEE Photonics J. 14(6), 1–6 (2022). [CrossRef]
16. A. Foroozesh and L. Shafai, “Application of combined electric- And magnetic-conductor ground planes for antenna performance enhancement,” Can. J. Electr. Comput. Eng. 33(2), 87–98 (2008). [CrossRef]
17. S. Velan, E. F. Sundarsingh, M. Kanagasabai, A. K. Sarma, C. Raviteja, R. Sivasamy, and J. K. Pakkathillam, “Dual-band EBG integrated monopole antenna deploying fractal geometry for wearable applications,” Antennas Wirel. Propag. Lett. 14, 249–252 (2015). [CrossRef]
18. G. Lu, F. Yan, K. Zhang, Y. Zhao, L. Zhang, Z. Shang, C. Diao, and X. Zhou, “A dual-band high-gain subwavelength cavity antenna with artificial magnetic conductor metamaterial microstructures,” Micromachines 13(1), 58 (2021). [CrossRef]
19. H. Malekpoor, A. Abolmasoumi, and M. Hamidkhani, “High gain, high isolation, and low-profile two-element MIMO array loaded by the Giuseppe Peano AMC reflector for wireless communication systems,” IET Microwaves Antenna & Prop 16(1), 46–61 (2022). [CrossRef]
20. M. Kokolia and Z. Raida, “Textile-integrated microwave components based on artificial magnetic conductor,” Int J Numer Model 34(4), e2864 (2021). [CrossRef]
21. A. P. Feresidis, G. Goussetis, S. Wang, and J. C. Vardaxoglou, “Artificial magnetic conductor surfaces and their application to low-profile high-gain planar antennas,” IEEE Trans. Antennas Propag. 53(1), 209–215 (2005). [CrossRef]
22. V. Giannini, J. A. Sánchez-Gil, J. V. García-Ramos, and E. R. Méndez, “Collective electromagnetic emission from molecular layers on metal nanostructures mediated by surface plasmons,” Phys. Rev. B 75(23), 235447 (2007). [CrossRef]
23. Y. Yu, Z. Akhter, and A. Shamim, “Ultra-thin artificial magnetic conductor for gain-enhancement of antenna-on-chip,” IEEE Trans. Antennas Propag. 70(6), 4319–4330 (2022). [CrossRef]
24. Z. Shen, X. Ji, Y. Liao, K. Wang, B. Jin, and F. Yan, “Resonant polysilicon antenna for terahertz detection,” IEEE Photonics J. 11(4), 1–8 (2019). [CrossRef]
25. R. Huang, X. Ji, Y. Liao, J. Peng, K. Wang, Y. Xu, and F. Yan, “Dual-frequency CMOS terahertz detector with silicon-based plasmonic antenna,” Opt. Express 27(16), 23250 (2019). [CrossRef]
26. C. Jungemann, T. Linn, K. Bittner, and H. G. Brachtendorf, “Numerical investigation of plasma effects in silicon MOSFETs for THz-wave detection,” Solid-State Electron. 128, 129–134 (2017). [CrossRef]
27. B. S. Cook and A. Shamim, “Utilizing wideband AMC structures for high-gain inkjet-printed antennas on lossy paper substrate,” Antennas Wirel. Propag. Lett. 12, 76–79 (2013). [CrossRef]
28. Y. Mishima, M. Hirose, and Y. Osaka, “Optical determination of mobility and carrier concentration in heavily doped polycrystalline silicon,” J. Appl. Phys. 51(2), 1157–1159 (1980). [CrossRef]
29. A. L. P. Rotondaro, M. R. Visokay, A. Shanware, J. J. Chambers, and L. Colombo, “Carrier mobility in MOSFETs fabricated with Hf-Si-O-N gate dielectric, polysilicon gate electrode, and self-aligned source and drain,” IEEE Electron Device Lett. 23(10), 603–605 (2002). [CrossRef]
30. A. Berrier, R. Ulbricht, D. Polke, P. H. Bolivar, M. Bonn, and J. G. Rivas, “THz plasmonic antennas: from metals to semiconductors,” IRMMW-THz 2010 - 35th Int. Conf. Infrared, Millimeter, Terahertz Waves, Conf. Guid. (2010).
31. J. Peng, J. Leng, D. Cao, X. He, F. Lin, and F. Liu, “Graphene-supported tunable bidirectional terahertz metamaterials absorbers,” Appl. Opt. 60(22), 6520 (2021). [CrossRef]
32. D. Sievenpiper and E. Yablonovitch, “High-impedance electromagnetic surfaces,” IEEE Trans. Microwave Theory Tech. 27(11), 2059–2074 (1999). [CrossRef]
33. X. Y. Bao, Y. X. Guo, and Y. Z. Xiong, “60-GHz AMC-based circularly polarized on-chip antenna using standard 0.18-µ m CMOS technology,” IEEE Trans. Antennas Propag. 60(5), 2234–2241 (2012). [CrossRef]
34. C. Jungemann, K. Bittner, and H. G. Brachtendorf, “Simulation of plasma resonances in MOSFETs for THz-signal detection,” 2016 Jt. Int. EUROSOI Work. Int. Conf. Ultim. Integr. Silicon, EUROSOI-ULIS 201648–51 (2016).
