1. Introduction

Terahertz spectroscopy systems can reveal the absorption signatures of various materials due to their unique molecular vibrational and rotational resonances, enabling chemical identification, material characterization, gas monitoring, and hyperspectral imaging [1–11]. Unlike terahertz time-domain spectroscopy (THz-TDS) systems, frequency-domain spectroscopy (THz-FDS) systems typically use a pair of continuous-wave (CW) lasers with a terahertz frequency difference to generate and detect CW terahertz radiation through photomixing, which is frequency-tunable in the band of interest. THz-FDS systems can provide high frequency resolution determined by the linewidth and stability of the CW lasers [12–19], which is very hard to realize in THz-TDS systems due to limitations in laser repetition rate and/or range of optical delay line. Other advantages of THz-FDS over THz-TDS are compactness and cost effectiveness, because THz-FDS systems do not require femtosecond lasers used in THz-TDS systems. In fact, many THz-FDS systems operating in the telecommunications wavelength range (∼ 1550 nm) provide small system footprints and highly reliable operation [15–19].

To date, most THz-FDS systems have been realized using short-carrier-lifetime photoconductive antennas (especially for detectors) based on various semiconductor defect introduction techniques [12–19]. While short-carrier-lifetime photoconductors provide sub-picosecond photoconductive response times that enable efficient operation at terahertz frequencies, they are known to degrade the photoconductive gain, carrier mobility, and thermal conductivity. In addition, the realization of short-carrier-lifetime photoconductors require non-standard semiconductor growth processes and/or doping elements that are often hard to access in many semiconductor manufacturing facilities. It was recently shown that the use of plasmonic contact electrodes and nanocavities can significantly reduce the response time of photoconductive antennas even in the absence of short-carrier-lifetime photoconductors [20–29]. Upon optical excitation, plasmonic nanostructures are known to significantly enhance the optical intensity close to the metal/semiconductor interface [29–40]. As a result, by tight confinement of the optical pump photons in close proximity to the plasmonic terahertz antenna, the transport path distance of the majority of the photogenerated carriers is substantially reduced, enabling efficient operation at terahertz frequencies. Therefore, the combination of high-mobility intrinsic photoconductors and largely reduced photocarrier transit times opens tremendous potential for efficient terahertz operation without relying on short-carrier-lifetime photoconductors. Here we present the experimental demonstration of a complete THz-FDS system based on long-carrier-lifetime plasmonic photoconductive antennas, offering more than a 95 dB peak dynamic range and an operation bandwidth over 2.5 THz.

2. Device design and experimental results

Figure 1 shows the schematic diagram and operation principles of the THz-FDS system, which utilizes a delayed self-heterodyning scheme based on linearly frequency-modulated CW (FMCW) terahertz radiation to achieve fast spectral scanning [41,42]. The combination of optical beams from a fixed-frequency laser and a swept-frequency laser with a linear optical frequency sweep feeds two plasmonic photoconductive antennas serving as the source/detector of the THz-FDS system to generate/detect CW terahertz radiation at the optical beat frequency (Fig. 1(a)). Different fiber lengths are used for the source and detector branches to introduce a constant time delay, ΔT, between the optical beam pumping the source and detector. This time delay translates to different optical beat frequencies at the source, ${f_{TX}}$, and detector, ${f_{RX}}$, leading to an inherent modulation of the detector output photocurrent at an intermediate frequency (IF), ${f_{IF}} = |{{f_{TX}} - {f_{RX}}} |$, which is equal to the product of the terahertz frequency sweep rate and ΔT (Fig. 1(b)). By a careful selection of system parameters, ${f_{IF}}$ is set such that the IF signal is detected with high signal-to-noise ratio (SNR) using mainstream electronics and lock-in amplification algorithms. By recording the detected IF signal as a function of time, the terahertz spectrum is resolved as a function of terahertz frequency.

 

Fig. 1. (a) Schematic diagram of the THz-FDS system based on delayed self-heterodyne detection of a linearly frequency-modulated CW terahertz radiation. By combining the optical beams from a static laser and a linearly swept laser, a linearly varying terahertz beatnote is achieved as a function of time. (b) By introducing a constant time-delay, ΔT, between the optical beams pumping two photoconductive antennas serving as the terahertz source and detector, different optical beat frequencies excite the source, ${f_{TX}}$, and detector, ${f_{RX}}$, with a constant intermediate frequency, ${f_{IF}}$, which is equal to the product of the terahertz frequency sweep rate and ΔT. (c) The plasmonic photoconductive antennas used in this THz-FDS system.

Download Full Size | PDF

The plasmonic photoconductive antennas used in this THz-FDS system are based on a broadband 1.5-turn logarithmic spiral antenna with a 460 µm diameter, equipped with Ti/Au plasmonic contact electrodes fabricated on an In_0.53Ga_0.47As/AlAs mesa structure, as shown in the inset of Fig. 1(c). The 200-nm-thick AlAs layer, which serves as a high resistivity buffer, and the 200-nm-thick undoped In_0.53Ga_0.47As layer, which serves as a high mobility photoabsorbing layer at a 1550 nm optical wavelength, are both grown on a semi-insulating (SI) GaAs substrate using molecular beam epitaxy (MBE). The plasmonic electrodes, which are in the form of a grating pair with a 1 µm tip-to-tip gap, are designed to excite surface plasmon waves in the 1550 nm wavelength range under a transverse-magnetic (TM) polarization to significantly enhance optical intensity close to the metal/In_0.53Ga_0.47As interface [21]. For this purpose, geometry of the gratings is optimized to have a 460 nm periodicity, an 80 nm grating gap, a 3/77 nm Ti/Au height, and a 240 nm Si₃N₄ anti-reflection coating thickness. With this grating design, the photocarrier concentration is highly confined within tens of nanometers from the plasmonic electrodes, greatly reducing the transport distances and, hence, the transit time for the majority of the photocarriers to the antenna electrodes. Importantly, this facilitates the ultrafast photoconductive response required for terahertz operation. The numerical electromagnetic analysis of the plasmonic structures and the fabrication process of the plasmonic photoconductive antennas are detailed in Ref. [21]. Using a combination of an objective lens (Mitutoyo PLAN NIR 100x) and a plano-convex cylindrical lens (Thorlabs LJ1567RM-C), the optical beam is focused down to a narrow elliptical beam spot with 2 µm and 10 µm 1/e² diameters to cover the tip-to-tip gap of the antenna active area.

To investigate the impact of the plasmonic photoconductive antennas used as the source and detector, the performance of the THz-FDS system is characterized in two steps. In the first step, only one of the plasmonic photoconductive antennas is used as the terahertz detector and a PIN photoconductive antenna is used as the terahertz source. In the second step, the two plasmonic photoconductive antennas are used as both the terahertz source and detector of the THz-FDS system.

2.1 THz-FDS system with a PIN photoconductive antenna used as the terahertz source and a plasmonic photoconductive antenna used as the terahertz detector

For this implementation, the swept laser (Finisar WaveSource) and fixed laser (Santec 510) are combined through a 50/50 fiber coupler and amplified by an Erbium-doped fiber amplifier (EDFA, Thorlabs EDFA100P). The amplified beam is split into two branches to pump a plasmonic photoconductive antenna used as the terahertz detector and a PIN photoconductive antenna (Toptica THz-CW-Tx) used as the terahertz source. The PIN photoconductive antenna is reverse biased at 1.5 V and pumped at 30 mW optical power. The detector output photocurrent, which contains the amplitude and phase information of the IF signal throughout the swept terahertz band, is routed to a transimpedance amplifier (FEMTO DHPCA-100) before being digitized by a data card (AlazarTech ATS660) at a sampling rate of 10 MHz. To resolve the terahertz power spectrum, the digitized transimpedance amplifier output is post-processed using a custom-made lock-in demodulation algorithm (Fig. 2). First, the data points associated with deterministic laser mode transitions (indicated by a synchronous electric signal generated from the swept laser module) are removed to achieve a mod-hop-free terahertz spectrum. Then, the frequency ${f_{IF}}$ is estimated by locating the frequency at which the signal periodogram is maximized [43,44].

 

Fig. 2. Block diagram of the custom-made lock-in demodulation algorithm for resolving the terahertz power spectrum, with illustrative signal waveforms and the final spectrum shown at multiple nodes.

Download Full Size | PDF

Figure 3(a) shows the power spectra obtained under various optical pump power levels incident on the detector after averaging 10,000 traces captured over 3 minutes. 30 mW optical power is used for pumping the terahertz source in these measurements. The detected peak terahertz electric field and noise power extracted from the resolved spectra are plotted in Fig. 3(b). The detected terahertz field, which is proportional to the detector output photocurrent, is expected to increase linearly as a function of the optical power. The measurement results are in close agreement with the theoretical predictions, except at high optical powers (> 30 mW) at which the detected terahertz field shows a small sign of saturation. This saturation behavior is due to the carrier screening effect as a result of the separation of photogenerated electrons and holes, creating an opposing electric field that resists further increase of the induced photocurrent. As theoretically expected, the measured noise power is linearly proportional to the optical power, suggesting that the dominant noise mechanism is Johnson-Nyquist noise. Due to the saturation effect, there is no significant improvement in the peak dynamic range by increasing the optical power from 30 mW to 50 mW (Fig. 3(c)).

 

Fig. 3. (a) Resolved terahertz power spectra under various optical power levels incident on the detector after averaging 10,000 traces. 30 mW optical power is used for pumping the terahertz source in these measurements. (b) The detected peak terahertz electric field and noise power extracted from the power spectra plotted as a function of the optical power. (c) Extracted peak dynamic range from the power spectra plotted as a function of the optical power.

Download Full Size | PDF

To further extend the bandwidth of the THz-FDS system, the terahertz power spectrum is resolved under 50 mW optical power incident on the detector over a 5000-second data acquisition time during which 280,000 traces are averaged, leading to a 124 dB peak dynamic range and a 2.5 THz bandwidth (Fig. 4(a)). To show the effectiveness of data averaging, the peak dynamic range of the resolved terahertz spectrum as a function of the acquisition time is shown in Fig. 4(b). A 10 dB increase in the peak dynamic range is observed for every 10-fold increase in the data acquisition time, due to the suppression of the uncorrelated noise captured with each trace. Similar to its performance in a THz-TDS system [21], the terahertz detector based on plasmonic photoconductive antenna fabricated on a long-carrier-lifetime photoconductor offers higher peak dynamic range compared to its short-carrier-lifetime counterparts when used in a THz-FDS system. The higher dynamic range is due to the incorporation of plasmonic contact electrodes and the use of a high-mobility, high-responsivity photoconductive active layer. However, the enhancement in the peak dynamic range is accompanied by a reduction in the resolved spectral bandwidth due to the impact of the slow photocarriers that are generated far from the plasmonic contact electrodes. We anticipate that further optimization of the device structure, e.g. the use of thinner In_0.53Ga_0.47As photoconductive layer embedded in a plasmonic nanocavity [22–24], could significantly reduce the slow photocarriers and extend the resolved spectral bandwidth while maintaining the high dynamic range.

 

Fig. 4. (a) The resolved terahertz power spectrum under 50 mW optical power after averaging 280,000 traces within a 5000-second data acquisition time. (b) The measured peak dynamic range as a function of the data acquisition time, showing a 10 dB gain in the dynamic range for every 10-fold increase in the data acquisition time.

Download Full Size | PDF

2.2 THz-FDS system with two plasmonic photoconductive antennas used as the terahertz source and detector

For this implementation, two similar plasmonic photoconductive antennas are used as the terahertz source and detector (as shown in Fig. 1). The only difference between the two plasmonic photoconductive antennas is that their logarithmic spiral antennas have opposite polarization handedness to enable efficient reception of the transmitted circularly polarized terahertz radiation. In order to operate at high terahertz powers while suppressing Joule heating, the THz-FDS system is operated at a low duty-cycle to boost the peak signal levels [33]. For this purpose, an electro-optic modulator (Thorlabs LN81S-FC) is used to modulate the intensity of the combined laser beams at a modulation frequency of ${f_M}$ = 5 MHz. The pulse-modulated optical beam is then amplified by the EDFA (Amonic AEDFA-PA-35). In addition, the bias voltage applied to the plasmonic photoconductive antenna source is pulse-modulated in synchronization with the optical modulation voltage using a 2-channel function generator (Siglent SDG6032X), so that the undesired dark current of the photoconductive antenna is minimized. As a result, with a well-chosen time-delay, ΔT, between the optical beams pumping the source and detector, the output photocurrent of the detector becomes a sinusoidal oscillation at ${f_{IF}}$ multiplied by a pulse train defined by the optical modulation, as illustrated in Fig. 5. In other words, the optical modulation frequency determines the analog sampling frequency of the IF photocurrent, which is set to be larger than $10 \times {f_{IF}}$ (${f_{IF}}$ ∼ 340 kHz in this setup) to allow for accurate frequency and phase estimation during the data post-processing. The output photocurrent of the detector is routed to a transimpedance amplifier (FEMTO DHPCA-100) before being digitized by a data card (AlazarTech ATS660) at a sampling rate of 50 MHz. The sinusoidal signal envelope (black dashed curve in Fig. 5) is then extracted from the raw digitized data before going through the lock-in demodulation algorithm (Fig. 2).

 

Fig. 5. Illustration of the detector output photocurrent in the THz-FDS system, when the optical beams pumping the source and detector are modulated with a low duty-cycle, showing a sinusoidal oscillation at IF multiplied by a pulse train at the optical modulation frequency.

Download Full Size | PDF

Performance of the THz-FDS system is characterized under different modulation duty-cycles, while maintaining an average optical power level of 10/30 mW incident on the source/detector. In addition, the bias voltage (∼ 4 V) of the source at each duty-cycle is set to maintain a 0.5 mA average photocurrent, beyond which thermal degradation starts to occur. Since the peak optical power is inversely proportional to the duty-cycle, higher peak terahertz radiation power and detected photocurrent levels are expected at lower duty-cycles, which is experimentally verified. Figure 6(a) shows the resolved terahertz spectra at different duty-cycles after averaging 10,000 traces captured over 3 minutes. As illustrated in Fig. 6(b), a peak dynamic range enhancement of 15 dB is observed when reducing the duty-cycle from 30% to 5%. A reduction in the dynamic range is observed for duty-cycles below 5%, which is attributed to be caused by the limited analog bandwidth (65 MHz) of the data card used for recording the detector output.

 

Fig. 6. (a) The resolved terahertz spectra under various optical modulation duty-cycles after averaging 10,000 traces captured over 3 minutes. (b) The peak dynamic range as a function of the duty-cycle.

Download Full Size | PDF

A longer THz-FDS data acquisition is performed at the optimal duty-cycle of 5%. By averaging 100,000 traces averaged over 30 minutes, a peak dynamic range of more than 95 dB and a spectral bandwidth of 2.5 THz are achieved, as shown in Fig. 7. It should be noted that the achieved dynamic range and spectral bandwidth are both limited by the analog bandwidth of the data card used for recording the detector output. The use of faster data acquisition techniques allows further enhancement in the dynamic range and spectral bandwidth at duty-cycles below 5%. This demonstration shows the successful implementation of a high-sweep-rate THz-FDS system using a pair of plasmonics-enhanced terahertz antennas fabricated on a long-carrier-lifetime photoconductor. Realizing ultrafast photoconductive antennas through plasmonics-enabled carrier transit time reduction instead of semiconductor defect-introduced carrier lifetime reduction opens up the possibility for using many other semiconductors and lasers for implementing THz-FDS systems.

 

Fig. 7. The resolved terahertz power spectrum at a 5% duty-cycle after averaging 100,000 traces captured over 30 minutes.

Download Full Size | PDF

3. Conclusion

In conclusion, we demonstrate telecommunication-compatible THz-FDS using photoconductive terahertz antennas that utilize plasmonic electrodes to introduce sub-picosecond carrier transit times rather than using short-carrier-lifetime photoconductors. Excitation of surface plasmon waves along the plasmonic electrodes enables tight concentration of a large fraction of the photogenerated carriers at the metal/photoconductor interface, significantly reducing their transit time to the terahertz antenna. Through this scheme, photoconductive antennas can be fabricated on long-carrier-lifetime photoconductors with higher mobility and responsivity compared to short-carrier-lifetime photoconductors, which possess high density of dislocation states and defects in their lattice. Therefore, plasmonic photoconductive antennas pave the way for high-performance THz-FDS systems realized using different semiconductor types and lasers, which were not possible before due to the limited accessibility of short-carrier-lifetime photo-absorbing semiconductors at different optical wavelengths. Using a PIN photoconductive antenna as the terahertz source and a plasmonic photoconductive antenna as the terahertz detector, both fabricated on long-carrier-lifetime photoconductors, we demonstrate THz-FDS with a peak dynamic range of 124 dB and a spectral bandwidth of 2.5 THz, exhibiting higher dynamic range compared to THz-FDS systems using short-carrier-lifetime photoconductive antennas at the cost of a narrower spectral bandwidth [41,42]. Using two plasmonic photoconductive antennas fabricated on a long-carrier-lifetime photoconductor as the source and detector, we demonstrate THz-FDS with a peak dynamic range of more than 95 dB and a spectral bandwidth of 2.5 THz, with the data readout electronics limiting the spectral bandwidth.

Appendix A: Specifications of the logarithmic spiral antenna integrated with the plasmonic photoconductor to realize the plasmonic photoconductive terahertz sources and detectors used in this work

Figure 8 shows the geometry, calculated impedance, current density distribution at 1 THz, and radiation pattern at 1 THz, using an electromagnetic solver (Ansys HFSS).

 

Fig. 8. The calculated antenna impedance, current density distribution at 1 THz, and radiation pattern at 1 THz for the logarithmic spiral antenna used in this work are shown in (a), (b), and (c) respectively. Clearly, most of the terahertz radiation propagates toward the substrate (θ = 180°).

Download Full Size | PDF

Funding

Institution of Engineering and Technology (A. F. Harvey Prize); U.S. Department of Energy (DE-SC0016925); Office of Naval Research (N00014-22-1-2531).

Acknowledgments

This work was supported by the Office of Naval Research (grant # N00014-22-1-2531) and the A. F. Harvey Engineering Research Prize from the Institution of Engineering and Technology. Ping Keng Lu was supported by the Department of Energy (grant # DE-SC0016925). We acknowledge contributions of Dr. Baolai Liang from California NanoSystems Institute for the MBE growth of the substrates and Parimi V. Muralikrishna for antenna simulations.

Disclosures

The authors declare no conflicts of interest.

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

References

1. M. Van Exter, C. Fattinger, and D. Grischkowsky, “Terahertz time-domain spectroscopy of water vapor,” Opt. Lett. 14(20), 1128–1130 (1989). [CrossRef]  

2. A. Markelz, S. Whitmire, J. Hillebrecht, and R. Birge, “THz time domain spectroscopy of biomolecular conformational modes,” Phys. Med. Biol. 47(21), 3797–3805 (2002). [CrossRef]  

3. Y. C. Shen, T. Lo, P. F. Taday, B. E. Cole, W. R. Tribe, and M. C. Kemp, “Detection and identification of explosives using terahertz pulsed spectroscopic imaging,” Appl. Phys. Lett. 86(24), 241116 (2005). [CrossRef]  

4. D. M. Mittleman, R. H. Jacobsen, R. Neelamani, R. G. Baraniuk, and M. C. Nuss, “Gas sensing using terahertz time-domain spectroscopy,” Appl. Phys. B 67(3), 379–390 (1998). [CrossRef]  

5. M. Kato, S. R. Tripathi, K. Murate, K. Imayama, and K. Kawase, “Non-destructive drug inspection in covering materials using a terahertz spectral imaging system with injection-seeded terahertz parametric generation and detection,” Opt. Express 24(6), 6425–6432 (2016). [CrossRef]  

6. A. Redo-Sanchez, B. Heshmat, A. Aghasi, S. Naqvi, M. Zhang, J. Romberg, and R. Raskar, “Terahertz time-gated spectral imaging for content extraction through layered structures,” Nat. Commun. 7(1), 1–7 (2016). [CrossRef]  

7. M. C. Kemp, P. F. Taday, B. E. Cole, J. A. Cluff, A. J. Fitzgerald, and W. R. Tribe, “Security applications of terahertz technology,” Terahertz for Military and Security Applications 5070, 44–52 (2003). [CrossRef]  

8. S. E. Ralph, S. Perkowitz, N. Katzenellenbogen, and D. Grischkowsky, “Terahertz spectroscopy of optically thick multilayered semiconductor structures,” J. Opt. Soc. Am. B 11(12), 2528–2532 (1994). [CrossRef]  

9. N. A. Aghamiri, F. Huth, A. J. Huber, A. Fali, R. Hillenbrand, and Y. Abate, “Hyperspectral time-domain terahertz nano-imaging,” Opt. Express 27(17), 24231–24242 (2019). [CrossRef]  

10. L. Olivieri, J. S. T. Gongora, L. Peters, V. Cecconi, A. Cutrona, J. Tunesi, R. Tucker, A. Pasquazi, and M. Peccianti, “Hyperspectral terahertz microscopy via nonlinear ghost imaging,” Optica 7(2), 186–191 (2020). [CrossRef]  

11. L. A. Sterczewski, J. Westberg, Y. Yang, D. Burghoff, J. Reno, Q. Hu, and G. Wysocki, “Terahertz hyperspectral imaging with dual chip-scale combs,” Optica 6(6), 766–771 (2019). [CrossRef]  

12. F. Hindle, A. Cuisset, R. Bocquet, and G. Mouret, “Continuous-wave terahertz by photomixing: applications to gas phase pollutant detection and quantification,” C. R. Phys. 9(2), 262–275 (2008). [CrossRef]  

13. C. Baker, I. S. Gregory, M. J. Evans, W. R. Tribe, E. H. Linfield, and M. Missous, “All-optoelectronic terahertz system using low-temperature-grown InGaAs photomixers,” Opt. Express 13(23), 9639–9644 (2005). [CrossRef]  

14. M. M. Aller, H. Lu, A. C. Gossard, and S. Preu, “Continuous-Wave Terahertz System with 50 dB Dynamic Range at 1 THz Using a ni-pn-ip Superlattice Photomixer and an ErAs:InGaAs Photoconductor Operated at 1550 nm,” in 43rd International Conference on Infrared, Millimeter, and Terahertz Waves (IRMMW-THz) (2018), pp. 1–2.

15. D. Stanze, A. Deninger, A. Roggenbuck, S. Schindler, M. Schlak, and B. Sartorius, “Compact cw terahertz spectrometer pumped at 1.5 µm wavelength,” J. Infrared, Millimeter, Terahertz Waves 32(2), 225–232 (2011). [CrossRef]  

16. L. Liebermeister, S. Nellen, R. B. Kohlhaas, S. Breuer, M. Schell, and B. Globisch, “Ultra-fast, high-bandwidth coherent cw THz spectrometer for non-destructive testing,” J. Infrared, Millimeter, Terahertz Waves 40(3), 288–296 (2019). [CrossRef]  

17. T. Göbel, D. Stanze, B. Globisch, R. J. Dietz, H. Roehle, and M. Schell, “Telecom technology based continuous wave terahertz photomixing system with 105 decibel signal-to-noise ratio and 3.5 terahertz bandwidth,” Opt. Lett. 38(20), 4197–4199 (2013). [CrossRef]  

18. S. Hisatake, Y. Koda, R. Nakamura, N. Hamada, and T. Nagatsuma, “Terahertz balanced self-heterodyne spectrometer with SNR-limited phase-measurement sensitivity,” Opt. Express 23(20), 26689–26695 (2015). [CrossRef]  

19. M. Deumer, S. Breuer, R. B. Kohlhaas, S. Nellen, L. Liebermeister, S. Lauck, M. Schell, and B. Globisch, “Continuous wave terahertz receivers with 4.5 THz bandwidth and 112 dB dynamic range,” Opt. Express 29(25), 41819–41826 (2021). [CrossRef]  

20. D. Turan, N. T. Yardimci, and M. Jarrahi, “Plasmonics-enhanced photoconductive terahertz detector pumped by Ytterbium-doped fiber laser,” Opt. Express 28(3), 3835–3845 (2020). [CrossRef]  

21. P.-K. Lu, D. Turan, and M. Jarrahi, “High-sensitivity telecommunication-compatible photoconductive terahertz detection through carrier transit time reduction,” Opt. Express 28(18), 26324–26335 (2020). [CrossRef]  

22. N. T. Yardimci, D. Turan, S. Cakmakyapan, and M. Jarrahi, “A high-responsivity and broadband photoconductive terahertz detector based on a plasmonic nanocavity,” Appl. Phys. Lett. 113(25), 251102 (2018). [CrossRef]  

23. N. T. Yardimci, D. Turan, and M. Jarrahi, “Efficient photoconductive terahertz detection through photon trapping in plasmonic nanocavities,” APL Photonics 6(8), 080802 (2021). [CrossRef]  

24. N. T. Yardimci, S. Cakmakyapan, S. Hemmati, and M. Jarrahi, “A high-power broadband terahertz source enabled by three-dimensional light confinement in a plasmonic nanocavity,” Sci. Rep. 7(1), 1–8 (2017). [CrossRef]  

25. P.-K. Lu and M. Jarrahi, “A continuous-wave terahertz self-heterodyne spectroscopy system without using short-carrier-lifetime photoconductors,” in 46th International Conference on Infrared, Millimeter and Terahertz Waves (IRMMW-THz) (2021), pp. 1–2.

26. P.-K. Lu, X. Jiang, Y. Zhao, D. Turan, and M. Jarrahi, “Bias-Free Terahertz Generation from a Silicon-Compatible Photoconductive Emitter Operating at Telecommunication Wavelengths,” Appl. Phys. Lett. 120(26), 261107 (2022). [CrossRef]  

27. P.-K. Lu, D. Turan, and M. Jarrahi, “High-Power Terahertz Generation from Bias-Free Nanoantennas on Graded Composition InGaAs Structures,” Opt. Express 30(2), 1584–1598 (2022). [CrossRef]  

28. P.-K. Lu, A. F. Olvera, D. Turan, T. S. Seifert, N. T. Yardimci, T. Kampfrath, S. Preu, and M. Jarrahi, “Ultrafast carrier dynamics in terahertz photoconductors and photomixers: beyond short-carrier-lifetime semiconductors,” Nanophotonics 11(11), 2661–2691 (2022). [CrossRef]  

29. D. Turan, P.-K. Lu, N. T. Yardimci, Z. Liu, L. Luo, J.-M. Park, U. Nandi, J. Wang, S. Preu, and M. Jarrahi, “Wavelength conversion through plasmon-coupled surface states,” Nat. Commun. 12(1), 1–8 (2021). [CrossRef]  

30. J. A. Schuller, E. S. Barnard, W. Cai, Y. C. Jun, J. S. White, and M. L. Brongersma, “Plasmonics for extreme light concentration and manipulation,” Nat. Mater. 9(3), 193–204 (2010). [CrossRef]  

31. C. W. Berry, N. Wang, M. R. Hashemi, M. Unlu, and M. Jarrahi, “Significant performance enhancement in photoconductive terahertz optoelectronics by incorporating plasmonic contact electrodes,” Nat. Commun. 4(1), 1–10 (2013). [CrossRef]  

32. N. T. Yardimci and M. Jarrahi, “Nanostructure-enhanced photoconductive terahertz emission and detection,” Small 14(44), 1802437 (2018). [CrossRef]  

33. C. W. Berry, M. R. Hashemi, S. Preu, H. Lu, A. C. Gossard, and M. Jarrahi, “High power terahertz generation using 1550 nm plasmonic photomixers,” Appl. Phys. Lett. 105(1), 011121 (2014). [CrossRef]  

34. S. G. Park, Y. Choi, Y.-J. Oh, and K.-H. Jeong, “Terahertz photoconductive antenna with metal nanoislands,” Opt. Express 20(23), 25530–25535 (2012). [CrossRef]  

35. A. Jooshesh, L. Smith, M. Masnadi-Shirazi, V. Bahrami-Yekta, T. Tiedje, T. E. Darcie, and R. Gordon, “Nanoplasmonics enhanced terahertz sources,” Opt. Express 22(23), 27992–28001 (2014). [CrossRef]  

36. N. T. Yardimci, H. Lu, and M. Jarrahi, “High power telecommunication-compatible photoconductive terahertz emitters based on plasmonic nano-antenna arrays,” Appl. Phys. Lett. 109(19), 191103–191119 (2016). [CrossRef]  

37. D. Turan, S. C. Corzo-Garcia, N. T. Yardimci, E. Castro-Camus, and M. Jarrahi, “Impact of the metal adhesion layer on the radiation power of plasmonic photoconductive terahertz sources,” J. Infrared, Millimeter, Terahertz Waves 38(12), 1448–1456 (2017). [CrossRef]  

38. S. H. Yang, M. R. Hashemi, C. W. Berry, and M. Jarrahi, “7.5% optical-to-terahertz conversion efficiency offered by photoconductive emitters with three-dimensional plasmonic contact electrodes,” IEEE Trans. Terahertz Sci. Technol. 4(5), 575–581 (2014). [CrossRef]  

39. N. T. Yardimci, S. H. Yang, C. W. Berry, and M. Jarrahi, “High-power terahertz generation using large-area plasmonic photoconductive emitters,” IEEE Trans. Terahertz Sci. Technol. 5(2), 223–229 (2015). [CrossRef]  

40. N. Wang, S. Cakmakyapan, Y.-J. Lin, H. Javadi, and M. Jarrahi, “Room-temperature heterodyne terahertz detection with quantum-level sensitivity,” Nat. Astron 3(11), 977–982 (2019). [CrossRef]  

41. L. Liebermeister, S. Nellen, R. B. Kohlhaas, S. Lauck, M. Deumer, S. Breuer, M. Schell, and B. Globisch, “Optoelectronic frequency-modulated continuous-wave terahertz spectroscopy with 4 THz bandwidth,” Nat. Commun. 12(1), 1–10 (2021). [CrossRef]  

42. J. Kutz, L. Liebermeister, N. Vieweg, K. Wenzel, R. B. Kohlhaas, and M. Naftaly, “A Terahertz Fast-Sweep Optoelectronic Frequency-Domain Spectrometer: Calibration, Performance Tests, and Comparison with TDS and FDS,” Appl. Sci. 12(16), 8257 (2022). [CrossRef]  

43. D. C. B. P. Rife and R. Boorstyn, “Single tone parameter estimation from discrete-time observations,” IEEE Trans. Inf. Theory 20(5), 591–598 (1974). [CrossRef]  

44. P. Händel and I. Kiss, “On correlation based single tone frequency estimation,” in Finnish Signal Processing Symposium (1997), pp. 32–36.