Open Access
Add to BibSonomyAbstract
We report here a photoconductive material for THz detection with sub-picosecond carrier lifetime made by C12 (Carbon) irradiation on commercially available semi-insulating (SI) GaAs. We are able to reduce the carrier lifetime of SI-GaAs down to sub-picosecond by irradiating it with various irradiation dosages of Carbon (C12) ions. With an increase of the irradiation dose from ~1012 /cm2 to ~1015 /cm2 the carrier lifetime of SI-GaAs monotonously decreases to 0.55 picosecond, whereas that of usual non-irradiated SI-GaAs is ~70 picosecond. This decreased carrier lifetime has resulted in a strong improvement in THz pulse detection compared with normal SI-GaAs. Improvement in signal to noise ratio as well as in detection bandwidth is observed. Carbon irradiated SI-GaAs appears to be an economical alternative to low temperature grown GaAs for fabrication of THz devices.
© 2015 Optical Society of America
1. Introduction
Electromagnetic radiation with frequencies in the Terahertz (THz) range (1THz = 1012 Hz) has potential applications in security imaging, bio-sensing, chemical identification, material characterization etc., mostly because of its special absorption properties in gases and solids [1–4]. As molecules exhibit rotational and vibrational frequencies in the THz range, THz absorption spectroscopy allows identification of hazardous or poisonous substances, e.g.. The facts that THz radiation is (i) non-ionizing for biological tissue and (ii) many materials, like paper and fabrics, are transparent for it, make THz imaging very appealing for medical as well as security scanning applications. Due to the relatively short wavelengths (λ = 300 μm at 1 THz) the spatial resolution of imaging is still sufficient for most of these applications.
Frequencies in the range of 1 THz are too high for conventional electronic devices. At the same time the photon energies in the range of a few meV (for ν = 1 THz, hν = 4.2 meV, e.g.) are too low for conventional optical devices. Therefore, both approaches are facing enormous challenges for generating or detecting THz radiation, in particular at room temperature. Among the semiconductor based techniques for generation and detection of THz radiation the photoconductive technique (actually combining the electronic with an optical approach) is one of the most popular ones [5–7]. It is relatively simple and inexpensive, quite efficient and the devices are compact. For photoconductive detection of THz pulses a transient photoconductivity with a, typically exponential decay, characterized by the photo-carrier recombination lifetime τrec, is generated in the detector by a laser pulse with a variable time delay τd relative to the incoming THz pulse. The transient photoconductivity samples the time dependent THz field, resulting in a photocurrent pulse, whose magnitude depends on τd. An ideal detector for THz pulses is supposed (1) to reproduce the time dependence of the THz field as a function of τd and (2) to exhibit high detectivity at the same time. This requires a short sampling time τrec, much smaller than the pulse duration τpu (i.e. τrec << τpu), for a “direct sampling detector” [8]. Another option is a very long sampling time, τrec >> τpu. In this case the detector is operating as an “integrating detector” [8] and the time dependence of the THz field is obtained from a time derivative of the measured current. In practice, both of these extreme cases result in low detectivity either due to small signals (direct sampling) or large noise (integrating detector). It has been shown in [8], by Castro-Camus et al that sampling times comparable to the pulse duration (i.e. τrec ~τpu) yield the highest signal to noise ratios (S/N). In this case, the time dependent detector signal reflects basically the time dependence of the THz field, however slightly broadened due to a convolution with the transient conductivity signal. Indeed, we will observe in our measurements below that the signal to noise ratio reaches its largest values if, with increasing carbon irradiation dose, τrec approaches τpu.
Unfortunately, the lack of suitable semiconductor materials represents a major restriction for the implementation of photoconductive THz detectors. In spite of being a direct bandgap material with high absorption in the near infrared (NIR) range, standard GaAs does not represent a good choice for photoconductive devices at THz frequencies because of its relatively large (n- or p-type) dark conductivity and its long recombination lifetimes for photo-generated carriers (τrec ~ns). Low temperature grown GaAs (LT-GaAs [9],), however, has turned out to be the most suitable material for fabrication of THz photoconductive sources and detectors because of its high resistivity and very short (sub-picosecond) carrier lifetimes [10]. Improving the efficiency and reducing the manufacturing cost of these sources and detectors is one major goal of current research in the area. Growth of LT-GaAs is itself a difficult and expensive technology. Semi-insulating GaAs (SI-GaAs) is an economical alternative to LT-GaAs but it has the disadvantage of a much higher carrier lifetime (~100 ps) which makes it suitable only for THz pulse detection in the integrating mode. In an earlier paper [11], we have reported on the advantages of carbon ion irradiated SI-GaAs for photoconductive THz sources compared with those made from standard SI-GaAs. The irradiation damage is resulting in a large density of defect states with energies close to midgap, yielding both, extremely low dark conductivity due to (close to) midgap Fermi level pinning and short recombination lifetimes by acting as efficient electron hole recombination centers. Here, we present a drastic improvement regarding the detection of THz pulses using these irradiated substrates. We have observed a monotonous reduction of the carrier lifetime of SI-GaAs from ~70 ps down to ~0.55 ps by irradiation dosages of Carbon (C12) ions ranging from ~1012/cm2 to ~1015/cm2. As a result the detection signal becomes an increasingly faithful reproduction of the THz pulse. At the same time, the amplitude of the detection signal, the signal-to-noise ratio and the detection bandwidth increase drastically compared with normal SI-GaAs.
2. The device and photoconductive detection
The irradiation was carried out using a 33.5 MeV beam of C12 from the Pelletron Linac Facility, Mumbai. The beam was passed through a ~10 μm thin gold foil which acted as an energy degrader and also generated the energy spread. Optimization of the gold foil thickness and the incident beam energy was done using the software SRIM [www.srim.org] to get a nearly uniform distribution of defects up to ~2 μm depth inside the SI-GaAs crystal. It may be noted that the energy spread and thickness non-uniformity of the gold foil will result in a spread in penetration depth in the SI-GaAs. Two parallel metal electrodes separated by 25 μmwere fabricated using standard photolithography technique on (un-annealed) irradiated and non-irradiated substrates of SI-GaAs. A schematic diagram is shown in Fig. 1. In photoconductive detection of THz pulses, the near infra-red (NIR) and THz pulses are focused onto the photo-conducting substrate with a variable relative delay time τd (see schematic diagram depicted in Fig. 2). The NIR pulses generate charge carriers, which move under the influence of the electric field caused by the THz pulse. The (integrated) photocurrent is collected by the electrodes. The measured photocurrent, I(τd) depends on the delay time τd between the NIR and the THz pulse. It is given by the integral of the product of the time dependent carrier density, n(t-τd), and the time dependent electric field of the THz pulse E(t), i.e.
As the NIR pulse is very short (~10 fs << 1 ps), there is a quasi-step like increase in the charge carrier density n in the photoconductor. However, depending on the carrier lifetime of the material, n(t-τd) will decay either slowly (like in SI-GaAs) or rapidly (like in LT-GaAs). In the case of very rapid decay, the response function n(t-τd) assumes a delta-function-like form compared to the relatively slowly varying THz field E(t). Hence, the photo current I(τd) measured in the case of a delay τd between arrival of the THz field and the probe pulse will be directly proportional to the electric field of the THz pulse, E(τd), present at the photoconductor surface at the delayed carrier excitation by the NIR probe pulse. By changing the delay between arrival time of the NIR pulse relative to the THz pulse the whole THz pulse profile, E(τd), is recorded. If the carrier lifetime is comparable to the THz field pulse or even longer, the measured photocurrent will no longer be directly proportional to the electric field of the THz pulse and further mathematical processing is required to get the actual pulse shape. The reconstruction technique to get the actual THz pulse shape E(τd) from the recorded current I(τd) with a slow carrier decay photoconductor is explained in [8] by Castro-Camus, et al. There have been attempts to improve the THz detection ability of SI-GaAs as a photoconductor by implanting it with different ions, like Ar, N [10, 12]. However, in most cases these implantations lead to doping of the material and then, due to the dark conductivity of the residual carriers present in the material, the THz detection efficiency is reduced.
Fig. 1 Schematic diagram of the electrodes on a photoconductor used for THz pulse detection. The THz pulse to be detected will be focused on the same spot of optical carrier excitation with almost matching arrival time of the 800 nm pulse. The photogenerated charge carriers will move in the presence of the electric field of the THz pulse and a current flow will be measured at the electrodes.
Fig. 2 Schematic diagram of THz Time Domain Spectroscopy set up used to characterize the photoconductive THz detectors.
3. Results
When using photoconductors for THz detection, it is important to know the carrier recombination lifetime τrec of the photoconductor as the recorded pulse shape depends on τrec. To estimate the carrier lifetime τrec, optical pump-probe reflection measurements were performed and reported in [11]. The carrier lifetime of non-irradiated SI-GaAs was found to be τrec ~70 ps, whereas that of irradiated samples were 6 ps, 2.2 ps, 1.8 ps and 0.55 ps for estimated dosages 1012, 1013, 1014 and 1015 cm−2, respectively. In Ref [11]. we have shown that the dark resistance of the irradiated material becomes very high (Fig. 1). The photoconductive response was found to be about proportional to the recombination lifetime (Fig. 4) indicating that the transport properties, in particular the “effective high-field mobilities” are otherwise not affected by the high density of defect states caused by the carbon irradiation. The photoconducting THz detectors were tested in a standard THz- time domain spectroscopy (TDS) setup. A schematic diagram is shown in Fig. 2. A commercial photoconductive antenna source (Batop-iPCA) made on LT-GaAs was used as THz source. The generated THz pulse was focused on these detectors using off-axis parabolic mirrors (PM2- PM5) and a small fraction (~5 mW) of the 300 mW, 10 fs, 800 nm laser pulse was also focused on these detectors after passing through a delay stage. The photocurrents recorded as a function of delay times are plotted in Fig. 3 for different detectors. Lock-in detection technique was used with an electronic chopping frequency of ~30 kHz, a time constant of 30 ms and a delay sweeping velocity of 40 μm/s. The THz pulse recorded with the electro optic technique, using a ~1 mm thick <110> ZnTe crystal, is also shown (as the black curve) in Fig. 3(f). To a good approximation, this signal can be considered as the actual electric field profile E(t) of the THz pulse. Theoretical photocurrent curves I(τd;τrec) for a given recombination lifetime can then be obtained by a convolution of this E(t) with the carrier density response function
according to Eq. (1). In Figs. 3(a)-3(e) the recorded experimental I(τd) from the photodetector with different irradiation dosages (full lines) are plotted together with the corresponding theoretical curves I(τd ;τrec) using the corresponding carrier lifetimes τrec obtained from the pump-probe reflection measurements from our Ref [11]. (Fig. 3; dashed lines). The agreement between measured and theoretical curves is very good. As expected, the photocurrent pulseshape recorded from non-irradiated SI-GaAs (τrec~70 ps) is way different from the actual THz pulse shape recorded from ZnTe. The noise level is also high because of the long lived charge carriers. As we increase the irradiation dose, the decreased carrier life time improves therecorded photo current pulse shape (Figs. 3(a)-3(e)); from Fig. 3(f) the increasingly faithful shape is best visible. Also, the signal to noise ratio (S/N) increases from ~2-3 to ~50 for the maximum dose. We have estimated the S/N by dividing the peak to peak value of the recorded signal pulse by its noise floor value. The measured parameters are listed in table 1. To get the frequency dependent response of these detectors, fast Fourier transforms (FFTs) of the recorded pulses are plotted in Figs. 4(a)-4(c). The noise levels are indicated by horizontal dashed lines and the S/N for different frequencies can be easily observed.Fig. 3 THz pulses recorded from different carbon irradiation dose photoconductive antennas (PCAs) with their theoretically predicted pulse shapes (a-e). In (f) the THz pulse shape recorded from the 3 highest dosages PCA detector is plotted together with the pulse shape recorded from the ZnTe electro optic crystal (continuous black curve).
Table 1. Carrier lifetimes and detection parameters of SI-GaAs and irradiated SI-GaAs
Fig. 4 FFTs of the THz pulses recorded from different carbon irradiation dose photoconductive antennas. The dotted lines are indicating the noise level.
4. Conclusion
In summary, carbon irradiation of SI-GaAs drastically improves its properties as a material for THz photoconductive detection. It reduces the carrier lifetimes, which has resulted in an improvement of the detection of THz pulses. The detectors fabricated from irradiated SI-GaAs exhibit a strongly enhanced S/N ratio and a more accurate pulse shape. Also, they are able to detect higher THz frequency components in comparison to the non-irradiated SI-GaAs detectors. Hence SI-GaAs, which is not as common as LT-GaAs for THz pulse detection, although it is cheaper and easily accessible, can represent, after carbon ion irradiation, an appealing alternative to LT-GaAs.
References and links
1. M. Tonouchi, “Cutting-edge terahertz technology,” Nat. Photonics 1(2), 97–105 (2007). [CrossRef]
2. areS. M. Kim, F. Hatami, J. S. Harris, A. W. Kurian, J. Ford, D. King, G. Scalari, M. Giovannini, N. Hoyler, J. Faist, and G. Harris, “Biomedical terahertz imaging with a quantum cascade laser,” Appl. Phys. Lett. 88(15), 153903 (2006). [CrossRef]
3. A. Gowen, C. O’Sullivan, and C. P. O’Donnell, “Terahertz time domain spectroscopy and imaging: Emerging techniques for food process monitoring and quality control,” Trends Food Sci. Technol. 25(1), 40–46 (2012). [CrossRef]
4. S. Borri, P. Patimisco, A. Sampaolo, H. E. Beere, D. A. Ritchie, M. S. Vitiello, G. Scamarcio, and V. Spagnolo, “Terahertz quartz enhanced photo-acoustic sensor,” Appl. Phys. Lett. 103(2), 021105 (2013). [CrossRef]
5. S. Winnerl, “Scalable microstructured photoconductive terahertz emitters,” J. Infrared Millim. THz Waves 33(4), 431–454 (2012). [CrossRef]
6. C. A. Schmuttenmaer, “Exploring dynamics in the far-infrared with terahertz spectroscopy,” Chem. Rev. 104(4), 1759–1780 (2004). [CrossRef] [PubMed]
7. D. H. Auston, “Picosecond optoelectronic switching and gating in silicon,” Appl. Phys. Lett. 26(3), 101 (1975). [CrossRef]
8. E. Castro-Camus, L. Fu, J. Lloyd-Hughes, H. H. Tan, C. Jagadish, and M. B. Johnston, “Photoconductive response correction for detectors of terahertz radiation,” J. Appl. Phys. 104(5), 053113 (2008). [CrossRef]
9. Y. Cai, I. Brener, J. Lopata, J. Wynn, L. Pfeiffer, and J. Federici, “Design and performance of singular electric field terahertz photoconducting antennas,” Appl. Phys. Lett. 71(15), 2076 (1997). [CrossRef]
10. S. Winnerl, F. Peter, S. Nitsche, A. Dreyhaupt, B. Zimmermann, M. Wagner, H. Schneider, M. Helm, and K. Köhler, “Generation and detection of THz radiation with scalable antennas based on GaAs substrates with different carrier lifetimes,” IEEE J. Sel. Top. Quantum Electron. 14(2), 449–457 (2008). [CrossRef]
11. A. Singh, S. Pal, H. Surdi, S. S. Prabhu, V. Nanal, and R. G. Pillay, “Highly efficient and electrically robust carbon irradiated semi-insulating GaAs based photoconductive terahertz emitters,” Appl. Phys. Lett. 104(6), 063501 (2014). [CrossRef]
12. T. Liu, M. Tani, M. Nakajima, M. Hangyo, and C. Pan, “Ultrabroadband terahertz field detection by photoconductive antennas based on multi-energy arsenic-ion-implanted GaAs and semi-insulating GaAs,” Appl. Phys. Lett. 83(7), 1322 (2003). [CrossRef]
