Open Access
Add to BibSonomy
Abstract
Intense terahertz (THz) emission in high quality GaAs film upon ultrafast excitation is demonstrated. Results showed that high quality GaAs grown by molecular beam epitaxy with a thin n-doped buffer can rival existing intense bare semiconductor THz surface emitters. The incorporation of a 0.2 µm n-GaAs buffer proved effective in enhancing the THz emission of GaAs by 281% and 295% in reflection and transmission THz time-domain excitation geometries, respectively. The GaAs film was of high crystallinity with or without the n-doped buffer layer as confirmed from X-ray diffraction and Raman scattering. The similar crystalline quality of the two samples was further exemplified by their comparable built-in field strength as measured by photoreflectance spectroscopy. The distinguishable difference in GaAs with and without the doped buffer was observed via low temperature photoluminescence (PL) spectroscopy. The GaAs film with the n-doped buffer exhibited intense GaAs PL while the GaAs film without the n-doped buffer exhibited prominent carbon impurity-related PL. THz enhancement was inferred to be due to the decrease in shallow defects in GaAs with n-doped buffer.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
The generation of electromagnetic radiation in the terahertz (THz) frequency regime from bare semiconductors has led to various works on THz surface emitters since its demonstration more than two decades ago [1–9]. Bare semiconductors offer convenience over other THz emitters as they do not necessitate device fabrication for external applied bias. To date, efficient bare semiconductor THz surface emitters are mostly narrow bandgap semiconductors. The most intense bare semiconductor THz surface emitter is still InAs [3,5,8,10]. Some works to further enhance the THz intensity in InAs include the addition of a heavily-doped InAs layer to act as plasma reflection layer [11]; and the introduction of a GaSb hot-electron injector as well as a thin InAs cap layer to strengthen the surge current and control the surface electric field, respectively [12]. However, even with the popularity of InAs as an intense emitter, it is less transparent to THz radiation [13] as compared with other bare semiconductors and has restricted epitaxial growth [14]. These could limit device design and device fabrication protocols for InAs epitaxial layers.
It has been shown that low temperature-grown GaAs (LTG-GaAs), a semiconductor with relatively high defect density [15–17], can rival p-InAs in terms of THz emission intensity and possible THz device applications. By incorporating a thin n-GaAs buffer in growing annealed LTG-GaAs by molecular beam epitaxy (MBE), viability of the film as a strong THz surface emitter has been extended in both reflection and transmission THz time-domain spectroscopy (THz-TDS) geometries [18]. Significant increase in the built-in field by the addition of the thin n-GaAs buffer has resulted in enhanced THz emission in LTG-GaAs comparable with that of p-InAs [18]. The current results demonstrate that the same technique is similarly applicable to other semiconductors such as high quality GaAs films, grown by MBE at conventional substrate temperature between 580°C – 600°C [19].
A high quality GaAs film with an optimum layer design can be an efficient, reliable and practical rival of existing intense bare semiconductor THz surface emitters [3–5], for both reflection and transmission geometries. Several works on GaAs and LTG-GaAs layers have already provided evidence of THz enhancement by the addition of n-doped layers [20–23]. These works, however, concentrated on incorporating 1.5-3.0 µm-thick n-GaAs layers limiting THz radiation especially in transmission geometry because of free-carrier absorption. In this work, we intend to generate THz radiation more effectively in high quality MBE-grown GaAs, in both reflection and transmission geometries, by incorporating only a thin 0.2 µm n-GaAs buffer. The advantage of using high quality GaAs is that the growth protocols can be easily implemented and adjusted to suit a desired application. Moreover, the material is highly reproducible and the device fabrication protocols are well established, thus GaAs epitaxial layers with thin n-GaAs buffer may still prove to be an efficient substrate for THz devices even with a doped layer.
2. Experimental details
All samples were grown using a Riber 32P MBE. Shown in Fig. 1 is the schematic of the samples. GaAs films were grown on semi-insulating GaAs (SI-GaAs) substrate with undoped (Sample A) and doped (Sample B) GaAs buffer. For comparison, LTG-GaAs with doped GaAs buffer were grown on SI-GaAs (Sample C) and n +-GaAs (Sample D) substrates. The GaAs layers were grown at substrate temperature (Ts) of 580°C while the LTG-GaAs layers were grown at Ts=400°C and in-situ annealed at Ts=600°C for 5 minutes. Following a growth rate of 1 µm/hr, the total thickness of the epitaxial layers was maintained at 1 µm in consideration with the penetration depth of the 800 nm pump beam. The n-GaAs buffer layers were silicon (Si)-doped with experimentally determined doping concentration of 3.0 × 1018 cm−3.
THz-TDS was performed in reflection and transmission excitation geometries. A p-polarized mode-locked Ti:sapphire laser with 100 fs pulse width, 800 nm central wavelength and 80MHz repetition rate was the pump beam used to excite the samples via standard THz-TDS. The modulation of the pump beam was by an optical chopper set at frequency of 2 kHz. The pump beam angle was set at 45° (60°) with respect to the sample-surface normal in reflection (transmission) geometry. The average pump power was at 120 mW (130 mW) while the average probe power was at 5 mW (10 mW). Shown in Figs. 2(a) and 2(b) is the sample orientation relative to the pump beam and detector in both geometries. A commercial dipole LTG-GaAs photoconductive antenna (PCA), in conjuction with appropriate optics, was used for the detection of the generated THz. The same fs-laser excitation source was used to optically gate the PCA. A paper placed between the paraboloid mirror in reflection geometry ensured that the pump beam did not contribute to the detected THz signal. The excitation beam diameter, focal lengths of paraboloid optics and other experimental parameters were all kept constant. Prior to THz-TDS characterization, all samples were mechanically lapped to reduce the thickness of the substrate to 300 µm.
Fig. 2 Sample orientation with respect to the THz-TDS pump beam and detector in (a) reflection and (b) transmission excitation geometries. (c) Sample orientation with respect to the probe, laser and detector in the PR set-up.
Photoreflectance (PR) spectroscopy was used to verify if there had been a modification in the built-in field. The modulated photoreflectance (∆R/R) of the samples was probed with a Tungsten-halogen lamp incident on the sample at 45° angle with respect to the sample-surface normal. The lamp was dispersed by a Spex 500M monochromator with slit width maintained at 400 µm. A mechanically chopped Argon (Ar)-ion laser mode-locked at 488 nm was used for the built-in field modulation in the samples. The power density and chopping frequency were set at ~5 mW/cm2 and 200 Hz, respectively. A Si photodiode was used for the detection of the reflected light. A 620 nm high pass filter at the detector ensured the non-detection of the laser beam. Shown in Fig. 2(c) is the sample orientation relative to the probe, laser and detector in the PR set-up. The built-in field of the samples was determined from the extrema of the oscillatory lineshape above the GaAs bandgap in the PR signal, described as the Franz-Keldysh oscillations (FKO) [24].
3. Results and discussion
Tabulated in Table 1 are the peak-to-peak THz current of the samples, including that of a bulk p-InAs, in both reflection and transmission excitation geometries. There was an observed 281% and 295% increase in the THz emission of sample B, which had a thin n-GaAs buffer, in reflection and transmission geometries, respectively, relative to sample A, which had an undoped GaAs buffer.
Table 1. Peak-to-Peak THz Current of the Samples.
Shown in Fig. 3 is the THz waveform of the samples in reflection geometry. Apart from the significant increase in the THz emission of sample B compared with sample A, the THz emission of sample B was also comparable with the THz emission of LTG-GaAs with n-GaAs buffer grown on SI-GaAs (sample C) and n +-GaAs substrates (sample D) as well as with a bulk p-InAs sample, all of which had been shown to emit intense THz radiation [5,18].
Shown in Fig. 4 is the THz waveform of the samples in transmission geometry. In general, the THz emission decreased as compared with the THz emission in reflection geometry. This was attributed to absorption losses. Even in transmission geometry, however, there was an increase in the THz emission of sample B as compared with sample A indicating that free-carrier absorption was offset due to the very thin n-GaAs buffer. It can be observed that sample D grown on n +-GaAs substrate had the least intense THz emission because of free-carrier absorption by the thick n-doped GaAs substrate. The oscillations after the main pulse in samples A to C were due to multiple reflections from the back and front surfaces of the samples. These oscillations were not observed in sample D because of absorption losses. Sample B exhibited the most intense THz emission in transmission geometry.
In the frequency spectrum shown in the inset of Figs. 3 and 4, taken from the Fourier transform of the THz waveforms, the samples had frequency bandwidth of ~1.2 THz and ~0.8 THz in reflection and transmission geometries, respectively. The frequency bandwidth did not vary much for the GaAs and LTG-GaAs samples. The observed decrease in the frequency bandwidth of the samples in transmission geometry was associated with free-carrier absorption [25–29].
Previously, it was shown that the increase in the THz emission of LTG-GaAs with n-GaAs buffer was due to the increase in the built-in field of the sample [18]. In Fig. 3, samples C and D exhibited intense THz radiation because of the enhanced field in the LTG-GaAs/n-GaAs interface. This was not the case, however, for the GaAs with n-GaAs buffer. Shown in Fig. 5 (left) is the PR signal of the samples. From the linearized plot of the FKO extrema against the FKO index, shown in Fig. 5 (Right), the built-in fields of samples A (13 kV/cm) and B (16 kV/cm) were determined to be low and comparable. The THz enhancement in sample B cannot be solely due to its resulting built-in field unlike that of samples C (147 kV/cm) and D (129 kV/cm). Experimentally determined built-in fields were consistent with theoretical values calculated using standard depletion region band structure models for degenerate, homogenous and abrupt semiconductor junctions [30]. The proportionality of the field strength with the THz emission was not straightforward for GaAs.
Fig. 5 (Left) PR signal of the samples. (Right) Linearized plot of the FKO extrema against the FKO index.
X-ray diffraction (XRD), Raman scattering and photoluminescence (PL) of samples A and B were therefore determined to verify whether the quality of the layers played a significant role in the efficiency of THz generation. The XRD rocking curves around the 004 Bragg angle of GaAs were obtained using a commercial Rigaku SmartLab X-ray diffractometer. For Raman spectroscopy, a 532 nm solid state laser source was used with the power kept at 10 mW. The Raman signal was fiber fed to a liquid nitrogen cooled CCD camera attached to a monochromator. The PL spectroscopy was performed at low temperature (11 K) using standard lock-in technique. The excitation source was a 488 nm Ar-ion laser mechanically chopped at 200 Hz. The corresponding PL was collected by a Spex 500M spectrometer and detected by a GaAs photomultiplier tube. The slit widths were maintained at 100 µm.
Shown in Fig. 6(a) is the x-ray diffraction of samples A and B. The rocking curves were sharp, centered at 33.022° and 33.020° for samples A and B, respectively. The negligible peak position and full width at half maximum (FWHM) variations in samples A (FWHM = 0.080°) and B (FWHM = 0.128°) were indicative of a consistent crystal quality. Similar result was observed via Raman spectroscopy. Shown in Fig. 6(b) is the Raman shift in samples A and B. There was no significant difference in the GaAs LO phonon in samples A (291.3 cm−1) and B (292.7 cm−1) confirming the comparable crystal quality of the two samples.
The distinguishable difference in samples A and B was observed in the low temperature PL spectra as shown in Fig. 7. In sample A, it can be seen that peak 1 (1.4946 eV) corresponding to a carbon acceptor-related transition was much higher than peak 2 (1.5165 eV) corresponding to a GaAs excitonic luminescence [31]. On the contrary, in sample B, apart from a much higher PL intensity than sample A, the carbon PL peak (1.4924 eV) was less pronounced than the GaAs PL peak (1.5141 eV). Carbon is a common background impurity that is inherently incorporated during MBE growth [19]. In GaAs films, Arsenic (As) vacancies allow for the incorporation of carbon acceptors [31]. Given the high GaAs PL intensity in sample B, it can be deduced that the incorporation of carbon acceptors was greatly reduced by Si-doping with the Si atom occupying possible vacant As sites [32,33]. This resulted to the passivation of carbon-related impurities in the n-GaAs buffer of sample B improving the optical characteristics of the subsequent GaAs epitaxial layer.
Fig. 7 (Left) PL spectra of samples A and B performed at 11K. (Right) Corresponding peak positions. The PL intensity is normalized with respect to to the PL peak of sample B.
In PL spectroscopy, carbon impurities only manifest at low temperature [31]. In relation to the resulting THz emission of samples A and B, it can be inferred that the transient occupation of carbon state that should have thermalized at room temperature may have adversely affected the ballistic transport of photocarriers, resulting in a less efficient THz emission in sample A as compared to sample B. Although the built-in field should have primarily dictated the photocurrent surge in GaAs as with that of the LTG-GaAs samples [34,35], the high crystallinity of samples A and B resulted in low built-in field. Given the comparable built-in field of samples A and B, the difference in the concentration of shallow defects may have come into play especially in the ultrafast regime thereby affecting the resulting transient photocurrent of accelerating carriers.
4. Summary and conclusions
The effectiveness of incorporating a thin n-GaAs buffer to enhance the THz emission in high quality MBE-grown GaAs film has been demonstrated. THz enhancement of 281% and 295% were measured in reflection and transmission geometries, respectively, in GaAs with 0.2 µm n-doped GaAs buffer. The resulting THz emission was comparable with bulk p-InAs and other intense bare semiconductor THz surface emitters. The GaAs with doped buffer was of high crystal quality thus the built-in field was low and comparable to GaAs with undoped buffer. The increase in the THz emission was inferred to be due to the lower concentration of shallow defects in GaAs with doped buffer leading to a more efficient generation of THz radiation.
Funding
Department of Science and Technology - Philippine Council for Industry, Energy and Emerging Technology Research and Development (DOST-PCIEERD); DOST-PCIEERD Grants-in-Aid (04001); Commission on Higher Education Philippine - California Advanced Research Institutes (CHED-PCARI) (IIID-2015-13).
Acknowledgments
AA Salvador and ES Estacio would like to thank DOST-PCIEERD and CHED-PCARI. MHM Balgos acknowledges the support of the International Program Associate from the RIKEN “Program for Junior Scientists".
References and links
1. X. Zhang and D. H. Auston, “Optoelectronic measurement of semiconductor surfaces and interfaces with femtosecond optics,” J. Appl. Physics 71(1), 326–338 (1992). [CrossRef]
2. M. B. Johnston, A. Dowd, D. M. Whittaker, A. Corchia, A. G. Davies, and E. H. Linfield, “Enhancement of THz emission from semiconductor surfaces,” in Proceedings of IEEE Tenth International Conference on Terahertz Electronics, (IEEE, 2002), pp. 48–51.
3. P. Gu, M. Tani, S. Kono, K. Sakai, and X.-C. Zhang, “Study of terahertz radiation from InAs and InSb,” J. Appl. Physics 91(9), 5533–5537 (2002). [CrossRef]
4. M. Suzuki, M. Tonouchi, K.-I. Fujii, H. Ohtake, and T. Hirosumi, “Excitation wavelength dependence of terahertz emission from semiconductor surface,” Appl. Phys. Lett. 89(9), 091111 (2006). [CrossRef]
5. V. L. Malevich, R. Adomavicius, and A. Krotkus, “THz emission from semiconductor surfaces,” Comptes Rendus Physique 9(2), 130–141 (2008). [CrossRef]
6. R. Inoue, K. Takayama, and M. Tonouchi, “Angular dependence of terahertz emission from semiconductor surfaces photoexcited by femtosecond optical pulses,” J. Opt. Soc. Am. B 26(9), A14–A22 (2009). [CrossRef]
7. C. T. Que, T. Edamura, M. Nakajima, M. Tani, and M. Hangyo, “Terahertz Radiation from InAs Films on Silicon Substrates Excited by Femtosecond Laser Pulses,” Japanese J. Appl. Physics 48(1R), 010211 (2009). [CrossRef]
8. A. Krotkus, A. Arlauskas, R. Adomavicius, I. Nevinskas, and V. L. Malevich, “Physical mechanisms of surface terahertz emission from semiconductors,” in Terahertz Emitters, Receivers, and Applications VII, Proc. SPIE9934, 993405 (2016).
9. M. Alfaro-Gomez and E. Castro-Camus, “The role of bandgap energy excess in surface emission of terahertz radiation from semiconductors,” Appl. Phys. Lett. 110(4), 042101 (2017). [CrossRef]
10. S. Sasa, S. Umino, Y. Ishibashi, T. Maemoto, M. Inoue, K. Takeya, and M. Tonouchi, “Intense Terahertz Radiation from InAs Thin Films,” J. Infrared, Millimeter, and Terahertz Waves 32(5), 646–654 (2011). [CrossRef]
11. M. Kozub, K. Nishisaka, T. Maemoto, S. Sasa, K. Takayama, and M. Tonouchi, “Reflection Layer Mediated Enhancement of Terahertz Radiation Utilizing Heavily-Doped InAs Thin FilmsM,” J. Infrared, Millimeter, and Terahertz Waves 36(5), 423–429 (2015). [CrossRef]
12. S. Sasa, Y. Kinoshita, M. Tatsumi, M. Koyama, T. Maemoto, S. Hamauchi, I. Kawayama, and M. Tonouchi, “Study for Enhancement of Terahertz Radiation Using GaSb/InAs Heterostructures,” J. Phys.: Conference Series 906(1), 012015 (2017).
13. E. Estacio, S. Takatori, M. H. Pham, T. Yoshioka, T. Nakazato, M. Cadatal-Raduban, T. Shimizu, N. Sarukura, M. Hangyo, C. T. Que, M. Tani, T. Edamura, M. Nakajima, J. V. Misa, R. Jaculbia, A. Somintac, and A. Salvador, “Intense terahertz emission from undoped GaAs/n-type GaAs and InAs/AlSb structures grown on Si substrates in the transmission-geometry excitation,” Appl. Phys. B 103(4), 825–829 (2011). [CrossRef]
14. C. P. Sadia, J. Muldera, E. S. Estacio, A. S. Somintac, A. A. Salvador, C. T. Que, K. Yamamoto, and M. Tani, “Interruption-assisted epitaxy of faceted p-InAs on buffered GaSb for terahertz emitters,” Appl. Phys. Express 8, 035501 (2015). [CrossRef]
15. M. Kaminska, Z. Liliental-Weber, E. R. Weber, T. George, J. B. Kortright, F. W. Smith, B. Tsaur, and A. R. Calawa, “Structural properties of As-rich GaAs grown by molecular beam epitaxy at low temperatures,” Appl. Phys. Lett. 54(19), 1881–1883 (1989). [CrossRef]
16. X. Liu, A. Prasad, W. M. Chen, A. Kurpiewski, A. Stoschek, Z. Liliental-Weber, and E. R. Weber, “Mechanism responsible for the semi-insulating properties of low-temperature-grown GaAs,” Appl. Phys. Lett. 65(23), 3002–3004 (1994). [CrossRef]
17. M. Luysberg, H. Sohn, A. Prasad, P. Specht, Z. Liliental- Weber, E. R. Weber, J. Gebauer, and R. Krause-Rehberg, “Effects of the growth temperature and As/Ga flux ratio on the incorporation of excess As into low temperature grown GaAs,” J. Appl. Physics 83(1), 561–566 (1998). [CrossRef]
18. E. A. P. Prieto, S. A. B. Vizcara, A. S. Somintac, A. A. Salvador, E. S. Estacio, C. T. Que, K. Yamamoto, and M. Tani, “Terahertz emission enhancement in low-temperature-grown GaAs with an n-GaAs buffer in reflection and transmission excitation geometries,” J. Opt. Soc. Am. B 31(2), 291 (2014). [CrossRef]
19. R. F. C. Farrow, Molecular Beam Epitaxy (Noyes Publications, 1995).
20. J. Darmo, G. Strasser, T. Muller, R. Bratschitsch, and K. Unterrainer, “Surface-modified GaAs terahertz plasmon emitter,” Appl. Phys. Lett. 81(5), 871–873 (2002). [CrossRef]
21. K. Liu, A. Krotkus, K. Bertulis, J. Xu, and X.-C. Zhang, “Terahertz radiation from n-type GaAs with Be-doped low-temperature-grown GaAs surface layers,” J. Appl. Physics 94(5), 3651–3653 (2003). [CrossRef]
22. H. Takeuchi, J. Yanagisawa, T. Hasegawa, and M. Nakayama, “Enhancement of terahertz electromagnetic wave emission from an undoped GaAs/n-type GaAs epitaxial layer structure,” Appl. Phys. Lett. 93(8), 081916 (2008). [CrossRef]
23. S. Tsuruta, H. Takeuchi, H. Yamada, M. Hata, and M. Nakayama, “Enhancement mechanism of terahertz radiation from coherent longitudinal optical phonons in undoped GaAs/n-type GaAs epitaxial structures,” J. Appl. Physics 113(14), 143502 (2013). [CrossRef]
24. M. Sydor, J. Angelo, J. J. Wilson, W. C. Mitchel, and M. Y. Yen, “Photoreflectance from GaAs and GaAs/GaAs interfaces,” Phys. Rev. B 40, 8473–8484 (1989). [CrossRef]
25. W. G. Spitzer and H. Y. Fan, “Determination of optical constants and carrier effective mass of semiconductors,” Phys. Rev. 106, 882–890 (1957). [CrossRef]
26. K. Osamura and Y. Murakami, “Free carrier absorption in n-GaAs,” Japanese J. Appl. Physics 11, 365 (1972). [CrossRef]
27. H. C. Casey, D. D. Sell, and K. W. Wecht, “Concentration dependence of the absorption coefficient for n- and p-type GaAs between 1.3 and 1.6 eV,” J. Appl. Physics 46, 250–257 (1975). [CrossRef]
28. C.-C. Wu and C.-J. Lin, “Free-carrier absorption in n-type gallium arsenide films for polar optical phonon scattering,” J. Appl. Physics 79, 781–785 (1996). [CrossRef]
29. X. L. Lei and S. Y. Liu, “Nonlinear free-carrier absorption of intense THz radiation in semiconductors,” J. Physics: Condensed Matter 12, 4655 (2000).
30. S. M. Sze and K. K. Ng, Physics of Semiconductor Devices (John Wiley and Sons, Inc., 2007).
31. F. Briones and D. M. Collins, “Low temperature photoluminescence of lightly Si-doped and undoped MBE GaAs,” J. Electronic Materials 11(4), 847–866 (1982). [CrossRef]
32. S. Schuppler, D. L. Adler, L. N. Pfeiffer, K. W. West, E. E. Chaban, and P. H. Citrin, “Identifying and quantifying point defects in semiconductors using x-ray-absorption spectroscopy: Si-doped GaAs,” Phys. Rev. B 51(16), 10527–10538 (1995). [CrossRef]
33. Y. Ohno, T. Taishi, I. Yonenaga, and S. Takeda, “Atomistic structure of stacking faults in a commercial GaAs:Si wafer revealed by cross-sectional scanning tunneling microscopy,” Physica B: Condensed Matter 401–402, 230–233 (2007). [CrossRef]
34. K. Sakai, Terahertz Optoelectronics (Springer-VerlagBerlin Heidelberg, 2005). [CrossRef]
35. Y.-S. Lee, Principles of Terahertz Science and Technology (Springer Science+Business Media, LLC, 2009).
