Open Access
Add to BibSonomy
Abstract
Photoconductive antenna (PCA) is a widely used terahertz (THz) radiation source, but its low radiated power limits the signal-to-noise ratio and bandwidth in THz imaging and spectroscopy applications. Here, we achieved significant PCA power enhancement through etching nanograting directly on the surface of the PCA substrate. The integrated nanograting not only maximizes the generation of photocarriers, but also benefits the bias electric field loaded on the photocarriers. Comparing with the conventional PCA, our PCA realizes a frequency independent THz power enhancement of 3.92 times in the range of 0.05-1.6 THz. Our results reported here not only provide a new method for increasing the THz power of PCAs, but also reveal another way that artificial nanostructures affect the PCAs, which paves the way for the subsequent researches of next-generation PCAs.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Photoconductive antennas (PCAs) have always been valued by the research community because of their outstanding advantages, such as compact in size, low cost, fiber technology compatible and room temperature operation. They are critical components to THz technology and have been widely used in many important applications such as THz spectroscopy, material characterization, biological imaging and hazardous substances inspection. However, the THz power radiated by PCAs still fails to meet the rapidly growing demand in THz technology and applications thus improving PCAs’ power has always been a challenging task [1–7]. In 2007, O'Hara et al. take the lead in integrating sub-wavelength microstructures onto the PCA to enhance its power [8], and several groups later attempted to increase the THz power through adding metallic nanostructures which locally enhance the interaction between the femtosecond laser and the photoconductive substrate [9–14]. Though the improved emission behavior of nanostructure-assisted PCAs arouses enormous interests, metallic nanostructures have inherent drawbacks such as Ohmic loss and poor heat resistance. On the other hand, dielectric nanostructures are free of these problems thus have attracted more and more attentions in recent years: Khorshidi M. et al. theoretically analyzed a dielectric nanostructure that can transmit more optical pump power into the substrate of the PCA [15]; Rana, G. et al. discussed the polarization-dependent behavior of a PCA with hexagonal GaAs nanopores [16]. In addition to PCA emitters, recently two sorts of dielectric nanostructures have also been reported to enhance the sensitivity of the PCAs based detectors [17,18], which showed the great potential of dielectric nanostructures in the manipulation of PCA’s performance. However, until now the introduced nanostructures, either metallic or dielectric, were only designed for manipulating the femtosecond laser pump [19,20], while the influences of nanostructures on the PCA’s bias field has not been investigated. Bias electric field distribution is just as critical to PCA’s performance as the optical pump, and is almost affected by all kinds of nanostructures. Taking the nanostructure proposed in [9] and [11] as examples, which are nanoislands distributed in the antenna gap and metallic nanograting connected to the PCA respectively, the bias field distribution will be inevitably altered comparing with the original PCA because of the high conductivity of metal. However, how to take this side effect of nanostructures into account in the design of novel PCAs are remain unanswered.
Therefore, in this letter, by optimizing the effect of nanostructures on the femtosecond laser and bias electric field at the same time, we proposed and demonstrated an all-dielectric nanograting-assisted PCA transmitter, achieving a nearly uniform THz power enhancement in a broad bandwidth. Unlike previous works, the optical absorption improvement is due to multiple beam inference in nanograting rather than the Mie or plasmonic resonances in a unit lattice, which facilities to fabricate the nanograting by dry etching the substrate itself, thus relaxing the choice of the photoconductive layer's thickness [18] and avoiding the metal-semiconductor contact problems [21,22]. More importantly, we took the lead in revealing the influence of nanostructures on PCA’s bias field, which has not been explored in previous reports to our knowledge. The findings here may provide a new method for developing next generation PCAs.
2. Concept and simulation results
Firstly, the nanograting was mainly introduced to suppress the reflectance around the center wavelength (780 nm) of the femtosecond laser, thereby increasing the photocarriers generated in the GaAs substrate. In previous works [12,17,18], Mie resonance was proposed to realize the anti-reflection effect. In this way the nanostructure is required to form a closed cavity, meaning that the material of the photoconductive layer should be different from the substrate and its thickness are limited to hundreds of nanometers [18]. In fact, besides the Mie resonance, multiple beam interference of dielectric nanostructure can suppress the reflection of incident light as well, which is adopted in this article. The nanograting was directly etched on the top of the substrate and near the anode transmission line, as illustrated in Figs. 1(b) and 1(c). The nanograting can be considered as an effective homogeneous thin layer upon the GaAs substrate, its refractive index neff and height required for zero reflectance are determined by sqrt(nGaAs) and (2m-1)λ/(4neff) respectively, where m is any positive integer [23]. To obtain the requested neff and height, we carried out numerical studies on the interaction between the nanograting and the incident light, where the duty ratio and the height of the nanograting were scanned in the time-domain solver of CST Microwave Studio. In the simulation, we constructed a unit cell of the nanograting and applied the periodic boundary on it. Planewave was normal incident on the unit cell with the polarization perpendicular to the nanograting and the reflectance was obtained by setting a probe in front of the nanograting. To investigate the electric field distribution around the structure, we also added field monitor at the anti-reflection wavelength. Figures 1(d) and 1(e) present the electric field distributions at 780 nm for a bare substrate and the nanograting, respectively. Figure 1(e) clearly shows that there is only weak pump light localized inside the nanograting which is essentially different from the Mie-resonance. The electric field leaks to the substrate and meets its maximum at the air gap between the unit cells of the nanograting. The simulated anti-reflectance effect of the nanograting is displayed in Fig. 1(f). Obviously, nanograting can strongly inhibit the reflection of the incidence over a wide wavelength range. Moreover, the central wavelength of the reflection inhibition can be controlled by adjusting the occupancy ratio of the nanograting (i.e. the width of the grating).
Fig. 1. (a) Schematic diagram of the nanograting-assisted PCA. The width of the coplanar line is 20 µm. Schematic diagram of the unit cells of (b) the Gy and (c) the Gx with the corresponding incidence polarizations, respectively, in which p = 350 nm and the height of the unit cell is 200 nm. The areas of the nanograting in (b) and (c) are 80×80 µm2 and 50×80 µm2 respectively. Electric field distribution of (d) the substrate and (e) an optimized unit cell at the working wavelength (λ∼780 nm) when w = 185 nm. The green dashed line represents the envelops of the bare substrate and the nanograting unit cells, respectively. (f) Simulated anti-reflection effect of the nanogratings with varied w = 245, 230, 215, 200, and 185 nm, respectively. The y-axis is the ratio of the reflectance of the nanograting to its reference.
Next, the influence of the nanograting on the bias electric field was numerically studied and we found that this mainly determines the orientation of the nanograting arranged on the PCA. As shown in Figs. 1(b) and 1(c), consider two nanograting with orientations parallel and perpendicular to the coplanar line, they are named as Gx and Gy according to the polarization of the pump light, respectively. According to the continuity of the electrical displacement vector on the boundary of the unit cell and the adjacent air gap, the x component of the bias electric field in the grating of the Gx should be less than the field in the air gap due to the grating has higher refractive index. In contrast, the structure of the Gy is continuous along x direction, so the x component of its electrical field should be more similar as the bare substrate. The simulated x component of field distributions in the Gx, the Gy and the bare GaAs substrate confirmed our deduction as shown in Fig. 2. In the bare GaAs substrate (Fig. 2(a)), the Ex is mainly localized around the border of the anode coplanar line and the substrate, and that in the Gy2 (Fig. 2(b)) is quite similar to Fig. 1(a). Refer to the Gx (Fig. 2(c)), it is clear that the Ex is mainly localized in the gaps between the gratings and the field in the GaAs grating has an amplitude only about 12% of that in the air gaps.
Fig. 2. The Ex distributions for (a) the bare substrate, (b) the Gy2, and (c) the Gx at 100 nm below the air-substrate interface under a bias voltage of 15 V. Green dashed lines represent the border of the anode coplanar line and the substrate.
Combining the anti-reflectance ability and the desired bias field distribution, the final design of the nanograting-assisted PCA is schematically shown in Fig. 1(a). The PCA is comprised of a semi-insulating gallium arsenide (SI-GaAs) substrate with a thickness of 650 µm and a pair of coplanar lines separated by 80 µm. Considering the processing errors, nanograting with various widths (w = 245, 230, 215, 200, and 185 nm) were designed along the inner side of the anode coplanar line, separated 900 µm away from each other and named as Gy1 to Gy5, respectively. A Gx (w = 230 nm) pattern was also added between Gy2 and Gy3 for experimental investigation of the nanograting’s influence on the bias field.
Fabrication of the PCA was classified as two steps: Firstly, the nanograting was directly fabricated on the 650-μm-thick semi-insulating GaAs substrate. The nanograting patterns were defined by E-beam lithography followed by an ICP etching with Cl2 and BCl3 gases to form the GaAs nanograting. In the second step, the PCA pattern was transferred by photolithography under precise alignment with the nanograting. Then the nanograting-assisted PCA was accomplished by thermally depositing 200-nm-thick gold coplanar lines onto the substrate with 10-nm-thick titanium as the attach layer. One completed nanograting-assisted PCA sample has an area of 25×10 mm2 and was assembled onto a PCB to apply the bias voltage with SMA connector. The typical photos of a fabricated nanograting-assisted PCA sample and its details are shown in Fig. 3.
Fig. 3. (a) Photo of an assembled nanograting-assisted PCA. (b) Microscopic photo of a fabricated nanograting-assisted PCA. SEM images of the dry etched (c) Gy and (d) Gx.
3. Experimental results and analysis
The optical reflectance of the nanograting was measured in a home-built reflectance test setup, in which a lamp (wavelength range: 360–2600 nm) was used to generate broadband illuminance. The reflected light from the nanograting is collected by the fiber detector of a spectrometer (wavelength range: 200-1100 nm). The optical reflectance ratio were calculated by the following formula: Rs(λ)=(S(λ)-B(λ))/(R(λ)-B(λ))×100%, where S(λ) and R(λ) represent the reflected light of the nanograting and the bare substrate, respectively, and B(λ) is the background signal of the setup. The reflectance ratios of the nanograting to its reference are shown in Fig. 4(a), where we can see that under perpendicular polarized incidence, all the nanogratings have very low reflectance ratio around 780 nm with a minimum down to 2.53%. The wavelength of the reflectance minimum varies with the width of the nanograting, which is in good accordance with the evaluation based on the multiple beam interference as well as the simulation results (Fig. 1(f)). In contrast, the dotted line (Gy2-x) in Fig. 4(a) shows the reflectance of the Gy2 under parallel polarized incidence, which is much higher than those under perpendicular polarization. The anti-reflection results shown here mean that more photocarriers should be generated in nanograting-assisted PCA than the bare GaAs substrate.
Fig. 4. (a) Measured optical anti-reflectance effect of the nanograting. Solid lines show anti-reflectance effect of the nanograting under the perpendicular polarized incidence while the dotted line shows the reflectance radio of the Gy2 to its reference under the parallel polarized incidence. (b) Schematic diagram of the 4F THz time-domain spectroscopy based characterization system and the added reflective microscopic imaging system. LD: laser diode, BP: beam splitter, M: mirror, PM: parabolic mirror, L: lens, LP: linear polarizer.
The nanograting-assisted PCA samples were tested in a 4F THz time-domain spectroscopy system as shown in Fig. 4(b), which was excited by incident femtosecond light with 780 nm central wavelength, 120 fs pulse duration and polarization perpendicular to the nanogratings. The incidence power on the silicon-on-sapphire based PCA detector is 10 mW. In order to ensure the accuracy and comparability of the measurement, we made two improvements to the system: firstly, a reflective microscopic imaging system was added to monitor the alignment of the PCA as there are several nanogratings along the anode coplanar line. The second improvement is we separately installed a 3-dimensional translation stage for the hyperspherical silicon lens and prepared a new holder to press the silicon lens onto the rear surface of the PCA. Through this way, the position of the silicon lens will not be affected when switching the nanogratings on the same sample. In addition, we have found that the THz signal varies as the pump focal spot slides up and down along the anode coplanar line, so we took the average value of the substrate-emitted THz signals at 150 µm above and below a certain nanograting as the corresponding reference, which minimizes the error introduced by the position of the nanograting along the coplanar line and ensures the comparability among the nanogratings.
The measured time-domain signal of the Gy2 that has the most announced power enhancement is shown in Fig. 5(a). It is obvious that the magnitude of the Gy2 is noticeably greater than that of its reference. The corresponding Fourier transform shown in Fig. 5(b) reveals more spectral information about this enhancement, where we calculated the frequency dependent power enhancement factor by dividing the THz power spectrum of the Gy to that of its reference. As shown in Fig. 5(b), (a) maximal power enhancement factor up to 5.64 is achieved at 1.39 THz. More importantly, in the whole interested bandwidth that from 0.05 to 1.6 THz, the THz power is enhanced uniformly with an average factor of 3.92. This even enhancement is a distinguishing feature of our nanograting-assisted PCA comparing to previous designs [9–14] and may find unique application for ultra-broadband PCA design. As the comparison, the corresponding results of the Gx are shown In Figs. 5(c) and 5(d), where we found the average terahertz power enhancement factor is only 1.17 at 0.05 - 1.6 THz, with a maximum up to 1.38 at 0.1 THz, much lower than that of the Gy2. Since the optical reflectance of the Gy2 and the Gx are very similar (Fig. 4), the outstanding performance of the Gy2 proves the substantial role of the bias field distribution in the design of nanostructure-assisted PCA (Fig. 2). For the Gy2, its orientation that perpendicular to the coplanar lines increases the photocarriers density and does not disturb the bias field much. While for the Gx, its orientation that parallel to the coplanar lines weakens the bias field loaded on the photocarriers, which cancels the positive contribution brought by photocarrier increasing.
Fig. 5. (a) THz time-domain pulses of the Gy2 (black line) and its reference (REF2) (red line) with the bias voltage of 15 V and the laser pump power of 8 mW. (b) THz power spectra of the Gy2 (black line) and its reference (red line), as well as the frequency dependent power enhancement factor (purple line). (c) THz time-domain pulses of the Gx (black line) and its reference (REFx) (red line) with the bias voltage of 15 V and the laser pump power of 6 mW. (d) THz power spectra of the Gx (black line) and its reference (red line), as well as the frequency dependent power enhancement factor (purple line). The insets of (a) and (c) show the direction of bias voltage and the pump field applied to the Gy2 and Gx, respectively.
The results discussed above were obtained at certain optical pump and bias voltage. To check whether the power enhancement function of the nanograting assisted PCA will be influenced by the bias and pump, we compared the peak-to-peak value of the time-domain THz field of the PCAs under various bias voltages (5–25 V) and optical pump powers (1–8 mW), as presented in Fig. 6. As expected, the peak-to-peak magnitudes of the Gy2 and its reference grow up as the bias voltage and optical pump increases, respectively. And the peak-to-peak enhancement factor of the Gy2 stays around 1.9 quite steady for all the optical pump powers under 15 volts bias. When the bias voltage changes under fixed 6 mW optical pump, the enhancement factor only slightly varies around 1.75 and has a maximum of 2 at 15 V. It can be seen that the nanograting-assisted PCA has the ability to enhance THz power under the commonly used bias voltage and optical pump conditions.
Fig. 6. (a) Peak to peak magnitudes of the THz time-domain pulses of the Gy2 (black line) and its reference (red line) with bias voltage of 15 V and varied optical pump powers from 1 to 8 mW. (b) Peak to peak magnitudes of the THz time-domain pulses of the Gy2 (black line) and its reference (red line) with varied bias voltages from 5 to 25 V and optical pump powers of 6 mW. The corresponding enhancement factors (purple line) were also plotted in the (a) and (b).
4. Conclusion
In summary, we proved that the integrated nanostructures can affect the PCA’s performance through altering the bias electrical field distribution, and based on this, demonstrated a PCA with all-dielectric nanograting that has enhanced THz radiation power. By adjusting the geometric parameters of the nanograting, a reflectivity of the incident femtosecond light as low as 2.53% was achieved. We found that the nanograting with different orientations have opposite influences on the bias electric field distribution which determines the power enhancement effect of the integrated nanograting. With the optimized nanograting, the THz power spectrum of the PCA achieves an even increment of 3.92 times in the frequency range of 0.05-1.6 THz. And this power enhancement is stable over wide ranges of the pump light power and the bias voltage. This study has revealed another way that artificial nanostructures affect the performance of PCAs, which provides a new method for the development of next-generation PCAs.
Funding
National Natural Science Foundation of China (61307125, 61420106006, 61422509, 61575141, 61622505, 61975143); National Science Foundation (ECCS-1232081).
Disclosures
The authors declare that there are no conflicts of interest related to this article.
References
1. D. H. Auston, K. P. Cheung, and P. R. Smith, “Picosecond photoconducting Hertzian dipoles,” Appl. Phys. Lett. 45(3), 284–286 (1984). [CrossRef]
2. S. Preu, G. H. Dohler, S. Malzer, L. J. Wang, and A. C. Gossard, “Tunable, continuous-wave terahertz photomixer sources and applications,” J. Appl. Phys. 109(6), 061301 (2011). [CrossRef]
3. S. Matsuura, M. Tani, and K. Sakai, “Generation of coherent terahertz radiation by photomixing in dipole photoconductive antennas,” Appl. Phys. Lett. 70(5), 559–561 (1997). [CrossRef]
4. A. Nahata, A. S. Weling, and T. F. Heinz, “A wideband coherent terahertz spectroscopy system using optical rectification and electro-optic sampling,” Appl. Phys. Lett. 69(16), 2321–2323 (1996). [CrossRef]
5. M. B. Ketchen, “Generation of subpicosecond electrical pulses on coplanar transmission lines,” Appl. Phys. Lett. 48(12), 751–753 (1986). [CrossRef]
6. P. U. Jepsen, R. H. Jacobsen, and S. R. Keiding, “Generation and detection of terahertz pulses from biased semiconductor antennas,” J. Opt. Soc. Am. B 13(11), 2424–2436 (1996). [CrossRef]
7. M. Van Exter and D. R. Grischkowsky, “Characterization of an optoelectronic terahertz beam system,” IEEE Trans. Microwave Theory Tech. 38(11), 1684–1691 (1990). [CrossRef]
8. J. F. O’Hara, H. Chen, A. J. Taylor, R. D. Averitt, and W. J. Padilla, “Split-Ring Resonator Enhanced Terahertz Antenna,” Nonlinear Optics: Materials, Fundamentals and Applications (Optical Society of America, 2007), TuB2.
9. 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]
10. 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]
11. 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), 1622 (2013). [CrossRef]
12. M. Bashirpour, M. Forouzmehr, S. E. Hosseininejad, M. Kolahdouz, and M. Neshat, “Improvement of terahertz photoconductive Antenna using optical Antenna Array of ZnO Nanorods,” Sci. Rep. 9(1), 1414 (2019). [CrossRef]
13. S. Lepeshov, A. Gorodetsky, A. Krasnok, N. Toropov, T. A. Vartanyan, P. Belov, A. Alú, and E. U. Rafailov, “Boosting terahertz photoconductive antenna performance with optimized plasmonic nanostructures,” Sci. Rep. 8(1), 6624 (2018). [CrossRef]
14. N. M. Burford, M. J. Evans, and M. O. El-Shenawee, “Plasmonic Nanodisk Thin-Film Terahertz Photoconductive Antenna,” IEEE Trans. Terahertz Sci. Technol. 8(2), 237–247 (2018). [CrossRef]
15. M. Khorshidi and G. J. Dadashzadeh, “Dielectric structure with periodic strips for increasing radiation power of photoconductive antennas: theoretical analysis,” J. Infrared, Millimeter, Terahertz Waves 38(5), 609–629 (2017). [CrossRef]
16. G. Rana, A. Bhattacharya, A. Gupta, D. Ghindani, R. Jain, S. P. Duttagupta, and S. S. Prabhu, “A Polarization-Resolved Study of Nanopatterned Photoconductive Antenna for Enhanced Terahertz Emission,” IEEE Trans. Terahertz Sci. Technol. 9(2), 193–199 (2019). [CrossRef]
17. O. Mitrofanov, T. Siday, R. J. Thompson, T. S. Luk, I. Brener, and J. L. Reno, “Efficient photoconductive terahertz detector with all-dielectric optical metasurface,” APL Photonics 3(5), 051703 (2018). [CrossRef]
18. S. T. Siday, P. P. Vabishchevich, L. Hale, C. T. Harris, T. S. Luk, J. L. Reno, I. Brener, and O. Mitrofanov, “Terahertz Detection with Perfectly-Absorbing Photoconductive Metasurface,” Nano Lett. 19(5), 2888–2896 (2019). [CrossRef]
19. A. E. Yachmenev, D. V. Lavrukhin, I. A. Glinskiy, N. V. Zenchenko, Y. G. Goncharov, I. E. Spektor, R. A. khabibullin, T. Otsuji, and D. S. Ponomarev, “Metallic and dielectric metasurfaces in photoconductive terahertz devices: a review,” Opt. Eng. 59(06), 1 (2019). [CrossRef]
20. N. T. Yardimci and M. Jarrahi, “Nanostructure-Enhanced Photoconductive Terahertz Emission and Detection,” Small 14(44), 1802437 (2018). [CrossRef]
21. E. H. Rhoderick, “Metal-semiconductor contacts,” IEE Proc., Part I: Solid-State Electron Devices 129(1), 1–14 (1982). [CrossRef]
22. W. Mönch, “Mechanisms of Schottky-barrier formation in metal–semiconductor contacts,” J. Vac. Sci. Technol., B: Microelectron. Process. Phenom. 6(4), 1270–1276 (1988). [CrossRef]
23. R. Bräuer and O. Bryngdahl, “Design of antireflection gratings with approximate and rigorous methods,” Appl. Opt. 33(34), 7875–7882 (1994). [CrossRef]
